Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-cpumask into merge-rr-cpumask

Conflicts:
	arch/x86/kernel/io_apic.c
	kernel/rcuclassic.c
	kernel/sched.c
	kernel/time/tick-sched.c

Signed-off-by: Mike Travis <travis@sgi.com>
[ mingo@elte.hu: backmerged typo fix for io_apic.c ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>

69 files changed, 3151 insertions, 3322 deletions

diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 1cad9318d217..d364df03c1d6 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -12,6 +12,7 @@ CFLAGS_REMOVE_tsc.o = -pg
 CFLAGS_REMOVE_rtc.o = -pg
 CFLAGS_REMOVE_paravirt-spinlocks.o = -pg
 CFLAGS_REMOVE_ftrace.o = -pg
+CFLAGS_REMOVE_early_printk.o = -pg
 endif
 
 #
@@ -23,7 +24,7 @@ CFLAGS_vsyscall_64.o	:= $(PROFILING) -g0 $(nostackp)
 CFLAGS_hpet.o		:= $(nostackp)
 CFLAGS_tsc.o		:= $(nostackp)
 
-obj-y			:= process_$(BITS).o signal_$(BITS).o entry_$(BITS).o
+obj-y			:= process_$(BITS).o signal.o entry_$(BITS).o
 obj-y			+= traps.o irq.o irq_$(BITS).o dumpstack_$(BITS).o
 obj-y			+= time_$(BITS).o ioport.o ldt.o dumpstack.o
 obj-y			+= setup.o i8259.o irqinit_$(BITS).o setup_percpu.o
@@ -106,6 +107,10 @@ microcode-$(CONFIG_MICROCODE_INTEL)	+= microcode_intel.o
 microcode-$(CONFIG_MICROCODE_AMD)	+= microcode_amd.o
 obj-$(CONFIG_MICROCODE)			+= microcode.o
 
+obj-$(CONFIG_X86_CHECK_BIOS_CORRUPTION) += check.o
+
+obj-$(CONFIG_SWIOTLB)			+= pci-swiotlb_64.o # NB rename without _64
+
 ###
 # 64 bit specific files
 ifeq ($(CONFIG_X86_64),y)
@@ -119,7 +124,6 @@ ifeq ($(CONFIG_X86_64),y)
         obj-$(CONFIG_GART_IOMMU)	+= pci-gart_64.o aperture_64.o
         obj-$(CONFIG_CALGARY_IOMMU)	+= pci-calgary_64.o tce_64.o
         obj-$(CONFIG_AMD_IOMMU)		+= amd_iommu_init.o amd_iommu.o
-        obj-$(CONFIG_SWIOTLB)		+= pci-swiotlb_64.o
 
         obj-$(CONFIG_PCI_MMCONFIG)	+= mmconf-fam10h_64.o
 endif
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index a7b6dec6fc3f..2e2da717b350 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -24,6 +24,7 @@
 #include <linux/iommu-helper.h>
 #include <asm/proto.h>
 #include <asm/iommu.h>
+#include <asm/gart.h>
 #include <asm/amd_iommu_types.h>
 #include <asm/amd_iommu.h>
 
@@ -235,8 +236,9 @@ static int iommu_completion_wait(struct amd_iommu *iommu)
 	status &= ~MMIO_STATUS_COM_WAIT_INT_MASK;
 	writel(status, iommu->mmio_base + MMIO_STATUS_OFFSET);
 
-	if (unlikely((i == EXIT_LOOP_COUNT) && printk_ratelimit()))
-		printk(KERN_WARNING "AMD IOMMU: Completion wait loop failed\n");
+	if (unlikely(i == EXIT_LOOP_COUNT))
+		panic("AMD IOMMU: Completion wait loop failed\n");
+
 out:
 	spin_unlock_irqrestore(&iommu->lock, flags);
 
diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
index 30ae2701b3df..c625800c55ca 100644
--- a/arch/x86/kernel/amd_iommu_init.c
+++ b/arch/x86/kernel/amd_iommu_init.c
@@ -28,6 +28,7 @@
 #include <asm/amd_iommu_types.h>
 #include <asm/amd_iommu.h>
 #include <asm/iommu.h>
+#include <asm/gart.h>
 
 /*
  * definitions for the ACPI scanning code
@@ -427,6 +428,10 @@ static u8 * __init alloc_command_buffer(struct amd_iommu *iommu)
 	memcpy_toio(iommu->mmio_base + MMIO_CMD_BUF_OFFSET,
 			&entry, sizeof(entry));
 
+	/* set head and tail to zero manually */
+	writel(0x00, iommu->mmio_base + MMIO_CMD_HEAD_OFFSET);
+	writel(0x00, iommu->mmio_base + MMIO_CMD_TAIL_OFFSET);
+
 	iommu_feature_enable(iommu, CONTROL_CMDBUF_EN);
 
 	return cmd_buf;
@@ -1074,7 +1079,8 @@ int __init amd_iommu_init(void)
 		goto free;
 
 	/* IOMMU rlookup table - find the IOMMU for a specific device */
-	amd_iommu_rlookup_table = (void *)__get_free_pages(GFP_KERNEL,
+	amd_iommu_rlookup_table = (void *)__get_free_pages(
+			GFP_KERNEL | __GFP_ZERO,
 			get_order(rlookup_table_size));
 	if (amd_iommu_rlookup_table == NULL)
 		goto free;
diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c
index 9a32b37ee2ee..676debfc1702 100644
--- a/arch/x86/kernel/aperture_64.c
+++ b/arch/x86/kernel/aperture_64.c
@@ -1,8 +1,9 @@
 /*
  * Firmware replacement code.
  *
- * Work around broken BIOSes that don't set an aperture or only set the
- * aperture in the AGP bridge.
+ * Work around broken BIOSes that don't set an aperture, only set the
+ * aperture in the AGP bridge, or set too small aperture.
+ *
  * If all fails map the aperture over some low memory.  This is cheaper than
  * doing bounce buffering. The memory is lost. This is done at early boot
  * because only the bootmem allocator can allocate 32+MB.
diff --git a/arch/x86/kernel/apic.c b/arch/x86/kernel/apic.c
index b9019271af62..6b7f824db160 100644
--- a/arch/x86/kernel/apic.c
+++ b/arch/x86/kernel/apic.c
@@ -30,6 +30,7 @@
 #include <linux/module.h>
 #include <linux/dmi.h>
 #include <linux/dmar.h>
+#include <linux/ftrace.h>
 
 #include <asm/atomic.h>
 #include <asm/smp.h>
@@ -775,11 +776,7 @@ static void local_apic_timer_interrupt(void)
 	/*
 	 * the NMI deadlock-detector uses this.
 	 */
-#ifdef CONFIG_X86_64
-	add_pda(apic_timer_irqs, 1);
-#else
-	per_cpu(irq_stat, cpu).apic_timer_irqs++;
-#endif
+	inc_irq_stat(apic_timer_irqs);
 
 	evt->event_handler(evt);
 }
@@ -792,7 +789,7 @@ static void local_apic_timer_interrupt(void)
  * [ if a single-CPU system runs an SMP kernel then we call the local
  *   interrupt as well. Thus we cannot inline the local irq ... ]
  */
-void smp_apic_timer_interrupt(struct pt_regs *regs)
+void __irq_entry smp_apic_timer_interrupt(struct pt_regs *regs)
 {
 	struct pt_regs *old_regs = set_irq_regs(regs);
 
@@ -806,9 +803,7 @@ void smp_apic_timer_interrupt(struct pt_regs *regs)
 	 * Besides, if we don't timer interrupts ignore the global
 	 * interrupt lock, which is the WrongThing (tm) to do.
 	 */
-#ifdef CONFIG_X86_64
 	exit_idle();
-#endif
 	irq_enter();
 	local_apic_timer_interrupt();
 	irq_exit();
@@ -1666,9 +1661,7 @@ void smp_spurious_interrupt(struct pt_regs *regs)
 {
 	u32 v;
 
-#ifdef CONFIG_X86_64
 	exit_idle();
-#endif
 	irq_enter();
 	/*
 	 * Check if this really is a spurious interrupt and ACK it
@@ -1679,14 +1672,11 @@ void smp_spurious_interrupt(struct pt_regs *regs)
 	if (v & (1 << (SPURIOUS_APIC_VECTOR & 0x1f)))
 		ack_APIC_irq();
 
-#ifdef CONFIG_X86_64
-	add_pda(irq_spurious_count, 1);
-#else
+	inc_irq_stat(irq_spurious_count);
+
 	/* see sw-dev-man vol 3, chapter 7.4.13.5 */
 	pr_info("spurious APIC interrupt on CPU#%d, "
 		"should never happen.\n", smp_processor_id());
-	__get_cpu_var(irq_stat).irq_spurious_count++;
-#endif
 	irq_exit();
 }
 
@@ -1697,9 +1687,7 @@ void smp_error_interrupt(struct pt_regs *regs)
 {
 	u32 v, v1;
 
-#ifdef CONFIG_X86_64
 	exit_idle();
-#endif
 	irq_enter();
 	/* First tickle the hardware, only then report what went on. -- REW */
 	v = apic_read(APIC_ESR);
diff --git a/arch/x86/kernel/asm-offsets_32.c b/arch/x86/kernel/asm-offsets_32.c
index 6649d09ad88f..ee4df08feee6 100644
--- a/arch/x86/kernel/asm-offsets_32.c
+++ b/arch/x86/kernel/asm-offsets_32.c
@@ -11,7 +11,7 @@
 #include <linux/suspend.h>
 #include <linux/kbuild.h>
 #include <asm/ucontext.h>
-#include "sigframe.h"
+#include <asm/sigframe.h>
 #include <asm/pgtable.h>
 #include <asm/fixmap.h>
 #include <asm/processor.h>
diff --git a/arch/x86/kernel/asm-offsets_64.c b/arch/x86/kernel/asm-offsets_64.c
index 7fcf63d22f8b..1d41d3f1edbc 100644
--- a/arch/x86/kernel/asm-offsets_64.c
+++ b/arch/x86/kernel/asm-offsets_64.c
@@ -20,6 +20,8 @@
 
 #include <xen/interface/xen.h>
 
+#include <asm/sigframe.h>
+
 #define __NO_STUBS 1
 #undef __SYSCALL
 #undef _ASM_X86_UNISTD_64_H
@@ -87,7 +89,7 @@ int main(void)
 	BLANK();
 #undef ENTRY
 	DEFINE(IA32_RT_SIGFRAME_sigcontext,
-	       offsetof (struct rt_sigframe32, uc.uc_mcontext));
+	       offsetof (struct rt_sigframe_ia32, uc.uc_mcontext));
 	BLANK();
 #endif
 	DEFINE(pbe_address, offsetof(struct pbe, address));
diff --git a/arch/x86/kernel/bios_uv.c b/arch/x86/kernel/bios_uv.c
index f0dfe6f17e7e..2a0a2a3cac26 100644
--- a/arch/x86/kernel/bios_uv.c
+++ b/arch/x86/kernel/bios_uv.c
@@ -69,10 +69,10 @@ s64 uv_bios_call_reentrant(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3,
 
 long sn_partition_id;
 EXPORT_SYMBOL_GPL(sn_partition_id);
-long uv_coherency_id;
-EXPORT_SYMBOL_GPL(uv_coherency_id);
-long uv_region_size;
-EXPORT_SYMBOL_GPL(uv_region_size);
+long sn_coherency_id;
+EXPORT_SYMBOL_GPL(sn_coherency_id);
+long sn_region_size;
+EXPORT_SYMBOL_GPL(sn_region_size);
 int uv_type;
 
 
@@ -100,6 +100,56 @@ s64 uv_bios_get_sn_info(int fc, int *uvtype, long *partid, long *coher,
 	return ret;
 }
 
+int
+uv_bios_mq_watchlist_alloc(int blade, unsigned long addr, unsigned int mq_size,
+			   unsigned long *intr_mmr_offset)
+{
+	union uv_watchlist_u size_blade;
+	u64 watchlist;
+	s64 ret;
+
+	size_blade.size = mq_size;
+	size_blade.blade = blade;
+
+	/*
+	 * bios returns watchlist number or negative error number.
+	 */
+	ret = (int)uv_bios_call_irqsave(UV_BIOS_WATCHLIST_ALLOC, addr,
+			size_blade.val, (u64)intr_mmr_offset,
+			(u64)&watchlist, 0);
+	if (ret < BIOS_STATUS_SUCCESS)
+		return ret;
+
+	return watchlist;
+}
+EXPORT_SYMBOL_GPL(uv_bios_mq_watchlist_alloc);
+
+int
+uv_bios_mq_watchlist_free(int blade, int watchlist_num)
+{
+	return (int)uv_bios_call_irqsave(UV_BIOS_WATCHLIST_FREE,
+				blade, watchlist_num, 0, 0, 0);
+}
+EXPORT_SYMBOL_GPL(uv_bios_mq_watchlist_free);
+
+s64
+uv_bios_change_memprotect(u64 paddr, u64 len, enum uv_memprotect perms)
+{
+	return uv_bios_call_irqsave(UV_BIOS_MEMPROTECT, paddr, len,
+					perms, 0, 0);
+}
+EXPORT_SYMBOL_GPL(uv_bios_change_memprotect);
+
+s64
+uv_bios_reserved_page_pa(u64 buf, u64 *cookie, u64 *addr, u64 *len)
+{
+	s64 ret;
+
+	ret = uv_bios_call_irqsave(UV_BIOS_GET_PARTITION_ADDR, (u64)cookie,
+					(u64)addr, buf, (u64)len, 0);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(uv_bios_reserved_page_pa);
 
 s64 uv_bios_freq_base(u64 clock_type, u64 *ticks_per_second)
 {
diff --git a/arch/x86/kernel/check.c b/arch/x86/kernel/check.c
new file mode 100644
index 000000000000..2ac0ab71412a
--- /dev/null
+++ b/arch/x86/kernel/check.c
@@ -0,0 +1,161 @@
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/kthread.h>
+#include <linux/workqueue.h>
+#include <asm/e820.h>
+#include <asm/proto.h>
+
+/*
+ * Some BIOSes seem to corrupt the low 64k of memory during events
+ * like suspend/resume and unplugging an HDMI cable.  Reserve all
+ * remaining free memory in that area and fill it with a distinct
+ * pattern.
+ */
+#define MAX_SCAN_AREAS	8
+
+static int __read_mostly memory_corruption_check = -1;
+
+static unsigned __read_mostly corruption_check_size = 64*1024;
+static unsigned __read_mostly corruption_check_period = 60; /* seconds */
+
+static struct e820entry scan_areas[MAX_SCAN_AREAS];
+static int num_scan_areas;
+
+
+static __init int set_corruption_check(char *arg)
+{
+	char *end;
+
+	memory_corruption_check = simple_strtol(arg, &end, 10);
+
+	return (*end == 0) ? 0 : -EINVAL;
+}
+early_param("memory_corruption_check", set_corruption_check);
+
+static __init int set_corruption_check_period(char *arg)
+{
+	char *end;
+
+	corruption_check_period = simple_strtoul(arg, &end, 10);
+
+	return (*end == 0) ? 0 : -EINVAL;
+}
+early_param("memory_corruption_check_period", set_corruption_check_period);
+
+static __init int set_corruption_check_size(char *arg)
+{
+	char *end;
+	unsigned size;
+
+	size = memparse(arg, &end);
+
+	if (*end == '\0')
+		corruption_check_size = size;
+
+	return (size == corruption_check_size) ? 0 : -EINVAL;
+}
+early_param("memory_corruption_check_size", set_corruption_check_size);
+
+
+void __init setup_bios_corruption_check(void)
+{
+	u64 addr = PAGE_SIZE;	/* assume first page is reserved anyway */
+
+	if (memory_corruption_check == -1) {
+		memory_corruption_check =
+#ifdef CONFIG_X86_BOOTPARAM_MEMORY_CORRUPTION_CHECK
+			1
+#else
+			0
+#endif
+			;
+	}
+
+	if (corruption_check_size == 0)
+		memory_corruption_check = 0;
+
+	if (!memory_corruption_check)
+		return;
+
+	corruption_check_size = round_up(corruption_check_size, PAGE_SIZE);
+
+	while (addr < corruption_check_size && num_scan_areas < MAX_SCAN_AREAS) {
+		u64 size;
+		addr = find_e820_area_size(addr, &size, PAGE_SIZE);
+
+		if (addr == 0)
+			break;
+
+		if ((addr + size) > corruption_check_size)
+			size = corruption_check_size - addr;
+
+		if (size == 0)
+			break;
+
+		e820_update_range(addr, size, E820_RAM, E820_RESERVED);
+		scan_areas[num_scan_areas].addr = addr;
+		scan_areas[num_scan_areas].size = size;
+		num_scan_areas++;
+
+		/* Assume we've already mapped this early memory */
+		memset(__va(addr), 0, size);
+
+		addr += size;
+	}
+
+	printk(KERN_INFO "Scanning %d areas for low memory corruption\n",
+	       num_scan_areas);
+	update_e820();
+}
+
+
+void check_for_bios_corruption(void)
+{
+	int i;
+	int corruption = 0;
+
+	if (!memory_corruption_check)
+		return;
+
+	for (i = 0; i < num_scan_areas; i++) {
+		unsigned long *addr = __va(scan_areas[i].addr);
+		unsigned long size = scan_areas[i].size;
+
+		for (; size; addr++, size -= sizeof(unsigned long)) {
+			if (!*addr)
+				continue;
+			printk(KERN_ERR "Corrupted low memory at %p (%lx phys) = %08lx\n",
+			       addr, __pa(addr), *addr);
+			corruption = 1;
+			*addr = 0;
+		}
+	}
+
+	WARN_ONCE(corruption, KERN_ERR "Memory corruption detected in low memory\n");
+}
+
+static void check_corruption(struct work_struct *dummy);
+static DECLARE_DELAYED_WORK(bios_check_work, check_corruption);
+
+static void check_corruption(struct work_struct *dummy)
+{
+	check_for_bios_corruption();
+	schedule_delayed_work(&bios_check_work,
+		round_jiffies_relative(corruption_check_period*HZ)); 
+}
+
+static int start_periodic_check_for_corruption(void)
+{
+	if (!memory_corruption_check || corruption_check_period == 0)
+		return 0;
+
+	printk(KERN_INFO "Scanning for low memory corruption every %d seconds\n",
+	       corruption_check_period);
+
+	/* First time we run the checks right away */
+	schedule_delayed_work(&bios_check_work, 0);
+	return 0;
+}
+
+module_init(start_periodic_check_for_corruption);
+
diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile
index 82ec6075c057..82db7f45e2de 100644
--- a/arch/x86/kernel/cpu/Makefile
+++ b/arch/x86/kernel/cpu/Makefile
@@ -2,8 +2,14 @@
 # Makefile for x86-compatible CPU details and quirks
 #
 
+# Don't trace early stages of a secondary CPU boot
+ifdef CONFIG_FUNCTION_TRACER
+CFLAGS_REMOVE_common.o = -pg
+endif
+
 obj-y			:= intel_cacheinfo.o addon_cpuid_features.o
 obj-y			+= proc.o capflags.o powerflags.o common.o
+obj-y			+= vmware.o hypervisor.o
 
 obj-$(CONFIG_X86_32)	+= bugs.o cmpxchg.o
 obj-$(CONFIG_X86_64)	+= bugs_64.o
diff --git a/arch/x86/kernel/cpu/addon_cpuid_features.c b/arch/x86/kernel/cpu/addon_cpuid_features.c
index ef8f831af823..2cf23634b6d9 100644
--- a/arch/x86/kernel/cpu/addon_cpuid_features.c
+++ b/arch/x86/kernel/cpu/addon_cpuid_features.c
@@ -120,9 +120,17 @@ void __cpuinit detect_extended_topology(struct cpuinfo_x86 *c)
 	c->cpu_core_id = phys_pkg_id(c->initial_apicid, ht_mask_width)
 						 & core_select_mask;
 	c->phys_proc_id = phys_pkg_id(c->initial_apicid, core_plus_mask_width);
+	/*
+	 * Reinit the apicid, now that we have extended initial_apicid.
+	 */
+	c->apicid = phys_pkg_id(c->initial_apicid, 0);
 #else
 	c->cpu_core_id = phys_pkg_id(ht_mask_width) & core_select_mask;
 	c->phys_proc_id = phys_pkg_id(core_plus_mask_width);
+	/*
+	 * Reinit the apicid, now that we have extended initial_apicid.
+	 */
+	c->apicid = phys_pkg_id(0);
 #endif
 	c->x86_max_cores = (core_level_siblings / smp_num_siblings);
 
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 8f1e31db2ad5..7c878f6aa919 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -283,9 +283,14 @@ static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
 {
 	early_init_amd_mc(c);
 
-	/* c->x86_power is 8000_0007 edx. Bit 8 is constant TSC */
-	if (c->x86_power & (1<<8))
+	/*
+	 * c->x86_power is 8000_0007 edx. Bit 8 is TSC runs at constant rate
+	 * with P/T states and does not stop in deep C-states
+	 */
+	if (c->x86_power & (1 << 8)) {
 		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
+		set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC);
+	}
 
 #ifdef CONFIG_X86_64
 	set_cpu_cap(c, X86_FEATURE_SYSCALL32);
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index b9c9ea0217a9..42e0853030cb 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -36,6 +36,7 @@
 #include <asm/proto.h>
 #include <asm/sections.h>
 #include <asm/setup.h>
+#include <asm/hypervisor.h>
 
 #include "cpu.h"
 
@@ -703,6 +704,7 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 	detect_ht(c);
 #endif
 
+	init_hypervisor(c);
 	/*
 	 * On SMP, boot_cpu_data holds the common feature set between
 	 * all CPUs; so make sure that we indicate which features are
@@ -862,7 +864,7 @@ EXPORT_SYMBOL(_cpu_pda);
 
 struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
 
-char boot_cpu_stack[IRQSTACKSIZE] __page_aligned_bss;
+static char boot_cpu_stack[IRQSTACKSIZE] __page_aligned_bss;
 
 void __cpuinit pda_init(int cpu)
 {
@@ -903,8 +905,8 @@ void __cpuinit pda_init(int cpu)
 	}
 }
 
-char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ +
-			   DEBUG_STKSZ] __page_aligned_bss;
+static char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ +
+				  DEBUG_STKSZ] __page_aligned_bss;
 
 extern asmlinkage void ignore_sysret(void);
 
diff --git a/arch/x86/kernel/cpu/hypervisor.c b/arch/x86/kernel/cpu/hypervisor.c
new file mode 100644
index 000000000000..fb5b86af0b01
--- /dev/null
+++ b/arch/x86/kernel/cpu/hypervisor.c
@@ -0,0 +1,58 @@
+/*
+ * Common hypervisor code
+ *
+ * Copyright (C) 2008, VMware, Inc.
+ * Author : Alok N Kataria <akataria@vmware.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.
+ *
+ * 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, GOOD TITLE or
+ * NON INFRINGEMENT.  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.
+ *
+ */
+
+#include <asm/processor.h>
+#include <asm/vmware.h>
+#include <asm/hypervisor.h>
+
+static inline void __cpuinit
+detect_hypervisor_vendor(struct cpuinfo_x86 *c)
+{
+	if (vmware_platform()) {
+		c->x86_hyper_vendor = X86_HYPER_VENDOR_VMWARE;
+	} else {
+		c->x86_hyper_vendor = X86_HYPER_VENDOR_NONE;
+	}
+}
+
+unsigned long get_hypervisor_tsc_freq(void)
+{
+	if (boot_cpu_data.x86_hyper_vendor == X86_HYPER_VENDOR_VMWARE)
+		return vmware_get_tsc_khz();
+	return 0;
+}
+
+static inline void __cpuinit
+hypervisor_set_feature_bits(struct cpuinfo_x86 *c)
+{
+	if (boot_cpu_data.x86_hyper_vendor == X86_HYPER_VENDOR_VMWARE) {
+		vmware_set_feature_bits(c);
+		return;
+	}
+}
+
+void __cpuinit init_hypervisor(struct cpuinfo_x86 *c)
+{
+	detect_hypervisor_vendor(c);
+	hypervisor_set_feature_bits(c);
+}
diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index 816f27f289b1..8ea6929e974c 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -11,7 +11,6 @@
 #include <asm/pgtable.h>
 #include <asm/msr.h>
 #include <asm/uaccess.h>
-#include <asm/ptrace.h>
 #include <asm/ds.h>
 #include <asm/bugs.h>
 
@@ -41,6 +40,16 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c)
 	if (c->x86 == 15 && c->x86_cache_alignment == 64)
 		c->x86_cache_alignment = 128;
 #endif
+
+	/*
+	 * c->x86_power is 8000_0007 edx. Bit 8 is TSC runs at constant rate
+	 * with P/T states and does not stop in deep C-states
+	 */
+	if (c->x86_power & (1 << 8)) {
+		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
+		set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC);
+	}
+
 }
 
 #ifdef CONFIG_X86_32
@@ -242,6 +251,13 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 
 	intel_workarounds(c);
 
+	/*
+	 * Detect the extended topology information if available. This
+	 * will reinitialise the initial_apicid which will be used
+	 * in init_intel_cacheinfo()
+	 */
+	detect_extended_topology(c);
+
 	l2 = init_intel_cacheinfo(c);
 	if (c->cpuid_level > 9) {
 		unsigned eax = cpuid_eax(10);
@@ -309,10 +325,6 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 		set_cpu_cap(c, X86_FEATURE_P3);
 #endif
 
-	if (cpu_has_bts)
-		ptrace_bts_init_intel(c);
-
-	detect_extended_topology(c);
 	if (!cpu_has(c, X86_FEATURE_XTOPOLOGY)) {
 		/*
 		 * let's use the legacy cpuid vector 0x1 and 0x4 for topology
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 7bd00a565672..48533d77be78 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -641,20 +641,17 @@ static inline ssize_t show_shared_cpu_list(struct _cpuid4_info *leaf, char *buf)
 	return show_shared_cpu_map_func(leaf, 1, buf);
 }
 
-static ssize_t show_type(struct _cpuid4_info *this_leaf, char *buf) {
-	switch(this_leaf->eax.split.type) {
-	    case CACHE_TYPE_DATA:
+static ssize_t show_type(struct _cpuid4_info *this_leaf, char *buf)
+{
+	switch (this_leaf->eax.split.type) {
+	case CACHE_TYPE_DATA:
 		return sprintf(buf, "Data\n");
-		break;
-	    case CACHE_TYPE_INST:
+	case CACHE_TYPE_INST:
 		return sprintf(buf, "Instruction\n");
-		break;
-	    case CACHE_TYPE_UNIFIED:
+	case CACHE_TYPE_UNIFIED:
 		return sprintf(buf, "Unified\n");
-		break;
-	    default:
+	default:
 		return sprintf(buf, "Unknown\n");
-		break;
 	}
 }
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c
index 4b031a4ac856..1c838032fd37 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -510,12 +510,9 @@ static void __cpuinit mce_cpu_features(struct cpuinfo_x86 *c)
  */
 void __cpuinit mcheck_init(struct cpuinfo_x86 *c)
 {
-	static cpumask_t mce_cpus = CPU_MASK_NONE;
-
 	mce_cpu_quirks(c);
 
 	if (mce_dont_init ||
-	    cpu_test_and_set(smp_processor_id(), mce_cpus) ||
 	    !mce_available(c))
 		return;
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
index a1de80f368f1..a5a5e0530370 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
@@ -248,7 +248,7 @@ asmlinkage void mce_threshold_interrupt(void)
 		}
 	}
 out:
-	add_pda(irq_threshold_count, 1);
+	inc_irq_stat(irq_threshold_count);
 	irq_exit();
 }
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c
index c17eaf5dd6dd..4b48f251fd39 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c
@@ -26,7 +26,7 @@ asmlinkage void smp_thermal_interrupt(void)
 	if (therm_throt_process(msr_val & 1))
 		mce_log_therm_throt_event(smp_processor_id(), msr_val);
 
-	add_pda(irq_thermal_count, 1);
+	inc_irq_stat(irq_thermal_count);
 	irq_exit();
 }
 
diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c
index c78c04821ea1..1159e269e596 100644
--- a/arch/x86/kernel/cpu/mtrr/main.c
+++ b/arch/x86/kernel/cpu/mtrr/main.c
@@ -803,6 +803,7 @@ x86_get_mtrr_mem_range(struct res_range *range, int nr_range,
 }
 
 static struct res_range __initdata range[RANGE_NUM];
+static int __initdata nr_range;
 
 #ifdef CONFIG_MTRR_SANITIZER
 
@@ -1206,39 +1207,43 @@ struct mtrr_cleanup_result {
 #define PSHIFT		(PAGE_SHIFT - 10)
 
 static struct mtrr_cleanup_result __initdata result[NUM_RESULT];
-static struct res_range __initdata range_new[RANGE_NUM];
 static unsigned long __initdata min_loss_pfn[RANGE_NUM];
 
-static int __init mtrr_cleanup(unsigned address_bits)
+static void __init print_out_mtrr_range_state(void)
 {
-	unsigned long extra_remove_base, extra_remove_size;
-	unsigned long base, size, def, dummy;
-	mtrr_type type;
-	int nr_range, nr_range_new;
-	u64 chunk_size, gran_size;
-	unsigned long range_sums, range_sums_new;
-	int index_good;
-	int num_reg_good;
 	int i;
+	char start_factor = 'K', size_factor = 'K';
+	unsigned long start_base, size_base;
+	mtrr_type type;
 
-	/* extra one for all 0 */
-	int num[MTRR_NUM_TYPES + 1];
+	for (i = 0; i < num_var_ranges; i++) {
 
-	if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1)
-		return 0;
-	rdmsr(MTRRdefType_MSR, def, dummy);
-	def &= 0xff;
-	if (def != MTRR_TYPE_UNCACHABLE)
-		return 0;
+		size_base = range_state[i].size_pfn << (PAGE_SHIFT - 10);
+		if (!size_base)
+			continue;
 
-	/* get it and store it aside */
-	memset(range_state, 0, sizeof(range_state));
-	for (i = 0; i < num_var_ranges; i++) {
-		mtrr_if->get(i, &base, &size, &type);
-		range_state[i].base_pfn = base;
-		range_state[i].size_pfn = size;
-		range_state[i].type = type;
+		size_base = to_size_factor(size_base, &size_factor),
+		start_base = range_state[i].base_pfn << (PAGE_SHIFT - 10);
+		start_base = to_size_factor(start_base, &start_factor),
+		type = range_state[i].type;
+
+		printk(KERN_DEBUG "reg %d, base: %ld%cB, range: %ld%cB, type %s\n",
+			i, start_base, start_factor,
+			size_base, size_factor,
+			(type == MTRR_TYPE_UNCACHABLE) ? "UC" :
+			    ((type == MTRR_TYPE_WRPROT) ? "WP" :
+			     ((type == MTRR_TYPE_WRBACK) ? "WB" : "Other"))
+			);
 	}
+}
+
+static int __init mtrr_need_cleanup(void)
+{
+	int i;
+	mtrr_type type;
+	unsigned long size;
+	/* extra one for all 0 */
+	int num[MTRR_NUM_TYPES + 1];
 
 	/* check entries number */
 	memset(num, 0, sizeof(num));
@@ -1263,29 +1268,133 @@ static int __init mtrr_cleanup(unsigned address_bits)
 		num_var_ranges - num[MTRR_NUM_TYPES])
 		return 0;
 
-	/* print original var MTRRs at first, for debugging: */
-	printk(KERN_DEBUG "original variable MTRRs\n");
-	for (i = 0; i < num_var_ranges; i++) {
-		char start_factor = 'K', size_factor = 'K';
-		unsigned long start_base, size_base;
+	return 1;
+}
 
-		size_base = range_state[i].size_pfn << (PAGE_SHIFT - 10);
-		if (!size_base)
-			continue;
+static unsigned long __initdata range_sums;
+static void __init mtrr_calc_range_state(u64 chunk_size, u64 gran_size,
+					 unsigned long extra_remove_base,
+					 unsigned long extra_remove_size,
+					 int i)
+{
+	int num_reg;
+	static struct res_range range_new[RANGE_NUM];
+	static int nr_range_new;
+	unsigned long range_sums_new;
+
+	/* convert ranges to var ranges state */
+	num_reg = x86_setup_var_mtrrs(range, nr_range,
+						chunk_size, gran_size);
+
+	/* we got new setting in range_state, check it */
+	memset(range_new, 0, sizeof(range_new));
+	nr_range_new = x86_get_mtrr_mem_range(range_new, 0,
+				extra_remove_base, extra_remove_size);
+	range_sums_new = sum_ranges(range_new, nr_range_new);
+
+	result[i].chunk_sizek = chunk_size >> 10;
+	result[i].gran_sizek = gran_size >> 10;
+	result[i].num_reg = num_reg;
+	if (range_sums < range_sums_new) {
+		result[i].lose_cover_sizek =
+			(range_sums_new - range_sums) << PSHIFT;
+		result[i].bad = 1;
+	} else
+		result[i].lose_cover_sizek =
+			(range_sums - range_sums_new) << PSHIFT;
 
-		size_base = to_size_factor(size_base, &size_factor),
-		start_base = range_state[i].base_pfn << (PAGE_SHIFT - 10);
-		start_base = to_size_factor(start_base, &start_factor),
-		type = range_state[i].type;
+	/* double check it */
+	if (!result[i].bad && !result[i].lose_cover_sizek) {
+		if (nr_range_new != nr_range ||
+			memcmp(range, range_new, sizeof(range)))
+				result[i].bad = 1;
+	}
 
-		printk(KERN_DEBUG "reg %d, base: %ld%cB, range: %ld%cB, type %s\n",
-			i, start_base, start_factor,
-			size_base, size_factor,
-			(type == MTRR_TYPE_UNCACHABLE) ? "UC" :
-			    ((type == MTRR_TYPE_WRPROT) ? "WP" :
-			     ((type == MTRR_TYPE_WRBACK) ? "WB" : "Other"))
-			);
+	if (!result[i].bad && (range_sums - range_sums_new <
+				min_loss_pfn[num_reg])) {
+		min_loss_pfn[num_reg] =
+			range_sums - range_sums_new;
 	}
+}
+
+static void __init mtrr_print_out_one_result(int i)
+{
+	char gran_factor, chunk_factor, lose_factor;
+	unsigned long gran_base, chunk_base, lose_base;
+
+	gran_base = to_size_factor(result[i].gran_sizek, &gran_factor),
+	chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor),
+	lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor),
+	printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t",
+			result[i].bad ? "*BAD*" : " ",
+			gran_base, gran_factor, chunk_base, chunk_factor);
+	printk(KERN_CONT "num_reg: %d  \tlose cover RAM: %s%ld%c\n",
+			result[i].num_reg, result[i].bad ? "-" : "",
+			lose_base, lose_factor);
+}
+
+static int __init mtrr_search_optimal_index(void)
+{
+	int i;
+	int num_reg_good;
+	int index_good;
+
+	if (nr_mtrr_spare_reg >= num_var_ranges)
+		nr_mtrr_spare_reg = num_var_ranges - 1;
+	num_reg_good = -1;
+	for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) {
+		if (!min_loss_pfn[i])
+			num_reg_good = i;
+	}
+
+	index_good = -1;
+	if (num_reg_good != -1) {
+		for (i = 0; i < NUM_RESULT; i++) {
+			if (!result[i].bad &&
+			    result[i].num_reg == num_reg_good &&
+			    !result[i].lose_cover_sizek) {
+				index_good = i;
+				break;
+			}
+		}
+	}
+
+	return index_good;
+}
+
+
+static int __init mtrr_cleanup(unsigned address_bits)
+{
+	unsigned long extra_remove_base, extra_remove_size;
+	unsigned long base, size, def, dummy;
+	mtrr_type type;
+	u64 chunk_size, gran_size;
+	int index_good;
+	int i;
+
+	if (!is_cpu(INTEL) || enable_mtrr_cleanup < 1)
+		return 0;
+	rdmsr(MTRRdefType_MSR, def, dummy);
+	def &= 0xff;
+	if (def != MTRR_TYPE_UNCACHABLE)
+		return 0;
+
+	/* get it and store it aside */
+	memset(range_state, 0, sizeof(range_state));
+	for (i = 0; i < num_var_ranges; i++) {
+		mtrr_if->get(i, &base, &size, &type);
+		range_state[i].base_pfn = base;
+		range_state[i].size_pfn = size;
+		range_state[i].type = type;
+	}
+
+	/* check if we need handle it and can handle it */
+	if (!mtrr_need_cleanup())
+		return 0;
+
+	/* print original var MTRRs at first, for debugging: */
+	printk(KERN_DEBUG "original variable MTRRs\n");
+	print_out_mtrr_range_state();
 
 	memset(range, 0, sizeof(range));
 	extra_remove_size = 0;
@@ -1309,176 +1418,64 @@ static int __init mtrr_cleanup(unsigned address_bits)
 	       range_sums >> (20 - PAGE_SHIFT));
 
 	if (mtrr_chunk_size && mtrr_gran_size) {
-		int num_reg;
-		char gran_factor, chunk_factor, lose_factor;
-		unsigned long gran_base, chunk_base, lose_base;
-
-		debug_print++;
-		/* convert ranges to var ranges state */
-		num_reg = x86_setup_var_mtrrs(range, nr_range, mtrr_chunk_size,
-					      mtrr_gran_size);
+		i = 0;
+		mtrr_calc_range_state(mtrr_chunk_size, mtrr_gran_size,
+				      extra_remove_base, extra_remove_size, i);
 
-		/* we got new setting in range_state, check it */
-		memset(range_new, 0, sizeof(range_new));
-		nr_range_new = x86_get_mtrr_mem_range(range_new, 0,
-						      extra_remove_base,
-						      extra_remove_size);
-		range_sums_new = sum_ranges(range_new, nr_range_new);
+		mtrr_print_out_one_result(i);
 
-		i = 0;
-		result[i].chunk_sizek = mtrr_chunk_size >> 10;
-		result[i].gran_sizek = mtrr_gran_size >> 10;
-		result[i].num_reg = num_reg;
-		if (range_sums < range_sums_new) {
-			result[i].lose_cover_sizek =
-				(range_sums_new - range_sums) << PSHIFT;
-			result[i].bad = 1;
-		} else
-			result[i].lose_cover_sizek =
-				(range_sums - range_sums_new) << PSHIFT;
-
-		gran_base = to_size_factor(result[i].gran_sizek, &gran_factor),
-		chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor),
-		lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor),
-		printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t",
-			 result[i].bad?"*BAD*":" ",
-			 gran_base, gran_factor, chunk_base, chunk_factor);
-		printk(KERN_CONT "num_reg: %d  \tlose cover RAM: %s%ld%c\n",
-			 result[i].num_reg, result[i].bad?"-":"",
-			 lose_base, lose_factor);
 		if (!result[i].bad) {
 			set_var_mtrr_all(address_bits);
 			return 1;
 		}
 		printk(KERN_INFO "invalid mtrr_gran_size or mtrr_chunk_size, "
 		       "will find optimal one\n");
-		debug_print--;
-		memset(result, 0, sizeof(result[0]));
 	}
 
 	i = 0;
 	memset(min_loss_pfn, 0xff, sizeof(min_loss_pfn));
 	memset(result, 0, sizeof(result));
 	for (gran_size = (1ULL<<16); gran_size < (1ULL<<32); gran_size <<= 1) {
-		char gran_factor;
-		unsigned long gran_base;
-
-		if (debug_print)
-			gran_base = to_size_factor(gran_size >> 10, &gran_factor);
 
 		for (chunk_size = gran_size; chunk_size < (1ULL<<32);
 		     chunk_size <<= 1) {
-			int num_reg;
 
-			if (debug_print) {
-				char chunk_factor;
-				unsigned long chunk_base;
-
-				chunk_base = to_size_factor(chunk_size>>10, &chunk_factor),
-				printk(KERN_INFO "\n");
-				printk(KERN_INFO "gran_size: %ld%c   chunk_size: %ld%c \n",
-				       gran_base, gran_factor, chunk_base, chunk_factor);
-			}
 			if (i >= NUM_RESULT)
 				continue;
 
-			/* convert ranges to var ranges state */
-			num_reg = x86_setup_var_mtrrs(range, nr_range,
-							 chunk_size, gran_size);
-
-			/* we got new setting in range_state, check it */
-			memset(range_new, 0, sizeof(range_new));
-			nr_range_new = x86_get_mtrr_mem_range(range_new, 0,
-					 extra_remove_base, extra_remove_size);
-			range_sums_new = sum_ranges(range_new, nr_range_new);
-
-			result[i].chunk_sizek = chunk_size >> 10;
-			result[i].gran_sizek = gran_size >> 10;
-			result[i].num_reg = num_reg;
-			if (range_sums < range_sums_new) {
-				result[i].lose_cover_sizek =
-					(range_sums_new - range_sums) << PSHIFT;
-				result[i].bad = 1;
-			} else
-				result[i].lose_cover_sizek =
-					(range_sums - range_sums_new) << PSHIFT;
-
-			/* double check it */
-			if (!result[i].bad && !result[i].lose_cover_sizek) {
-				if (nr_range_new != nr_range ||
-					memcmp(range, range_new, sizeof(range)))
-						result[i].bad = 1;
+			mtrr_calc_range_state(chunk_size, gran_size,
+				      extra_remove_base, extra_remove_size, i);
+			if (debug_print) {
+				mtrr_print_out_one_result(i);
+				printk(KERN_INFO "\n");
 			}
 
-			if (!result[i].bad && (range_sums - range_sums_new <
-					       min_loss_pfn[num_reg])) {
-				min_loss_pfn[num_reg] =
-					range_sums - range_sums_new;
-			}
 			i++;
 		}
 	}
 
-	/* print out all */
-	for (i = 0; i < NUM_RESULT; i++) {
-		char gran_factor, chunk_factor, lose_factor;
-		unsigned long gran_base, chunk_base, lose_base;
-
-		gran_base = to_size_factor(result[i].gran_sizek, &gran_factor),
-		chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor),
-		lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor),
-		printk(KERN_INFO "%sgran_size: %ld%c \tchunk_size: %ld%c \t",
-			 result[i].bad?"*BAD*":" ",
-			 gran_base, gran_factor, chunk_base, chunk_factor);
-		printk(KERN_CONT "num_reg: %d  \tlose cover RAM: %s%ld%c\n",
-			 result[i].num_reg, result[i].bad?"-":"",
-			 lose_base, lose_factor);
-	}
-
 	/* try to find the optimal index */
-	if (nr_mtrr_spare_reg >= num_var_ranges)
-		nr_mtrr_spare_reg = num_var_ranges - 1;
-	num_reg_good = -1;
-	for (i = num_var_ranges - nr_mtrr_spare_reg; i > 0; i--) {
-		if (!min_loss_pfn[i])
-			num_reg_good = i;
-	}
-
-	index_good = -1;
-	if (num_reg_good != -1) {
-		for (i = 0; i < NUM_RESULT; i++) {
-			if (!result[i].bad &&
-			    result[i].num_reg == num_reg_good &&
-			    !result[i].lose_cover_sizek) {
-				index_good = i;
-				break;
-			}
-		}
-	}
+	index_good = mtrr_search_optimal_index();
 
 	if (index_good != -1) {
-		char gran_factor, chunk_factor, lose_factor;
-		unsigned long gran_base, chunk_base, lose_base;
-
 		printk(KERN_INFO "Found optimal setting for mtrr clean up\n");
 		i = index_good;
-		gran_base = to_size_factor(result[i].gran_sizek, &gran_factor),
-		chunk_base = to_size_factor(result[i].chunk_sizek, &chunk_factor),
-		lose_base = to_size_factor(result[i].lose_cover_sizek, &lose_factor),
-		printk(KERN_INFO "gran_size: %ld%c \tchunk_size: %ld%c \t",
-			 gran_base, gran_factor, chunk_base, chunk_factor);
-		printk(KERN_CONT "num_reg: %d  \tlose RAM: %ld%c\n",
-			 result[i].num_reg, lose_base, lose_factor);
+		mtrr_print_out_one_result(i);
+
 		/* convert ranges to var ranges state */
 		chunk_size = result[i].chunk_sizek;
 		chunk_size <<= 10;
 		gran_size = result[i].gran_sizek;
 		gran_size <<= 10;
-		debug_print++;
 		x86_setup_var_mtrrs(range, nr_range, chunk_size, gran_size);
-		debug_print--;
 		set_var_mtrr_all(address_bits);
+		printk(KERN_DEBUG "New variable MTRRs\n");
+		print_out_mtrr_range_state();
 		return 1;
+	} else {
+		/* print out all */
+		for (i = 0; i < NUM_RESULT; i++)
+			mtrr_print_out_one_result(i);
 	}
 
 	printk(KERN_INFO "mtrr_cleanup: can not find optimal value\n");
@@ -1562,7 +1559,6 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn)
 {
 	unsigned long i, base, size, highest_pfn = 0, def, dummy;
 	mtrr_type type;
-	int nr_range;
 	u64 total_trim_size;
 
 	/* extra one for all 0 */
diff --git a/arch/x86/kernel/cpu/vmware.c b/arch/x86/kernel/cpu/vmware.c
new file mode 100644
index 000000000000..284c399e3234
--- /dev/null
+++ b/arch/x86/kernel/cpu/vmware.c
@@ -0,0 +1,112 @@
+/*
+ * VMware Detection code.
+ *
+ * Copyright (C) 2008, VMware, Inc.
+ * Author : Alok N Kataria <akataria@vmware.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.
+ *
+ * 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, GOOD TITLE or
+ * NON INFRINGEMENT.  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.
+ *
+ */
+
+#include <linux/dmi.h>
+#include <asm/div64.h>
+#include <asm/vmware.h>
+
+#define CPUID_VMWARE_INFO_LEAF	0x40000000
+#define VMWARE_HYPERVISOR_MAGIC	0x564D5868
+#define VMWARE_HYPERVISOR_PORT	0x5658
+
+#define VMWARE_PORT_CMD_GETVERSION	10
+#define VMWARE_PORT_CMD_GETHZ		45
+
+#define VMWARE_PORT(cmd, eax, ebx, ecx, edx)				\
+	__asm__("inl (%%dx)" :						\
+			"=a"(eax), "=c"(ecx), "=d"(edx), "=b"(ebx) :	\
+			"0"(VMWARE_HYPERVISOR_MAGIC),			\
+			"1"(VMWARE_PORT_CMD_##cmd),			\
+			"2"(VMWARE_HYPERVISOR_PORT), "3"(UINT_MAX) :	\
+			"memory");
+
+static inline int __vmware_platform(void)
+{
+	uint32_t eax, ebx, ecx, edx;
+	VMWARE_PORT(GETVERSION, eax, ebx, ecx, edx);
+	return eax != (uint32_t)-1 && ebx == VMWARE_HYPERVISOR_MAGIC;
+}
+
+static unsigned long __vmware_get_tsc_khz(void)
+{
+        uint64_t tsc_hz;
+        uint32_t eax, ebx, ecx, edx;
+
+        VMWARE_PORT(GETHZ, eax, ebx, ecx, edx);
+
+        if (ebx == UINT_MAX)
+                return 0;
+        tsc_hz = eax | (((uint64_t)ebx) << 32);
+        do_div(tsc_hz, 1000);
+        BUG_ON(tsc_hz >> 32);
+        return tsc_hz;
+}
+
+/*
+ * While checking the dmi string infomation, just checking the product
+ * serial key should be enough, as this will always have a VMware
+ * specific string when running under VMware hypervisor.
+ */
+int vmware_platform(void)
+{
+	if (cpu_has_hypervisor) {
+		unsigned int eax, ebx, ecx, edx;
+		char hyper_vendor_id[13];
+
+		cpuid(CPUID_VMWARE_INFO_LEAF, &eax, &ebx, &ecx, &edx);
+		memcpy(hyper_vendor_id + 0, &ebx, 4);
+		memcpy(hyper_vendor_id + 4, &ecx, 4);
+		memcpy(hyper_vendor_id + 8, &edx, 4);
+		hyper_vendor_id[12] = '\0';
+		if (!strcmp(hyper_vendor_id, "VMwareVMware"))
+			return 1;
+	} else if (dmi_available && dmi_name_in_serial("VMware") &&
+		   __vmware_platform())
+		return 1;
+
+	return 0;
+}
+
+unsigned long vmware_get_tsc_khz(void)
+{
+	BUG_ON(!vmware_platform());
+	return __vmware_get_tsc_khz();
+}
+
+/*
+ * VMware hypervisor takes care of exporting a reliable TSC to the guest.
+ * Still, due to timing difference when running on virtual cpus, the TSC can
+ * be marked as unstable in some cases. For example, the TSC sync check at
+ * bootup can fail due to a marginal offset between vcpus' TSCs (though the
+ * TSCs do not drift from each other).  Also, the ACPI PM timer clocksource
+ * is not suitable as a watchdog when running on a hypervisor because the
+ * kernel may miss a wrap of the counter if the vcpu is descheduled for a
+ * long time. To skip these checks at runtime we set these capability bits,
+ * so that the kernel could just trust the hypervisor with providing a
+ * reliable virtual TSC that is suitable for timekeeping.
+ */
+void __cpuinit vmware_set_feature_bits(struct cpuinfo_x86 *c)
+{
+	set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
+	set_cpu_cap(c, X86_FEATURE_TSC_RELIABLE);
+}
diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c
index 19a8c2c0389f..da91701a2348 100644
--- a/arch/x86/kernel/ds.c
+++ b/arch/x86/kernel/ds.c
@@ -6,13 +6,13 @@
  * precise-event based sampling (PEBS).
  *
  * It manages:
- * - per-thread and per-cpu allocation of BTS and PEBS
+ * - DS and BTS hardware configuration
  * - buffer overflow handling (to be done)
  * - buffer access
  *
- * It assumes:
- * - get_task_struct on all traced tasks
- * - current is allowed to trace tasks
+ * It does not do:
+ * - security checking (is the caller allowed to trace the task)
+ * - buffer allocation (memory accounting)
  *
  *
  * Copyright (C) 2007-2008 Intel Corporation.
@@ -34,15 +34,30 @@
  * The configuration for a particular DS hardware implementation.
  */
 struct ds_configuration {
-	/* the size of the DS structure in bytes */
-	unsigned char  sizeof_ds;
-	/* the size of one pointer-typed field in the DS structure in bytes;
-	   this covers the first 8 fields related to buffer management. */
+	/* the name of the configuration */
+	const char *name;
+	/* the size of one pointer-typed field in the DS structure and
+	   in the BTS and PEBS buffers in bytes;
+	   this covers the first 8 DS fields related to buffer management. */
 	unsigned char  sizeof_field;
 	/* the size of a BTS/PEBS record in bytes */
 	unsigned char  sizeof_rec[2];
+	/* a series of bit-masks to control various features indexed
+	 * by enum ds_feature */
+	unsigned long ctl[dsf_ctl_max];
 };
-static struct ds_configuration ds_cfg;
+static DEFINE_PER_CPU(struct ds_configuration, ds_cfg_array);
+
+#define ds_cfg per_cpu(ds_cfg_array, smp_processor_id())
+
+#define MAX_SIZEOF_DS (12 * 8)	/* maximal size of a DS configuration */
+#define MAX_SIZEOF_BTS (3 * 8)	/* maximal size of a BTS record */
+#define DS_ALIGNMENT (1 << 3)	/* BTS and PEBS buffer alignment */
+
+#define BTS_CONTROL \
+ (ds_cfg.ctl[dsf_bts] | ds_cfg.ctl[dsf_bts_kernel] | ds_cfg.ctl[dsf_bts_user] |\
+  ds_cfg.ctl[dsf_bts_overflow])
+
 
 /*
  * A BTS or PEBS tracer.
@@ -61,6 +76,8 @@ struct ds_tracer {
 struct bts_tracer {
 	/* the common DS part */
 	struct ds_tracer ds;
+	/* the trace including the DS configuration */
+	struct bts_trace trace;
 	/* buffer overflow notification function */
 	bts_ovfl_callback_t ovfl;
 };
@@ -68,6 +85,8 @@ struct bts_tracer {
 struct pebs_tracer {
 	/* the common DS part */
 	struct ds_tracer ds;
+	/* the trace including the DS configuration */
+	struct pebs_trace trace;
 	/* buffer overflow notification function */
 	pebs_ovfl_callback_t ovfl;
 };
@@ -134,13 +153,11 @@ static inline void ds_set(unsigned char *base, enum ds_qualifier qual,
 	(*(unsigned long *)base) = value;
 }
 
-#define DS_ALIGNMENT (1 << 3)	/* BTS and PEBS buffer alignment */
-
 
 /*
  * Locking is done only for allocating BTS or PEBS resources.
  */
-static spinlock_t ds_lock = __SPIN_LOCK_UNLOCKED(ds_lock);
+static DEFINE_SPINLOCK(ds_lock);
 
 
 /*
@@ -156,27 +173,32 @@ static spinlock_t ds_lock = __SPIN_LOCK_UNLOCKED(ds_lock);
  *   >0  number of per-thread tracers
  *   <0  number of per-cpu tracers
  *
- * The below functions to get and put tracers and to check the
- * allocation type require the ds_lock to be held by the caller.
- *
  * Tracers essentially gives the number of ds contexts for a certain
  * type of allocation.
  */
-static long tracers;
+static atomic_t tracers = ATOMIC_INIT(0);
 
 static inline void get_tracer(struct task_struct *task)
 {
-	tracers += (task ? 1 : -1);
+	if (task)
+		atomic_inc(&tracers);
+	else
+		atomic_dec(&tracers);
 }
 
 static inline void put_tracer(struct task_struct *task)
 {
-	tracers -= (task ? 1 : -1);
+	if (task)
+		atomic_dec(&tracers);
+	else
+		atomic_inc(&tracers);
 }
 
 static inline int check_tracer(struct task_struct *task)
 {
-	return (task ? (tracers >= 0) : (tracers <= 0));
+	return task ?
+		(atomic_read(&tracers) >= 0) :
+		(atomic_read(&tracers) <= 0);
 }
 
 
@@ -190,46 +212,66 @@ static inline int check_tracer(struct task_struct *task)
  * Contexts are use-counted. They are allocated on first access and
  * deallocated when the last user puts the context.
  */
-static DEFINE_PER_CPU(struct ds_context *, system_context);
+struct ds_context {
+	/* pointer to the DS configuration; goes into MSR_IA32_DS_AREA */
+	unsigned char ds[MAX_SIZEOF_DS];
+	/* the owner of the BTS and PEBS configuration, respectively */
+	struct bts_tracer *bts_master;
+	struct pebs_tracer *pebs_master;
+	/* use count */
+	unsigned long count;
+	/* a pointer to the context location inside the thread_struct
+	 * or the per_cpu context array */
+	struct ds_context **this;
+	/* a pointer to the task owning this context, or NULL, if the
+	 * context is owned by a cpu */
+	struct task_struct *task;
+};
+
+static DEFINE_PER_CPU(struct ds_context *, system_context_array);
+
+#define system_context per_cpu(system_context_array, smp_processor_id())
 
-#define this_system_context per_cpu(system_context, smp_processor_id())
 
 static inline struct ds_context *ds_get_context(struct task_struct *task)
 {
 	struct ds_context **p_context =
-		(task ? &task->thread.ds_ctx : &this_system_context);
-	struct ds_context *context = *p_context;
+		(task ? &task->thread.ds_ctx : &system_context);
+	struct ds_context *context = NULL;
+	struct ds_context *new_context = NULL;
 	unsigned long irq;
 
-	if (!context) {
-		context = kzalloc(sizeof(*context), GFP_KERNEL);
-		if (!context)
-			return NULL;
+	/* Chances are small that we already have a context. */
+	new_context = kzalloc(sizeof(*new_context), GFP_KERNEL);
+	if (!new_context)
+		return NULL;
 
-		spin_lock_irqsave(&ds_lock, irq);
+	spin_lock_irqsave(&ds_lock, irq);
 
-		if (*p_context) {
-			kfree(context);
+	context = *p_context;
+	if (!context) {
+		context = new_context;
 
-			context = *p_context;
-		} else {
-			*p_context = context;
+		context->this = p_context;
+		context->task = task;
+		context->count = 0;
 
-			context->this = p_context;
-			context->task = task;
+		if (task)
+			set_tsk_thread_flag(task, TIF_DS_AREA_MSR);
 
-			if (task)
-				set_tsk_thread_flag(task, TIF_DS_AREA_MSR);
+		if (!task || (task == current))
+			wrmsrl(MSR_IA32_DS_AREA, (unsigned long)context->ds);
 
-			if (!task || (task == current))
-				wrmsrl(MSR_IA32_DS_AREA,
-				       (unsigned long)context->ds);
-		}
-		spin_unlock_irqrestore(&ds_lock, irq);
+		*p_context = context;
 	}
 
 	context->count++;
 
+	spin_unlock_irqrestore(&ds_lock, irq);
+
+	if (context != new_context)
+		kfree(new_context);
+
 	return context;
 }
 
@@ -242,8 +284,10 @@ static inline void ds_put_context(struct ds_context *context)
 
 	spin_lock_irqsave(&ds_lock, irq);
 
-	if (--context->count)
-		goto out;
+	if (--context->count) {
+		spin_unlock_irqrestore(&ds_lock, irq);
+		return;
+	}
 
 	*(context->this) = NULL;
 
@@ -253,14 +297,14 @@ static inline void ds_put_context(struct ds_context *context)
 	if (!context->task || (context->task == current))
 		wrmsrl(MSR_IA32_DS_AREA, 0);
 
-	kfree(context);
- out:
 	spin_unlock_irqrestore(&ds_lock, irq);
+
+	kfree(context);
 }
 
 
 /*
- * Handle a buffer overflow
+ * Call the tracer's callback on a buffer overflow.
  *
  * context: the ds context
  * qual: the buffer type
@@ -268,30 +312,247 @@ static inline void ds_put_context(struct ds_context *context)
 static void ds_overflow(struct ds_context *context, enum ds_qualifier qual)
 {
 	switch (qual) {
-	case ds_bts: {
-		struct bts_tracer *tracer =
-			container_of(context->owner[qual],
-				     struct bts_tracer, ds);
-		if (tracer->ovfl)
-			tracer->ovfl(tracer);
-	}
+	case ds_bts:
+		if (context->bts_master &&
+		    context->bts_master->ovfl)
+			context->bts_master->ovfl(context->bts_master);
 		break;
-	case ds_pebs: {
-		struct pebs_tracer *tracer =
-			container_of(context->owner[qual],
-				     struct pebs_tracer, ds);
-		if (tracer->ovfl)
-			tracer->ovfl(tracer);
+	case ds_pebs:
+		if (context->pebs_master &&
+		    context->pebs_master->ovfl)
+			context->pebs_master->ovfl(context->pebs_master);
+		break;
+	}
+}
+
+
+/*
+ * Write raw data into the BTS or PEBS buffer.
+ *
+ * The remainder of any partially written record is zeroed out.
+ *
+ * context: the DS context
+ * qual: the buffer type
+ * record: the data to write
+ * size: the size of the data
+ */
+static int ds_write(struct ds_context *context, enum ds_qualifier qual,
+		    const void *record, size_t size)
+{
+	int bytes_written = 0;
+
+	if (!record)
+		return -EINVAL;
+
+	while (size) {
+		unsigned long base, index, end, write_end, int_th;
+		unsigned long write_size, adj_write_size;
+
+		/*
+		 * write as much as possible without producing an
+		 * overflow interrupt.
+		 *
+		 * interrupt_threshold must either be
+		 * - bigger than absolute_maximum or
+		 * - point to a record between buffer_base and absolute_maximum
+		 *
+		 * index points to a valid record.
+		 */
+		base   = ds_get(context->ds, qual, ds_buffer_base);
+		index  = ds_get(context->ds, qual, ds_index);
+		end    = ds_get(context->ds, qual, ds_absolute_maximum);
+		int_th = ds_get(context->ds, qual, ds_interrupt_threshold);
+
+		write_end = min(end, int_th);
+
+		/* if we are already beyond the interrupt threshold,
+		 * we fill the entire buffer */
+		if (write_end <= index)
+			write_end = end;
+
+		if (write_end <= index)
+			break;
+
+		write_size = min((unsigned long) size, write_end - index);
+		memcpy((void *)index, record, write_size);
+
+		record = (const char *)record + write_size;
+		size -= write_size;
+		bytes_written += write_size;
+
+		adj_write_size = write_size / ds_cfg.sizeof_rec[qual];
+		adj_write_size *= ds_cfg.sizeof_rec[qual];
+
+		/* zero out trailing bytes */
+		memset((char *)index + write_size, 0,
+		       adj_write_size - write_size);
+		index += adj_write_size;
+
+		if (index >= end)
+			index = base;
+		ds_set(context->ds, qual, ds_index, index);
+
+		if (index >= int_th)
+			ds_overflow(context, qual);
+	}
+
+	return bytes_written;
+}
+
+
+/*
+ * Branch Trace Store (BTS) uses the following format. Different
+ * architectures vary in the size of those fields.
+ * - source linear address
+ * - destination linear address
+ * - flags
+ *
+ * Later architectures use 64bit pointers throughout, whereas earlier
+ * architectures use 32bit pointers in 32bit mode.
+ *
+ * We compute the base address for the first 8 fields based on:
+ * - the field size stored in the DS configuration
+ * - the relative field position
+ *
+ * In order to store additional information in the BTS buffer, we use
+ * a special source address to indicate that the record requires
+ * special interpretation.
+ *
+ * Netburst indicated via a bit in the flags field whether the branch
+ * was predicted; this is ignored.
+ *
+ * We use two levels of abstraction:
+ * - the raw data level defined here
+ * - an arch-independent level defined in ds.h
+ */
+
+enum bts_field {
+	bts_from,
+	bts_to,
+	bts_flags,
+
+	bts_qual = bts_from,
+	bts_jiffies = bts_to,
+	bts_pid = bts_flags,
+
+	bts_qual_mask = (bts_qual_max - 1),
+	bts_escape = ((unsigned long)-1 & ~bts_qual_mask)
+};
+
+static inline unsigned long bts_get(const char *base, enum bts_field field)
+{
+	base += (ds_cfg.sizeof_field * field);
+	return *(unsigned long *)base;
+}
+
+static inline void bts_set(char *base, enum bts_field field, unsigned long val)
+{
+	base += (ds_cfg.sizeof_field * field);;
+	(*(unsigned long *)base) = val;
+}
+
+
+/*
+ * The raw BTS data is architecture dependent.
+ *
+ * For higher-level users, we give an arch-independent view.
+ * - ds.h defines struct bts_struct
+ * - bts_read translates one raw bts record into a bts_struct
+ * - bts_write translates one bts_struct into the raw format and
+ *   writes it into the top of the parameter tracer's buffer.
+ *
+ * return: bytes read/written on success; -Eerrno, otherwise
+ */
+static int bts_read(struct bts_tracer *tracer, const void *at,
+		    struct bts_struct *out)
+{
+	if (!tracer)
+		return -EINVAL;
+
+	if (at < tracer->trace.ds.begin)
+		return -EINVAL;
+
+	if (tracer->trace.ds.end < (at + tracer->trace.ds.size))
+		return -EINVAL;
+
+	memset(out, 0, sizeof(*out));
+	if ((bts_get(at, bts_qual) & ~bts_qual_mask) == bts_escape) {
+		out->qualifier = (bts_get(at, bts_qual) & bts_qual_mask);
+		out->variant.timestamp.jiffies = bts_get(at, bts_jiffies);
+		out->variant.timestamp.pid = bts_get(at, bts_pid);
+	} else {
+		out->qualifier = bts_branch;
+		out->variant.lbr.from = bts_get(at, bts_from);
+		out->variant.lbr.to   = bts_get(at, bts_to);
+
+		if (!out->variant.lbr.from && !out->variant.lbr.to)
+			out->qualifier = bts_invalid;
 	}
+
+	return ds_cfg.sizeof_rec[ds_bts];
+}
+
+static int bts_write(struct bts_tracer *tracer, const struct bts_struct *in)
+{
+	unsigned char raw[MAX_SIZEOF_BTS];
+
+	if (!tracer)
+		return -EINVAL;
+
+	if (MAX_SIZEOF_BTS < ds_cfg.sizeof_rec[ds_bts])
+		return -EOVERFLOW;
+
+	switch (in->qualifier) {
+	case bts_invalid:
+		bts_set(raw, bts_from, 0);
+		bts_set(raw, bts_to, 0);
+		bts_set(raw, bts_flags, 0);
+		break;
+	case bts_branch:
+		bts_set(raw, bts_from, in->variant.lbr.from);
+		bts_set(raw, bts_to,   in->variant.lbr.to);
+		bts_set(raw, bts_flags, 0);
 		break;
+	case bts_task_arrives:
+	case bts_task_departs:
+		bts_set(raw, bts_qual, (bts_escape | in->qualifier));
+		bts_set(raw, bts_jiffies, in->variant.timestamp.jiffies);
+		bts_set(raw, bts_pid, in->variant.timestamp.pid);
+		break;
+	default:
+		return -EINVAL;
 	}
+
+	return ds_write(tracer->ds.context, ds_bts, raw,
+			ds_cfg.sizeof_rec[ds_bts]);
 }
 
 
-static void ds_install_ds_config(struct ds_context *context,
-				 enum ds_qualifier qual,
-				 void *base, size_t size, size_t ith)
+static void ds_write_config(struct ds_context *context,
+			    struct ds_trace *cfg, enum ds_qualifier qual)
 {
+	unsigned char *ds = context->ds;
+
+	ds_set(ds, qual, ds_buffer_base, (unsigned long)cfg->begin);
+	ds_set(ds, qual, ds_index, (unsigned long)cfg->top);
+	ds_set(ds, qual, ds_absolute_maximum, (unsigned long)cfg->end);
+	ds_set(ds, qual, ds_interrupt_threshold, (unsigned long)cfg->ith);
+}
+
+static void ds_read_config(struct ds_context *context,
+			   struct ds_trace *cfg, enum ds_qualifier qual)
+{
+	unsigned char *ds = context->ds;
+
+	cfg->begin = (void *)ds_get(ds, qual, ds_buffer_base);
+	cfg->top = (void *)ds_get(ds, qual, ds_index);
+	cfg->end = (void *)ds_get(ds, qual, ds_absolute_maximum);
+	cfg->ith = (void *)ds_get(ds, qual, ds_interrupt_threshold);
+}
+
+static void ds_init_ds_trace(struct ds_trace *trace, enum ds_qualifier qual,
+			     void *base, size_t size, size_t ith,
+			     unsigned int flags) {
 	unsigned long buffer, adj;
 
 	/* adjust the buffer address and size to meet alignment
@@ -308,32 +569,30 @@ static void ds_install_ds_config(struct ds_context *context,
 	buffer += adj;
 	size   -= adj;
 
-	size /= ds_cfg.sizeof_rec[qual];
-	size *= ds_cfg.sizeof_rec[qual];
+	trace->n = size / ds_cfg.sizeof_rec[qual];
+	trace->size = ds_cfg.sizeof_rec[qual];
 
-	ds_set(context->ds, qual, ds_buffer_base, buffer);
-	ds_set(context->ds, qual, ds_index, buffer);
-	ds_set(context->ds, qual, ds_absolute_maximum, buffer + size);
+	size = (trace->n * trace->size);
 
+	trace->begin = (void *)buffer;
+	trace->top = trace->begin;
+	trace->end = (void *)(buffer + size);
 	/* The value for 'no threshold' is -1, which will set the
 	 * threshold outside of the buffer, just like we want it.
 	 */
-	ds_set(context->ds, qual,
-	       ds_interrupt_threshold, buffer + size - ith);
+	trace->ith = (void *)(buffer + size - ith);
+
+	trace->flags = flags;
 }
 
-static int ds_request(struct ds_tracer *tracer, enum ds_qualifier qual,
-		      struct task_struct *task,
-		      void *base, size_t size, size_t th)
+
+static int ds_request(struct ds_tracer *tracer, struct ds_trace *trace,
+		      enum ds_qualifier qual, struct task_struct *task,
+		      void *base, size_t size, size_t th, unsigned int flags)
 {
 	struct ds_context *context;
-	unsigned long irq;
 	int error;
 
-	error = -EOPNOTSUPP;
-	if (!ds_cfg.sizeof_ds)
-		goto out;
-
 	error = -EINVAL;
 	if (!base)
 		goto out;
@@ -360,43 +619,26 @@ static int ds_request(struct ds_tracer *tracer, enum ds_qualifier qual,
 		goto out;
 	tracer->context = context;
 
+	ds_init_ds_trace(trace, qual, base, size, th, flags);
 
-	spin_lock_irqsave(&ds_lock, irq);
-
-	error = -EPERM;
-	if (!check_tracer(task))
-		goto out_unlock;
-	get_tracer(task);
-
-	error = -EPERM;
-	if (context->owner[qual])
-		goto out_put_tracer;
-	context->owner[qual] = tracer;
-
-	spin_unlock_irqrestore(&ds_lock, irq);
-
-
-	ds_install_ds_config(context, qual, base, size, th);
-
-	return 0;
-
- out_put_tracer:
-	put_tracer(task);
- out_unlock:
-	spin_unlock_irqrestore(&ds_lock, irq);
-	ds_put_context(context);
-	tracer->context = NULL;
+	error = 0;
  out:
 	return error;
 }
 
 struct bts_tracer *ds_request_bts(struct task_struct *task,
 				  void *base, size_t size,
-				  bts_ovfl_callback_t ovfl, size_t th)
+				  bts_ovfl_callback_t ovfl, size_t th,
+				  unsigned int flags)
 {
 	struct bts_tracer *tracer;
+	unsigned long irq;
 	int error;
 
+	error = -EOPNOTSUPP;
+	if (!ds_cfg.ctl[dsf_bts])
+		goto out;
+
 	/* buffer overflow notification is not yet implemented */
 	error = -EOPNOTSUPP;
 	if (ovfl)
@@ -408,12 +650,40 @@ struct bts_tracer *ds_request_bts(struct task_struct *task,
 		goto out;
 	tracer->ovfl = ovfl;
 
-	error = ds_request(&tracer->ds, ds_bts, task, base, size, th);
+	error = ds_request(&tracer->ds, &tracer->trace.ds,
+			   ds_bts, task, base, size, th, flags);
 	if (error < 0)
 		goto out_tracer;
 
+
+	spin_lock_irqsave(&ds_lock, irq);
+
+	error = -EPERM;
+	if (!check_tracer(task))
+		goto out_unlock;
+	get_tracer(task);
+
+	error = -EPERM;
+	if (tracer->ds.context->bts_master)
+		goto out_put_tracer;
+	tracer->ds.context->bts_master = tracer;
+
+	spin_unlock_irqrestore(&ds_lock, irq);
+
+
+	tracer->trace.read  = bts_read;
+	tracer->trace.write = bts_write;
+
+	ds_write_config(tracer->ds.context, &tracer->trace.ds, ds_bts);
+	ds_resume_bts(tracer);
+
 	return tracer;
 
+ out_put_tracer:
+	put_tracer(task);
+ out_unlock:
+	spin_unlock_irqrestore(&ds_lock, irq);
+	ds_put_context(tracer->ds.context);
  out_tracer:
 	kfree(tracer);
  out:
@@ -422,9 +692,11 @@ struct bts_tracer *ds_request_bts(struct task_struct *task,
 
 struct pebs_tracer *ds_request_pebs(struct task_struct *task,
 				    void *base, size_t size,
-				    pebs_ovfl_callback_t ovfl, size_t th)
+				    pebs_ovfl_callback_t ovfl, size_t th,
+				    unsigned int flags)
 {
 	struct pebs_tracer *tracer;
+	unsigned long irq;
 	int error;
 
 	/* buffer overflow notification is not yet implemented */
@@ -438,300 +710,171 @@ struct pebs_tracer *ds_request_pebs(struct task_struct *task,
 		goto out;
 	tracer->ovfl = ovfl;
 
-	error = ds_request(&tracer->ds, ds_pebs, task, base, size, th);
+	error = ds_request(&tracer->ds, &tracer->trace.ds,
+			   ds_pebs, task, base, size, th, flags);
 	if (error < 0)
 		goto out_tracer;
 
+	spin_lock_irqsave(&ds_lock, irq);
+
+	error = -EPERM;
+	if (!check_tracer(task))
+		goto out_unlock;
+	get_tracer(task);
+
+	error = -EPERM;
+	if (tracer->ds.context->pebs_master)
+		goto out_put_tracer;
+	tracer->ds.context->pebs_master = tracer;
+
+	spin_unlock_irqrestore(&ds_lock, irq);
+
+	ds_write_config(tracer->ds.context, &tracer->trace.ds, ds_bts);
+	ds_resume_pebs(tracer);
+
 	return tracer;
 
+ out_put_tracer:
+	put_tracer(task);
+ out_unlock:
+	spin_unlock_irqrestore(&ds_lock, irq);
+	ds_put_context(tracer->ds.context);
  out_tracer:
 	kfree(tracer);
  out:
 	return ERR_PTR(error);
 }
 
-static void ds_release(struct ds_tracer *tracer, enum ds_qualifier qual)
-{
-	BUG_ON(tracer->context->owner[qual] != tracer);
-	tracer->context->owner[qual] = NULL;
-
-	put_tracer(tracer->context->task);
-	ds_put_context(tracer->context);
-}
-
-int ds_release_bts(struct bts_tracer *tracer)
+void ds_release_bts(struct bts_tracer *tracer)
 {
 	if (!tracer)
-		return -EINVAL;
+		return;
 
-	ds_release(&tracer->ds, ds_bts);
-	kfree(tracer);
+	ds_suspend_bts(tracer);
 
-	return 0;
-}
+	WARN_ON_ONCE(tracer->ds.context->bts_master != tracer);
+	tracer->ds.context->bts_master = NULL;
 
-int ds_release_pebs(struct pebs_tracer *tracer)
-{
-	if (!tracer)
-		return -EINVAL;
+	put_tracer(tracer->ds.context->task);
+	ds_put_context(tracer->ds.context);
 
-	ds_release(&tracer->ds, ds_pebs);
 	kfree(tracer);
-
-	return 0;
 }
 
-static size_t ds_get_index(struct ds_context *context, enum ds_qualifier qual)
+void ds_suspend_bts(struct bts_tracer *tracer)
 {
-	unsigned long base, index;
-
-	base  = ds_get(context->ds, qual, ds_buffer_base);
-	index = ds_get(context->ds, qual, ds_index);
-
-	return (index - base) / ds_cfg.sizeof_rec[qual];
-}
+	struct task_struct *task;
 
-int ds_get_bts_index(struct bts_tracer *tracer, size_t *pos)
-{
 	if (!tracer)
-		return -EINVAL;
-
-	if (!pos)
-		return -EINVAL;
-
-	*pos = ds_get_index(tracer->ds.context, ds_bts);
-
-	return 0;
-}
+		return;
 
-int ds_get_pebs_index(struct pebs_tracer *tracer, size_t *pos)
-{
-	if (!tracer)
-		return -EINVAL;
+	task = tracer->ds.context->task;
 
-	if (!pos)
-		return -EINVAL;
+	if (!task || (task == current))
+		update_debugctlmsr(get_debugctlmsr() & ~BTS_CONTROL);
 
-	*pos = ds_get_index(tracer->ds.context, ds_pebs);
+	if (task) {
+		task->thread.debugctlmsr &= ~BTS_CONTROL;
 
-	return 0;
+		if (!task->thread.debugctlmsr)
+			clear_tsk_thread_flag(task, TIF_DEBUGCTLMSR);
+	}
 }
 
-static size_t ds_get_end(struct ds_context *context, enum ds_qualifier qual)
+void ds_resume_bts(struct bts_tracer *tracer)
 {
-	unsigned long base, max;
+	struct task_struct *task;
+	unsigned long control;
 
-	base = ds_get(context->ds, qual, ds_buffer_base);
-	max  = ds_get(context->ds, qual, ds_absolute_maximum);
-
-	return (max - base) / ds_cfg.sizeof_rec[qual];
-}
-
-int ds_get_bts_end(struct bts_tracer *tracer, size_t *pos)
-{
 	if (!tracer)
-		return -EINVAL;
-
-	if (!pos)
-		return -EINVAL;
-
-	*pos = ds_get_end(tracer->ds.context, ds_bts);
-
-	return 0;
-}
-
-int ds_get_pebs_end(struct pebs_tracer *tracer, size_t *pos)
-{
-	if (!tracer)
-		return -EINVAL;
-
-	if (!pos)
-		return -EINVAL;
-
-	*pos = ds_get_end(tracer->ds.context, ds_pebs);
-
-	return 0;
-}
-
-static int ds_access(struct ds_context *context, enum ds_qualifier qual,
-		     size_t index, const void **record)
-{
-	unsigned long base, idx;
-
-	if (!record)
-		return -EINVAL;
-
-	base = ds_get(context->ds, qual, ds_buffer_base);
-	idx = base + (index * ds_cfg.sizeof_rec[qual]);
-
-	if (idx > ds_get(context->ds, qual, ds_absolute_maximum))
-		return -EINVAL;
+		return;
 
-	*record = (const void *)idx;
+	task = tracer->ds.context->task;
 
-	return ds_cfg.sizeof_rec[qual];
-}
+	control = ds_cfg.ctl[dsf_bts];
+	if (!(tracer->trace.ds.flags & BTS_KERNEL))
+		control |= ds_cfg.ctl[dsf_bts_kernel];
+	if (!(tracer->trace.ds.flags & BTS_USER))
+		control |= ds_cfg.ctl[dsf_bts_user];
 
-int ds_access_bts(struct bts_tracer *tracer, size_t index,
-		  const void **record)
-{
-	if (!tracer)
-		return -EINVAL;
+	if (task) {
+		task->thread.debugctlmsr |= control;
+		set_tsk_thread_flag(task, TIF_DEBUGCTLMSR);
+	}
 
-	return ds_access(tracer->ds.context, ds_bts, index, record);
+	if (!task || (task == current))
+		update_debugctlmsr(get_debugctlmsr() | control);
 }
 
-int ds_access_pebs(struct pebs_tracer *tracer, size_t index,
-		   const void **record)
+void ds_release_pebs(struct pebs_tracer *tracer)
 {
 	if (!tracer)
-		return -EINVAL;
-
-	return ds_access(tracer->ds.context, ds_pebs, index, record);
-}
-
-static int ds_write(struct ds_context *context, enum ds_qualifier qual,
-		    const void *record, size_t size)
-{
-	int bytes_written = 0;
-
-	if (!record)
-		return -EINVAL;
-
-	while (size) {
-		unsigned long base, index, end, write_end, int_th;
-		unsigned long write_size, adj_write_size;
-
-		/*
-		 * write as much as possible without producing an
-		 * overflow interrupt.
-		 *
-		 * interrupt_threshold must either be
-		 * - bigger than absolute_maximum or
-		 * - point to a record between buffer_base and absolute_maximum
-		 *
-		 * index points to a valid record.
-		 */
-		base   = ds_get(context->ds, qual, ds_buffer_base);
-		index  = ds_get(context->ds, qual, ds_index);
-		end    = ds_get(context->ds, qual, ds_absolute_maximum);
-		int_th = ds_get(context->ds, qual, ds_interrupt_threshold);
-
-		write_end = min(end, int_th);
-
-		/* if we are already beyond the interrupt threshold,
-		 * we fill the entire buffer */
-		if (write_end <= index)
-			write_end = end;
-
-		if (write_end <= index)
-			break;
-
-		write_size = min((unsigned long) size, write_end - index);
-		memcpy((void *)index, record, write_size);
-
-		record = (const char *)record + write_size;
-		size -= write_size;
-		bytes_written += write_size;
-
-		adj_write_size = write_size / ds_cfg.sizeof_rec[qual];
-		adj_write_size *= ds_cfg.sizeof_rec[qual];
-
-		/* zero out trailing bytes */
-		memset((char *)index + write_size, 0,
-		       adj_write_size - write_size);
-		index += adj_write_size;
-
-		if (index >= end)
-			index = base;
-		ds_set(context->ds, qual, ds_index, index);
+		return;
 
-		if (index >= int_th)
-			ds_overflow(context, qual);
-	}
+	ds_suspend_pebs(tracer);
 
-	return bytes_written;
-}
+	WARN_ON_ONCE(tracer->ds.context->pebs_master != tracer);
+	tracer->ds.context->pebs_master = NULL;
 
-int ds_write_bts(struct bts_tracer *tracer, const void *record, size_t size)
-{
-	if (!tracer)
-		return -EINVAL;
+	put_tracer(tracer->ds.context->task);
+	ds_put_context(tracer->ds.context);
 
-	return ds_write(tracer->ds.context, ds_bts, record, size);
+	kfree(tracer);
 }
 
-int ds_write_pebs(struct pebs_tracer *tracer, const void *record, size_t size)
+void ds_suspend_pebs(struct pebs_tracer *tracer)
 {
-	if (!tracer)
-		return -EINVAL;
 
-	return ds_write(tracer->ds.context, ds_pebs, record, size);
 }
 
-static void ds_reset_or_clear(struct ds_context *context,
-			      enum ds_qualifier qual, int clear)
+void ds_resume_pebs(struct pebs_tracer *tracer)
 {
-	unsigned long base, end;
-
-	base = ds_get(context->ds, qual, ds_buffer_base);
-	end  = ds_get(context->ds, qual, ds_absolute_maximum);
 
-	if (clear)
-		memset((void *)base, 0, end - base);
-
-	ds_set(context->ds, qual, ds_index, base);
 }
 
-int ds_reset_bts(struct bts_tracer *tracer)
+const struct bts_trace *ds_read_bts(struct bts_tracer *tracer)
 {
 	if (!tracer)
-		return -EINVAL;
+		return NULL;
 
-	ds_reset_or_clear(tracer->ds.context, ds_bts, /* clear = */ 0);
-
-	return 0;
+	ds_read_config(tracer->ds.context, &tracer->trace.ds, ds_bts);
+	return &tracer->trace;
 }
 
-int ds_reset_pebs(struct pebs_tracer *tracer)
+const struct pebs_trace *ds_read_pebs(struct pebs_tracer *tracer)
 {
 	if (!tracer)
-		return -EINVAL;
+		return NULL;
 
-	ds_reset_or_clear(tracer->ds.context, ds_pebs, /* clear = */ 0);
+	ds_read_config(tracer->ds.context, &tracer->trace.ds, ds_pebs);
+	tracer->trace.reset_value =
+		*(u64 *)(tracer->ds.context->ds + (ds_cfg.sizeof_field * 8));
 
-	return 0;
+	return &tracer->trace;
 }
 
-int ds_clear_bts(struct bts_tracer *tracer)
+int ds_reset_bts(struct bts_tracer *tracer)
 {
 	if (!tracer)
 		return -EINVAL;
 
-	ds_reset_or_clear(tracer->ds.context, ds_bts, /* clear = */ 1);
-
-	return 0;
-}
-
-int ds_clear_pebs(struct pebs_tracer *tracer)
-{
-	if (!tracer)
-		return -EINVAL;
+	tracer->trace.ds.top = tracer->trace.ds.begin;
 
-	ds_reset_or_clear(tracer->ds.context, ds_pebs, /* clear = */ 1);
+	ds_set(tracer->ds.context->ds, ds_bts, ds_index,
+	       (unsigned long)tracer->trace.ds.top);
 
 	return 0;
 }
 
-int ds_get_pebs_reset(struct pebs_tracer *tracer, u64 *value)
+int ds_reset_pebs(struct pebs_tracer *tracer)
 {
 	if (!tracer)
 		return -EINVAL;
 
-	if (!value)
-		return -EINVAL;
+	tracer->trace.ds.top = tracer->trace.ds.begin;
 
-	*value = *(u64 *)(tracer->ds.context->ds + (ds_cfg.sizeof_field * 8));
+	ds_set(tracer->ds.context->ds, ds_bts, ds_index,
+	       (unsigned long)tracer->trace.ds.top);
 
 	return 0;
 }
@@ -746,35 +889,59 @@ int ds_set_pebs_reset(struct pebs_tracer *tracer, u64 value)
 	return 0;
 }
 
-static const struct ds_configuration ds_cfg_var = {
-	.sizeof_ds    = sizeof(long) * 12,
-	.sizeof_field = sizeof(long),
-	.sizeof_rec[ds_bts]   = sizeof(long) * 3,
+static const struct ds_configuration ds_cfg_netburst = {
+	.name = "netburst",
+	.ctl[dsf_bts]		= (1 << 2) | (1 << 3),
+	.ctl[dsf_bts_kernel]	= (1 << 5),
+	.ctl[dsf_bts_user]	= (1 << 6),
+
+	.sizeof_field		= sizeof(long),
+	.sizeof_rec[ds_bts]	= sizeof(long) * 3,
 #ifdef __i386__
-	.sizeof_rec[ds_pebs]  = sizeof(long) * 10
+	.sizeof_rec[ds_pebs]	= sizeof(long) * 10,
 #else
-	.sizeof_rec[ds_pebs]  = sizeof(long) * 18
+	.sizeof_rec[ds_pebs]	= sizeof(long) * 18,
 #endif
 };
-static const struct ds_configuration ds_cfg_64 = {
-	.sizeof_ds    = 8 * 12,
-	.sizeof_field = 8,
-	.sizeof_rec[ds_bts]   = 8 * 3,
+static const struct ds_configuration ds_cfg_pentium_m = {
+	.name = "pentium m",
+	.ctl[dsf_bts]		= (1 << 6) | (1 << 7),
+
+	.sizeof_field		= sizeof(long),
+	.sizeof_rec[ds_bts]	= sizeof(long) * 3,
 #ifdef __i386__
-	.sizeof_rec[ds_pebs]  = 8 * 10
+	.sizeof_rec[ds_pebs]	= sizeof(long) * 10,
 #else
-	.sizeof_rec[ds_pebs]  = 8 * 18
+	.sizeof_rec[ds_pebs]	= sizeof(long) * 18,
 #endif
 };
+static const struct ds_configuration ds_cfg_core2 = {
+	.name = "core 2",
+	.ctl[dsf_bts]		= (1 << 6) | (1 << 7),
+	.ctl[dsf_bts_kernel]	= (1 << 9),
+	.ctl[dsf_bts_user]	= (1 << 10),
+
+	.sizeof_field		= 8,
+	.sizeof_rec[ds_bts]	= 8 * 3,
+	.sizeof_rec[ds_pebs]	= 8 * 18,
+};
 
-static inline void
+static void
 ds_configure(const struct ds_configuration *cfg)
 {
+	memset(&ds_cfg, 0, sizeof(ds_cfg));
 	ds_cfg = *cfg;
 
-	printk(KERN_INFO "DS available\n");
+	printk(KERN_INFO "[ds] using %s configuration\n", ds_cfg.name);
+
+	if (!cpu_has_bts) {
+		ds_cfg.ctl[dsf_bts] = 0;
+		printk(KERN_INFO "[ds] bts not available\n");
+	}
+	if (!cpu_has_pebs)
+		printk(KERN_INFO "[ds] pebs not available\n");
 
-	BUG_ON(MAX_SIZEOF_DS < ds_cfg.sizeof_ds);
+	WARN_ON_ONCE(MAX_SIZEOF_DS < (12 * ds_cfg.sizeof_field));
 }
 
 void __cpuinit ds_init_intel(struct cpuinfo_x86 *c)
@@ -787,10 +954,10 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c)
 			break;
 		case 0xD:
 		case 0xE: /* Pentium M */
-			ds_configure(&ds_cfg_var);
+			ds_configure(&ds_cfg_pentium_m);
 			break;
 		default: /* Core2, Atom, ... */
-			ds_configure(&ds_cfg_64);
+			ds_configure(&ds_cfg_core2);
 			break;
 		}
 		break;
@@ -799,7 +966,7 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c)
 		case 0x0:
 		case 0x1:
 		case 0x2: /* Netburst */
-			ds_configure(&ds_cfg_var);
+			ds_configure(&ds_cfg_netburst);
 			break;
 		default:
 			/* sorry, don't know about them */
@@ -812,14 +979,52 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c)
 	}
 }
 
-void ds_free(struct ds_context *context)
+/*
+ * Change the DS configuration from tracing prev to tracing next.
+ */
+void ds_switch_to(struct task_struct *prev, struct task_struct *next)
 {
-	/* This is called when the task owning the parameter context
-	 * is dying. There should not be any user of that context left
-	 * to disturb us, anymore. */
-	unsigned long leftovers = context->count;
-	while (leftovers--) {
-		put_tracer(context->task);
-		ds_put_context(context);
+	struct ds_context *prev_ctx = prev->thread.ds_ctx;
+	struct ds_context *next_ctx = next->thread.ds_ctx;
+
+	if (prev_ctx) {
+		update_debugctlmsr(0);
+
+		if (prev_ctx->bts_master &&
+		    (prev_ctx->bts_master->trace.ds.flags & BTS_TIMESTAMPS)) {
+			struct bts_struct ts = {
+				.qualifier = bts_task_departs,
+				.variant.timestamp.jiffies = jiffies_64,
+				.variant.timestamp.pid = prev->pid
+			};
+			bts_write(prev_ctx->bts_master, &ts);
+		}
+	}
+
+	if (next_ctx) {
+		if (next_ctx->bts_master &&
+		    (next_ctx->bts_master->trace.ds.flags & BTS_TIMESTAMPS)) {
+			struct bts_struct ts = {
+				.qualifier = bts_task_arrives,
+				.variant.timestamp.jiffies = jiffies_64,
+				.variant.timestamp.pid = next->pid
+			};
+			bts_write(next_ctx->bts_master, &ts);
+		}
+
+		wrmsrl(MSR_IA32_DS_AREA, (unsigned long)next_ctx->ds);
 	}
+
+	update_debugctlmsr(next->thread.debugctlmsr);
+}
+
+void ds_copy_thread(struct task_struct *tsk, struct task_struct *father)
+{
+	clear_tsk_thread_flag(tsk, TIF_DS_AREA_MSR);
+	tsk->thread.ds_ctx = NULL;
+}
+
+void ds_exit_thread(struct task_struct *tsk)
+{
+	WARN_ON(tsk->thread.ds_ctx);
 }
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 7aafeb5263ef..65a13943e098 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -677,22 +677,6 @@ struct early_res {
 };
 static struct early_res early_res[MAX_EARLY_RES] __initdata = {
 	{ 0, PAGE_SIZE, "BIOS data page" },	/* BIOS data page */
-#if defined(CONFIG_X86_64) && defined(CONFIG_X86_TRAMPOLINE)
-	{ TRAMPOLINE_BASE, TRAMPOLINE_BASE + 2 * PAGE_SIZE, "TRAMPOLINE" },
-#endif
-#if defined(CONFIG_X86_32) && defined(CONFIG_SMP)
-	/*
-	 * But first pinch a few for the stack/trampoline stuff
-	 * FIXME: Don't need the extra page at 4K, but need to fix
-	 * trampoline before removing it. (see the GDT stuff)
-	 */
-	{ PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, "EX TRAMPOLINE" },
-	/*
-	 * Has to be in very low memory so we can execute
-	 * real-mode AP code.
-	 */
-	{ TRAMPOLINE_BASE, TRAMPOLINE_BASE + PAGE_SIZE, "TRAMPOLINE" },
-#endif
 	{}
 };
 
diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
index 1b894b72c0f5..744aa7fc49d5 100644
--- a/arch/x86/kernel/early-quirks.c
+++ b/arch/x86/kernel/early-quirks.c
@@ -17,6 +17,7 @@
 #include <asm/io_apic.h>
 #include <asm/apic.h>
 #include <asm/iommu.h>
+#include <asm/gart.h>
 
 static void __init fix_hypertransport_config(int num, int slot, int func)
 {
diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c
index 34ad997d3834..23b138e31e9c 100644
--- a/arch/x86/kernel/early_printk.c
+++ b/arch/x86/kernel/early_printk.c
@@ -875,49 +875,6 @@ static struct console early_dbgp_console = {
 };
 #endif
 
-/* Console interface to a host file on AMD's SimNow! */
-
-static int simnow_fd;
-
-enum {
-	MAGIC1 = 0xBACCD00A,
-	MAGIC2 = 0xCA110000,
-	XOPEN = 5,
-	XWRITE = 4,
-};
-
-static noinline long simnow(long cmd, long a, long b, long c)
-{
-	long ret;
-
-	asm volatile("cpuid" :
-		     "=a" (ret) :
-		     "b" (a), "c" (b), "d" (c), "0" (MAGIC1), "D" (cmd + MAGIC2));
-	return ret;
-}
-
-static void __init simnow_init(char *str)
-{
-	char *fn = "klog";
-
-	if (*str == '=')
-		fn = ++str;
-	/* error ignored */
-	simnow_fd = simnow(XOPEN, (unsigned long)fn, O_WRONLY|O_APPEND|O_CREAT, 0644);
-}
-
-static void simnow_write(struct console *con, const char *s, unsigned n)
-{
-	simnow(XWRITE, simnow_fd, (unsigned long)s, n);
-}
-
-static struct console simnow_console = {
-	.name =		"simnow",
-	.write =	simnow_write,
-	.flags =	CON_PRINTBUFFER,
-	.index =	-1,
-};
-
 /* Direct interface for emergencies */
 static struct console *early_console = &early_vga_console;
 static int __initdata early_console_initialized;
@@ -960,10 +917,6 @@ static int __init setup_early_printk(char *buf)
 		max_ypos = boot_params.screen_info.orig_video_lines;
 		current_ypos = boot_params.screen_info.orig_y;
 		early_console = &early_vga_console;
-	} else if (!strncmp(buf, "simnow", 6)) {
-		simnow_init(buf + 6);
-		early_console = &simnow_console;
-		keep_early = 1;
 #ifdef CONFIG_EARLY_PRINTK_DBGP
 	} else if (!strncmp(buf, "dbgp", 4)) {
 		if (early_dbgp_init(buf+4) < 0)
diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
index 43ceb3f454bf..d6f0490a7391 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -619,28 +619,37 @@ END(syscall_badsys)
 27:;
 
 /*
- * Build the entry stubs and pointer table with
- * some assembler magic.
+ * Build the entry stubs and pointer table with some assembler magic.
+ * We pack 7 stubs into a single 32-byte chunk, which will fit in a
+ * single cache line on all modern x86 implementations.
  */
-.section .rodata,"a"
+.section .init.rodata,"a"
 ENTRY(interrupt)
 .text
-
+	.p2align 5
+	.p2align CONFIG_X86_L1_CACHE_SHIFT
 ENTRY(irq_entries_start)
 	RING0_INT_FRAME
-vector=0
-.rept NR_VECTORS
-	ALIGN
- .if vector
+vector=FIRST_EXTERNAL_VECTOR
+.rept (NR_VECTORS-FIRST_EXTERNAL_VECTOR+6)/7
+	.balign 32
+  .rept	7
+    .if vector < NR_VECTORS
+      .if vector <> FIRST_EXTERNAL_VECTOR
 	CFI_ADJUST_CFA_OFFSET -4
- .endif
-1:	pushl $~(vector)
+      .endif
+1:	pushl $(~vector+0x80)	/* Note: always in signed byte range */
 	CFI_ADJUST_CFA_OFFSET 4
-	jmp common_interrupt
- .previous
+      .if ((vector-FIRST_EXTERNAL_VECTOR)%7) <> 6
+	jmp 2f
+      .endif
+      .previous
 	.long 1b
- .text
+      .text
 vector=vector+1
+    .endif
+  .endr
+2:	jmp common_interrupt
 .endr
 END(irq_entries_start)
 
@@ -652,8 +661,9 @@ END(interrupt)
  * the CPU automatically disables interrupts when executing an IRQ vector,
  * so IRQ-flags tracing has to follow that:
  */
-	ALIGN
+	.p2align CONFIG_X86_L1_CACHE_SHIFT
 common_interrupt:
+	addl $-0x80,(%esp)	/* Adjust vector into the [-256,-1] range */
 	SAVE_ALL
 	TRACE_IRQS_OFF
 	movl %esp,%eax
@@ -678,65 +688,6 @@ ENDPROC(name)
 /* The include is where all of the SMP etc. interrupts come from */
 #include "entry_arch.h"
 
-KPROBE_ENTRY(page_fault)
-	RING0_EC_FRAME
-	pushl $do_page_fault
-	CFI_ADJUST_CFA_OFFSET 4
-	ALIGN
-error_code:
-	/* the function address is in %fs's slot on the stack */
-	pushl %es
-	CFI_ADJUST_CFA_OFFSET 4
-	/*CFI_REL_OFFSET es, 0*/
-	pushl %ds
-	CFI_ADJUST_CFA_OFFSET 4
-	/*CFI_REL_OFFSET ds, 0*/
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET eax, 0
-	pushl %ebp
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET ebp, 0
-	pushl %edi
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET edi, 0
-	pushl %esi
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET esi, 0
-	pushl %edx
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET edx, 0
-	pushl %ecx
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET ecx, 0
-	pushl %ebx
-	CFI_ADJUST_CFA_OFFSET 4
-	CFI_REL_OFFSET ebx, 0
-	cld
-	pushl %fs
-	CFI_ADJUST_CFA_OFFSET 4
-	/*CFI_REL_OFFSET fs, 0*/
-	movl $(__KERNEL_PERCPU), %ecx
-	movl %ecx, %fs
-	UNWIND_ESPFIX_STACK
-	popl %ecx
-	CFI_ADJUST_CFA_OFFSET -4
-	/*CFI_REGISTER es, ecx*/
-	movl PT_FS(%esp), %edi		# get the function address
-	movl PT_ORIG_EAX(%esp), %edx	# get the error code
-	movl $-1, PT_ORIG_EAX(%esp)	# no syscall to restart
-	mov  %ecx, PT_FS(%esp)
-	/*CFI_REL_OFFSET fs, ES*/
-	movl $(__USER_DS), %ecx
-	movl %ecx, %ds
-	movl %ecx, %es
-	TRACE_IRQS_OFF
-	movl %esp,%eax			# pt_regs pointer
-	call *%edi
-	jmp ret_from_exception
-	CFI_ENDPROC
-KPROBE_END(page_fault)
-
 ENTRY(coprocessor_error)
 	RING0_INT_FRAME
 	pushl $0
@@ -767,140 +718,6 @@ ENTRY(device_not_available)
 	CFI_ENDPROC
 END(device_not_available)
 
-/*
- * Debug traps and NMI can happen at the one SYSENTER instruction
- * that sets up the real kernel stack. Check here, since we can't
- * allow the wrong stack to be used.
- *
- * "TSS_sysenter_sp0+12" is because the NMI/debug handler will have
- * already pushed 3 words if it hits on the sysenter instruction:
- * eflags, cs and eip.
- *
- * We just load the right stack, and push the three (known) values
- * by hand onto the new stack - while updating the return eip past
- * the instruction that would have done it for sysenter.
- */
-#define FIX_STACK(offset, ok, label)		\
-	cmpw $__KERNEL_CS,4(%esp);		\
-	jne ok;					\
-label:						\
-	movl TSS_sysenter_sp0+offset(%esp),%esp;	\
-	CFI_DEF_CFA esp, 0;			\
-	CFI_UNDEFINED eip;			\
-	pushfl;					\
-	CFI_ADJUST_CFA_OFFSET 4;		\
-	pushl $__KERNEL_CS;			\
-	CFI_ADJUST_CFA_OFFSET 4;		\
-	pushl $sysenter_past_esp;		\
-	CFI_ADJUST_CFA_OFFSET 4;		\
-	CFI_REL_OFFSET eip, 0
-
-KPROBE_ENTRY(debug)
-	RING0_INT_FRAME
-	cmpl $ia32_sysenter_target,(%esp)
-	jne debug_stack_correct
-	FIX_STACK(12, debug_stack_correct, debug_esp_fix_insn)
-debug_stack_correct:
-	pushl $-1			# mark this as an int
-	CFI_ADJUST_CFA_OFFSET 4
-	SAVE_ALL
-	TRACE_IRQS_OFF
-	xorl %edx,%edx			# error code 0
-	movl %esp,%eax			# pt_regs pointer
-	call do_debug
-	jmp ret_from_exception
-	CFI_ENDPROC
-KPROBE_END(debug)
-
-/*
- * NMI is doubly nasty. It can happen _while_ we're handling
- * a debug fault, and the debug fault hasn't yet been able to
- * clear up the stack. So we first check whether we got  an
- * NMI on the sysenter entry path, but after that we need to
- * check whether we got an NMI on the debug path where the debug
- * fault happened on the sysenter path.
- */
-KPROBE_ENTRY(nmi)
-	RING0_INT_FRAME
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
-	movl %ss, %eax
-	cmpw $__ESPFIX_SS, %ax
-	popl %eax
-	CFI_ADJUST_CFA_OFFSET -4
-	je nmi_espfix_stack
-	cmpl $ia32_sysenter_target,(%esp)
-	je nmi_stack_fixup
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
-	movl %esp,%eax
-	/* Do not access memory above the end of our stack page,
-	 * it might not exist.
-	 */
-	andl $(THREAD_SIZE-1),%eax
-	cmpl $(THREAD_SIZE-20),%eax
-	popl %eax
-	CFI_ADJUST_CFA_OFFSET -4
-	jae nmi_stack_correct
-	cmpl $ia32_sysenter_target,12(%esp)
-	je nmi_debug_stack_check
-nmi_stack_correct:
-	/* We have a RING0_INT_FRAME here */
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
-	SAVE_ALL
-	TRACE_IRQS_OFF
-	xorl %edx,%edx		# zero error code
-	movl %esp,%eax		# pt_regs pointer
-	call do_nmi
-	jmp restore_nocheck_notrace
-	CFI_ENDPROC
-
-nmi_stack_fixup:
-	RING0_INT_FRAME
-	FIX_STACK(12,nmi_stack_correct, 1)
-	jmp nmi_stack_correct
-
-nmi_debug_stack_check:
-	/* We have a RING0_INT_FRAME here */
-	cmpw $__KERNEL_CS,16(%esp)
-	jne nmi_stack_correct
-	cmpl $debug,(%esp)
-	jb nmi_stack_correct
-	cmpl $debug_esp_fix_insn,(%esp)
-	ja nmi_stack_correct
-	FIX_STACK(24,nmi_stack_correct, 1)
-	jmp nmi_stack_correct
-
-nmi_espfix_stack:
-	/* We have a RING0_INT_FRAME here.
-	 *
-	 * create the pointer to lss back
-	 */
-	pushl %ss
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl %esp
-	CFI_ADJUST_CFA_OFFSET 4
-	addw $4, (%esp)
-	/* copy the iret frame of 12 bytes */
-	.rept 3
-	pushl 16(%esp)
-	CFI_ADJUST_CFA_OFFSET 4
-	.endr
-	pushl %eax
-	CFI_ADJUST_CFA_OFFSET 4
-	SAVE_ALL
-	TRACE_IRQS_OFF
-	FIXUP_ESPFIX_STACK		# %eax == %esp
-	xorl %edx,%edx			# zero error code
-	call do_nmi
-	RESTORE_REGS
-	lss 12+4(%esp), %esp		# back to espfix stack
-	CFI_ADJUST_CFA_OFFSET -24
-	jmp irq_return
-	CFI_ENDPROC
-KPROBE_END(nmi)
-
 #ifdef CONFIG_PARAVIRT
 ENTRY(native_iret)
 	iret
@@ -916,19 +733,6 @@ ENTRY(native_irq_enable_sysexit)
 END(native_irq_enable_sysexit)
 #endif
 
-KPROBE_ENTRY(int3)
-	RING0_INT_FRAME
-	pushl $-1			# mark this as an int
-	CFI_ADJUST_CFA_OFFSET 4
-	SAVE_ALL
-	TRACE_IRQS_OFF
-	xorl %edx,%edx		# zero error code
-	movl %esp,%eax		# pt_regs pointer
-	call do_int3
-	jmp ret_from_exception
-	CFI_ENDPROC
-KPROBE_END(int3)
-
 ENTRY(overflow)
 	RING0_INT_FRAME
 	pushl $0
@@ -993,14 +797,6 @@ ENTRY(stack_segment)
 	CFI_ENDPROC
 END(stack_segment)
 
-KPROBE_ENTRY(general_protection)
-	RING0_EC_FRAME
-	pushl $do_general_protection
-	CFI_ADJUST_CFA_OFFSET 4
-	jmp error_code
-	CFI_ENDPROC
-KPROBE_END(general_protection)
-
 ENTRY(alignment_check)
 	RING0_EC_FRAME
 	pushl $do_alignment_check
@@ -1051,6 +847,7 @@ ENTRY(kernel_thread_helper)
 	push %eax
 	CFI_ADJUST_CFA_OFFSET 4
 	call do_exit
+	ud2			# padding for call trace
 	CFI_ENDPROC
 ENDPROC(kernel_thread_helper)
 
@@ -1259,3 +1056,227 @@ return_to_handler:
 #include "syscall_table_32.S"
 
 syscall_table_size=(.-sys_call_table)
+
+/*
+ * Some functions should be protected against kprobes
+ */
+	.pushsection .kprobes.text, "ax"
+
+ENTRY(page_fault)
+	RING0_EC_FRAME
+	pushl $do_page_fault
+	CFI_ADJUST_CFA_OFFSET 4
+	ALIGN
+error_code:
+	/* the function address is in %fs's slot on the stack */
+	pushl %es
+	CFI_ADJUST_CFA_OFFSET 4
+	/*CFI_REL_OFFSET es, 0*/
+	pushl %ds
+	CFI_ADJUST_CFA_OFFSET 4
+	/*CFI_REL_OFFSET ds, 0*/
+	pushl %eax
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET eax, 0
+	pushl %ebp
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ebp, 0
+	pushl %edi
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edi, 0
+	pushl %esi
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET esi, 0
+	pushl %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edx, 0
+	pushl %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx, 0
+	pushl %ebx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ebx, 0
+	cld
+	pushl %fs
+	CFI_ADJUST_CFA_OFFSET 4
+	/*CFI_REL_OFFSET fs, 0*/
+	movl $(__KERNEL_PERCPU), %ecx
+	movl %ecx, %fs
+	UNWIND_ESPFIX_STACK
+	popl %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	/*CFI_REGISTER es, ecx*/
+	movl PT_FS(%esp), %edi		# get the function address
+	movl PT_ORIG_EAX(%esp), %edx	# get the error code
+	movl $-1, PT_ORIG_EAX(%esp)	# no syscall to restart
+	mov  %ecx, PT_FS(%esp)
+	/*CFI_REL_OFFSET fs, ES*/
+	movl $(__USER_DS), %ecx
+	movl %ecx, %ds
+	movl %ecx, %es
+	TRACE_IRQS_OFF
+	movl %esp,%eax			# pt_regs pointer
+	call *%edi
+	jmp ret_from_exception
+	CFI_ENDPROC
+END(page_fault)
+
+/*
+ * Debug traps and NMI can happen at the one SYSENTER instruction
+ * that sets up the real kernel stack. Check here, since we can't
+ * allow the wrong stack to be used.
+ *
+ * "TSS_sysenter_sp0+12" is because the NMI/debug handler will have
+ * already pushed 3 words if it hits on the sysenter instruction:
+ * eflags, cs and eip.
+ *
+ * We just load the right stack, and push the three (known) values
+ * by hand onto the new stack - while updating the return eip past
+ * the instruction that would have done it for sysenter.
+ */
+#define FIX_STACK(offset, ok, label)		\
+	cmpw $__KERNEL_CS,4(%esp);		\
+	jne ok;					\
+label:						\
+	movl TSS_sysenter_sp0+offset(%esp),%esp;	\
+	CFI_DEF_CFA esp, 0;			\
+	CFI_UNDEFINED eip;			\
+	pushfl;					\
+	CFI_ADJUST_CFA_OFFSET 4;		\
+	pushl $__KERNEL_CS;			\
+	CFI_ADJUST_CFA_OFFSET 4;		\
+	pushl $sysenter_past_esp;		\
+	CFI_ADJUST_CFA_OFFSET 4;		\
+	CFI_REL_OFFSET eip, 0
+
+ENTRY(debug)
+	RING0_INT_FRAME
+	cmpl $ia32_sysenter_target,(%esp)
+	jne debug_stack_correct
+	FIX_STACK(12, debug_stack_correct, debug_esp_fix_insn)
+debug_stack_correct:
+	pushl $-1			# mark this as an int
+	CFI_ADJUST_CFA_OFFSET 4
+	SAVE_ALL
+	TRACE_IRQS_OFF
+	xorl %edx,%edx			# error code 0
+	movl %esp,%eax			# pt_regs pointer
+	call do_debug
+	jmp ret_from_exception
+	CFI_ENDPROC
+END(debug)
+
+/*
+ * NMI is doubly nasty. It can happen _while_ we're handling
+ * a debug fault, and the debug fault hasn't yet been able to
+ * clear up the stack. So we first check whether we got  an
+ * NMI on the sysenter entry path, but after that we need to
+ * check whether we got an NMI on the debug path where the debug
+ * fault happened on the sysenter path.
+ */
+ENTRY(nmi)
+	RING0_INT_FRAME
+	pushl %eax
+	CFI_ADJUST_CFA_OFFSET 4
+	movl %ss, %eax
+	cmpw $__ESPFIX_SS, %ax
+	popl %eax
+	CFI_ADJUST_CFA_OFFSET -4
+	je nmi_espfix_stack
+	cmpl $ia32_sysenter_target,(%esp)
+	je nmi_stack_fixup
+	pushl %eax
+	CFI_ADJUST_CFA_OFFSET 4
+	movl %esp,%eax
+	/* Do not access memory above the end of our stack page,
+	 * it might not exist.
+	 */
+	andl $(THREAD_SIZE-1),%eax
+	cmpl $(THREAD_SIZE-20),%eax
+	popl %eax
+	CFI_ADJUST_CFA_OFFSET -4
+	jae nmi_stack_correct
+	cmpl $ia32_sysenter_target,12(%esp)
+	je nmi_debug_stack_check
+nmi_stack_correct:
+	/* We have a RING0_INT_FRAME here */
+	pushl %eax
+	CFI_ADJUST_CFA_OFFSET 4
+	SAVE_ALL
+	TRACE_IRQS_OFF
+	xorl %edx,%edx		# zero error code
+	movl %esp,%eax		# pt_regs pointer
+	call do_nmi
+	jmp restore_nocheck_notrace
+	CFI_ENDPROC
+
+nmi_stack_fixup:
+	RING0_INT_FRAME
+	FIX_STACK(12,nmi_stack_correct, 1)
+	jmp nmi_stack_correct
+
+nmi_debug_stack_check:
+	/* We have a RING0_INT_FRAME here */
+	cmpw $__KERNEL_CS,16(%esp)
+	jne nmi_stack_correct
+	cmpl $debug,(%esp)
+	jb nmi_stack_correct
+	cmpl $debug_esp_fix_insn,(%esp)
+	ja nmi_stack_correct
+	FIX_STACK(24,nmi_stack_correct, 1)
+	jmp nmi_stack_correct
+
+nmi_espfix_stack:
+	/* We have a RING0_INT_FRAME here.
+	 *
+	 * create the pointer to lss back
+	 */
+	pushl %ss
+	CFI_ADJUST_CFA_OFFSET 4
+	pushl %esp
+	CFI_ADJUST_CFA_OFFSET 4
+	addw $4, (%esp)
+	/* copy the iret frame of 12 bytes */
+	.rept 3
+	pushl 16(%esp)
+	CFI_ADJUST_CFA_OFFSET 4
+	.endr
+	pushl %eax
+	CFI_ADJUST_CFA_OFFSET 4
+	SAVE_ALL
+	TRACE_IRQS_OFF
+	FIXUP_ESPFIX_STACK		# %eax == %esp
+	xorl %edx,%edx			# zero error code
+	call do_nmi
+	RESTORE_REGS
+	lss 12+4(%esp), %esp		# back to espfix stack
+	CFI_ADJUST_CFA_OFFSET -24
+	jmp irq_return
+	CFI_ENDPROC
+END(nmi)
+
+ENTRY(int3)
+	RING0_INT_FRAME
+	pushl $-1			# mark this as an int
+	CFI_ADJUST_CFA_OFFSET 4
+	SAVE_ALL
+	TRACE_IRQS_OFF
+	xorl %edx,%edx		# zero error code
+	movl %esp,%eax		# pt_regs pointer
+	call do_int3
+	jmp ret_from_exception
+	CFI_ENDPROC
+END(int3)
+
+ENTRY(general_protection)
+	RING0_EC_FRAME
+	pushl $do_general_protection
+	CFI_ADJUST_CFA_OFFSET 4
+	jmp error_code
+	CFI_ENDPROC
+END(general_protection)
+
+/*
+ * End of kprobes section
+ */
+	.popsection
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index 54e0bbdccb99..e28c7a987793 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -11,15 +11,15 @@
  *
  * NOTE: This code handles signal-recognition, which happens every time
  * after an interrupt and after each system call.
- * 
- * Normal syscalls and interrupts don't save a full stack frame, this is 
+ *
+ * Normal syscalls and interrupts don't save a full stack frame, this is
  * only done for syscall tracing, signals or fork/exec et.al.
- * 
- * A note on terminology:	 
- * - top of stack: Architecture defined interrupt frame from SS to RIP 
- * at the top of the kernel process stack.	
+ *
+ * A note on terminology:
+ * - top of stack: Architecture defined interrupt frame from SS to RIP
+ * at the top of the kernel process stack.
  * - partial stack frame: partially saved registers upto R11.
- * - full stack frame: Like partial stack frame, but all register saved. 
+ * - full stack frame: Like partial stack frame, but all register saved.
  *
  * Some macro usage:
  * - CFI macros are used to generate dwarf2 unwind information for better
@@ -60,7 +60,6 @@
 #define __AUDIT_ARCH_LE	   0x40000000
 
 	.code64
-
 #ifdef CONFIG_FUNCTION_TRACER
 #ifdef CONFIG_DYNAMIC_FTRACE
 ENTRY(mcount)
@@ -71,15 +70,7 @@ ENTRY(ftrace_caller)
 	cmpl $0, function_trace_stop
 	jne  ftrace_stub
 
-	/* 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)
+	MCOUNT_SAVE_FRAME
 
 	movq 0x38(%rsp), %rdi
 	movq 8(%rbp), %rsi
@@ -89,14 +80,7 @@ ENTRY(ftrace_caller)
 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
+	MCOUNT_RESTORE_FRAME
 
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 .globl ftrace_graph_call
@@ -130,15 +114,7 @@ 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)
+	MCOUNT_SAVE_FRAME
 
 	movq 0x38(%rsp), %rdi
 	movq 8(%rbp), %rsi
@@ -146,14 +122,7 @@ trace:
 
 	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
+	MCOUNT_RESTORE_FRAME
 
 	jmp ftrace_stub
 END(mcount)
@@ -165,14 +134,7 @@ ENTRY(ftrace_graph_caller)
 	cmpl $0, function_trace_stop
 	jne ftrace_stub
 
-	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)
+	MCOUNT_SAVE_FRAME
 
 	leaq 8(%rbp), %rdi
 	movq 0x38(%rsp), %rsi
@@ -180,14 +142,8 @@ ENTRY(ftrace_graph_caller)
 
 	call	prepare_ftrace_return
 
-	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
+	MCOUNT_RESTORE_FRAME
+
 	retq
 END(ftrace_graph_caller)
 
@@ -225,7 +181,7 @@ return_to_handler:
 
 #ifndef CONFIG_PREEMPT
 #define retint_kernel retint_restore_args
-#endif	
+#endif
 
 #ifdef CONFIG_PARAVIRT
 ENTRY(native_usergs_sysret64)
@@ -244,29 +200,29 @@ ENTRY(native_usergs_sysret64)
 .endm
 
 /*
- * C code is not supposed to know about undefined top of stack. Every time 
- * a C function with an pt_regs argument is called from the SYSCALL based 
+ * C code is not supposed to know about undefined top of stack. Every time
+ * a C function with an pt_regs argument is called from the SYSCALL based
  * fast path FIXUP_TOP_OF_STACK is needed.
  * RESTORE_TOP_OF_STACK syncs the syscall state after any possible ptregs
  * manipulation.
- */        	
-		
-	/* %rsp:at FRAMEEND */ 
-	.macro FIXUP_TOP_OF_STACK tmp
-	movq	%gs:pda_oldrsp,\tmp
-	movq  	\tmp,RSP(%rsp)
-	movq    $__USER_DS,SS(%rsp)
-	movq    $__USER_CS,CS(%rsp)
-	movq 	$-1,RCX(%rsp)
-	movq	R11(%rsp),\tmp  /* get eflags */
-	movq	\tmp,EFLAGS(%rsp)
+ */
+
+	/* %rsp:at FRAMEEND */
+	.macro FIXUP_TOP_OF_STACK tmp offset=0
+	movq %gs:pda_oldrsp,\tmp
+	movq \tmp,RSP+\offset(%rsp)
+	movq $__USER_DS,SS+\offset(%rsp)
+	movq $__USER_CS,CS+\offset(%rsp)
+	movq $-1,RCX+\offset(%rsp)
+	movq R11+\offset(%rsp),\tmp  /* get eflags */
+	movq \tmp,EFLAGS+\offset(%rsp)
 	.endm
 
-	.macro RESTORE_TOP_OF_STACK tmp,offset=0
-	movq   RSP-\offset(%rsp),\tmp
-	movq   \tmp,%gs:pda_oldrsp
-	movq   EFLAGS-\offset(%rsp),\tmp
-	movq   \tmp,R11-\offset(%rsp)
+	.macro RESTORE_TOP_OF_STACK tmp offset=0
+	movq RSP+\offset(%rsp),\tmp
+	movq \tmp,%gs:pda_oldrsp
+	movq EFLAGS+\offset(%rsp),\tmp
+	movq \tmp,R11+\offset(%rsp)
 	.endm
 
 	.macro FAKE_STACK_FRAME child_rip
@@ -278,7 +234,7 @@ ENTRY(native_usergs_sysret64)
 	pushq %rax /* rsp */
 	CFI_ADJUST_CFA_OFFSET	8
 	CFI_REL_OFFSET	rsp,0
-	pushq $(1<<9) /* eflags - interrupts on */
+	pushq $X86_EFLAGS_IF /* eflags - interrupts on */
 	CFI_ADJUST_CFA_OFFSET	8
 	/*CFI_REL_OFFSET	rflags,0*/
 	pushq $__KERNEL_CS /* cs */
@@ -296,62 +252,184 @@ ENTRY(native_usergs_sysret64)
 	CFI_ADJUST_CFA_OFFSET	-(6*8)
 	.endm
 
-	.macro	CFI_DEFAULT_STACK start=1
+/*
+ * initial frame state for interrupts (and exceptions without error code)
+ */
+	.macro EMPTY_FRAME start=1 offset=0
 	.if \start
-	CFI_STARTPROC	simple
+	CFI_STARTPROC simple
 	CFI_SIGNAL_FRAME
-	CFI_DEF_CFA	rsp,SS+8
+	CFI_DEF_CFA rsp,8+\offset
 	.else
-	CFI_DEF_CFA_OFFSET SS+8
+	CFI_DEF_CFA_OFFSET 8+\offset
 	.endif
-	CFI_REL_OFFSET	r15,R15
-	CFI_REL_OFFSET	r14,R14
-	CFI_REL_OFFSET	r13,R13
-	CFI_REL_OFFSET	r12,R12
-	CFI_REL_OFFSET	rbp,RBP
-	CFI_REL_OFFSET	rbx,RBX
-	CFI_REL_OFFSET	r11,R11
-	CFI_REL_OFFSET	r10,R10
-	CFI_REL_OFFSET	r9,R9
-	CFI_REL_OFFSET	r8,R8
-	CFI_REL_OFFSET	rax,RAX
-	CFI_REL_OFFSET	rcx,RCX
-	CFI_REL_OFFSET	rdx,RDX
-	CFI_REL_OFFSET	rsi,RSI
-	CFI_REL_OFFSET	rdi,RDI
-	CFI_REL_OFFSET	rip,RIP
-	/*CFI_REL_OFFSET	cs,CS*/
-	/*CFI_REL_OFFSET	rflags,EFLAGS*/
-	CFI_REL_OFFSET	rsp,RSP
-	/*CFI_REL_OFFSET	ss,SS*/
 	.endm
+
 /*
- * A newly forked process directly context switches into this.
- */ 	
-/* rdi:	prev */	
+ * initial frame state for interrupts (and exceptions without error code)
+ */
+	.macro INTR_FRAME start=1 offset=0
+	EMPTY_FRAME \start, SS+8+\offset-RIP
+	/*CFI_REL_OFFSET ss, SS+\offset-RIP*/
+	CFI_REL_OFFSET rsp, RSP+\offset-RIP
+	/*CFI_REL_OFFSET rflags, EFLAGS+\offset-RIP*/
+	/*CFI_REL_OFFSET cs, CS+\offset-RIP*/
+	CFI_REL_OFFSET rip, RIP+\offset-RIP
+	.endm
+
+/*
+ * initial frame state for exceptions with error code (and interrupts
+ * with vector already pushed)
+ */
+	.macro XCPT_FRAME start=1 offset=0
+	INTR_FRAME \start, RIP+\offset-ORIG_RAX
+	/*CFI_REL_OFFSET orig_rax, ORIG_RAX-ORIG_RAX*/
+	.endm
+
+/*
+ * frame that enables calling into C.
+ */
+	.macro PARTIAL_FRAME start=1 offset=0
+	XCPT_FRAME \start, ORIG_RAX+\offset-ARGOFFSET
+	CFI_REL_OFFSET rdi, RDI+\offset-ARGOFFSET
+	CFI_REL_OFFSET rsi, RSI+\offset-ARGOFFSET
+	CFI_REL_OFFSET rdx, RDX+\offset-ARGOFFSET
+	CFI_REL_OFFSET rcx, RCX+\offset-ARGOFFSET
+	CFI_REL_OFFSET rax, RAX+\offset-ARGOFFSET
+	CFI_REL_OFFSET r8, R8+\offset-ARGOFFSET
+	CFI_REL_OFFSET r9, R9+\offset-ARGOFFSET
+	CFI_REL_OFFSET r10, R10+\offset-ARGOFFSET
+	CFI_REL_OFFSET r11, R11+\offset-ARGOFFSET
+	.endm
+
+/*
+ * frame that enables passing a complete pt_regs to a C function.
+ */
+	.macro DEFAULT_FRAME start=1 offset=0
+	PARTIAL_FRAME \start, R11+\offset-R15
+	CFI_REL_OFFSET rbx, RBX+\offset
+	CFI_REL_OFFSET rbp, RBP+\offset
+	CFI_REL_OFFSET r12, R12+\offset
+	CFI_REL_OFFSET r13, R13+\offset
+	CFI_REL_OFFSET r14, R14+\offset
+	CFI_REL_OFFSET r15, R15+\offset
+	.endm
+
+/* save partial stack frame */
+ENTRY(save_args)
+	XCPT_FRAME
+	cld
+	movq_cfi rdi, RDI+16-ARGOFFSET
+	movq_cfi rsi, RSI+16-ARGOFFSET
+	movq_cfi rdx, RDX+16-ARGOFFSET
+	movq_cfi rcx, RCX+16-ARGOFFSET
+	movq_cfi rax, RAX+16-ARGOFFSET
+	movq_cfi  r8,  R8+16-ARGOFFSET
+	movq_cfi  r9,  R9+16-ARGOFFSET
+	movq_cfi r10, R10+16-ARGOFFSET
+	movq_cfi r11, R11+16-ARGOFFSET
+
+	leaq -ARGOFFSET+16(%rsp),%rdi	/* arg1 for handler */
+	movq_cfi rbp, 8		/* push %rbp */
+	leaq 8(%rsp), %rbp		/* mov %rsp, %ebp */
+	testl $3, CS(%rdi)
+	je 1f
+	SWAPGS
+	/*
+	 * irqcount is used to check if a CPU is already on an interrupt stack
+	 * or not. While this is essentially redundant with preempt_count it is
+	 * a little cheaper to use a separate counter in the PDA (short of
+	 * moving irq_enter into assembly, which would be too much work)
+	 */
+1:	incl %gs:pda_irqcount
+	jne 2f
+	popq_cfi %rax			/* move return address... */
+	mov %gs:pda_irqstackptr,%rsp
+	EMPTY_FRAME 0
+	pushq_cfi %rax			/* ... to the new stack */
+	/*
+	 * We entered an interrupt context - irqs are off:
+	 */
+2:	TRACE_IRQS_OFF
+	ret
+	CFI_ENDPROC
+END(save_args)
+
+ENTRY(save_rest)
+	PARTIAL_FRAME 1 REST_SKIP+8
+	movq 5*8+16(%rsp), %r11	/* save return address */
+	movq_cfi rbx, RBX+16
+	movq_cfi rbp, RBP+16
+	movq_cfi r12, R12+16
+	movq_cfi r13, R13+16
+	movq_cfi r14, R14+16
+	movq_cfi r15, R15+16
+	movq %r11, 8(%rsp)	/* return address */
+	FIXUP_TOP_OF_STACK %r11, 16
+	ret
+	CFI_ENDPROC
+END(save_rest)
+
+/* save complete stack frame */
+ENTRY(save_paranoid)
+	XCPT_FRAME 1 RDI+8
+	cld
+	movq_cfi rdi, RDI+8
+	movq_cfi rsi, RSI+8
+	movq_cfi rdx, RDX+8
+	movq_cfi rcx, RCX+8
+	movq_cfi rax, RAX+8
+	movq_cfi r8, R8+8
+	movq_cfi r9, R9+8
+	movq_cfi r10, R10+8
+	movq_cfi r11, R11+8
+	movq_cfi rbx, RBX+8
+	movq_cfi rbp, RBP+8
+	movq_cfi r12, R12+8
+	movq_cfi r13, R13+8
+	movq_cfi r14, R14+8
+	movq_cfi r15, R15+8
+	movl $1,%ebx
+	movl $MSR_GS_BASE,%ecx
+	rdmsr
+	testl %edx,%edx
+	js 1f	/* negative -> in kernel */
+	SWAPGS
+	xorl %ebx,%ebx
+1:	ret
+	CFI_ENDPROC
+END(save_paranoid)
+
+/*
+ * A newly forked process directly context switches into this address.
+ *
+ * rdi: prev task we switched from
+ */
 ENTRY(ret_from_fork)
-	CFI_DEFAULT_STACK
+	DEFAULT_FRAME
+
 	push kernel_eflags(%rip)
 	CFI_ADJUST_CFA_OFFSET 8
-	popf				# reset kernel eflags
+	popf					# reset kernel eflags
 	CFI_ADJUST_CFA_OFFSET -8
-	call schedule_tail
+
+	call schedule_tail			# rdi: 'prev' task parameter
+
 	GET_THREAD_INFO(%rcx)
-	testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),TI_flags(%rcx)
-	jnz rff_trace
-rff_action:	
+
+	CFI_REMEMBER_STATE
 	RESTORE_REST
-	testl $3,CS-ARGOFFSET(%rsp)	# from kernel_thread?
+
+	testl $3, CS-ARGOFFSET(%rsp)		# from kernel_thread?
 	je   int_ret_from_sys_call
-	testl $_TIF_IA32,TI_flags(%rcx)
+
+	testl $_TIF_IA32, TI_flags(%rcx)	# 32-bit compat task needs IRET
 	jnz  int_ret_from_sys_call
-	RESTORE_TOP_OF_STACK %rdi,ARGOFFSET
-	jmp ret_from_sys_call
-rff_trace:
-	movq %rsp,%rdi
-	call syscall_trace_leave
-	GET_THREAD_INFO(%rcx)	
-	jmp rff_action
+
+	RESTORE_TOP_OF_STACK %rdi, -ARGOFFSET
+	jmp ret_from_sys_call			# go to the SYSRET fastpath
+
+	CFI_RESTORE_STATE
 	CFI_ENDPROC
 END(ret_from_fork)
 
@@ -361,20 +439,20 @@ END(ret_from_fork)
  * SYSCALL does not save anything on the stack and does not change the
  * stack pointer.
  */
-		
+
 /*
- * Register setup:	
+ * Register setup:
  * rax  system call number
  * rdi  arg0
- * rcx  return address for syscall/sysret, C arg3 
+ * rcx  return address for syscall/sysret, C arg3
  * rsi  arg1
- * rdx  arg2	
+ * rdx  arg2
  * r10  arg3 	(--> moved to rcx for C)
  * r8   arg4
  * r9   arg5
  * r11  eflags for syscall/sysret, temporary for C
- * r12-r15,rbp,rbx saved by C code, not touched. 		
- * 
+ * r12-r15,rbp,rbx saved by C code, not touched.
+ *
  * Interrupts are off on entry.
  * Only called from user space.
  *
@@ -384,7 +462,7 @@ END(ret_from_fork)
  * When user can change the frames always force IRET. That is because
  * it deals with uncanonical addresses better. SYSRET has trouble
  * with them due to bugs in both AMD and Intel CPUs.
- */ 			 		
+ */
 
 ENTRY(system_call)
 	CFI_STARTPROC	simple
@@ -400,7 +478,7 @@ ENTRY(system_call)
 	 */
 ENTRY(system_call_after_swapgs)
 
-	movq	%rsp,%gs:pda_oldrsp 
+	movq	%rsp,%gs:pda_oldrsp
 	movq	%gs:pda_kernelstack,%rsp
 	/*
 	 * No need to follow this irqs off/on section - it's straight
@@ -408,7 +486,7 @@ ENTRY(system_call_after_swapgs)
 	 */
 	ENABLE_INTERRUPTS(CLBR_NONE)
 	SAVE_ARGS 8,1
-	movq  %rax,ORIG_RAX-ARGOFFSET(%rsp) 
+	movq  %rax,ORIG_RAX-ARGOFFSET(%rsp)
 	movq  %rcx,RIP-ARGOFFSET(%rsp)
 	CFI_REL_OFFSET rip,RIP-ARGOFFSET
 	GET_THREAD_INFO(%rcx)
@@ -422,19 +500,19 @@ system_call_fastpath:
 	movq %rax,RAX-ARGOFFSET(%rsp)
 /*
  * Syscall return path ending with SYSRET (fast path)
- * Has incomplete stack frame and undefined top of stack. 
- */		
+ * Has incomplete stack frame and undefined top of stack.
+ */
 ret_from_sys_call:
 	movl $_TIF_ALLWORK_MASK,%edi
 	/* edi:	flagmask */
-sysret_check:		
+sysret_check:
 	LOCKDEP_SYS_EXIT
 	GET_THREAD_INFO(%rcx)
 	DISABLE_INTERRUPTS(CLBR_NONE)
 	TRACE_IRQS_OFF
 	movl TI_flags(%rcx),%edx
 	andl %edi,%edx
-	jnz  sysret_careful 
+	jnz  sysret_careful
 	CFI_REMEMBER_STATE
 	/*
 	 * sysretq will re-enable interrupts:
@@ -449,7 +527,7 @@ sysret_check:
 
 	CFI_RESTORE_STATE
 	/* Handle reschedules */
-	/* edx:	work, edi: workmask */	
+	/* edx:	work, edi: workmask */
 sysret_careful:
 	bt $TIF_NEED_RESCHED,%edx
 	jnc sysret_signal
@@ -462,7 +540,7 @@ sysret_careful:
 	CFI_ADJUST_CFA_OFFSET -8
 	jmp sysret_check
 
-	/* Handle a signal */ 
+	/* Handle a signal */
 sysret_signal:
 	TRACE_IRQS_ON
 	ENABLE_INTERRUPTS(CLBR_NONE)
@@ -471,17 +549,20 @@ sysret_signal:
 	jc sysret_audit
 #endif
 	/* edx:	work flags (arg3) */
-	leaq do_notify_resume(%rip),%rax
 	leaq -ARGOFFSET(%rsp),%rdi # &pt_regs -> arg1
 	xorl %esi,%esi # oldset -> arg2
-	call ptregscall_common
+	SAVE_REST
+	FIXUP_TOP_OF_STACK %r11
+	call do_notify_resume
+	RESTORE_TOP_OF_STACK %r11
+	RESTORE_REST
 	movl $_TIF_WORK_MASK,%edi
 	/* Use IRET because user could have changed frame. This
 	   works because ptregscall_common has called FIXUP_TOP_OF_STACK. */
 	DISABLE_INTERRUPTS(CLBR_NONE)
 	TRACE_IRQS_OFF
 	jmp int_with_check
-	
+
 badsys:
 	movq $-ENOSYS,RAX-ARGOFFSET(%rsp)
 	jmp ret_from_sys_call
@@ -520,7 +601,7 @@ sysret_audit:
 #endif	/* CONFIG_AUDITSYSCALL */
 
 	/* Do syscall tracing */
-tracesys:			 
+tracesys:
 #ifdef CONFIG_AUDITSYSCALL
 	testl $(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%rcx)
 	jz auditsys
@@ -543,8 +624,8 @@ tracesys:
 	call *sys_call_table(,%rax,8)
 	movq %rax,RAX-ARGOFFSET(%rsp)
 	/* Use IRET because user could have changed frame */
-		
-/* 
+
+/*
  * Syscall return path ending with IRET.
  * Has correct top of stack, but partial stack frame.
  */
@@ -588,18 +669,18 @@ int_very_careful:
 	TRACE_IRQS_ON
 	ENABLE_INTERRUPTS(CLBR_NONE)
 	SAVE_REST
-	/* Check for syscall exit trace */	
+	/* Check for syscall exit trace */
 	testl $_TIF_WORK_SYSCALL_EXIT,%edx
 	jz int_signal
 	pushq %rdi
 	CFI_ADJUST_CFA_OFFSET 8
-	leaq 8(%rsp),%rdi	# &ptregs -> arg1	
+	leaq 8(%rsp),%rdi	# &ptregs -> arg1
 	call syscall_trace_leave
 	popq %rdi
 	CFI_ADJUST_CFA_OFFSET -8
 	andl $~(_TIF_WORK_SYSCALL_EXIT|_TIF_SYSCALL_EMU),%edi
 	jmp int_restore_rest
-	
+
 int_signal:
 	testl $_TIF_DO_NOTIFY_MASK,%edx
 	jz 1f
@@ -614,22 +695,24 @@ int_restore_rest:
 	jmp int_with_check
 	CFI_ENDPROC
 END(system_call)
-		
-/* 
+
+/*
  * Certain special system calls that need to save a complete full stack frame.
- */ 								
-	
+ */
 	.macro PTREGSCALL label,func,arg
-	.globl \label
-\label:
-	leaq	\func(%rip),%rax
-	leaq    -ARGOFFSET+8(%rsp),\arg /* 8 for return address */
-	jmp	ptregscall_common
+ENTRY(\label)
+	PARTIAL_FRAME 1 8		/* offset 8: return address */
+	subq $REST_SKIP, %rsp
+	CFI_ADJUST_CFA_OFFSET REST_SKIP
+	call save_rest
+	DEFAULT_FRAME 0 8		/* offset 8: return address */
+	leaq 8(%rsp), \arg	/* pt_regs pointer */
+	call \func
+	jmp ptregscall_common
+	CFI_ENDPROC
 END(\label)
 	.endm
 
-	CFI_STARTPROC
-
 	PTREGSCALL stub_clone, sys_clone, %r8
 	PTREGSCALL stub_fork, sys_fork, %rdi
 	PTREGSCALL stub_vfork, sys_vfork, %rdi
@@ -637,25 +720,18 @@ END(\label)
 	PTREGSCALL stub_iopl, sys_iopl, %rsi
 
 ENTRY(ptregscall_common)
-	popq %r11
-	CFI_ADJUST_CFA_OFFSET -8
-	CFI_REGISTER rip, r11
-	SAVE_REST
-	movq %r11, %r15
-	CFI_REGISTER rip, r15
-	FIXUP_TOP_OF_STACK %r11
-	call *%rax
-	RESTORE_TOP_OF_STACK %r11
-	movq %r15, %r11
-	CFI_REGISTER rip, r11
-	RESTORE_REST
-	pushq %r11
-	CFI_ADJUST_CFA_OFFSET 8
-	CFI_REL_OFFSET rip, 0
-	ret
+	DEFAULT_FRAME 1 8	/* offset 8: return address */
+	RESTORE_TOP_OF_STACK %r11, 8
+	movq_cfi_restore R15+8, r15
+	movq_cfi_restore R14+8, r14
+	movq_cfi_restore R13+8, r13
+	movq_cfi_restore R12+8, r12
+	movq_cfi_restore RBP+8, rbp
+	movq_cfi_restore RBX+8, rbx
+	ret $REST_SKIP		/* pop extended registers */
 	CFI_ENDPROC
 END(ptregscall_common)
-	
+
 ENTRY(stub_execve)
 	CFI_STARTPROC
 	popq %r11
@@ -671,11 +747,11 @@ ENTRY(stub_execve)
 	jmp int_ret_from_sys_call
 	CFI_ENDPROC
 END(stub_execve)
-	
+
 /*
  * sigreturn is special because it needs to restore all registers on return.
  * This cannot be done with SYSRET, so use the IRET return path instead.
- */                
+ */
 ENTRY(stub_rt_sigreturn)
 	CFI_STARTPROC
 	addq $8, %rsp
@@ -691,70 +767,70 @@ ENTRY(stub_rt_sigreturn)
 END(stub_rt_sigreturn)
 
 /*
- * initial frame state for interrupts and exceptions
+ * Build the entry stubs and pointer table with some assembler magic.
+ * We pack 7 stubs into a single 32-byte chunk, which will fit in a
+ * single cache line on all modern x86 implementations.
  */
-	.macro _frame ref
-	CFI_STARTPROC simple
-	CFI_SIGNAL_FRAME
-	CFI_DEF_CFA rsp,SS+8-\ref
-	/*CFI_REL_OFFSET ss,SS-\ref*/
-	CFI_REL_OFFSET rsp,RSP-\ref
-	/*CFI_REL_OFFSET rflags,EFLAGS-\ref*/
-	/*CFI_REL_OFFSET cs,CS-\ref*/
-	CFI_REL_OFFSET rip,RIP-\ref
-	.endm
+	.section .init.rodata,"a"
+ENTRY(interrupt)
+	.text
+	.p2align 5
+	.p2align CONFIG_X86_L1_CACHE_SHIFT
+ENTRY(irq_entries_start)
+	INTR_FRAME
+vector=FIRST_EXTERNAL_VECTOR
+.rept (NR_VECTORS-FIRST_EXTERNAL_VECTOR+6)/7
+	.balign 32
+  .rept	7
+    .if vector < NR_VECTORS
+      .if vector <> FIRST_EXTERNAL_VECTOR
+	CFI_ADJUST_CFA_OFFSET -8
+      .endif
+1:	pushq $(~vector+0x80)	/* Note: always in signed byte range */
+	CFI_ADJUST_CFA_OFFSET 8
+      .if ((vector-FIRST_EXTERNAL_VECTOR)%7) <> 6
+	jmp 2f
+      .endif
+      .previous
+	.quad 1b
+      .text
+vector=vector+1
+    .endif
+  .endr
+2:	jmp common_interrupt
+.endr
+	CFI_ENDPROC
+END(irq_entries_start)
 
-/* initial frame state for interrupts (and exceptions without error code) */
-#define INTR_FRAME _frame RIP
-/* initial frame state for exceptions with error code (and interrupts with
-   vector already pushed) */
-#define XCPT_FRAME _frame ORIG_RAX
+.previous
+END(interrupt)
+.previous
 
-/* 
+/*
  * Interrupt entry/exit.
  *
  * Interrupt entry points save only callee clobbered registers in fast path.
- *	
- * Entry runs with interrupts off.	
- */ 
+ *
+ * Entry runs with interrupts off.
+ */
 
-/* 0(%rsp): interrupt number */ 
+/* 0(%rsp): ~(interrupt number) */
 	.macro interrupt func
-	cld
-	SAVE_ARGS
-	leaq -ARGOFFSET(%rsp),%rdi	# arg1 for handler
-	pushq %rbp
-	/*
-	 * Save rbp twice: One is for marking the stack frame, as usual, and the
-	 * other, to fill pt_regs properly. This is because bx comes right
-	 * before the last saved register in that structure, and not bp. If the
-	 * base pointer were in the place bx is today, this would not be needed.
-	 */
-	movq %rbp, -8(%rsp)
-	CFI_ADJUST_CFA_OFFSET	8
-	CFI_REL_OFFSET		rbp, 0
-	movq %rsp,%rbp
-	CFI_DEF_CFA_REGISTER	rbp
-	testl $3,CS(%rdi)
-	je 1f
-	SWAPGS
-	/* irqcount is used to check if a CPU is already on an interrupt
-	   stack or not. While this is essentially redundant with preempt_count
-	   it is a little cheaper to use a separate counter in the PDA
-	   (short of moving irq_enter into assembly, which would be too
-	    much work) */
-1:	incl	%gs:pda_irqcount
-	cmoveq %gs:pda_irqstackptr,%rsp
-	push    %rbp			# backlink for old unwinder
-	/*
-	 * We entered an interrupt context - irqs are off:
-	 */
-	TRACE_IRQS_OFF
+	subq $10*8, %rsp
+	CFI_ADJUST_CFA_OFFSET 10*8
+	call save_args
+	PARTIAL_FRAME 0
 	call \func
 	.endm
 
-ENTRY(common_interrupt)
+	/*
+	 * The interrupt stubs push (~vector+0x80) onto the stack and
+	 * then jump to common_interrupt.
+	 */
+	.p2align CONFIG_X86_L1_CACHE_SHIFT
+common_interrupt:
 	XCPT_FRAME
+	addq $-0x80,(%rsp)		/* Adjust vector to [-256,-1] range */
 	interrupt do_IRQ
 	/* 0(%rsp): oldrsp-ARGOFFSET */
 ret_from_intr:
@@ -768,12 +844,12 @@ exit_intr:
 	GET_THREAD_INFO(%rcx)
 	testl $3,CS-ARGOFFSET(%rsp)
 	je retint_kernel
-	
+
 	/* Interrupt came from user space */
 	/*
 	 * Has a correct top of stack, but a partial stack frame
 	 * %rcx: thread info. Interrupts off.
-	 */		
+	 */
 retint_with_reschedule:
 	movl $_TIF_WORK_MASK,%edi
 retint_check:
@@ -846,20 +922,20 @@ retint_careful:
 	pushq %rdi
 	CFI_ADJUST_CFA_OFFSET	8
 	call  schedule
-	popq %rdi		
+	popq %rdi
 	CFI_ADJUST_CFA_OFFSET	-8
 	GET_THREAD_INFO(%rcx)
 	DISABLE_INTERRUPTS(CLBR_NONE)
 	TRACE_IRQS_OFF
 	jmp retint_check
-	
+
 retint_signal:
 	testl $_TIF_DO_NOTIFY_MASK,%edx
 	jz    retint_swapgs
 	TRACE_IRQS_ON
 	ENABLE_INTERRUPTS(CLBR_NONE)
 	SAVE_REST
-	movq $-1,ORIG_RAX(%rsp) 			
+	movq $-1,ORIG_RAX(%rsp)
 	xorl %esi,%esi		# oldset
 	movq %rsp,%rdi		# &pt_regs
 	call do_notify_resume
@@ -881,324 +957,211 @@ ENTRY(retint_kernel)
 	jnc  retint_restore_args
 	call preempt_schedule_irq
 	jmp exit_intr
-#endif	
+#endif
 
 	CFI_ENDPROC
 END(common_interrupt)
-	
+
 /*
  * APIC interrupts.
- */		
-	.macro apicinterrupt num,func
+ */
+.macro apicinterrupt num sym do_sym
+ENTRY(\sym)
 	INTR_FRAME
 	pushq $~(\num)
 	CFI_ADJUST_CFA_OFFSET 8
-	interrupt \func
+	interrupt \do_sym
 	jmp ret_from_intr
 	CFI_ENDPROC
-	.endm
-
-ENTRY(thermal_interrupt)
-	apicinterrupt THERMAL_APIC_VECTOR,smp_thermal_interrupt
-END(thermal_interrupt)
-
-ENTRY(threshold_interrupt)
-	apicinterrupt THRESHOLD_APIC_VECTOR,mce_threshold_interrupt
-END(threshold_interrupt)
-
-#ifdef CONFIG_SMP	
-ENTRY(reschedule_interrupt)
-	apicinterrupt RESCHEDULE_VECTOR,smp_reschedule_interrupt
-END(reschedule_interrupt)
-
-	.macro INVALIDATE_ENTRY num
-ENTRY(invalidate_interrupt\num)
-	apicinterrupt INVALIDATE_TLB_VECTOR_START+\num,smp_invalidate_interrupt	
-END(invalidate_interrupt\num)
-	.endm
+END(\sym)
+.endm
 
-	INVALIDATE_ENTRY 0
-	INVALIDATE_ENTRY 1
-	INVALIDATE_ENTRY 2
-	INVALIDATE_ENTRY 3
-	INVALIDATE_ENTRY 4
-	INVALIDATE_ENTRY 5
-	INVALIDATE_ENTRY 6
-	INVALIDATE_ENTRY 7
-
-ENTRY(call_function_interrupt)
-	apicinterrupt CALL_FUNCTION_VECTOR,smp_call_function_interrupt
-END(call_function_interrupt)
-ENTRY(call_function_single_interrupt)
-	apicinterrupt CALL_FUNCTION_SINGLE_VECTOR,smp_call_function_single_interrupt
-END(call_function_single_interrupt)
-ENTRY(irq_move_cleanup_interrupt)
-	apicinterrupt IRQ_MOVE_CLEANUP_VECTOR,smp_irq_move_cleanup_interrupt
-END(irq_move_cleanup_interrupt)
+#ifdef CONFIG_SMP
+apicinterrupt IRQ_MOVE_CLEANUP_VECTOR \
+	irq_move_cleanup_interrupt smp_irq_move_cleanup_interrupt
 #endif
 
-ENTRY(apic_timer_interrupt)
-	apicinterrupt LOCAL_TIMER_VECTOR,smp_apic_timer_interrupt
-END(apic_timer_interrupt)
+apicinterrupt UV_BAU_MESSAGE \
+	uv_bau_message_intr1 uv_bau_message_interrupt
+apicinterrupt LOCAL_TIMER_VECTOR \
+	apic_timer_interrupt smp_apic_timer_interrupt
+
+#ifdef CONFIG_SMP
+apicinterrupt INVALIDATE_TLB_VECTOR_START+0 \
+	invalidate_interrupt0 smp_invalidate_interrupt
+apicinterrupt INVALIDATE_TLB_VECTOR_START+1 \
+	invalidate_interrupt1 smp_invalidate_interrupt
+apicinterrupt INVALIDATE_TLB_VECTOR_START+2 \
+	invalidate_interrupt2 smp_invalidate_interrupt
+apicinterrupt INVALIDATE_TLB_VECTOR_START+3 \
+	invalidate_interrupt3 smp_invalidate_interrupt
+apicinterrupt INVALIDATE_TLB_VECTOR_START+4 \
+	invalidate_interrupt4 smp_invalidate_interrupt
+apicinterrupt INVALIDATE_TLB_VECTOR_START+5 \
+	invalidate_interrupt5 smp_invalidate_interrupt
+apicinterrupt INVALIDATE_TLB_VECTOR_START+6 \
+	invalidate_interrupt6 smp_invalidate_interrupt
+apicinterrupt INVALIDATE_TLB_VECTOR_START+7 \
+	invalidate_interrupt7 smp_invalidate_interrupt
+#endif
 
-ENTRY(uv_bau_message_intr1)
-	apicinterrupt 220,uv_bau_message_interrupt
-END(uv_bau_message_intr1)
+apicinterrupt THRESHOLD_APIC_VECTOR \
+	threshold_interrupt mce_threshold_interrupt
+apicinterrupt THERMAL_APIC_VECTOR \
+	thermal_interrupt smp_thermal_interrupt
+
+#ifdef CONFIG_SMP
+apicinterrupt CALL_FUNCTION_SINGLE_VECTOR \
+	call_function_single_interrupt smp_call_function_single_interrupt
+apicinterrupt CALL_FUNCTION_VECTOR \
+	call_function_interrupt smp_call_function_interrupt
+apicinterrupt RESCHEDULE_VECTOR \
+	reschedule_interrupt smp_reschedule_interrupt
+#endif
 
-ENTRY(error_interrupt)
-	apicinterrupt ERROR_APIC_VECTOR,smp_error_interrupt
-END(error_interrupt)
+apicinterrupt ERROR_APIC_VECTOR \
+	error_interrupt smp_error_interrupt
+apicinterrupt SPURIOUS_APIC_VECTOR \
+	spurious_interrupt smp_spurious_interrupt
 
-ENTRY(spurious_interrupt)
-	apicinterrupt SPURIOUS_APIC_VECTOR,smp_spurious_interrupt
-END(spurious_interrupt)
-				
 /*
  * Exception entry points.
- */ 		
-	.macro zeroentry sym
+ */
+.macro zeroentry sym do_sym
+ENTRY(\sym)
 	INTR_FRAME
 	PARAVIRT_ADJUST_EXCEPTION_FRAME
-	pushq $0	/* push error code/oldrax */ 
-	CFI_ADJUST_CFA_OFFSET 8
-	pushq %rax	/* push real oldrax to the rdi slot */ 
-	CFI_ADJUST_CFA_OFFSET 8
-	CFI_REL_OFFSET rax,0
-	leaq  \sym(%rip),%rax
-	jmp error_entry
+	pushq_cfi $-1		/* ORIG_RAX: no syscall to restart */
+	subq $15*8,%rsp
+	CFI_ADJUST_CFA_OFFSET 15*8
+	call error_entry
+	DEFAULT_FRAME 0
+	movq %rsp,%rdi		/* pt_regs pointer */
+	xorl %esi,%esi		/* no error code */
+	call \do_sym
+	jmp error_exit		/* %ebx: no swapgs flag */
 	CFI_ENDPROC
-	.endm	
+END(\sym)
+.endm
 
-	.macro errorentry sym
-	XCPT_FRAME
+.macro paranoidzeroentry sym do_sym
+ENTRY(\sym)
+	INTR_FRAME
 	PARAVIRT_ADJUST_EXCEPTION_FRAME
-	pushq %rax
+	pushq $-1		/* ORIG_RAX: no syscall to restart */
 	CFI_ADJUST_CFA_OFFSET 8
-	CFI_REL_OFFSET rax,0
-	leaq  \sym(%rip),%rax
-	jmp error_entry
+	subq $15*8, %rsp
+	call save_paranoid
+	TRACE_IRQS_OFF
+	movq %rsp,%rdi		/* pt_regs pointer */
+	xorl %esi,%esi		/* no error code */
+	call \do_sym
+	jmp paranoid_exit	/* %ebx: no swapgs flag */
 	CFI_ENDPROC
-	.endm
+END(\sym)
+.endm
 
-	/* error code is on the stack already */
-	/* handle NMI like exceptions that can happen everywhere */
-	.macro paranoidentry sym, ist=0, irqtrace=1
-	SAVE_ALL
-	cld
-	movl $1,%ebx
-	movl  $MSR_GS_BASE,%ecx
-	rdmsr
-	testl %edx,%edx
-	js    1f
-	SWAPGS
-	xorl  %ebx,%ebx
-1:
-	.if \ist
-	movq	%gs:pda_data_offset, %rbp
-	.endif
-	.if \irqtrace
-	TRACE_IRQS_OFF
-	.endif
-	movq %rsp,%rdi
-	movq ORIG_RAX(%rsp),%rsi
-	movq $-1,ORIG_RAX(%rsp)
-	.if \ist
-	subq	$EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp)
-	.endif
-	call \sym
-	.if \ist
-	addq	$EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp)
-	.endif
-	DISABLE_INTERRUPTS(CLBR_NONE)
-	.if \irqtrace
+.macro paranoidzeroentry_ist sym do_sym ist
+ENTRY(\sym)
+	INTR_FRAME
+	PARAVIRT_ADJUST_EXCEPTION_FRAME
+	pushq $-1		/* ORIG_RAX: no syscall to restart */
+	CFI_ADJUST_CFA_OFFSET 8
+	subq $15*8, %rsp
+	call save_paranoid
 	TRACE_IRQS_OFF
-	.endif
-	.endm
+	movq %rsp,%rdi		/* pt_regs pointer */
+	xorl %esi,%esi		/* no error code */
+	movq %gs:pda_data_offset, %rbp
+	subq $EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp)
+	call \do_sym
+	addq $EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp)
+	jmp paranoid_exit	/* %ebx: no swapgs flag */
+	CFI_ENDPROC
+END(\sym)
+.endm
 
-	/*
- 	 * "Paranoid" exit path from exception stack.
-  	 * Paranoid because this is used by NMIs and cannot take
-	 * any kernel state for granted.
-	 * We don't do kernel preemption checks here, because only
-	 * NMI should be common and it does not enable IRQs and
-	 * cannot get reschedule ticks.
-	 *
-	 * "trace" is 0 for the NMI handler only, because irq-tracing
-	 * is fundamentally NMI-unsafe. (we cannot change the soft and
-	 * hard flags at once, atomically)
-	 */
-	.macro paranoidexit trace=1
-	/* ebx:	no swapgs flag */
-paranoid_exit\trace:
-	testl %ebx,%ebx				/* swapgs needed? */
-	jnz paranoid_restore\trace
-	testl $3,CS(%rsp)
-	jnz   paranoid_userspace\trace
-paranoid_swapgs\trace:
-	.if \trace
-	TRACE_IRQS_IRETQ 0
-	.endif
-	SWAPGS_UNSAFE_STACK
-paranoid_restore\trace:
-	RESTORE_ALL 8
-	jmp irq_return
-paranoid_userspace\trace:
-	GET_THREAD_INFO(%rcx)
-	movl TI_flags(%rcx),%ebx
-	andl $_TIF_WORK_MASK,%ebx
-	jz paranoid_swapgs\trace
-	movq %rsp,%rdi			/* &pt_regs */
-	call sync_regs
-	movq %rax,%rsp			/* switch stack for scheduling */
-	testl $_TIF_NEED_RESCHED,%ebx
-	jnz paranoid_schedule\trace
-	movl %ebx,%edx			/* arg3: thread flags */
-	.if \trace
-	TRACE_IRQS_ON
-	.endif
-	ENABLE_INTERRUPTS(CLBR_NONE)
-	xorl %esi,%esi 			/* arg2: oldset */
-	movq %rsp,%rdi 			/* arg1: &pt_regs */
-	call do_notify_resume
-	DISABLE_INTERRUPTS(CLBR_NONE)
-	.if \trace
-	TRACE_IRQS_OFF
-	.endif
-	jmp paranoid_userspace\trace
-paranoid_schedule\trace:
-	.if \trace
-	TRACE_IRQS_ON
-	.endif
-	ENABLE_INTERRUPTS(CLBR_ANY)
-	call schedule
-	DISABLE_INTERRUPTS(CLBR_ANY)
-	.if \trace
-	TRACE_IRQS_OFF
-	.endif
-	jmp paranoid_userspace\trace
+.macro errorentry sym do_sym
+ENTRY(\sym)
+	XCPT_FRAME
+	PARAVIRT_ADJUST_EXCEPTION_FRAME
+	subq $15*8,%rsp
+	CFI_ADJUST_CFA_OFFSET 15*8
+	call error_entry
+	DEFAULT_FRAME 0
+	movq %rsp,%rdi			/* pt_regs pointer */
+	movq ORIG_RAX(%rsp),%rsi	/* get error code */
+	movq $-1,ORIG_RAX(%rsp)		/* no syscall to restart */
+	call \do_sym
+	jmp error_exit			/* %ebx: no swapgs flag */
 	CFI_ENDPROC
-	.endm
+END(\sym)
+.endm
 
-/*
- * Exception entry point. This expects an error code/orig_rax on the stack
- * and the exception handler in %rax.	
- */ 		  				
-KPROBE_ENTRY(error_entry)
-	_frame RDI
-	CFI_REL_OFFSET rax,0
-	/* rdi slot contains rax, oldrax contains error code */
-	cld	
-	subq  $14*8,%rsp
-	CFI_ADJUST_CFA_OFFSET	(14*8)
-	movq %rsi,13*8(%rsp)
-	CFI_REL_OFFSET	rsi,RSI
-	movq 14*8(%rsp),%rsi	/* load rax from rdi slot */
-	CFI_REGISTER	rax,rsi
-	movq %rdx,12*8(%rsp)
-	CFI_REL_OFFSET	rdx,RDX
-	movq %rcx,11*8(%rsp)
-	CFI_REL_OFFSET	rcx,RCX
-	movq %rsi,10*8(%rsp)	/* store rax */ 
-	CFI_REL_OFFSET	rax,RAX
-	movq %r8, 9*8(%rsp)
-	CFI_REL_OFFSET	r8,R8
-	movq %r9, 8*8(%rsp)
-	CFI_REL_OFFSET	r9,R9
-	movq %r10,7*8(%rsp)
-	CFI_REL_OFFSET	r10,R10
-	movq %r11,6*8(%rsp)
-	CFI_REL_OFFSET	r11,R11
-	movq %rbx,5*8(%rsp) 
-	CFI_REL_OFFSET	rbx,RBX
-	movq %rbp,4*8(%rsp) 
-	CFI_REL_OFFSET	rbp,RBP
-	movq %r12,3*8(%rsp) 
-	CFI_REL_OFFSET	r12,R12
-	movq %r13,2*8(%rsp) 
-	CFI_REL_OFFSET	r13,R13
-	movq %r14,1*8(%rsp) 
-	CFI_REL_OFFSET	r14,R14
-	movq %r15,(%rsp) 
-	CFI_REL_OFFSET	r15,R15
-	xorl %ebx,%ebx	
-	testl $3,CS(%rsp)
-	je  error_kernelspace
-error_swapgs:	
-	SWAPGS
-error_sti:
-	TRACE_IRQS_OFF
-	movq %rdi,RDI(%rsp) 	
-	CFI_REL_OFFSET	rdi,RDI
-	movq %rsp,%rdi
-	movq ORIG_RAX(%rsp),%rsi	/* get error code */ 
-	movq $-1,ORIG_RAX(%rsp)
-	call *%rax
-	/* ebx:	no swapgs flag (1: don't need swapgs, 0: need it) */
-error_exit:
-	movl %ebx,%eax
-	RESTORE_REST
-	DISABLE_INTERRUPTS(CLBR_NONE)
+	/* error code is on the stack already */
+.macro paranoiderrorentry sym do_sym
+ENTRY(\sym)
+	XCPT_FRAME
+	PARAVIRT_ADJUST_EXCEPTION_FRAME
+	subq $15*8,%rsp
+	CFI_ADJUST_CFA_OFFSET 15*8
+	call save_paranoid
+	DEFAULT_FRAME 0
 	TRACE_IRQS_OFF
-	GET_THREAD_INFO(%rcx)	
-	testl %eax,%eax
-	jne  retint_kernel
-	LOCKDEP_SYS_EXIT_IRQ
-	movl  TI_flags(%rcx),%edx
-	movl  $_TIF_WORK_MASK,%edi
-	andl  %edi,%edx
-	jnz  retint_careful
-	jmp retint_swapgs
+	movq %rsp,%rdi			/* pt_regs pointer */
+	movq ORIG_RAX(%rsp),%rsi	/* get error code */
+	movq $-1,ORIG_RAX(%rsp)		/* no syscall to restart */
+	call \do_sym
+	jmp paranoid_exit		/* %ebx: no swapgs flag */
 	CFI_ENDPROC
+END(\sym)
+.endm
 
-error_kernelspace:
-	incl %ebx
-       /* There are two places in the kernel that can potentially fault with
-          usergs. Handle them here. The exception handlers after
-	   iret run with kernel gs again, so don't set the user space flag.
-	   B stepping K8s sometimes report an truncated RIP for IRET 
-	   exceptions returning to compat mode. Check for these here too. */
-	leaq irq_return(%rip),%rcx
-	cmpq %rcx,RIP(%rsp)
-	je   error_swapgs
-	movl %ecx,%ecx	/* zero extend */
-	cmpq %rcx,RIP(%rsp)
-	je   error_swapgs
-	cmpq $gs_change,RIP(%rsp)
-        je   error_swapgs
-	jmp  error_sti
-KPROBE_END(error_entry)
-	
-       /* Reload gs selector with exception handling */
-       /* edi:  new selector */ 
+zeroentry divide_error do_divide_error
+zeroentry overflow do_overflow
+zeroentry bounds do_bounds
+zeroentry invalid_op do_invalid_op
+zeroentry device_not_available do_device_not_available
+paranoiderrorentry double_fault do_double_fault
+zeroentry coprocessor_segment_overrun do_coprocessor_segment_overrun
+errorentry invalid_TSS do_invalid_TSS
+errorentry segment_not_present do_segment_not_present
+zeroentry spurious_interrupt_bug do_spurious_interrupt_bug
+zeroentry coprocessor_error do_coprocessor_error
+errorentry alignment_check do_alignment_check
+zeroentry simd_coprocessor_error do_simd_coprocessor_error
+
+	/* Reload gs selector with exception handling */
+	/* edi:  new selector */
 ENTRY(native_load_gs_index)
 	CFI_STARTPROC
 	pushf
 	CFI_ADJUST_CFA_OFFSET 8
 	DISABLE_INTERRUPTS(CLBR_ANY | ~(CLBR_RDI))
-        SWAPGS
-gs_change:     
-        movl %edi,%gs   
+	SWAPGS
+gs_change:
+	movl %edi,%gs
 2:	mfence		/* workaround */
 	SWAPGS
-        popf
+	popf
 	CFI_ADJUST_CFA_OFFSET -8
-        ret
+	ret
 	CFI_ENDPROC
-ENDPROC(native_load_gs_index)
-       
-        .section __ex_table,"a"
-        .align 8
-        .quad gs_change,bad_gs
-        .previous
-        .section .fixup,"ax"
+END(native_load_gs_index)
+
+	.section __ex_table,"a"
+	.align 8
+	.quad gs_change,bad_gs
+	.previous
+	.section .fixup,"ax"
 	/* running with kernelgs */
-bad_gs: 
+bad_gs:
 	SWAPGS			/* switch back to user gs */
 	xorl %eax,%eax
-        movl %eax,%gs
-        jmp  2b
-        .previous       
-	
+	movl %eax,%gs
+	jmp  2b
+	.previous
+
 /*
  * Create a kernel thread.
  *
@@ -1221,7 +1184,7 @@ ENTRY(kernel_thread)
 
 	xorl %r8d,%r8d
 	xorl %r9d,%r9d
-	
+
 	# clone now
 	call do_fork
 	movq %rax,RAX(%rsp)
@@ -1232,15 +1195,15 @@ ENTRY(kernel_thread)
 	 * so internally to the x86_64 port you can rely on kernel_thread()
 	 * not to reschedule the child before returning, this avoids the need
 	 * of hacks for example to fork off the per-CPU idle tasks.
-         * [Hopefully no generic code relies on the reschedule -AK]	
+	 * [Hopefully no generic code relies on the reschedule -AK]
 	 */
 	RESTORE_ALL
 	UNFAKE_STACK_FRAME
 	ret
 	CFI_ENDPROC
-ENDPROC(kernel_thread)
-	
-child_rip:
+END(kernel_thread)
+
+ENTRY(child_rip)
 	pushq $0		# fake return address
 	CFI_STARTPROC
 	/*
@@ -1253,8 +1216,9 @@ child_rip:
 	# exit
 	mov %eax, %edi
 	call do_exit
+	ud2			# padding for call trace
 	CFI_ENDPROC
-ENDPROC(child_rip)
+END(child_rip)
 
 /*
  * execve(). This function needs to use IRET, not SYSRET, to set up all state properly.
@@ -1274,10 +1238,10 @@ ENDPROC(child_rip)
 ENTRY(kernel_execve)
 	CFI_STARTPROC
 	FAKE_STACK_FRAME $0
-	SAVE_ALL	
+	SAVE_ALL
 	movq %rsp,%rcx
 	call sys_execve
-	movq %rax, RAX(%rsp)	
+	movq %rax, RAX(%rsp)
 	RESTORE_REST
 	testq %rax,%rax
 	je int_ret_from_sys_call
@@ -1285,129 +1249,7 @@ ENTRY(kernel_execve)
 	UNFAKE_STACK_FRAME
 	ret
 	CFI_ENDPROC
-ENDPROC(kernel_execve)
-
-KPROBE_ENTRY(page_fault)
-	errorentry do_page_fault
-KPROBE_END(page_fault)
-
-ENTRY(coprocessor_error)
-	zeroentry do_coprocessor_error
-END(coprocessor_error)
-
-ENTRY(simd_coprocessor_error)
-	zeroentry do_simd_coprocessor_error	
-END(simd_coprocessor_error)
-
-ENTRY(device_not_available)
-	zeroentry do_device_not_available
-END(device_not_available)
-
-	/* runs on exception stack */
-KPROBE_ENTRY(debug)
- 	INTR_FRAME
-	PARAVIRT_ADJUST_EXCEPTION_FRAME
-	pushq $0
-	CFI_ADJUST_CFA_OFFSET 8		
-	paranoidentry do_debug, DEBUG_STACK
-	paranoidexit
-KPROBE_END(debug)
-
-	/* runs on exception stack */	
-KPROBE_ENTRY(nmi)
-	INTR_FRAME
-	PARAVIRT_ADJUST_EXCEPTION_FRAME
-	pushq $-1
-	CFI_ADJUST_CFA_OFFSET 8
-	paranoidentry do_nmi, 0, 0
-#ifdef CONFIG_TRACE_IRQFLAGS
-	paranoidexit 0
-#else
-	jmp paranoid_exit1
- 	CFI_ENDPROC
-#endif
-KPROBE_END(nmi)
-
-KPROBE_ENTRY(int3)
- 	INTR_FRAME
-	PARAVIRT_ADJUST_EXCEPTION_FRAME
- 	pushq $0
- 	CFI_ADJUST_CFA_OFFSET 8
- 	paranoidentry do_int3, DEBUG_STACK
- 	jmp paranoid_exit1
- 	CFI_ENDPROC
-KPROBE_END(int3)
-
-ENTRY(overflow)
-	zeroentry do_overflow
-END(overflow)
-
-ENTRY(bounds)
-	zeroentry do_bounds
-END(bounds)
-
-ENTRY(invalid_op)
-	zeroentry do_invalid_op	
-END(invalid_op)
-
-ENTRY(coprocessor_segment_overrun)
-	zeroentry do_coprocessor_segment_overrun
-END(coprocessor_segment_overrun)
-
-	/* runs on exception stack */
-ENTRY(double_fault)
-	XCPT_FRAME
-	PARAVIRT_ADJUST_EXCEPTION_FRAME
-	paranoidentry do_double_fault
-	jmp paranoid_exit1
-	CFI_ENDPROC
-END(double_fault)
-
-ENTRY(invalid_TSS)
-	errorentry do_invalid_TSS
-END(invalid_TSS)
-
-ENTRY(segment_not_present)
-	errorentry do_segment_not_present
-END(segment_not_present)
-
-	/* runs on exception stack */
-ENTRY(stack_segment)
-	XCPT_FRAME
-	PARAVIRT_ADJUST_EXCEPTION_FRAME
-	paranoidentry do_stack_segment
-	jmp paranoid_exit1
-	CFI_ENDPROC
-END(stack_segment)
-
-KPROBE_ENTRY(general_protection)
-	errorentry do_general_protection
-KPROBE_END(general_protection)
-
-ENTRY(alignment_check)
-	errorentry do_alignment_check
-END(alignment_check)
-
-ENTRY(divide_error)
-	zeroentry do_divide_error
-END(divide_error)
-
-ENTRY(spurious_interrupt_bug)
-	zeroentry do_spurious_interrupt_bug
-END(spurious_interrupt_bug)
-
-#ifdef CONFIG_X86_MCE
-	/* runs on exception stack */
-ENTRY(machine_check)
-	INTR_FRAME
-	PARAVIRT_ADJUST_EXCEPTION_FRAME
-	pushq $0
-	CFI_ADJUST_CFA_OFFSET 8	
-	paranoidentry do_machine_check
-	jmp paranoid_exit1
-	CFI_ENDPROC
-END(machine_check)
-#endif
+END(kernel_execve)
 
 /* Call softirq on interrupt stack. Interrupts are off. */
 ENTRY(call_softirq)
@@ -1427,40 +1269,33 @@ ENTRY(call_softirq)
 	decl %gs:pda_irqcount
 	ret
 	CFI_ENDPROC
-ENDPROC(call_softirq)
-
-KPROBE_ENTRY(ignore_sysret)
-	CFI_STARTPROC
-	mov $-ENOSYS,%eax
-	sysret
-	CFI_ENDPROC
-ENDPROC(ignore_sysret)
+END(call_softirq)
 
 #ifdef CONFIG_XEN
-ENTRY(xen_hypervisor_callback)
-	zeroentry xen_do_hypervisor_callback
-END(xen_hypervisor_callback)
+zeroentry xen_hypervisor_callback xen_do_hypervisor_callback
 
 /*
-# A note on the "critical region" in our callback handler.
-# We want to avoid stacking callback handlers due to events occurring
-# during handling of the last event. To do this, we keep events disabled
-# until we've done all processing. HOWEVER, we must enable events before
-# popping the stack frame (can't be done atomically) and so it would still
-# be possible to get enough handler activations to overflow the stack.
-# Although unlikely, bugs of that kind are hard to track down, so we'd
-# like to avoid the possibility.
-# So, on entry to the handler we detect whether we interrupted an
-# existing activation in its critical region -- if so, we pop the current
-# activation and restart the handler using the previous one.
-*/
+ * A note on the "critical region" in our callback handler.
+ * We want to avoid stacking callback handlers due to events occurring
+ * during handling of the last event. To do this, we keep events disabled
+ * until we've done all processing. HOWEVER, we must enable events before
+ * popping the stack frame (can't be done atomically) and so it would still
+ * be possible to get enough handler activations to overflow the stack.
+ * Although unlikely, bugs of that kind are hard to track down, so we'd
+ * like to avoid the possibility.
+ * So, on entry to the handler we detect whether we interrupted an
+ * existing activation in its critical region -- if so, we pop the current
+ * activation and restart the handler using the previous one.
+ */
 ENTRY(xen_do_hypervisor_callback)   # do_hypervisor_callback(struct *pt_regs)
 	CFI_STARTPROC
-/* Since we don't modify %rdi, evtchn_do_upall(struct *pt_regs) will
-   see the correct pointer to the pt_regs */
+/*
+ * Since we don't modify %rdi, evtchn_do_upall(struct *pt_regs) will
+ * see the correct pointer to the pt_regs
+ */
 	movq %rdi, %rsp            # we don't return, adjust the stack frame
 	CFI_ENDPROC
-	CFI_DEFAULT_STACK
+	DEFAULT_FRAME
 11:	incl %gs:pda_irqcount
 	movq %rsp,%rbp
 	CFI_DEF_CFA_REGISTER rbp
@@ -1475,23 +1310,26 @@ ENTRY(xen_do_hypervisor_callback)   # do_hypervisor_callback(struct *pt_regs)
 END(do_hypervisor_callback)
 
 /*
-# Hypervisor uses this for application faults while it executes.
-# We get here for two reasons:
-#  1. Fault while reloading DS, ES, FS or GS
-#  2. Fault while executing IRET
-# Category 1 we do not need to fix up as Xen has already reloaded all segment
-# registers that could be reloaded and zeroed the others.
-# Category 2 we fix up by killing the current process. We cannot use the
-# normal Linux return path in this case because if we use the IRET hypercall
-# to pop the stack frame we end up in an infinite loop of failsafe callbacks.
-# We distinguish between categories by comparing each saved segment register
-# with its current contents: any discrepancy means we in category 1.
-*/
+ * Hypervisor uses this for application faults while it executes.
+ * We get here for two reasons:
+ *  1. Fault while reloading DS, ES, FS or GS
+ *  2. Fault while executing IRET
+ * Category 1 we do not need to fix up as Xen has already reloaded all segment
+ * registers that could be reloaded and zeroed the others.
+ * Category 2 we fix up by killing the current process. We cannot use the
+ * normal Linux return path in this case because if we use the IRET hypercall
+ * to pop the stack frame we end up in an infinite loop of failsafe callbacks.
+ * We distinguish between categories by comparing each saved segment register
+ * with its current contents: any discrepancy means we in category 1.
+ */
 ENTRY(xen_failsafe_callback)
-	framesz = (RIP-0x30)	/* workaround buggy gas */
-	_frame framesz
-	CFI_REL_OFFSET rcx, 0
-	CFI_REL_OFFSET r11, 8
+	INTR_FRAME 1 (6*8)
+	/*CFI_REL_OFFSET gs,GS*/
+	/*CFI_REL_OFFSET fs,FS*/
+	/*CFI_REL_OFFSET es,ES*/
+	/*CFI_REL_OFFSET ds,DS*/
+	CFI_REL_OFFSET r11,8
+	CFI_REL_OFFSET rcx,0
 	movw %ds,%cx
 	cmpw %cx,0x10(%rsp)
 	CFI_REMEMBER_STATE
@@ -1512,12 +1350,9 @@ ENTRY(xen_failsafe_callback)
 	CFI_RESTORE r11
 	addq $0x30,%rsp
 	CFI_ADJUST_CFA_OFFSET -0x30
-	pushq $0
-	CFI_ADJUST_CFA_OFFSET 8
-	pushq %r11
-	CFI_ADJUST_CFA_OFFSET 8
-	pushq %rcx
-	CFI_ADJUST_CFA_OFFSET 8
+	pushq_cfi $0	/* RIP */
+	pushq_cfi %r11
+	pushq_cfi %rcx
 	jmp general_protection
 	CFI_RESTORE_STATE
 1:	/* Segment mismatch => Category 1 (Bad segment). Retry the IRET. */
@@ -1527,11 +1362,223 @@ ENTRY(xen_failsafe_callback)
 	CFI_RESTORE r11
 	addq $0x30,%rsp
 	CFI_ADJUST_CFA_OFFSET -0x30
-	pushq $0
-	CFI_ADJUST_CFA_OFFSET 8
+	pushq_cfi $0
 	SAVE_ALL
 	jmp error_exit
 	CFI_ENDPROC
 END(xen_failsafe_callback)
 
 #endif /* CONFIG_XEN */
+
+/*
+ * Some functions should be protected against kprobes
+ */
+	.pushsection .kprobes.text, "ax"
+
+paranoidzeroentry_ist debug do_debug DEBUG_STACK
+paranoidzeroentry_ist int3 do_int3 DEBUG_STACK
+paranoiderrorentry stack_segment do_stack_segment
+errorentry general_protection do_general_protection
+errorentry page_fault do_page_fault
+#ifdef CONFIG_X86_MCE
+paranoidzeroentry machine_check do_machine_check
+#endif
+
+	/*
+	 * "Paranoid" exit path from exception stack.
+	 * Paranoid because this is used by NMIs and cannot take
+	 * any kernel state for granted.
+	 * We don't do kernel preemption checks here, because only
+	 * NMI should be common and it does not enable IRQs and
+	 * cannot get reschedule ticks.
+	 *
+	 * "trace" is 0 for the NMI handler only, because irq-tracing
+	 * is fundamentally NMI-unsafe. (we cannot change the soft and
+	 * hard flags at once, atomically)
+	 */
+
+	/* ebx:	no swapgs flag */
+ENTRY(paranoid_exit)
+	INTR_FRAME
+	DISABLE_INTERRUPTS(CLBR_NONE)
+	TRACE_IRQS_OFF
+	testl %ebx,%ebx				/* swapgs needed? */
+	jnz paranoid_restore
+	testl $3,CS(%rsp)
+	jnz   paranoid_userspace
+paranoid_swapgs:
+	TRACE_IRQS_IRETQ 0
+	SWAPGS_UNSAFE_STACK
+paranoid_restore:
+	RESTORE_ALL 8
+	jmp irq_return
+paranoid_userspace:
+	GET_THREAD_INFO(%rcx)
+	movl TI_flags(%rcx),%ebx
+	andl $_TIF_WORK_MASK,%ebx
+	jz paranoid_swapgs
+	movq %rsp,%rdi			/* &pt_regs */
+	call sync_regs
+	movq %rax,%rsp			/* switch stack for scheduling */
+	testl $_TIF_NEED_RESCHED,%ebx
+	jnz paranoid_schedule
+	movl %ebx,%edx			/* arg3: thread flags */
+	TRACE_IRQS_ON
+	ENABLE_INTERRUPTS(CLBR_NONE)
+	xorl %esi,%esi 			/* arg2: oldset */
+	movq %rsp,%rdi 			/* arg1: &pt_regs */
+	call do_notify_resume
+	DISABLE_INTERRUPTS(CLBR_NONE)
+	TRACE_IRQS_OFF
+	jmp paranoid_userspace
+paranoid_schedule:
+	TRACE_IRQS_ON
+	ENABLE_INTERRUPTS(CLBR_ANY)
+	call schedule
+	DISABLE_INTERRUPTS(CLBR_ANY)
+	TRACE_IRQS_OFF
+	jmp paranoid_userspace
+	CFI_ENDPROC
+END(paranoid_exit)
+
+/*
+ * Exception entry point. This expects an error code/orig_rax on the stack.
+ * returns in "no swapgs flag" in %ebx.
+ */
+ENTRY(error_entry)
+	XCPT_FRAME
+	CFI_ADJUST_CFA_OFFSET 15*8
+	/* oldrax contains error code */
+	cld
+	movq_cfi rdi, RDI+8
+	movq_cfi rsi, RSI+8
+	movq_cfi rdx, RDX+8
+	movq_cfi rcx, RCX+8
+	movq_cfi rax, RAX+8
+	movq_cfi  r8,  R8+8
+	movq_cfi  r9,  R9+8
+	movq_cfi r10, R10+8
+	movq_cfi r11, R11+8
+	movq_cfi rbx, RBX+8
+	movq_cfi rbp, RBP+8
+	movq_cfi r12, R12+8
+	movq_cfi r13, R13+8
+	movq_cfi r14, R14+8
+	movq_cfi r15, R15+8
+	xorl %ebx,%ebx
+	testl $3,CS+8(%rsp)
+	je error_kernelspace
+error_swapgs:
+	SWAPGS
+error_sti:
+	TRACE_IRQS_OFF
+	ret
+	CFI_ENDPROC
+
+/*
+ * There are two places in the kernel that can potentially fault with
+ * usergs. Handle them here. The exception handlers after iret run with
+ * kernel gs again, so don't set the user space flag. B stepping K8s
+ * sometimes report an truncated RIP for IRET exceptions returning to
+ * compat mode. Check for these here too.
+ */
+error_kernelspace:
+	incl %ebx
+	leaq irq_return(%rip),%rcx
+	cmpq %rcx,RIP+8(%rsp)
+	je error_swapgs
+	movl %ecx,%ecx	/* zero extend */
+	cmpq %rcx,RIP+8(%rsp)
+	je error_swapgs
+	cmpq $gs_change,RIP+8(%rsp)
+	je error_swapgs
+	jmp error_sti
+END(error_entry)
+
+
+/* ebx:	no swapgs flag (1: don't need swapgs, 0: need it) */
+ENTRY(error_exit)
+	DEFAULT_FRAME
+	movl %ebx,%eax
+	RESTORE_REST
+	DISABLE_INTERRUPTS(CLBR_NONE)
+	TRACE_IRQS_OFF
+	GET_THREAD_INFO(%rcx)
+	testl %eax,%eax
+	jne retint_kernel
+	LOCKDEP_SYS_EXIT_IRQ
+	movl TI_flags(%rcx),%edx
+	movl $_TIF_WORK_MASK,%edi
+	andl %edi,%edx
+	jnz retint_careful
+	jmp retint_swapgs
+	CFI_ENDPROC
+END(error_exit)
+
+
+	/* runs on exception stack */
+ENTRY(nmi)
+	INTR_FRAME
+	PARAVIRT_ADJUST_EXCEPTION_FRAME
+	pushq_cfi $-1
+	subq $15*8, %rsp
+	CFI_ADJUST_CFA_OFFSET 15*8
+	call save_paranoid
+	DEFAULT_FRAME 0
+	/* paranoidentry do_nmi, 0; without TRACE_IRQS_OFF */
+	movq %rsp,%rdi
+	movq $-1,%rsi
+	call do_nmi
+#ifdef CONFIG_TRACE_IRQFLAGS
+	/* paranoidexit; without TRACE_IRQS_OFF */
+	/* ebx:	no swapgs flag */
+	DISABLE_INTERRUPTS(CLBR_NONE)
+	testl %ebx,%ebx				/* swapgs needed? */
+	jnz nmi_restore
+	testl $3,CS(%rsp)
+	jnz nmi_userspace
+nmi_swapgs:
+	SWAPGS_UNSAFE_STACK
+nmi_restore:
+	RESTORE_ALL 8
+	jmp irq_return
+nmi_userspace:
+	GET_THREAD_INFO(%rcx)
+	movl TI_flags(%rcx),%ebx
+	andl $_TIF_WORK_MASK,%ebx
+	jz nmi_swapgs
+	movq %rsp,%rdi			/* &pt_regs */
+	call sync_regs
+	movq %rax,%rsp			/* switch stack for scheduling */
+	testl $_TIF_NEED_RESCHED,%ebx
+	jnz nmi_schedule
+	movl %ebx,%edx			/* arg3: thread flags */
+	ENABLE_INTERRUPTS(CLBR_NONE)
+	xorl %esi,%esi 			/* arg2: oldset */
+	movq %rsp,%rdi 			/* arg1: &pt_regs */
+	call do_notify_resume
+	DISABLE_INTERRUPTS(CLBR_NONE)
+	jmp nmi_userspace
+nmi_schedule:
+	ENABLE_INTERRUPTS(CLBR_ANY)
+	call schedule
+	DISABLE_INTERRUPTS(CLBR_ANY)
+	jmp nmi_userspace
+	CFI_ENDPROC
+#else
+	jmp paranoid_exit
+	CFI_ENDPROC
+#endif
+END(nmi)
+
+ENTRY(ignore_sysret)
+	CFI_STARTPROC
+	mov $-ENOSYS,%eax
+	sysret
+	CFI_ENDPROC
+END(ignore_sysret)
+
+/*
+ * End of kprobes section
+ */
+	.popsection
diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c
index 0e88be11227d..b193e082f6ce 100644
--- a/arch/x86/kernel/genx2apic_uv_x.c
+++ b/arch/x86/kernel/genx2apic_uv_x.c
@@ -10,6 +10,7 @@
 
 #include <linux/kernel.h>
 #include <linux/threads.h>
+#include <linux/cpu.h>
 #include <linux/cpumask.h>
 #include <linux/string.h>
 #include <linux/ctype.h>
@@ -17,6 +18,9 @@
 #include <linux/sched.h>
 #include <linux/module.h>
 #include <linux/hardirq.h>
+#include <linux/timer.h>
+#include <linux/proc_fs.h>
+#include <asm/current.h>
 #include <asm/smp.h>
 #include <asm/ipi.h>
 #include <asm/genapic.h>
@@ -383,6 +387,103 @@ static __init void uv_rtc_init(void)
 }
 
 /*
+ * percpu heartbeat timer
+ */
+static void uv_heartbeat(unsigned long ignored)
+{
+	struct timer_list *timer = &uv_hub_info->scir.timer;
+	unsigned char bits = uv_hub_info->scir.state;
+
+	/* flip heartbeat bit */
+	bits ^= SCIR_CPU_HEARTBEAT;
+
+	/* is this cpu idle? */
+	if (idle_cpu(raw_smp_processor_id()))
+		bits &= ~SCIR_CPU_ACTIVITY;
+	else
+		bits |= SCIR_CPU_ACTIVITY;
+
+	/* update system controller interface reg */
+	uv_set_scir_bits(bits);
+
+	/* enable next timer period */
+	mod_timer(timer, jiffies + SCIR_CPU_HB_INTERVAL);
+}
+
+static void __cpuinit uv_heartbeat_enable(int cpu)
+{
+	if (!uv_cpu_hub_info(cpu)->scir.enabled) {
+		struct timer_list *timer = &uv_cpu_hub_info(cpu)->scir.timer;
+
+		uv_set_cpu_scir_bits(cpu, SCIR_CPU_HEARTBEAT|SCIR_CPU_ACTIVITY);
+		setup_timer(timer, uv_heartbeat, cpu);
+		timer->expires = jiffies + SCIR_CPU_HB_INTERVAL;
+		add_timer_on(timer, cpu);
+		uv_cpu_hub_info(cpu)->scir.enabled = 1;
+	}
+
+	/* check boot cpu */
+	if (!uv_cpu_hub_info(0)->scir.enabled)
+		uv_heartbeat_enable(0);
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+static void __cpuinit uv_heartbeat_disable(int cpu)
+{
+	if (uv_cpu_hub_info(cpu)->scir.enabled) {
+		uv_cpu_hub_info(cpu)->scir.enabled = 0;
+		del_timer(&uv_cpu_hub_info(cpu)->scir.timer);
+	}
+	uv_set_cpu_scir_bits(cpu, 0xff);
+}
+
+/*
+ * cpu hotplug notifier
+ */
+static __cpuinit int uv_scir_cpu_notify(struct notifier_block *self,
+				       unsigned long action, void *hcpu)
+{
+	long cpu = (long)hcpu;
+
+	switch (action) {
+	case CPU_ONLINE:
+		uv_heartbeat_enable(cpu);
+		break;
+	case CPU_DOWN_PREPARE:
+		uv_heartbeat_disable(cpu);
+		break;
+	default:
+		break;
+	}
+	return NOTIFY_OK;
+}
+
+static __init void uv_scir_register_cpu_notifier(void)
+{
+	hotcpu_notifier(uv_scir_cpu_notify, 0);
+}
+
+#else /* !CONFIG_HOTPLUG_CPU */
+
+static __init void uv_scir_register_cpu_notifier(void)
+{
+}
+
+static __init int uv_init_heartbeat(void)
+{
+	int cpu;
+
+	if (is_uv_system())
+		for_each_online_cpu(cpu)
+			uv_heartbeat_enable(cpu);
+	return 0;
+}
+
+late_initcall(uv_init_heartbeat);
+
+#endif /* !CONFIG_HOTPLUG_CPU */
+
+/*
  * Called on each cpu to initialize the per_cpu UV data area.
  * 	ZZZ hotplug not supported yet
  */
@@ -455,7 +556,7 @@ void __init uv_system_init(void)
 
 	uv_bios_init();
 	uv_bios_get_sn_info(0, &uv_type, &sn_partition_id,
-			    &uv_coherency_id, &uv_region_size);
+			    &sn_coherency_id, &sn_region_size);
 	uv_rtc_init();
 
 	for_each_present_cpu(cpu) {
@@ -466,8 +567,7 @@ void __init uv_system_init(void)
 		uv_blade_info[blade].nr_possible_cpus++;
 
 		uv_cpu_hub_info(cpu)->lowmem_remap_base = lowmem_redir_base;
-		uv_cpu_hub_info(cpu)->lowmem_remap_top =
-					lowmem_redir_base + lowmem_redir_size;
+		uv_cpu_hub_info(cpu)->lowmem_remap_top = lowmem_redir_size;
 		uv_cpu_hub_info(cpu)->m_val = m_val;
 		uv_cpu_hub_info(cpu)->n_val = m_val;
 		uv_cpu_hub_info(cpu)->numa_blade_id = blade;
@@ -477,7 +577,8 @@ void __init uv_system_init(void)
 		uv_cpu_hub_info(cpu)->gpa_mask = (1 << (m_val + n_val)) - 1;
 		uv_cpu_hub_info(cpu)->gnode_upper = gnode_upper;
 		uv_cpu_hub_info(cpu)->global_mmr_base = mmr_base;
-		uv_cpu_hub_info(cpu)->coherency_domain_number = uv_coherency_id;
+		uv_cpu_hub_info(cpu)->coherency_domain_number = sn_coherency_id;
+		uv_cpu_hub_info(cpu)->scir.offset = SCIR_LOCAL_MMR_BASE + lcpu;
 		uv_node_to_blade[nid] = blade;
 		uv_cpu_to_blade[cpu] = blade;
 		max_pnode = max(pnode, max_pnode);
@@ -494,4 +595,6 @@ void __init uv_system_init(void)
 	map_mmioh_high(max_pnode);
 
 	uv_cpu_init();
+	uv_scir_register_cpu_notifier();
+	proc_mkdir("sgi_uv", NULL);
 }
diff --git a/arch/x86/kernel/head.c b/arch/x86/kernel/head.c
index 1dcb0f13897e..3e66bd364a9d 100644
--- a/arch/x86/kernel/head.c
+++ b/arch/x86/kernel/head.c
@@ -35,7 +35,6 @@ void __init reserve_ebda_region(void)
 
 	/* start of EBDA area */
 	ebda_addr = get_bios_ebda();
-	printk(KERN_INFO "BIOS EBDA/lowmem at: %08x/%08x\n", ebda_addr, lowmem);
 
 	/* Fixup: bios puts an EBDA in the top 64K segment */
 	/* of conventional memory, but does not adjust lowmem. */
diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c
index fa1d25dd83e3..ac108d1fe182 100644
--- a/arch/x86/kernel/head32.c
+++ b/arch/x86/kernel/head32.c
@@ -12,9 +12,12 @@
 #include <asm/sections.h>
 #include <asm/e820.h>
 #include <asm/bios_ebda.h>
+#include <asm/trampoline.h>
 
 void __init i386_start_kernel(void)
 {
+	reserve_trampoline_memory();
+
 	reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS");
 
 #ifdef CONFIG_BLK_DEV_INITRD
diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index d16084f90649..388e05a5fc17 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -24,6 +24,7 @@
 #include <asm/kdebug.h>
 #include <asm/e820.h>
 #include <asm/bios_ebda.h>
+#include <asm/trampoline.h>
 
 /* boot cpu pda */
 static struct x8664_pda _boot_cpu_pda __read_mostly;
@@ -120,6 +121,8 @@ void __init x86_64_start_reservations(char *real_mode_data)
 {
 	copy_bootdata(__va(real_mode_data));
 
+	reserve_trampoline_memory();
+
 	reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS");
 
 #ifdef CONFIG_BLK_DEV_INITRD
diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index e76d7e272974..cd759ad90690 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -33,7 +33,9 @@
  * HPET address is set in acpi/boot.c, when an ACPI entry exists
  */
 unsigned long				hpet_address;
-unsigned long				hpet_num_timers;
+#ifdef CONFIG_PCI_MSI
+static unsigned long			hpet_num_timers;
+#endif
 static void __iomem			*hpet_virt_address;
 
 struct hpet_dev {
@@ -811,7 +813,7 @@ int __init hpet_enable(void)
 
 out_nohpet:
 	hpet_clear_mapping();
-	boot_hpet_disable = 1;
+	hpet_address = 0;
 	return 0;
 }
 
@@ -834,10 +836,11 @@ static __init int hpet_late_init(void)
 
 		hpet_address = force_hpet_address;
 		hpet_enable();
-		if (!hpet_virt_address)
-			return -ENODEV;
 	}
 
+	if (!hpet_virt_address)
+		return -ENODEV;
+
 	hpet_reserve_platform_timers(hpet_readl(HPET_ID));
 
 	for_each_online_cpu(cpu) {
diff --git a/arch/x86/kernel/init_task.c b/arch/x86/kernel/init_task.c
index a4f93b4120c1..d39918076bb4 100644
--- a/arch/x86/kernel/init_task.c
+++ b/arch/x86/kernel/init_task.c
@@ -14,7 +14,6 @@ static struct fs_struct init_fs = INIT_FS;
 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 */
 
 /*
  * Initial thread structure.
diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c
index 1cbf7c8d46e0..3e070bb961d7 100644
--- a/arch/x86/kernel/io_apic.c
+++ b/arch/x86/kernel/io_apic.c
@@ -2472,10 +2472,9 @@ static void set_ir_ioapic_affinity_irq(unsigned int irq,
 asmlinkage void smp_irq_move_cleanup_interrupt(void)
 {
 	unsigned vector, me;
+
 	ack_APIC_irq();
-#ifdef CONFIG_X86_64
 	exit_idle();
-#endif
 	irq_enter();
 
 	me = smp_processor_id();
@@ -2520,7 +2519,7 @@ static void irq_complete_move(struct irq_desc **descp)
 		if (likely(!cfg->move_desc_pending))
 			return;
 
-		/* domain is not change, but affinity is changed */
+		/* domain has not changed, but affinity did */
 		me = smp_processor_id();
 		if (cpu_isset(me, desc->affinity)) {
 			*descp = desc = move_irq_desc(desc, me);
diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c
index fca2991443f5..6383d50f82ea 100644
--- a/arch/x86/kernel/irq_64.c
+++ b/arch/x86/kernel/irq_64.c
@@ -13,12 +13,12 @@
 #include <linux/seq_file.h>
 #include <linux/module.h>
 #include <linux/delay.h>
+#include <linux/ftrace.h>
 #include <asm/uaccess.h>
 #include <asm/io_apic.h>
 #include <asm/idle.h>
 #include <asm/smp.h>
 
-#ifdef CONFIG_DEBUG_STACKOVERFLOW
 /*
  * Probabilistic stack overflow check:
  *
@@ -28,26 +28,25 @@
  */
 static inline void stack_overflow_check(struct pt_regs *regs)
 {
+#ifdef CONFIG_DEBUG_STACKOVERFLOW
 	u64 curbase = (u64)task_stack_page(current);
-	static unsigned long warned = -60*HZ;
-
-	if (regs->sp >= curbase && regs->sp <= curbase + THREAD_SIZE &&
-	    regs->sp <  curbase + sizeof(struct thread_info) + 128 &&
-	    time_after(jiffies, warned + 60*HZ)) {
-		printk("do_IRQ: %s near stack overflow (cur:%Lx,sp:%lx)\n",
-		       current->comm, curbase, regs->sp);
-		show_stack(NULL,NULL);
-		warned = jiffies;
-	}
-}
+
+	WARN_ONCE(regs->sp >= curbase &&
+		  regs->sp <= curbase + THREAD_SIZE &&
+		  regs->sp <  curbase + sizeof(struct thread_info) +
+					sizeof(struct pt_regs) + 128,
+
+		  "do_IRQ: %s near stack overflow (cur:%Lx,sp:%lx)\n",
+			current->comm, curbase, regs->sp);
 #endif
+}
 
 /*
  * do_IRQ handles all normal device IRQ's (the special
  * SMP cross-CPU interrupts have their own specific
  * handlers).
  */
-asmlinkage unsigned int do_IRQ(struct pt_regs *regs)
+asmlinkage unsigned int __irq_entry do_IRQ(struct pt_regs *regs)
 {
 	struct pt_regs *old_regs = set_irq_regs(regs);
 	struct irq_desc *desc;
@@ -60,9 +59,7 @@ asmlinkage unsigned int do_IRQ(struct pt_regs *regs)
 	irq_enter();
 	irq = __get_cpu_var(vector_irq)[vector];
 
-#ifdef CONFIG_DEBUG_STACKOVERFLOW
 	stack_overflow_check(regs);
-#endif
 
 	desc = irq_to_desc(irq);
 	if (likely(desc))
diff --git a/arch/x86/kernel/irqinit_32.c b/arch/x86/kernel/irqinit_32.c
index 61aa2a1004b5..84723295f88a 100644
--- a/arch/x86/kernel/irqinit_32.c
+++ b/arch/x86/kernel/irqinit_32.c
@@ -140,7 +140,7 @@ void __init native_init_IRQ(void)
 	for (i =  FIRST_EXTERNAL_VECTOR; i < NR_VECTORS; i++) {
 		/* SYSCALL_VECTOR was reserved in trap_init. */
 		if (i != SYSCALL_VECTOR)
-			set_intr_gate(i, interrupt[i]);
+			set_intr_gate(i, interrupt[i-FIRST_EXTERNAL_VECTOR]);
 	}
 
 
diff --git a/arch/x86/kernel/irqinit_64.c b/arch/x86/kernel/irqinit_64.c
index 1020919efe1c..31ebfe38e96c 100644
--- a/arch/x86/kernel/irqinit_64.c
+++ b/arch/x86/kernel/irqinit_64.c
@@ -24,41 +24,6 @@
 #include <asm/i8259.h>
 
 /*
- * Common place to define all x86 IRQ vectors
- *
- * This builds up the IRQ handler stubs using some ugly macros in irq.h
- *
- * These macros create the low-level assembly IRQ routines that save
- * register context and call do_IRQ(). do_IRQ() then does all the
- * operations that are needed to keep the AT (or SMP IOAPIC)
- * interrupt-controller happy.
- */
-
-#define IRQ_NAME2(nr) nr##_interrupt(void)
-#define IRQ_NAME(nr) IRQ_NAME2(IRQ##nr)
-
-/*
- *	SMP has a few special interrupts for IPI messages
- */
-
-#define BUILD_IRQ(nr)				\
-	asmlinkage void IRQ_NAME(nr);		\
-	asm("\n.text\n.p2align\n"		\
-	    "IRQ" #nr "_interrupt:\n\t"		\
-	    "push $~(" #nr ") ; "		\
-	    "jmp common_interrupt\n"		\
-	    ".previous");
-
-#define BI(x,y) \
-	BUILD_IRQ(x##y)
-
-#define BUILD_16_IRQS(x) \
-	BI(x,0) BI(x,1) BI(x,2) BI(x,3) \
-	BI(x,4) BI(x,5) BI(x,6) BI(x,7) \
-	BI(x,8) BI(x,9) BI(x,a) BI(x,b) \
-	BI(x,c) BI(x,d) BI(x,e) BI(x,f)
-
-/*
  * ISA PIC or low IO-APIC triggered (INTA-cycle or APIC) interrupts:
  * (these are usually mapped to vectors 0x30-0x3f)
  */
@@ -73,37 +38,6 @@
  *
  * (these are usually mapped into the 0x30-0xff vector range)
  */
-				      BUILD_16_IRQS(0x2) BUILD_16_IRQS(0x3)
-BUILD_16_IRQS(0x4) BUILD_16_IRQS(0x5) BUILD_16_IRQS(0x6) BUILD_16_IRQS(0x7)
-BUILD_16_IRQS(0x8) BUILD_16_IRQS(0x9) BUILD_16_IRQS(0xa) BUILD_16_IRQS(0xb)
-BUILD_16_IRQS(0xc) BUILD_16_IRQS(0xd) BUILD_16_IRQS(0xe) BUILD_16_IRQS(0xf)
-
-#undef BUILD_16_IRQS
-#undef BI
-
-
-#define IRQ(x,y) \
-	IRQ##x##y##_interrupt
-
-#define IRQLIST_16(x) \
-	IRQ(x,0), IRQ(x,1), IRQ(x,2), IRQ(x,3), \
-	IRQ(x,4), IRQ(x,5), IRQ(x,6), IRQ(x,7), \
-	IRQ(x,8), IRQ(x,9), IRQ(x,a), IRQ(x,b), \
-	IRQ(x,c), IRQ(x,d), IRQ(x,e), IRQ(x,f)
-
-/* for the irq vectors */
-static void (*__initdata interrupt[NR_VECTORS - FIRST_EXTERNAL_VECTOR])(void) = {
-					  IRQLIST_16(0x2), IRQLIST_16(0x3),
-	IRQLIST_16(0x4), IRQLIST_16(0x5), IRQLIST_16(0x6), IRQLIST_16(0x7),
-	IRQLIST_16(0x8), IRQLIST_16(0x9), IRQLIST_16(0xa), IRQLIST_16(0xb),
-	IRQLIST_16(0xc), IRQLIST_16(0xd), IRQLIST_16(0xe), IRQLIST_16(0xf)
-};
-
-#undef IRQ
-#undef IRQLIST_16
-
-
-
 
 /*
  * IRQ2 is cascade interrupt to second interrupt controller
diff --git a/arch/x86/kernel/microcode_amd.c b/arch/x86/kernel/microcode_amd.c
index 5f8e5d75a254..c25fdb382292 100644
--- a/arch/x86/kernel/microcode_amd.c
+++ b/arch/x86/kernel/microcode_amd.c
@@ -10,7 +10,7 @@
  *  This driver allows to upgrade microcode on AMD
  *  family 0x10 and 0x11 processors.
  *
- *  Licensed unter the terms of the GNU General Public
+ *  Licensed under the terms of the GNU General Public
  *  License version 2. See file COPYING for details.
 */
 
@@ -32,9 +32,9 @@
 #include <linux/platform_device.h>
 #include <linux/pci.h>
 #include <linux/pci_ids.h>
+#include <linux/uaccess.h>
 
 #include <asm/msr.h>
-#include <asm/uaccess.h>
 #include <asm/processor.h>
 #include <asm/microcode.h>
 
@@ -47,43 +47,38 @@ MODULE_LICENSE("GPL v2");
 #define UCODE_UCODE_TYPE           0x00000001
 
 struct equiv_cpu_entry {
-	unsigned int installed_cpu;
-	unsigned int fixed_errata_mask;
-	unsigned int fixed_errata_compare;
-	unsigned int equiv_cpu;
-};
+	u32	installed_cpu;
+	u32	fixed_errata_mask;
+	u32	fixed_errata_compare;
+	u16	equiv_cpu;
+	u16	res;
+} __attribute__((packed));
 
 struct microcode_header_amd {
-	unsigned int  data_code;
-	unsigned int  patch_id;
-	unsigned char mc_patch_data_id[2];
-	unsigned char mc_patch_data_len;
-	unsigned char init_flag;
-	unsigned int  mc_patch_data_checksum;
-	unsigned int  nb_dev_id;
-	unsigned int  sb_dev_id;
-	unsigned char processor_rev_id[2];
-	unsigned char nb_rev_id;
-	unsigned char sb_rev_id;
-	unsigned char bios_api_rev;
-	unsigned char reserved1[3];
-	unsigned int  match_reg[8];
-};
+	u32	data_code;
+	u32	patch_id;
+	u16	mc_patch_data_id;
+	u8	mc_patch_data_len;
+	u8	init_flag;
+	u32	mc_patch_data_checksum;
+	u32	nb_dev_id;
+	u32	sb_dev_id;
+	u16	processor_rev_id;
+	u8	nb_rev_id;
+	u8	sb_rev_id;
+	u8	bios_api_rev;
+	u8	reserved1[3];
+	u32	match_reg[8];
+} __attribute__((packed));
 
 struct microcode_amd {
 	struct microcode_header_amd hdr;
 	unsigned int mpb[0];
 };
 
-#define UCODE_MAX_SIZE          (2048)
-#define DEFAULT_UCODE_DATASIZE	(896)
-#define MC_HEADER_SIZE		(sizeof(struct microcode_header_amd))
-#define DEFAULT_UCODE_TOTALSIZE (DEFAULT_UCODE_DATASIZE + MC_HEADER_SIZE)
-#define DWSIZE			(sizeof(u32))
-/* For now we support a fixed ucode total size only */
-#define get_totalsize(mc) \
-	((((struct microcode_amd *)mc)->hdr.mc_patch_data_len * 28) \
-	 + MC_HEADER_SIZE)
+#define UCODE_MAX_SIZE			2048
+#define UCODE_CONTAINER_SECTION_HDR	8
+#define UCODE_CONTAINER_HEADER_SIZE	12
 
 /* serialize access to the physical write */
 static DEFINE_SPINLOCK(microcode_update_lock);
@@ -93,31 +88,24 @@ static struct equiv_cpu_entry *equiv_cpu_table;
 static int collect_cpu_info_amd(int cpu, struct cpu_signature *csig)
 {
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
+	u32 dummy;
 
 	memset(csig, 0, sizeof(*csig));
-
 	if (c->x86_vendor != X86_VENDOR_AMD || c->x86 < 0x10) {
-		printk(KERN_ERR "microcode: CPU%d not a capable AMD processor\n",
-		       cpu);
+		printk(KERN_WARNING "microcode: CPU%d: AMD CPU family 0x%x not "
+		       "supported\n", cpu, c->x86);
 		return -1;
 	}
-
-	asm volatile("movl %1, %%ecx; rdmsr"
-		     : "=a" (csig->rev)
-		     : "i" (0x0000008B) : "ecx");
-
-	printk(KERN_INFO "microcode: collect_cpu_info_amd : patch_id=0x%x\n",
-		csig->rev);
-
+	rdmsr(MSR_AMD64_PATCH_LEVEL, csig->rev, dummy);
+	printk(KERN_INFO "microcode: CPU%d: patch_level=0x%x\n", cpu, csig->rev);
 	return 0;
 }
 
 static int get_matching_microcode(int cpu, void *mc, int rev)
 {
 	struct microcode_header_amd *mc_header = mc;
-	struct pci_dev *nb_pci_dev, *sb_pci_dev;
 	unsigned int current_cpu_id;
-	unsigned int equiv_cpu_id = 0x00;
+	u16 equiv_cpu_id = 0;
 	unsigned int i = 0;
 
 	BUG_ON(equiv_cpu_table == NULL);
@@ -132,57 +120,25 @@ static int get_matching_microcode(int cpu, void *mc, int rev)
 	}
 
 	if (!equiv_cpu_id) {
-		printk(KERN_ERR "microcode: CPU%d cpu_id "
-		       "not found in equivalent cpu table \n", cpu);
+		printk(KERN_WARNING "microcode: CPU%d: cpu revision "
+		       "not listed in equivalent cpu table\n", cpu);
 		return 0;
 	}
 
-	if ((mc_header->processor_rev_id[0]) != (equiv_cpu_id & 0xff)) {
-		printk(KERN_ERR
-			"microcode: CPU%d patch does not match "
-			"(patch is %x, cpu extended is %x) \n",
-			cpu, mc_header->processor_rev_id[0],
-			(equiv_cpu_id & 0xff));
+	if (mc_header->processor_rev_id != equiv_cpu_id) {
+		printk(KERN_ERR	"microcode: CPU%d: patch mismatch "
+		       "(processor_rev_id: %x, equiv_cpu_id: %x)\n",
+		       cpu, mc_header->processor_rev_id, equiv_cpu_id);
 		return 0;
 	}
 
-	if ((mc_header->processor_rev_id[1]) != ((equiv_cpu_id >> 16) & 0xff)) {
-		printk(KERN_ERR "microcode: CPU%d patch does not match "
-			"(patch is %x, cpu base id is %x) \n",
-			cpu, mc_header->processor_rev_id[1],
-			((equiv_cpu_id >> 16) & 0xff));
-
+	/* ucode might be chipset specific -- currently we don't support this */
+	if (mc_header->nb_dev_id || mc_header->sb_dev_id) {
+		printk(KERN_ERR "microcode: CPU%d: loading of chipset "
+		       "specific code not yet supported\n", cpu);
 		return 0;
 	}
 
-	/* ucode may be northbridge specific */
-	if (mc_header->nb_dev_id) {
-		nb_pci_dev = pci_get_device(PCI_VENDOR_ID_AMD,
-					    (mc_header->nb_dev_id & 0xff),
-					    NULL);
-		if ((!nb_pci_dev) ||
-		    (mc_header->nb_rev_id != nb_pci_dev->revision)) {
-			printk(KERN_ERR "microcode: CPU%d NB mismatch \n", cpu);
-			pci_dev_put(nb_pci_dev);
-			return 0;
-		}
-		pci_dev_put(nb_pci_dev);
-	}
-
-	/* ucode may be southbridge specific */
-	if (mc_header->sb_dev_id) {
-		sb_pci_dev = pci_get_device(PCI_VENDOR_ID_AMD,
-					    (mc_header->sb_dev_id & 0xff),
-					    NULL);
-		if ((!sb_pci_dev) ||
-		    (mc_header->sb_rev_id != sb_pci_dev->revision)) {
-			printk(KERN_ERR "microcode: CPU%d SB mismatch \n", cpu);
-			pci_dev_put(sb_pci_dev);
-			return 0;
-		}
-		pci_dev_put(sb_pci_dev);
-	}
-
 	if (mc_header->patch_id <= rev)
 		return 0;
 
@@ -192,12 +148,10 @@ static int get_matching_microcode(int cpu, void *mc, int rev)
 static void apply_microcode_amd(int cpu)
 {
 	unsigned long flags;
-	unsigned int eax, edx;
-	unsigned int rev;
+	u32 rev, dummy;
 	int cpu_num = raw_smp_processor_id();
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num;
 	struct microcode_amd *mc_amd = uci->mc;
-	unsigned long addr;
 
 	/* We should bind the task to the CPU */
 	BUG_ON(cpu_num != cpu);
@@ -206,42 +160,34 @@ static void apply_microcode_amd(int cpu)
 		return;
 
 	spin_lock_irqsave(&microcode_update_lock, flags);
-
-	addr = (unsigned long)&mc_amd->hdr.data_code;
-	edx = (unsigned int)(((unsigned long)upper_32_bits(addr)));
-	eax = (unsigned int)(((unsigned long)lower_32_bits(addr)));
-
-	asm volatile("movl %0, %%ecx; wrmsr" :
-		     : "i" (0xc0010020), "a" (eax), "d" (edx) : "ecx");
-
+	wrmsrl(MSR_AMD64_PATCH_LOADER, (u64)(long)&mc_amd->hdr.data_code);
 	/* get patch id after patching */
-	asm volatile("movl %1, %%ecx; rdmsr"
-		     : "=a" (rev)
-		     : "i" (0x0000008B) : "ecx");
-
+	rdmsr(MSR_AMD64_PATCH_LEVEL, rev, dummy);
 	spin_unlock_irqrestore(&microcode_update_lock, flags);
 
 	/* check current patch id and patch's id for match */
 	if (rev != mc_amd->hdr.patch_id) {
-		printk(KERN_ERR "microcode: CPU%d update from revision "
-		       "0x%x to 0x%x failed\n", cpu_num,
-		       mc_amd->hdr.patch_id, rev);
+		printk(KERN_ERR "microcode: CPU%d: update failed "
+		       "(for patch_level=0x%x)\n", cpu, mc_amd->hdr.patch_id);
 		return;
 	}
 
-	printk(KERN_INFO "microcode: CPU%d updated from revision "
-	       "0x%x to 0x%x \n",
-	       cpu_num, uci->cpu_sig.rev, mc_amd->hdr.patch_id);
+	printk(KERN_INFO "microcode: CPU%d: updated (new patch_level=0x%x)\n",
+	       cpu, rev);
 
 	uci->cpu_sig.rev = rev;
 }
 
-static void * get_next_ucode(u8 *buf, unsigned int size,
-			int (*get_ucode_data)(void *, const void *, size_t),
-			unsigned int *mc_size)
+static int get_ucode_data(void *to, const u8 *from, size_t n)
+{
+	memcpy(to, from, n);
+	return 0;
+}
+
+static void *get_next_ucode(const u8 *buf, unsigned int size,
+			    unsigned int *mc_size)
 {
 	unsigned int total_size;
-#define UCODE_CONTAINER_SECTION_HDR	8
 	u8 section_hdr[UCODE_CONTAINER_SECTION_HDR];
 	void *mc;
 
@@ -249,39 +195,37 @@ static void * get_next_ucode(u8 *buf, unsigned int size,
 		return NULL;
 
 	if (section_hdr[0] != UCODE_UCODE_TYPE) {
-		printk(KERN_ERR "microcode: error! "
-		       "Wrong microcode payload type field\n");
+		printk(KERN_ERR "microcode: error: invalid type field in "
+		       "container file section header\n");
 		return NULL;
 	}
 
 	total_size = (unsigned long) (section_hdr[4] + (section_hdr[5] << 8));
 
-	printk(KERN_INFO "microcode: size %u, total_size %u\n",
-		size, total_size);
+	printk(KERN_DEBUG "microcode: size %u, total_size %u\n",
+	       size, total_size);
 
 	if (total_size > size || total_size > UCODE_MAX_SIZE) {
-		printk(KERN_ERR "microcode: error! Bad data in microcode data file\n");
+		printk(KERN_ERR "microcode: error: size mismatch\n");
 		return NULL;
 	}
 
 	mc = vmalloc(UCODE_MAX_SIZE);
 	if (mc) {
 		memset(mc, 0, UCODE_MAX_SIZE);
-		if (get_ucode_data(mc, buf + UCODE_CONTAINER_SECTION_HDR, total_size)) {
+		if (get_ucode_data(mc, buf + UCODE_CONTAINER_SECTION_HDR,
+				   total_size)) {
 			vfree(mc);
 			mc = NULL;
 		} else
 			*mc_size = total_size + UCODE_CONTAINER_SECTION_HDR;
 	}
-#undef UCODE_CONTAINER_SECTION_HDR
 	return mc;
 }
 
 
-static int install_equiv_cpu_table(u8 *buf,
-		int (*get_ucode_data)(void *, const void *, size_t))
+static int install_equiv_cpu_table(const u8 *buf)
 {
-#define UCODE_CONTAINER_HEADER_SIZE	12
 	u8 *container_hdr[UCODE_CONTAINER_HEADER_SIZE];
 	unsigned int *buf_pos = (unsigned int *)container_hdr;
 	unsigned long size;
@@ -292,14 +236,15 @@ static int install_equiv_cpu_table(u8 *buf,
 	size = buf_pos[2];
 
 	if (buf_pos[1] != UCODE_EQUIV_CPU_TABLE_TYPE || !size) {
-		printk(KERN_ERR "microcode: error! "
-		       "Wrong microcode equivalnet cpu table\n");
+		printk(KERN_ERR "microcode: error: invalid type field in "
+		       "container file section header\n");
 		return 0;
 	}
 
 	equiv_cpu_table = (struct equiv_cpu_entry *) vmalloc(size);
 	if (!equiv_cpu_table) {
-		printk(KERN_ERR "microcode: error, can't allocate memory for equiv CPU table\n");
+		printk(KERN_ERR "microcode: failed to allocate "
+		       "equivalent CPU table\n");
 		return 0;
 	}
 
@@ -310,7 +255,6 @@ static int install_equiv_cpu_table(u8 *buf,
 	}
 
 	return size + UCODE_CONTAINER_HEADER_SIZE; /* add header length */
-#undef UCODE_CONTAINER_HEADER_SIZE
 }
 
 static void free_equiv_cpu_table(void)
@@ -321,18 +265,20 @@ static void free_equiv_cpu_table(void)
 	}
 }
 
-static int generic_load_microcode(int cpu, void *data, size_t size,
-		int (*get_ucode_data)(void *, const void *, size_t))
+static int generic_load_microcode(int cpu, const u8 *data, size_t size)
 {
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
-	u8 *ucode_ptr = data, *new_mc = NULL, *mc;
+	const u8 *ucode_ptr = data;
+	void *new_mc = NULL;
+	void *mc;
 	int new_rev = uci->cpu_sig.rev;
 	unsigned int leftover;
 	unsigned long offset;
 
-	offset = install_equiv_cpu_table(ucode_ptr, get_ucode_data);
+	offset = install_equiv_cpu_table(ucode_ptr);
 	if (!offset) {
-		printk(KERN_ERR "microcode: installing equivalent cpu table failed\n");
+		printk(KERN_ERR "microcode: failed to create "
+		       "equivalent cpu table\n");
 		return -EINVAL;
 	}
 
@@ -343,7 +289,7 @@ static int generic_load_microcode(int cpu, void *data, size_t size,
 		unsigned int uninitialized_var(mc_size);
 		struct microcode_header_amd *mc_header;
 
-		mc = get_next_ucode(ucode_ptr, leftover, get_ucode_data, &mc_size);
+		mc = get_next_ucode(ucode_ptr, leftover, &mc_size);
 		if (!mc)
 			break;
 
@@ -353,7 +299,7 @@ static int generic_load_microcode(int cpu, void *data, size_t size,
 				vfree(new_mc);
 			new_rev = mc_header->patch_id;
 			new_mc  = mc;
-		} else 
+		} else
 			vfree(mc);
 
 		ucode_ptr += mc_size;
@@ -365,9 +311,9 @@ static int generic_load_microcode(int cpu, void *data, size_t size,
 			if (uci->mc)
 				vfree(uci->mc);
 			uci->mc = new_mc;
-			pr_debug("microcode: CPU%d found a matching microcode update with"
-				" version 0x%x (current=0x%x)\n",
-				cpu, new_rev, uci->cpu_sig.rev);
+			pr_debug("microcode: CPU%d found a matching microcode "
+				 "update with version 0x%x (current=0x%x)\n",
+				 cpu, new_rev, uci->cpu_sig.rev);
 		} else
 			vfree(new_mc);
 	}
@@ -377,12 +323,6 @@ static int generic_load_microcode(int cpu, void *data, size_t size,
 	return (int)leftover;
 }
 
-static int get_ucode_fw(void *to, const void *from, size_t n)
-{
-	memcpy(to, from, n);
-	return 0;
-}
-
 static int request_microcode_fw(int cpu, struct device *device)
 {
 	const char *fw_name = "amd-ucode/microcode_amd.bin";
@@ -394,12 +334,11 @@ static int request_microcode_fw(int cpu, struct device *device)
 
 	ret = request_firmware(&firmware, fw_name, device);
 	if (ret) {
-		printk(KERN_ERR "microcode: ucode data file %s load failed\n", fw_name);
+		printk(KERN_ERR "microcode: failed to load file %s\n", fw_name);
 		return ret;
 	}
 
-	ret = generic_load_microcode(cpu, (void*)firmware->data, firmware->size,
-			&get_ucode_fw);
+	ret = generic_load_microcode(cpu, firmware->data, firmware->size);
 
 	release_firmware(firmware);
 
@@ -408,8 +347,8 @@ static int request_microcode_fw(int cpu, struct device *device)
 
 static int request_microcode_user(int cpu, const void __user *buf, size_t size)
 {
-	printk(KERN_WARNING "microcode: AMD microcode update via /dev/cpu/microcode"
-			"is not supported\n");
+	printk(KERN_INFO "microcode: AMD microcode update via "
+	       "/dev/cpu/microcode not supported\n");
 	return -1;
 }
 
@@ -433,3 +372,4 @@ struct microcode_ops * __init init_amd_microcode(void)
 {
 	return &microcode_amd_ops;
 }
+
diff --git a/arch/x86/kernel/microcode_core.c b/arch/x86/kernel/microcode_core.c
index 82fb2809ce32..c9b721ba968c 100644
--- a/arch/x86/kernel/microcode_core.c
+++ b/arch/x86/kernel/microcode_core.c
@@ -99,7 +99,7 @@ MODULE_LICENSE("GPL");
 
 #define MICROCODE_VERSION 	"2.00"
 
-struct microcode_ops *microcode_ops;
+static struct microcode_ops *microcode_ops;
 
 /* no concurrent ->write()s are allowed on /dev/cpu/microcode */
 static DEFINE_MUTEX(microcode_mutex);
@@ -203,7 +203,7 @@ MODULE_ALIAS_MISCDEV(MICROCODE_MINOR);
 #endif
 
 /* fake device for request_firmware */
-struct platform_device *microcode_pdev;
+static struct platform_device *microcode_pdev;
 
 static ssize_t reload_store(struct sys_device *dev,
 			    struct sysdev_attribute *attr,
@@ -272,13 +272,18 @@ static struct attribute_group mc_attr_group = {
 	.name = "microcode",
 };
 
-static void microcode_fini_cpu(int cpu)
+static void __microcode_fini_cpu(int cpu)
 {
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
 
-	mutex_lock(&microcode_mutex);
 	microcode_ops->microcode_fini_cpu(cpu);
 	uci->valid = 0;
+}
+
+static void microcode_fini_cpu(int cpu)
+{
+	mutex_lock(&microcode_mutex);
+	__microcode_fini_cpu(cpu);
 	mutex_unlock(&microcode_mutex);
 }
 
@@ -306,12 +311,16 @@ static int microcode_resume_cpu(int cpu)
 	 * to this cpu (a bit of paranoia):
 	 */
 	if (microcode_ops->collect_cpu_info(cpu, &nsig)) {
-		microcode_fini_cpu(cpu);
+		__microcode_fini_cpu(cpu);
+		printk(KERN_ERR "failed to collect_cpu_info for resuming cpu #%d\n",
+				cpu);
 		return -1;
 	}
 
-	if (memcmp(&nsig, &uci->cpu_sig, sizeof(nsig))) {
-		microcode_fini_cpu(cpu);
+	if ((nsig.sig != uci->cpu_sig.sig) || (nsig.pf != uci->cpu_sig.pf)) {
+		__microcode_fini_cpu(cpu);
+		printk(KERN_ERR "cached ucode doesn't match the resuming cpu #%d\n",
+				cpu);
 		/* Should we look for a new ucode here? */
 		return 1;
 	}
@@ -319,7 +328,7 @@ static int microcode_resume_cpu(int cpu)
 	return 0;
 }
 
-void microcode_update_cpu(int cpu)
+static void microcode_update_cpu(int cpu)
 {
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
 	int err = 0;
diff --git a/arch/x86/kernel/microcode_intel.c b/arch/x86/kernel/microcode_intel.c
index 622dc4a21784..b7f4c929e615 100644
--- a/arch/x86/kernel/microcode_intel.c
+++ b/arch/x86/kernel/microcode_intel.c
@@ -155,6 +155,7 @@ static DEFINE_SPINLOCK(microcode_update_lock);
 static int collect_cpu_info(int cpu_num, struct cpu_signature *csig)
 {
 	struct cpuinfo_x86 *c = &cpu_data(cpu_num);
+	unsigned long flags;
 	unsigned int val[2];
 
 	memset(csig, 0, sizeof(*csig));
@@ -174,11 +175,16 @@ static int collect_cpu_info(int cpu_num, struct cpu_signature *csig)
 		csig->pf = 1 << ((val[1] >> 18) & 7);
 	}
 
+	/* serialize access to the physical write to MSR 0x79 */
+	spin_lock_irqsave(&microcode_update_lock, flags);
+
 	wrmsr(MSR_IA32_UCODE_REV, 0, 0);
 	/* see notes above for revision 1.07.  Apparent chip bug */
 	sync_core();
 	/* get the current revision from MSR 0x8B */
 	rdmsr(MSR_IA32_UCODE_REV, val[0], csig->rev);
+	spin_unlock_irqrestore(&microcode_update_lock, flags);
+
 	pr_debug("microcode: collect_cpu_info : sig=0x%x, pf=0x%x, rev=0x%x\n",
 			csig->sig, csig->pf, csig->rev);
 
@@ -465,7 +471,7 @@ static void microcode_fini_cpu(int cpu)
 	uci->mc = NULL;
 }
 
-struct microcode_ops microcode_intel_ops = {
+static struct microcode_ops microcode_intel_ops = {
 	.request_microcode_user		  = request_microcode_user,
 	.request_microcode_fw             = request_microcode_fw,
 	.collect_cpu_info                 = collect_cpu_info,
diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
index 13316cf57cdb..8bd1bf9622a7 100644
--- a/arch/x86/kernel/nmi.c
+++ b/arch/x86/kernel/nmi.c
@@ -208,12 +208,17 @@ static int __init setup_nmi_watchdog(char *str)
 		++str;
 	}
 
-	get_option(&str, &nmi);
-
-	if (nmi >= NMI_INVALID)
-		return 0;
+	if (!strncmp(str, "lapic", 5))
+		nmi_watchdog = NMI_LOCAL_APIC;
+	else if (!strncmp(str, "ioapic", 6))
+		nmi_watchdog = NMI_IO_APIC;
+	else {
+		get_option(&str, &nmi);
+		if (nmi >= NMI_INVALID)
+			return 0;
+		nmi_watchdog = nmi;
+	}
 
-	nmi_watchdog = nmi;
 	return 1;
 }
 __setup("nmi_watchdog=", setup_nmi_watchdog);
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 192624820217..19a1044a0cd9 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -6,6 +6,7 @@
 #include <asm/proto.h>
 #include <asm/dma.h>
 #include <asm/iommu.h>
+#include <asm/gart.h>
 #include <asm/calgary.h>
 #include <asm/amd_iommu.h>
 
@@ -30,11 +31,6 @@ int no_iommu __read_mostly;
 /* Set this to 1 if there is a HW IOMMU in the system */
 int iommu_detected __read_mostly = 0;
 
-/* This tells the BIO block layer to assume merging. Default to off
-   because we cannot guarantee merging later. */
-int iommu_bio_merge __read_mostly = 0;
-EXPORT_SYMBOL(iommu_bio_merge);
-
 dma_addr_t bad_dma_address __read_mostly = 0;
 EXPORT_SYMBOL(bad_dma_address);
 
@@ -105,11 +101,15 @@ static void __init dma32_free_bootmem(void)
 	dma32_bootmem_ptr = NULL;
 	dma32_bootmem_size = 0;
 }
+#endif
 
 void __init pci_iommu_alloc(void)
 {
+#ifdef CONFIG_X86_64
 	/* free the range so iommu could get some range less than 4G */
 	dma32_free_bootmem();
+#endif
+
 	/*
 	 * The order of these functions is important for
 	 * fall-back/fail-over reasons
@@ -125,15 +125,6 @@ void __init pci_iommu_alloc(void)
 	pci_swiotlb_init();
 }
 
-unsigned long iommu_nr_pages(unsigned long addr, unsigned long len)
-{
-	unsigned long size = roundup((addr & ~PAGE_MASK) + len, PAGE_SIZE);
-
-	return size >> PAGE_SHIFT;
-}
-EXPORT_SYMBOL(iommu_nr_pages);
-#endif
-
 void *dma_generic_alloc_coherent(struct device *dev, size_t size,
 				 dma_addr_t *dma_addr, gfp_t flag)
 {
@@ -188,7 +179,6 @@ static __init int iommu_setup(char *p)
 		}
 
 		if (!strncmp(p, "biomerge", 8)) {
-			iommu_bio_merge = 4096;
 			iommu_merge = 1;
 			force_iommu = 1;
 		}
@@ -300,8 +290,8 @@ fs_initcall(pci_iommu_init);
 static __devinit void via_no_dac(struct pci_dev *dev)
 {
 	if ((dev->class >> 8) == PCI_CLASS_BRIDGE_PCI && forbid_dac == 0) {
-		printk(KERN_INFO "PCI: VIA PCI bridge detected."
-				 "Disabling DAC.\n");
+		printk(KERN_INFO
+			"PCI: VIA PCI bridge detected. Disabling DAC.\n");
 		forbid_dac = 1;
 	}
 }
diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index ba7ad83e20a8..a35eaa379ff6 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -745,10 +745,8 @@ void __init gart_iommu_init(void)
 	unsigned long scratch;
 	long i;
 
-	if (cache_k8_northbridges() < 0 || num_k8_northbridges == 0) {
-		printk(KERN_INFO "PCI-GART: No AMD GART found.\n");
+	if (cache_k8_northbridges() < 0 || num_k8_northbridges == 0)
 		return;
-	}
 
 #ifndef CONFIG_AGP_AMD64
 	no_agp = 1;
diff --git a/arch/x86/kernel/pci-swiotlb_64.c b/arch/x86/kernel/pci-swiotlb_64.c
index 3c539d111abb..242c3440687f 100644
--- a/arch/x86/kernel/pci-swiotlb_64.c
+++ b/arch/x86/kernel/pci-swiotlb_64.c
@@ -3,6 +3,8 @@
 #include <linux/pci.h>
 #include <linux/cache.h>
 #include <linux/module.h>
+#include <linux/swiotlb.h>
+#include <linux/bootmem.h>
 #include <linux/dma-mapping.h>
 
 #include <asm/iommu.h>
@@ -11,6 +13,31 @@
 
 int swiotlb __read_mostly;
 
+void *swiotlb_alloc_boot(size_t size, unsigned long nslabs)
+{
+	return alloc_bootmem_low_pages(size);
+}
+
+void *swiotlb_alloc(unsigned order, unsigned long nslabs)
+{
+	return (void *)__get_free_pages(GFP_DMA | __GFP_NOWARN, order);
+}
+
+dma_addr_t swiotlb_phys_to_bus(phys_addr_t paddr)
+{
+	return paddr;
+}
+
+phys_addr_t swiotlb_bus_to_phys(dma_addr_t baddr)
+{
+	return baddr;
+}
+
+int __weak swiotlb_arch_range_needs_mapping(void *ptr, size_t size)
+{
+	return 0;
+}
+
 static dma_addr_t
 swiotlb_map_single_phys(struct device *hwdev, phys_addr_t paddr, size_t size,
 			int direction)
@@ -50,8 +77,10 @@ struct dma_mapping_ops swiotlb_dma_ops = {
 void __init pci_swiotlb_init(void)
 {
 	/* don't initialize swiotlb if iommu=off (no_iommu=1) */
+#ifdef CONFIG_X86_64
 	if (!iommu_detected && !no_iommu && max_pfn > MAX_DMA32_PFN)
 	       swiotlb = 1;
+#endif
 	if (swiotlb_force)
 		swiotlb = 1;
 	if (swiotlb) {
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 95d811a9594f..e68bb9e30864 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -1,6 +1,7 @@
 #include <linux/errno.h>
 #include <linux/kernel.h>
 #include <linux/mm.h>
+#include <asm/idle.h>
 #include <linux/smp.h>
 #include <linux/slab.h>
 #include <linux/sched.h>
@@ -302,7 +303,7 @@ static void c1e_idle(void)
 		rdmsr(MSR_K8_INT_PENDING_MSG, lo, hi);
 		if (lo & K8_INTP_C1E_ACTIVE_MASK) {
 			c1e_detected = 1;
-			if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC))
+			if (!boot_cpu_has(X86_FEATURE_NONSTOP_TSC))
 				mark_tsc_unstable("TSC halt in AMD C1E");
 			printk(KERN_INFO "System has AMD C1E enabled\n");
 			set_cpu_cap(&boot_cpu_data, X86_FEATURE_AMDC1E);
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 24c2276aa453..3ba155d24884 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -60,6 +60,7 @@
 #include <asm/idle.h>
 #include <asm/syscalls.h>
 #include <asm/smp.h>
+#include <asm/ds.h>
 
 asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
 
@@ -251,14 +252,8 @@ void exit_thread(void)
 		tss->x86_tss.io_bitmap_base = INVALID_IO_BITMAP_OFFSET;
 		put_cpu();
 	}
-#ifdef CONFIG_X86_DS
-	/* Free any DS contexts that have not been properly released. */
-	if (unlikely(current->thread.ds_ctx)) {
-		/* we clear debugctl to make sure DS is not used. */
-		update_debugctlmsr(0);
-		ds_free(current->thread.ds_ctx);
-	}
-#endif /* CONFIG_X86_DS */
+
+	ds_exit_thread(current);
 }
 
 void flush_thread(void)
@@ -340,6 +335,12 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
 		kfree(p->thread.io_bitmap_ptr);
 		p->thread.io_bitmap_max = 0;
 	}
+
+	ds_copy_thread(p, current);
+
+	clear_tsk_thread_flag(p, TIF_DEBUGCTLMSR);
+	p->thread.debugctlmsr = 0;
+
 	return err;
 }
 
@@ -420,48 +421,19 @@ int set_tsc_mode(unsigned int val)
 	return 0;
 }
 
-#ifdef CONFIG_X86_DS
-static int update_debugctl(struct thread_struct *prev,
-			struct thread_struct *next, unsigned long debugctl)
-{
-	unsigned long ds_prev = 0;
-	unsigned long ds_next = 0;
-
-	if (prev->ds_ctx)
-		ds_prev = (unsigned long)prev->ds_ctx->ds;
-	if (next->ds_ctx)
-		ds_next = (unsigned long)next->ds_ctx->ds;
-
-	if (ds_next != ds_prev) {
-		/* we clear debugctl to make sure DS
-		 * is not in use when we change it */
-		debugctl = 0;
-		update_debugctlmsr(0);
-		wrmsr(MSR_IA32_DS_AREA, ds_next, 0);
-	}
-	return debugctl;
-}
-#else
-static int update_debugctl(struct thread_struct *prev,
-			struct thread_struct *next, unsigned long debugctl)
-{
-	return debugctl;
-}
-#endif /* CONFIG_X86_DS */
-
 static noinline void
 __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 		 struct tss_struct *tss)
 {
 	struct thread_struct *prev, *next;
-	unsigned long debugctl;
 
 	prev = &prev_p->thread;
 	next = &next_p->thread;
 
-	debugctl = update_debugctl(prev, next, prev->debugctlmsr);
-
-	if (next->debugctlmsr != debugctl)
+	if (test_tsk_thread_flag(next_p, TIF_DS_AREA_MSR) ||
+	    test_tsk_thread_flag(prev_p, TIF_DS_AREA_MSR))
+		ds_switch_to(prev_p, next_p);
+	else if (next->debugctlmsr != prev->debugctlmsr)
 		update_debugctlmsr(next->debugctlmsr);
 
 	if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
@@ -483,15 +455,6 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
 			hard_enable_TSC();
 	}
 
-#ifdef CONFIG_X86_PTRACE_BTS
-	if (test_tsk_thread_flag(prev_p, TIF_BTS_TRACE_TS))
-		ptrace_bts_take_timestamp(prev_p, BTS_TASK_DEPARTS);
-
-	if (test_tsk_thread_flag(next_p, TIF_BTS_TRACE_TS))
-		ptrace_bts_take_timestamp(next_p, BTS_TASK_ARRIVES);
-#endif /* CONFIG_X86_PTRACE_BTS */
-
-
 	if (!test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) {
 		/*
 		 * Disable the bitmap via an invalid offset. We still cache
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index fbb321d53d34..416fb9282f4f 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -53,6 +53,7 @@
 #include <asm/ia32.h>
 #include <asm/idle.h>
 #include <asm/syscalls.h>
+#include <asm/ds.h>
 
 asmlinkage extern void ret_from_fork(void);
 
@@ -236,14 +237,8 @@ void exit_thread(void)
 		t->io_bitmap_max = 0;
 		put_cpu();
 	}
-#ifdef CONFIG_X86_DS
-	/* Free any DS contexts that have not been properly released. */
-	if (unlikely(t->ds_ctx)) {
-		/* we clear debugctl to make sure DS is not used. */
-		update_debugctlmsr(0);
-		ds_free(t->ds_ctx);
-	}
-#endif /* CONFIG_X86_DS */
+
+	ds_exit_thread(current);
 }
 
 void flush_thread(void)
@@ -373,6 +368,12 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
 		if (err)
 			goto out;
 	}
+
+	ds_copy_thread(p, me);
+
+	clear_tsk_thread_flag(p, TIF_DEBUGCTLMSR);
+	p->thread.debugctlmsr = 0;
+
 	err = 0;
 out:
 	if (err && p->thread.io_bitmap_ptr) {
@@ -471,35 +472,14 @@ static inline void __switch_to_xtra(struct task_struct *prev_p,
 				    struct tss_struct *tss)
 {
 	struct thread_struct *prev, *next;
-	unsigned long debugctl;
 
 	prev = &prev_p->thread,
 	next = &next_p->thread;
 
-	debugctl = prev->debugctlmsr;
-
-#ifdef CONFIG_X86_DS
-	{
-		unsigned long ds_prev = 0, ds_next = 0;
-
-		if (prev->ds_ctx)
-			ds_prev = (unsigned long)prev->ds_ctx->ds;
-		if (next->ds_ctx)
-			ds_next = (unsigned long)next->ds_ctx->ds;
-
-		if (ds_next != ds_prev) {
-			/*
-			 * We clear debugctl to make sure DS
-			 * is not in use when we change it:
-			 */
-			debugctl = 0;
-			update_debugctlmsr(0);
-			wrmsrl(MSR_IA32_DS_AREA, ds_next);
-		}
-	}
-#endif /* CONFIG_X86_DS */
-
-	if (next->debugctlmsr != debugctl)
+	if (test_tsk_thread_flag(next_p, TIF_DS_AREA_MSR) ||
+	    test_tsk_thread_flag(prev_p, TIF_DS_AREA_MSR))
+		ds_switch_to(prev_p, next_p);
+	else if (next->debugctlmsr != prev->debugctlmsr)
 		update_debugctlmsr(next->debugctlmsr);
 
 	if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
@@ -534,14 +514,6 @@ static inline void __switch_to_xtra(struct task_struct *prev_p,
 		 */
 		memset(tss->io_bitmap, 0xff, prev->io_bitmap_max);
 	}
-
-#ifdef CONFIG_X86_PTRACE_BTS
-	if (test_tsk_thread_flag(prev_p, TIF_BTS_TRACE_TS))
-		ptrace_bts_take_timestamp(prev_p, BTS_TASK_DEPARTS);
-
-	if (test_tsk_thread_flag(next_p, TIF_BTS_TRACE_TS))
-		ptrace_bts_take_timestamp(next_p, BTS_TASK_ARRIVES);
-#endif /* CONFIG_X86_PTRACE_BTS */
 }
 
 /*
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 2c8ec1ba75e6..0a5df5f82fb9 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -581,153 +581,91 @@ static int ioperm_get(struct task_struct *target,
 }
 
 #ifdef CONFIG_X86_PTRACE_BTS
-/*
- * The configuration for a particular BTS hardware implementation.
- */
-struct bts_configuration {
-	/* the size of a BTS record in bytes; at most BTS_MAX_RECORD_SIZE */
-	unsigned char  sizeof_bts;
-	/* the size of a field in the BTS record in bytes */
-	unsigned char  sizeof_field;
-	/* a bitmask to enable/disable BTS in DEBUGCTL MSR */
-	unsigned long debugctl_mask;
-};
-static struct bts_configuration bts_cfg;
-
-#define BTS_MAX_RECORD_SIZE (8 * 3)
-
-
-/*
- * Branch Trace Store (BTS) uses the following format. Different
- * architectures vary in the size of those fields.
- * - source linear address
- * - destination linear address
- * - flags
- *
- * Later architectures use 64bit pointers throughout, whereas earlier
- * architectures use 32bit pointers in 32bit mode.
- *
- * We compute the base address for the first 8 fields based on:
- * - the field size stored in the DS configuration
- * - the relative field position
- *
- * In order to store additional information in the BTS buffer, we use
- * a special source address to indicate that the record requires
- * special interpretation.
- *
- * Netburst indicated via a bit in the flags field whether the branch
- * was predicted; this is ignored.
- */
-
-enum bts_field {
-	bts_from = 0,
-	bts_to,
-	bts_flags,
-
-	bts_escape = (unsigned long)-1,
-	bts_qual = bts_to,
-	bts_jiffies = bts_flags
-};
-
-static inline unsigned long bts_get(const char *base, enum bts_field field)
-{
-	base += (bts_cfg.sizeof_field * field);
-	return *(unsigned long *)base;
-}
-
-static inline void bts_set(char *base, enum bts_field field, unsigned long val)
-{
-	base += (bts_cfg.sizeof_field * field);;
-	(*(unsigned long *)base) = val;
-}
-
-/*
- * Translate a BTS record from the raw format into the bts_struct format
- *
- * out (out): bts_struct interpretation
- * raw: raw BTS record
- */
-static void ptrace_bts_translate_record(struct bts_struct *out, const void *raw)
-{
-	memset(out, 0, sizeof(*out));
-	if (bts_get(raw, bts_from) == bts_escape) {
-		out->qualifier       = bts_get(raw, bts_qual);
-		out->variant.jiffies = bts_get(raw, bts_jiffies);
-	} else {
-		out->qualifier = BTS_BRANCH;
-		out->variant.lbr.from_ip = bts_get(raw, bts_from);
-		out->variant.lbr.to_ip   = bts_get(raw, bts_to);
-	}
-}
-
 static int ptrace_bts_read_record(struct task_struct *child, size_t index,
 				  struct bts_struct __user *out)
 {
-	struct bts_struct ret;
-	const void *bts_record;
-	size_t bts_index, bts_end;
+	const struct bts_trace *trace;
+	struct bts_struct bts;
+	const unsigned char *at;
 	int error;
 
-	error = ds_get_bts_end(child->bts, &bts_end);
-	if (error < 0)
-		return error;
+	trace = ds_read_bts(child->bts);
+	if (!trace)
+		return -EPERM;
 
-	if (bts_end <= index)
-		return -EINVAL;
+	at = trace->ds.top - ((index + 1) * trace->ds.size);
+	if ((void *)at < trace->ds.begin)
+		at += (trace->ds.n * trace->ds.size);
 
-	error = ds_get_bts_index(child->bts, &bts_index);
-	if (error < 0)
-		return error;
-
-	/* translate the ptrace bts index into the ds bts index */
-	bts_index += bts_end - (index + 1);
-	if (bts_end <= bts_index)
-		bts_index -= bts_end;
+	if (!trace->read)
+		return -EOPNOTSUPP;
 
-	error = ds_access_bts(child->bts, bts_index, &bts_record);
+	error = trace->read(child->bts, at, &bts);
 	if (error < 0)
 		return error;
 
-	ptrace_bts_translate_record(&ret, bts_record);
-
-	if (copy_to_user(out, &ret, sizeof(ret)))
+	if (copy_to_user(out, &bts, sizeof(bts)))
 		return -EFAULT;
 
-	return sizeof(ret);
+	return sizeof(bts);
 }
 
 static int ptrace_bts_drain(struct task_struct *child,
 			    long size,
 			    struct bts_struct __user *out)
 {
-	struct bts_struct ret;
-	const unsigned char *raw;
-	size_t end, i;
-	int error;
+	const struct bts_trace *trace;
+	const unsigned char *at;
+	int error, drained = 0;
 
-	error = ds_get_bts_index(child->bts, &end);
-	if (error < 0)
-		return error;
+	trace = ds_read_bts(child->bts);
+	if (!trace)
+		return -EPERM;
 
-	if (size < (end * sizeof(struct bts_struct)))
+	if (!trace->read)
+		return -EOPNOTSUPP;
+
+	if (size < (trace->ds.top - trace->ds.begin))
 		return -EIO;
 
-	error = ds_access_bts(child->bts, 0, (const void **)&raw);
-	if (error < 0)
-		return error;
+	for (at = trace->ds.begin; (void *)at < trace->ds.top;
+	     out++, drained++, at += trace->ds.size) {
+		struct bts_struct bts;
+		int error;
 
-	for (i = 0; i < end; i++, out++, raw += bts_cfg.sizeof_bts) {
-		ptrace_bts_translate_record(&ret, raw);
+		error = trace->read(child->bts, at, &bts);
+		if (error < 0)
+			return error;
 
-		if (copy_to_user(out, &ret, sizeof(ret)))
+		if (copy_to_user(out, &bts, sizeof(bts)))
 			return -EFAULT;
 	}
 
-	error = ds_clear_bts(child->bts);
+	memset(trace->ds.begin, 0, trace->ds.n * trace->ds.size);
+
+	error = ds_reset_bts(child->bts);
 	if (error < 0)
 		return error;
 
-	return end;
+	return drained;
+}
+
+static int ptrace_bts_allocate_buffer(struct task_struct *child, size_t size)
+{
+	child->bts_buffer = alloc_locked_buffer(size);
+	if (!child->bts_buffer)
+		return -ENOMEM;
+
+	child->bts_size = size;
+
+	return 0;
+}
+
+static void ptrace_bts_free_buffer(struct task_struct *child)
+{
+	free_locked_buffer(child->bts_buffer, child->bts_size);
+	child->bts_buffer = NULL;
+	child->bts_size = 0;
 }
 
 static int ptrace_bts_config(struct task_struct *child,
@@ -735,136 +673,86 @@ static int ptrace_bts_config(struct task_struct *child,
 			     const struct ptrace_bts_config __user *ucfg)
 {
 	struct ptrace_bts_config cfg;
-	int error = 0;
-
-	error = -EOPNOTSUPP;
-	if (!bts_cfg.sizeof_bts)
-		goto errout;
+	unsigned int flags = 0;
 
-	error = -EIO;
 	if (cfg_size < sizeof(cfg))
-		goto errout;
+		return -EIO;
 
-	error = -EFAULT;
 	if (copy_from_user(&cfg, ucfg, sizeof(cfg)))
-		goto errout;
-
-	error = -EINVAL;
-	if ((cfg.flags & PTRACE_BTS_O_SIGNAL) &&
-	    !(cfg.flags & PTRACE_BTS_O_ALLOC))
-		goto errout;
-
-	if (cfg.flags & PTRACE_BTS_O_ALLOC) {
-		bts_ovfl_callback_t ovfl = NULL;
-		unsigned int sig = 0;
-
-		error = -EINVAL;
-		if (cfg.size < (10 * bts_cfg.sizeof_bts))
-			goto errout;
+		return -EFAULT;
 
-		if (cfg.flags & PTRACE_BTS_O_SIGNAL) {
-			if (!cfg.signal)
-				goto errout;
+	if (child->bts) {
+		ds_release_bts(child->bts);
+		child->bts = NULL;
+	}
 
-			error = -EOPNOTSUPP;
-			goto errout;
+	if (cfg.flags & PTRACE_BTS_O_SIGNAL) {
+		if (!cfg.signal)
+			return -EINVAL;
 
-			sig  = cfg.signal;
-		}
+		return -EOPNOTSUPP;
 
-		if (child->bts) {
-			(void)ds_release_bts(child->bts);
-			kfree(child->bts_buffer);
+		child->thread.bts_ovfl_signal = cfg.signal;
+	}
 
-			child->bts = NULL;
-			child->bts_buffer = NULL;
-		}
+	if ((cfg.flags & PTRACE_BTS_O_ALLOC) &&
+	    (cfg.size != child->bts_size)) {
+		int error;
 
-		error = -ENOMEM;
-		child->bts_buffer = kzalloc(cfg.size, GFP_KERNEL);
-		if (!child->bts_buffer)
-			goto errout;
-
-		child->bts = ds_request_bts(child, child->bts_buffer, cfg.size,
-					    ovfl, /* th = */ (size_t)-1);
-		if (IS_ERR(child->bts)) {
-			error = PTR_ERR(child->bts);
-			kfree(child->bts_buffer);
-			child->bts = NULL;
-			child->bts_buffer = NULL;
-			goto errout;
-		}
+		ptrace_bts_free_buffer(child);
 
-		child->thread.bts_ovfl_signal = sig;
+		error = ptrace_bts_allocate_buffer(child, cfg.size);
+		if (error < 0)
+			return error;
 	}
 
-	error = -EINVAL;
-	if (!child->thread.ds_ctx && cfg.flags)
-		goto errout;
-
 	if (cfg.flags & PTRACE_BTS_O_TRACE)
-		child->thread.debugctlmsr |= bts_cfg.debugctl_mask;
-	else
-		child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask;
+		flags |= BTS_USER;
 
 	if (cfg.flags & PTRACE_BTS_O_SCHED)
-		set_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
-	else
-		clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
+		flags |= BTS_TIMESTAMPS;
 
-	error = sizeof(cfg);
+	child->bts = ds_request_bts(child, child->bts_buffer, child->bts_size,
+				    /* ovfl = */ NULL, /* th = */ (size_t)-1,
+				    flags);
+	if (IS_ERR(child->bts)) {
+		int error = PTR_ERR(child->bts);
 
-out:
-	if (child->thread.debugctlmsr)
-		set_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
-	else
-		clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
+		ptrace_bts_free_buffer(child);
+		child->bts = NULL;
 
-	return error;
+		return error;
+	}
 
-errout:
-	child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask;
-	clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
-	goto out;
+	return sizeof(cfg);
 }
 
 static int ptrace_bts_status(struct task_struct *child,
 			     long cfg_size,
 			     struct ptrace_bts_config __user *ucfg)
 {
+	const struct bts_trace *trace;
 	struct ptrace_bts_config cfg;
-	size_t end;
-	const void *base, *max;
-	int error;
 
 	if (cfg_size < sizeof(cfg))
 		return -EIO;
 
-	error = ds_get_bts_end(child->bts, &end);
-	if (error < 0)
-		return error;
-
-	error = ds_access_bts(child->bts, /* index = */ 0, &base);
-	if (error < 0)
-		return error;
-
-	error = ds_access_bts(child->bts, /* index = */ end, &max);
-	if (error < 0)
-		return error;
+	trace = ds_read_bts(child->bts);
+	if (!trace)
+		return -EPERM;
 
 	memset(&cfg, 0, sizeof(cfg));
-	cfg.size = (max - base);
+	cfg.size = trace->ds.end - trace->ds.begin;
 	cfg.signal = child->thread.bts_ovfl_signal;
 	cfg.bts_size = sizeof(struct bts_struct);
 
 	if (cfg.signal)
 		cfg.flags |= PTRACE_BTS_O_SIGNAL;
 
-	if (test_tsk_thread_flag(child, TIF_DEBUGCTLMSR) &&
-	    child->thread.debugctlmsr & bts_cfg.debugctl_mask)
+	if (trace->ds.flags & BTS_USER)
 		cfg.flags |= PTRACE_BTS_O_TRACE;
 
-	if (test_tsk_thread_flag(child, TIF_BTS_TRACE_TS))
+	if (trace->ds.flags & BTS_TIMESTAMPS)
 		cfg.flags |= PTRACE_BTS_O_SCHED;
 
 	if (copy_to_user(ucfg, &cfg, sizeof(cfg)))
@@ -873,106 +761,77 @@ static int ptrace_bts_status(struct task_struct *child,
 	return sizeof(cfg);
 }
 
-static int ptrace_bts_write_record(struct task_struct *child,
-				   const struct bts_struct *in)
+static int ptrace_bts_clear(struct task_struct *child)
 {
-	unsigned char bts_record[BTS_MAX_RECORD_SIZE];
+	const struct bts_trace *trace;
 
-	BUG_ON(BTS_MAX_RECORD_SIZE < bts_cfg.sizeof_bts);
+	trace = ds_read_bts(child->bts);
+	if (!trace)
+		return -EPERM;
 
-	memset(bts_record, 0, bts_cfg.sizeof_bts);
-	switch (in->qualifier) {
-	case BTS_INVALID:
-		break;
+	memset(trace->ds.begin, 0, trace->ds.n * trace->ds.size);
 
-	case BTS_BRANCH:
-		bts_set(bts_record, bts_from, in->variant.lbr.from_ip);
-		bts_set(bts_record, bts_to,   in->variant.lbr.to_ip);
-		break;
+	return ds_reset_bts(child->bts);
+}
 
-	case BTS_TASK_ARRIVES:
-	case BTS_TASK_DEPARTS:
-		bts_set(bts_record, bts_from,    bts_escape);
-		bts_set(bts_record, bts_qual,    in->qualifier);
-		bts_set(bts_record, bts_jiffies, in->variant.jiffies);
-		break;
+static int ptrace_bts_size(struct task_struct *child)
+{
+	const struct bts_trace *trace;
 
-	default:
-		return -EINVAL;
-	}
+	trace = ds_read_bts(child->bts);
+	if (!trace)
+		return -EPERM;
 
-	return ds_write_bts(child->bts, bts_record, bts_cfg.sizeof_bts);
+	return (trace->ds.top - trace->ds.begin) / trace->ds.size;
 }
 
-void ptrace_bts_take_timestamp(struct task_struct *tsk,
-			       enum bts_qualifier qualifier)
+static void ptrace_bts_fork(struct task_struct *tsk)
 {
-	struct bts_struct rec = {
-		.qualifier = qualifier,
-		.variant.jiffies = jiffies_64
-	};
-
-	ptrace_bts_write_record(tsk, &rec);
+	tsk->bts = NULL;
+	tsk->bts_buffer = NULL;
+	tsk->bts_size = 0;
+	tsk->thread.bts_ovfl_signal = 0;
 }
 
-static const struct bts_configuration bts_cfg_netburst = {
-	.sizeof_bts    = sizeof(long) * 3,
-	.sizeof_field  = sizeof(long),
-	.debugctl_mask = (1<<2)|(1<<3)|(1<<5)
-};
+static void ptrace_bts_untrace(struct task_struct *child)
+{
+	if (unlikely(child->bts)) {
+		ds_release_bts(child->bts);
+		child->bts = NULL;
 
-static const struct bts_configuration bts_cfg_pentium_m = {
-	.sizeof_bts    = sizeof(long) * 3,
-	.sizeof_field  = sizeof(long),
-	.debugctl_mask = (1<<6)|(1<<7)
-};
+		/* We cannot update total_vm and locked_vm since
+		   child's mm is already gone. But we can reclaim the
+		   memory. */
+		kfree(child->bts_buffer);
+		child->bts_buffer = NULL;
+		child->bts_size = 0;
+	}
+}
 
-static const struct bts_configuration bts_cfg_core2 = {
-	.sizeof_bts    = 8 * 3,
-	.sizeof_field  = 8,
-	.debugctl_mask = (1<<6)|(1<<7)|(1<<9)
-};
+static void ptrace_bts_detach(struct task_struct *child)
+{
+	if (unlikely(child->bts)) {
+		ds_release_bts(child->bts);
+		child->bts = NULL;
 
-static inline void bts_configure(const struct bts_configuration *cfg)
+		ptrace_bts_free_buffer(child);
+	}
+}
+#else
+static inline void ptrace_bts_fork(struct task_struct *tsk) {}
+static inline void ptrace_bts_detach(struct task_struct *child) {}
+static inline void ptrace_bts_untrace(struct task_struct *child) {}
+#endif /* CONFIG_X86_PTRACE_BTS */
+
+void x86_ptrace_fork(struct task_struct *child, unsigned long clone_flags)
 {
-	bts_cfg = *cfg;
+	ptrace_bts_fork(child);
 }
 
-void __cpuinit ptrace_bts_init_intel(struct cpuinfo_x86 *c)
+void x86_ptrace_untrace(struct task_struct *child)
 {
-	switch (c->x86) {
-	case 0x6:
-		switch (c->x86_model) {
-		case 0 ... 0xC:
-			/* sorry, don't know about them */
-			break;
-		case 0xD:
-		case 0xE: /* Pentium M */
-			bts_configure(&bts_cfg_pentium_m);
-			break;
-		default: /* Core2, Atom, ... */
-			bts_configure(&bts_cfg_core2);
-			break;
-		}
-		break;
-	case 0xF:
-		switch (c->x86_model) {
-		case 0x0:
-		case 0x1:
-		case 0x2: /* Netburst */
-			bts_configure(&bts_cfg_netburst);
-			break;
-		default:
-			/* sorry, don't know about them */
-			break;
-		}
-		break;
-	default:
-		/* sorry, don't know about them */
-		break;
-	}
+	ptrace_bts_untrace(child);
 }
-#endif /* CONFIG_X86_PTRACE_BTS */
 
 /*
  * Called by kernel/ptrace.c when detaching..
@@ -985,19 +844,7 @@ void ptrace_disable(struct task_struct *child)
 #ifdef TIF_SYSCALL_EMU
 	clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
 #endif
-#ifdef CONFIG_X86_PTRACE_BTS
-	if (child->bts) {
-		(void)ds_release_bts(child->bts);
-		kfree(child->bts_buffer);
-		child->bts_buffer = NULL;
-
-		child->thread.debugctlmsr &= ~bts_cfg.debugctl_mask;
-		if (!child->thread.debugctlmsr)
-			clear_tsk_thread_flag(child, TIF_DEBUGCTLMSR);
-
-		clear_tsk_thread_flag(child, TIF_BTS_TRACE_TS);
-	}
-#endif /* CONFIG_X86_PTRACE_BTS */
+	ptrace_bts_detach(child);
 }
 
 #if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
@@ -1128,16 +975,9 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 			(child, data, (struct ptrace_bts_config __user *)addr);
 		break;
 
-	case PTRACE_BTS_SIZE: {
-		size_t size;
-
-		ret = ds_get_bts_index(child->bts, &size);
-		if (ret == 0) {
-			BUG_ON(size != (int) size);
-			ret = (int) size;
-		}
+	case PTRACE_BTS_SIZE:
+		ret = ptrace_bts_size(child);
 		break;
-	}
 
 	case PTRACE_BTS_GET:
 		ret = ptrace_bts_read_record
@@ -1145,7 +985,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		break;
 
 	case PTRACE_BTS_CLEAR:
-		ret = ds_clear_bts(child->bts);
+		ret = ptrace_bts_clear(child);
 		break;
 
 	case PTRACE_BTS_DRAIN:
@@ -1408,6 +1248,14 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
 
 	case PTRACE_GET_THREAD_AREA:
 	case PTRACE_SET_THREAD_AREA:
+#ifdef CONFIG_X86_PTRACE_BTS
+	case PTRACE_BTS_CONFIG:
+	case PTRACE_BTS_STATUS:
+	case PTRACE_BTS_SIZE:
+	case PTRACE_BTS_GET:
+	case PTRACE_BTS_CLEAR:
+	case PTRACE_BTS_DRAIN:
+#endif /* CONFIG_X86_PTRACE_BTS */
 		return arch_ptrace(child, request, addr, data);
 
 	default:
diff --git a/arch/x86/kernel/quirks.c b/arch/x86/kernel/quirks.c
index 67465ed89310..309949e9e1c1 100644
--- a/arch/x86/kernel/quirks.c
+++ b/arch/x86/kernel/quirks.c
@@ -168,6 +168,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH7_31,
 			 ich_force_enable_hpet);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_1,
 			 ich_force_enable_hpet);
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_4,
+			 ich_force_enable_hpet);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH9_7,
 			 ich_force_enable_hpet);
 
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 32fe42f26e79..ae0d8042cf69 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -93,11 +93,13 @@
 #include <asm/desc.h>
 #include <asm/dma.h>
 #include <asm/iommu.h>
+#include <asm/gart.h>
 #include <asm/mmu_context.h>
 #include <asm/proto.h>
 
 #include <mach_apic.h>
 #include <asm/paravirt.h>
+#include <asm/hypervisor.h>
 
 #include <asm/percpu.h>
 #include <asm/topology.h>
@@ -448,6 +450,7 @@ static void __init reserve_early_setup_data(void)
  * @size: Size of the crashkernel memory to reserve.
  * Returns the base address on success, and -1ULL on failure.
  */
+static
 unsigned long long __init find_and_reserve_crashkernel(unsigned long long size)
 {
 	const unsigned long long alignment = 16<<20; 	/* 16M */
@@ -600,157 +603,7 @@ static struct x86_quirks default_x86_quirks __initdata = {
 
 struct x86_quirks *x86_quirks __initdata = &default_x86_quirks;
 
-/*
- * Some BIOSes seem to corrupt the low 64k of memory during events
- * like suspend/resume and unplugging an HDMI cable.  Reserve all
- * remaining free memory in that area and fill it with a distinct
- * pattern.
- */
-#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION
-#define MAX_SCAN_AREAS	8
-
-static int __read_mostly memory_corruption_check = -1;
-
-static unsigned __read_mostly corruption_check_size = 64*1024;
-static unsigned __read_mostly corruption_check_period = 60; /* seconds */
-
-static struct e820entry scan_areas[MAX_SCAN_AREAS];
-static int num_scan_areas;
-
-
-static int set_corruption_check(char *arg)
-{
-	char *end;
-
-	memory_corruption_check = simple_strtol(arg, &end, 10);
-
-	return (*end == 0) ? 0 : -EINVAL;
-}
-early_param("memory_corruption_check", set_corruption_check);
-
-static int set_corruption_check_period(char *arg)
-{
-	char *end;
-
-	corruption_check_period = simple_strtoul(arg, &end, 10);
-
-	return (*end == 0) ? 0 : -EINVAL;
-}
-early_param("memory_corruption_check_period", set_corruption_check_period);
-
-static int set_corruption_check_size(char *arg)
-{
-	char *end;
-	unsigned size;
-
-	size = memparse(arg, &end);
-
-	if (*end == '\0')
-		corruption_check_size = size;
-
-	return (size == corruption_check_size) ? 0 : -EINVAL;
-}
-early_param("memory_corruption_check_size", set_corruption_check_size);
-
-
-static void __init setup_bios_corruption_check(void)
-{
-	u64 addr = PAGE_SIZE;	/* assume first page is reserved anyway */
-
-	if (memory_corruption_check == -1) {
-		memory_corruption_check =
-#ifdef CONFIG_X86_BOOTPARAM_MEMORY_CORRUPTION_CHECK
-			1
-#else
-			0
-#endif
-			;
-	}
-
-	if (corruption_check_size == 0)
-		memory_corruption_check = 0;
-
-	if (!memory_corruption_check)
-		return;
-
-	corruption_check_size = round_up(corruption_check_size, PAGE_SIZE);
-
-	while(addr < corruption_check_size && num_scan_areas < MAX_SCAN_AREAS) {
-		u64 size;
-		addr = find_e820_area_size(addr, &size, PAGE_SIZE);
-
-		if (addr == 0)
-			break;
-
-		if ((addr + size) > corruption_check_size)
-			size = corruption_check_size - addr;
-
-		if (size == 0)
-			break;
-
-		e820_update_range(addr, size, E820_RAM, E820_RESERVED);
-		scan_areas[num_scan_areas].addr = addr;
-		scan_areas[num_scan_areas].size = size;
-		num_scan_areas++;
-
-		/* Assume we've already mapped this early memory */
-		memset(__va(addr), 0, size);
-
-		addr += size;
-	}
-
-	printk(KERN_INFO "Scanning %d areas for low memory corruption\n",
-	       num_scan_areas);
-	update_e820();
-}
-
-static struct timer_list periodic_check_timer;
-
-void check_for_bios_corruption(void)
-{
-	int i;
-	int corruption = 0;
-
-	if (!memory_corruption_check)
-		return;
-
-	for(i = 0; i < num_scan_areas; i++) {
-		unsigned long *addr = __va(scan_areas[i].addr);
-		unsigned long size = scan_areas[i].size;
-
-		for(; size; addr++, size -= sizeof(unsigned long)) {
-			if (!*addr)
-				continue;
-			printk(KERN_ERR "Corrupted low memory at %p (%lx phys) = %08lx\n",
-			       addr, __pa(addr), *addr);
-			corruption = 1;
-			*addr = 0;
-		}
-	}
-
-	WARN(corruption, KERN_ERR "Memory corruption detected in low memory\n");
-}
-
-static void periodic_check_for_corruption(unsigned long data)
-{
-	check_for_bios_corruption();
-	mod_timer(&periodic_check_timer, round_jiffies(jiffies + corruption_check_period*HZ));
-}
-
-void start_periodic_check_for_corruption(void)
-{
-	if (!memory_corruption_check || corruption_check_period == 0)
-		return;
-
-	printk(KERN_INFO "Scanning for low memory corruption every %d seconds\n",
-	       corruption_check_period);
-
-	init_timer(&periodic_check_timer);
-	periodic_check_timer.function = &periodic_check_for_corruption;
-	periodic_check_for_corruption(0);
-}
-#endif
-
+#ifdef CONFIG_X86_RESERVE_LOW_64K
 static int __init dmi_low_memory_corruption(const struct dmi_system_id *d)
 {
 	printk(KERN_NOTICE
@@ -762,6 +615,7 @@ static int __init dmi_low_memory_corruption(const struct dmi_system_id *d)
 
 	return 0;
 }
+#endif
 
 /* List of systems that have known low memory corruption BIOS problems */
 static struct dmi_system_id __initdata bad_bios_dmi_table[] = {
@@ -920,6 +774,12 @@ void __init setup_arch(char **cmdline_p)
 
 	dmi_check_system(bad_bios_dmi_table);
 
+	/*
+	 * VMware detection requires dmi to be available, so this
+	 * needs to be done after dmi_scan_machine, for the BP.
+	 */
+	init_hypervisor(&boot_cpu_data);
+
 #ifdef CONFIG_X86_32
 	probe_roms();
 #endif
diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c
index 0b63b08e7530..49f3f709ee1f 100644
--- a/arch/x86/kernel/setup_percpu.c
+++ b/arch/x86/kernel/setup_percpu.c
@@ -339,25 +339,25 @@ static const cpumask_t cpu_mask_none;
 /*
  * Returns a pointer to the bitmask of CPUs on Node 'node'.
  */
-const cpumask_t *_node_to_cpumask_ptr(int node)
+const cpumask_t *cpumask_of_node(int node)
 {
 	if (node_to_cpumask_map == NULL) {
 		printk(KERN_WARNING
-			"_node_to_cpumask_ptr(%d): no node_to_cpumask_map!\n",
+			"cpumask_of_node(%d): no node_to_cpumask_map!\n",
 			node);
 		dump_stack();
 		return (const cpumask_t *)&cpu_online_map;
 	}
 	if (node >= nr_node_ids) {
 		printk(KERN_WARNING
-			"_node_to_cpumask_ptr(%d): node > nr_node_ids(%d)\n",
+			"cpumask_of_node(%d): node > nr_node_ids(%d)\n",
 			node, nr_node_ids);
 		dump_stack();
 		return &cpu_mask_none;
 	}
 	return &node_to_cpumask_map[node];
 }
-EXPORT_SYMBOL(_node_to_cpumask_ptr);
+EXPORT_SYMBOL(cpumask_of_node);
 
 /*
  * Returns a bitmask of CPUs on Node 'node'.
diff --git a/arch/x86/kernel/sigframe.h b/arch/x86/kernel/sigframe.h
deleted file mode 100644
index cc673aa55ce4..000000000000
--- a/arch/x86/kernel/sigframe.h
+++ /dev/null
@@ -1,42 +0,0 @@
-#ifdef CONFIG_X86_32
-struct sigframe {
-	char __user *pretcode;
-	int sig;
-	struct sigcontext sc;
-	/*
-	 * fpstate is unused. fpstate is moved/allocated after
-	 * retcode[] below. This movement allows to have the FP state and the
-	 * future state extensions (xsave) stay together.
-	 * And at the same time retaining the unused fpstate, prevents changing
-	 * the offset of extramask[] in the sigframe and thus prevent any
-	 * legacy application accessing/modifying it.
-	 */
-	struct _fpstate fpstate_unused;
-	unsigned long extramask[_NSIG_WORDS-1];
-	char retcode[8];
-	/* fp state follows here */
-};
-
-struct rt_sigframe {
-	char __user *pretcode;
-	int sig;
-	struct siginfo __user *pinfo;
-	void __user *puc;
-	struct siginfo info;
-	struct ucontext uc;
-	char retcode[8];
-	/* fp state follows here */
-};
-#else
-struct rt_sigframe {
-	char __user *pretcode;
-	struct ucontext uc;
-	struct siginfo info;
-	/* fp state follows here */
-};
-
-int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
-		sigset_t *set, struct pt_regs *regs);
-int ia32_setup_frame(int sig, struct k_sigaction *ka,
-		sigset_t *set, struct pt_regs *regs);
-#endif
diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal.c
index d6dd057d0f22..89bb7668041d 100644
--- a/arch/x86/kernel/signal_32.c
+++ b/arch/x86/kernel/signal.c
@@ -1,36 +1,41 @@
 /*
  *  Copyright (C) 1991, 1992  Linus Torvalds
+ *  Copyright (C) 2000, 2001, 2002 Andi Kleen SuSE Labs
  *
  *  1997-11-28  Modified for POSIX.1b signals by Richard Henderson
  *  2000-06-20  Pentium III FXSR, SSE support by Gareth Hughes
+ *  2000-2002   x86-64 support by Andi Kleen
  */
-#include <linux/list.h>
 
-#include <linux/personality.h>
-#include <linux/binfmts.h>
-#include <linux/suspend.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/smp.h>
 #include <linux/kernel.h>
-#include <linux/ptrace.h>
 #include <linux/signal.h>
-#include <linux/stddef.h>
-#include <linux/unistd.h>
 #include <linux/errno.h>
-#include <linux/sched.h>
 #include <linux/wait.h>
+#include <linux/ptrace.h>
 #include <linux/tracehook.h>
-#include <linux/elf.h>
-#include <linux/smp.h>
-#include <linux/mm.h>
+#include <linux/unistd.h>
+#include <linux/stddef.h>
+#include <linux/personality.h>
+#include <linux/uaccess.h>
 
 #include <asm/processor.h>
 #include <asm/ucontext.h>
-#include <asm/uaccess.h>
 #include <asm/i387.h>
 #include <asm/vdso.h>
+
+#ifdef CONFIG_X86_64
+#include <asm/proto.h>
+#include <asm/ia32_unistd.h>
+#include <asm/mce.h>
+#endif /* CONFIG_X86_64 */
+
 #include <asm/syscall.h>
 #include <asm/syscalls.h>
 
-#include "sigframe.h"
+#include <asm/sigframe.h>
 
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
@@ -45,74 +50,6 @@
 # define FIX_EFLAGS	__FIX_EFLAGS
 #endif
 
-/*
- * Atomically swap in the new signal mask, and wait for a signal.
- */
-asmlinkage int
-sys_sigsuspend(int history0, int history1, old_sigset_t mask)
-{
-	mask &= _BLOCKABLE;
-	spin_lock_irq(&current->sighand->siglock);
-	current->saved_sigmask = current->blocked;
-	siginitset(&current->blocked, mask);
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	current->state = TASK_INTERRUPTIBLE;
-	schedule();
-	set_restore_sigmask();
-
-	return -ERESTARTNOHAND;
-}
-
-asmlinkage int
-sys_sigaction(int sig, const struct old_sigaction __user *act,
-	      struct old_sigaction __user *oact)
-{
-	struct k_sigaction new_ka, old_ka;
-	int ret;
-
-	if (act) {
-		old_sigset_t mask;
-
-		if (!access_ok(VERIFY_READ, act, sizeof(*act)) ||
-		    __get_user(new_ka.sa.sa_handler, &act->sa_handler) ||
-		    __get_user(new_ka.sa.sa_restorer, &act->sa_restorer))
-			return -EFAULT;
-
-		__get_user(new_ka.sa.sa_flags, &act->sa_flags);
-		__get_user(mask, &act->sa_mask);
-		siginitset(&new_ka.sa.sa_mask, mask);
-	}
-
-	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
-
-	if (!ret && oact) {
-		if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) ||
-		    __put_user(old_ka.sa.sa_handler, &oact->sa_handler) ||
-		    __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer))
-			return -EFAULT;
-
-		__put_user(old_ka.sa.sa_flags, &oact->sa_flags);
-		__put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask);
-	}
-
-	return ret;
-}
-
-asmlinkage int sys_sigaltstack(unsigned long bx)
-{
-	/*
-	 * This is needed to make gcc realize it doesn't own the
-	 * "struct pt_regs"
-	 */
-	struct pt_regs *regs = (struct pt_regs *)&bx;
-	const stack_t __user *uss = (const stack_t __user *)bx;
-	stack_t __user *uoss = (stack_t __user *)regs->cx;
-
-	return do_sigaltstack(uss, uoss, regs->sp);
-}
-
 #define COPY(x)			{		\
 	err |= __get_user(regs->x, &sc->x);	\
 }
@@ -123,7 +60,7 @@ asmlinkage int sys_sigaltstack(unsigned long bx)
 		regs->seg = tmp;			\
 }
 
-#define COPY_SEG_STRICT(seg)	{			\
+#define COPY_SEG_CPL3(seg)	{			\
 		unsigned short tmp;			\
 		err |= __get_user(tmp, &sc->seg);	\
 		regs->seg = tmp | 3;			\
@@ -135,9 +72,6 @@ asmlinkage int sys_sigaltstack(unsigned long bx)
 		loadsegment(seg, tmp);			\
 }
 
-/*
- * Do a signal return; undo the signal stack.
- */
 static int
 restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc,
 		   unsigned long *pax)
@@ -149,14 +83,36 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc,
 	/* Always make any pending restarted system calls return -EINTR */
 	current_thread_info()->restart_block.fn = do_no_restart_syscall;
 
+#ifdef CONFIG_X86_32
 	GET_SEG(gs);
 	COPY_SEG(fs);
 	COPY_SEG(es);
 	COPY_SEG(ds);
+#endif /* CONFIG_X86_32 */
+
 	COPY(di); COPY(si); COPY(bp); COPY(sp); COPY(bx);
 	COPY(dx); COPY(cx); COPY(ip);
-	COPY_SEG_STRICT(cs);
-	COPY_SEG_STRICT(ss);
+
+#ifdef CONFIG_X86_64
+	COPY(r8);
+	COPY(r9);
+	COPY(r10);
+	COPY(r11);
+	COPY(r12);
+	COPY(r13);
+	COPY(r14);
+	COPY(r15);
+#endif /* CONFIG_X86_64 */
+
+#ifdef CONFIG_X86_32
+	COPY_SEG_CPL3(cs);
+	COPY_SEG_CPL3(ss);
+#else /* !CONFIG_X86_32 */
+	/* Kernel saves and restores only the CS segment register on signals,
+	 * which is the bare minimum needed to allow mixed 32/64-bit code.
+	 * App's signal handler can save/restore other segments if needed. */
+	COPY_SEG_CPL3(cs);
+#endif /* CONFIG_X86_32 */
 
 	err |= __get_user(tmpflags, &sc->flags);
 	regs->flags = (regs->flags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS);
@@ -169,102 +125,24 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc,
 	return err;
 }
 
-asmlinkage unsigned long sys_sigreturn(unsigned long __unused)
-{
-	struct sigframe __user *frame;
-	struct pt_regs *regs;
-	unsigned long ax;
-	sigset_t set;
-
-	regs = (struct pt_regs *) &__unused;
-	frame = (struct sigframe __user *)(regs->sp - 8);
-
-	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
-		goto badframe;
-	if (__get_user(set.sig[0], &frame->sc.oldmask) || (_NSIG_WORDS > 1
-		&& __copy_from_user(&set.sig[1], &frame->extramask,
-				    sizeof(frame->extramask))))
-		goto badframe;
-
-	sigdelsetmask(&set, ~_BLOCKABLE);
-	spin_lock_irq(&current->sighand->siglock);
-	current->blocked = set;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	if (restore_sigcontext(regs, &frame->sc, &ax))
-		goto badframe;
-	return ax;
-
-badframe:
-	if (show_unhandled_signals && printk_ratelimit()) {
-		printk("%s%s[%d] bad frame in sigreturn frame:"
-			"%p ip:%lx sp:%lx oeax:%lx",
-		    task_pid_nr(current) > 1 ? KERN_INFO : KERN_EMERG,
-		    current->comm, task_pid_nr(current), frame, regs->ip,
-		    regs->sp, regs->orig_ax);
-		print_vma_addr(" in ", regs->ip);
-		printk(KERN_CONT "\n");
-	}
-
-	force_sig(SIGSEGV, current);
-
-	return 0;
-}
-
-static long do_rt_sigreturn(struct pt_regs *regs)
-{
-	struct rt_sigframe __user *frame;
-	unsigned long ax;
-	sigset_t set;
-
-	frame = (struct rt_sigframe __user *)(regs->sp - sizeof(long));
-	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
-		goto badframe;
-	if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
-		goto badframe;
-
-	sigdelsetmask(&set, ~_BLOCKABLE);
-	spin_lock_irq(&current->sighand->siglock);
-	current->blocked = set;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ax))
-		goto badframe;
-
-	if (do_sigaltstack(&frame->uc.uc_stack, NULL, regs->sp) == -EFAULT)
-		goto badframe;
-
-	return ax;
-
-badframe:
-	signal_fault(regs, frame, "rt_sigreturn");
-	return 0;
-}
-
-asmlinkage int sys_rt_sigreturn(unsigned long __unused)
-{
-	struct pt_regs *regs = (struct pt_regs *)&__unused;
-
-	return do_rt_sigreturn(regs);
-}
-
-/*
- * Set up a signal frame.
- */
 static int
 setup_sigcontext(struct sigcontext __user *sc, void __user *fpstate,
 		 struct pt_regs *regs, unsigned long mask)
 {
-	int tmp, err = 0;
+	int err = 0;
 
-	err |= __put_user(regs->fs, (unsigned int __user *)&sc->fs);
-	savesegment(gs, tmp);
-	err |= __put_user(tmp, (unsigned int __user *)&sc->gs);
+#ifdef CONFIG_X86_32
+	{
+		unsigned int tmp;
 
+		savesegment(gs, tmp);
+		err |= __put_user(tmp, (unsigned int __user *)&sc->gs);
+	}
+	err |= __put_user(regs->fs, (unsigned int __user *)&sc->fs);
 	err |= __put_user(regs->es, (unsigned int __user *)&sc->es);
 	err |= __put_user(regs->ds, (unsigned int __user *)&sc->ds);
+#endif /* CONFIG_X86_32 */
+
 	err |= __put_user(regs->di, &sc->di);
 	err |= __put_user(regs->si, &sc->si);
 	err |= __put_user(regs->bp, &sc->bp);
@@ -273,19 +151,33 @@ setup_sigcontext(struct sigcontext __user *sc, void __user *fpstate,
 	err |= __put_user(regs->dx, &sc->dx);
 	err |= __put_user(regs->cx, &sc->cx);
 	err |= __put_user(regs->ax, &sc->ax);
+#ifdef CONFIG_X86_64
+	err |= __put_user(regs->r8, &sc->r8);
+	err |= __put_user(regs->r9, &sc->r9);
+	err |= __put_user(regs->r10, &sc->r10);
+	err |= __put_user(regs->r11, &sc->r11);
+	err |= __put_user(regs->r12, &sc->r12);
+	err |= __put_user(regs->r13, &sc->r13);
+	err |= __put_user(regs->r14, &sc->r14);
+	err |= __put_user(regs->r15, &sc->r15);
+#endif /* CONFIG_X86_64 */
+
 	err |= __put_user(current->thread.trap_no, &sc->trapno);
 	err |= __put_user(current->thread.error_code, &sc->err);
 	err |= __put_user(regs->ip, &sc->ip);
+#ifdef CONFIG_X86_32
 	err |= __put_user(regs->cs, (unsigned int __user *)&sc->cs);
 	err |= __put_user(regs->flags, &sc->flags);
 	err |= __put_user(regs->sp, &sc->sp_at_signal);
 	err |= __put_user(regs->ss, (unsigned int __user *)&sc->ss);
+#else /* !CONFIG_X86_32 */
+	err |= __put_user(regs->flags, &sc->flags);
+	err |= __put_user(regs->cs, &sc->cs);
+	err |= __put_user(0, &sc->gs);
+	err |= __put_user(0, &sc->fs);
+#endif /* CONFIG_X86_32 */
 
-	tmp = save_i387_xstate(fpstate);
-	if (tmp < 0)
-		err = 1;
-	else
-		err |= __put_user(tmp ? fpstate : NULL, &sc->fpstate);
+	err |= __put_user(fpstate, &sc->fpstate);
 
 	/* non-iBCS2 extensions.. */
 	err |= __put_user(mask, &sc->oldmask);
@@ -295,6 +187,32 @@ setup_sigcontext(struct sigcontext __user *sc, void __user *fpstate,
 }
 
 /*
+ * Set up a signal frame.
+ */
+#ifdef CONFIG_X86_32
+static const struct {
+	u16 poplmovl;
+	u32 val;
+	u16 int80;
+} __attribute__((packed)) retcode = {
+	0xb858,		/* popl %eax; movl $..., %eax */
+	__NR_sigreturn,
+	0x80cd,		/* int $0x80 */
+};
+
+static const struct {
+	u8  movl;
+	u32 val;
+	u16 int80;
+	u8  pad;
+} __attribute__((packed)) rt_retcode = {
+	0xb8,		/* movl $..., %eax */
+	__NR_rt_sigreturn,
+	0x80cd,		/* int $0x80 */
+	0
+};
+
+/*
  * Determine which stack to use..
  */
 static inline void __user *
@@ -328,6 +246,8 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
 	if (used_math()) {
 		sp = sp - sig_xstate_size;
 		*fpstate = (struct _fpstate *) sp;
+		if (save_i387_xstate(*fpstate) < 0)
+			return (void __user *)-1L;
 	}
 
 	sp -= frame_size;
@@ -383,9 +303,7 @@ __setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
 	 * reasons and because gdb uses it as a signature to notice
 	 * signal handler stack frames.
 	 */
-	err |= __put_user(0xb858, (short __user *)(frame->retcode+0));
-	err |= __put_user(__NR_sigreturn, (int __user *)(frame->retcode+2));
-	err |= __put_user(0x80cd, (short __user *)(frame->retcode+6));
+	err |= __put_user(*((u64 *)&retcode), (u64 *)frame->retcode);
 
 	if (err)
 		return -EFAULT;
@@ -454,9 +372,7 @@ static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	 * reasons and because gdb uses it as a signature to notice
 	 * signal handler stack frames.
 	 */
-	err |= __put_user(0xb8, (char __user *)(frame->retcode+0));
-	err |= __put_user(__NR_rt_sigreturn, (int __user *)(frame->retcode+1));
-	err |= __put_user(0x80cd, (short __user *)(frame->retcode+5));
+	err |= __put_user(*((u64 *)&rt_retcode), (u64 *)frame->retcode);
 
 	if (err)
 		return -EFAULT;
@@ -475,23 +391,293 @@ static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 
 	return 0;
 }
+#else /* !CONFIG_X86_32 */
+/*
+ * Determine which stack to use..
+ */
+static void __user *
+get_stack(struct k_sigaction *ka, unsigned long sp, unsigned long size)
+{
+	/* Default to using normal stack - redzone*/
+	sp -= 128;
+
+	/* This is the X/Open sanctioned signal stack switching.  */
+	if (ka->sa.sa_flags & SA_ONSTACK) {
+		if (sas_ss_flags(sp) == 0)
+			sp = current->sas_ss_sp + current->sas_ss_size;
+	}
+
+	return (void __user *)round_down(sp - size, 64);
+}
+
+static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+			    sigset_t *set, struct pt_regs *regs)
+{
+	struct rt_sigframe __user *frame;
+	void __user *fp = NULL;
+	int err = 0;
+	struct task_struct *me = current;
+
+	if (used_math()) {
+		fp = get_stack(ka, regs->sp, sig_xstate_size);
+		frame = (void __user *)round_down(
+			(unsigned long)fp - sizeof(struct rt_sigframe), 16) - 8;
+
+		if (save_i387_xstate(fp) < 0)
+			return -EFAULT;
+	} else
+		frame = get_stack(ka, regs->sp, sizeof(struct rt_sigframe)) - 8;
+
+	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
+		return -EFAULT;
+
+	if (ka->sa.sa_flags & SA_SIGINFO) {
+		if (copy_siginfo_to_user(&frame->info, info))
+			return -EFAULT;
+	}
+
+	/* Create the ucontext.  */
+	if (cpu_has_xsave)
+		err |= __put_user(UC_FP_XSTATE, &frame->uc.uc_flags);
+	else
+		err |= __put_user(0, &frame->uc.uc_flags);
+	err |= __put_user(0, &frame->uc.uc_link);
+	err |= __put_user(me->sas_ss_sp, &frame->uc.uc_stack.ss_sp);
+	err |= __put_user(sas_ss_flags(regs->sp),
+			  &frame->uc.uc_stack.ss_flags);
+	err |= __put_user(me->sas_ss_size, &frame->uc.uc_stack.ss_size);
+	err |= setup_sigcontext(&frame->uc.uc_mcontext, fp, regs, set->sig[0]);
+	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
+
+	/* Set up to return from userspace.  If provided, use a stub
+	   already in userspace.  */
+	/* x86-64 should always use SA_RESTORER. */
+	if (ka->sa.sa_flags & SA_RESTORER) {
+		err |= __put_user(ka->sa.sa_restorer, &frame->pretcode);
+	} else {
+		/* could use a vstub here */
+		return -EFAULT;
+	}
+
+	if (err)
+		return -EFAULT;
+
+	/* Set up registers for signal handler */
+	regs->di = sig;
+	/* In case the signal handler was declared without prototypes */
+	regs->ax = 0;
+
+	/* This also works for non SA_SIGINFO handlers because they expect the
+	   next argument after the signal number on the stack. */
+	regs->si = (unsigned long)&frame->info;
+	regs->dx = (unsigned long)&frame->uc;
+	regs->ip = (unsigned long) ka->sa.sa_handler;
+
+	regs->sp = (unsigned long)frame;
+
+	/* Set up the CS register to run signal handlers in 64-bit mode,
+	   even if the handler happens to be interrupting 32-bit code. */
+	regs->cs = __USER_CS;
+
+	return 0;
+}
+#endif /* CONFIG_X86_32 */
+
+#ifdef CONFIG_X86_32
+/*
+ * Atomically swap in the new signal mask, and wait for a signal.
+ */
+asmlinkage int
+sys_sigsuspend(int history0, int history1, old_sigset_t mask)
+{
+	mask &= _BLOCKABLE;
+	spin_lock_irq(&current->sighand->siglock);
+	current->saved_sigmask = current->blocked;
+	siginitset(&current->blocked, mask);
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	current->state = TASK_INTERRUPTIBLE;
+	schedule();
+	set_restore_sigmask();
+
+	return -ERESTARTNOHAND;
+}
+
+asmlinkage int
+sys_sigaction(int sig, const struct old_sigaction __user *act,
+	      struct old_sigaction __user *oact)
+{
+	struct k_sigaction new_ka, old_ka;
+	int ret;
+
+	if (act) {
+		old_sigset_t mask;
+
+		if (!access_ok(VERIFY_READ, act, sizeof(*act)) ||
+		    __get_user(new_ka.sa.sa_handler, &act->sa_handler) ||
+		    __get_user(new_ka.sa.sa_restorer, &act->sa_restorer))
+			return -EFAULT;
+
+		__get_user(new_ka.sa.sa_flags, &act->sa_flags);
+		__get_user(mask, &act->sa_mask);
+		siginitset(&new_ka.sa.sa_mask, mask);
+	}
+
+	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
+
+	if (!ret && oact) {
+		if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) ||
+		    __put_user(old_ka.sa.sa_handler, &oact->sa_handler) ||
+		    __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer))
+			return -EFAULT;
+
+		__put_user(old_ka.sa.sa_flags, &oact->sa_flags);
+		__put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask);
+	}
+
+	return ret;
+}
+#endif /* CONFIG_X86_32 */
+
+#ifdef CONFIG_X86_32
+asmlinkage int sys_sigaltstack(unsigned long bx)
+{
+	/*
+	 * This is needed to make gcc realize it doesn't own the
+	 * "struct pt_regs"
+	 */
+	struct pt_regs *regs = (struct pt_regs *)&bx;
+	const stack_t __user *uss = (const stack_t __user *)bx;
+	stack_t __user *uoss = (stack_t __user *)regs->cx;
+
+	return do_sigaltstack(uss, uoss, regs->sp);
+}
+#else /* !CONFIG_X86_32 */
+asmlinkage long
+sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
+		struct pt_regs *regs)
+{
+	return do_sigaltstack(uss, uoss, regs->sp);
+}
+#endif /* CONFIG_X86_32 */
+
+/*
+ * Do a signal return; undo the signal stack.
+ */
+#ifdef CONFIG_X86_32
+asmlinkage unsigned long sys_sigreturn(unsigned long __unused)
+{
+	struct sigframe __user *frame;
+	struct pt_regs *regs;
+	unsigned long ax;
+	sigset_t set;
+
+	regs = (struct pt_regs *) &__unused;
+	frame = (struct sigframe __user *)(regs->sp - 8);
+
+	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
+		goto badframe;
+	if (__get_user(set.sig[0], &frame->sc.oldmask) || (_NSIG_WORDS > 1
+		&& __copy_from_user(&set.sig[1], &frame->extramask,
+				    sizeof(frame->extramask))))
+		goto badframe;
+
+	sigdelsetmask(&set, ~_BLOCKABLE);
+	spin_lock_irq(&current->sighand->siglock);
+	current->blocked = set;
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	if (restore_sigcontext(regs, &frame->sc, &ax))
+		goto badframe;
+	return ax;
+
+badframe:
+	signal_fault(regs, frame, "sigreturn");
+
+	return 0;
+}
+#endif /* CONFIG_X86_32 */
+
+static long do_rt_sigreturn(struct pt_regs *regs)
+{
+	struct rt_sigframe __user *frame;
+	unsigned long ax;
+	sigset_t set;
+
+	frame = (struct rt_sigframe __user *)(regs->sp - sizeof(long));
+	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
+		goto badframe;
+	if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
+		goto badframe;
+
+	sigdelsetmask(&set, ~_BLOCKABLE);
+	spin_lock_irq(&current->sighand->siglock);
+	current->blocked = set;
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ax))
+		goto badframe;
+
+	if (do_sigaltstack(&frame->uc.uc_stack, NULL, regs->sp) == -EFAULT)
+		goto badframe;
+
+	return ax;
+
+badframe:
+	signal_fault(regs, frame, "rt_sigreturn");
+	return 0;
+}
+
+#ifdef CONFIG_X86_32
+asmlinkage int sys_rt_sigreturn(struct pt_regs regs)
+{
+	return do_rt_sigreturn(&regs);
+}
+#else /* !CONFIG_X86_32 */
+asmlinkage long sys_rt_sigreturn(struct pt_regs *regs)
+{
+	return do_rt_sigreturn(regs);
+}
+#endif /* CONFIG_X86_32 */
 
 /*
  * OK, we're invoking a handler:
  */
 static int signr_convert(int sig)
 {
+#ifdef CONFIG_X86_32
 	struct thread_info *info = current_thread_info();
 
 	if (info->exec_domain && info->exec_domain->signal_invmap && sig < 32)
 		return info->exec_domain->signal_invmap[sig];
+#endif /* CONFIG_X86_32 */
 	return sig;
 }
 
+#ifdef CONFIG_X86_32
+
 #define is_ia32	1
 #define ia32_setup_frame	__setup_frame
 #define ia32_setup_rt_frame	__setup_rt_frame
 
+#else /* !CONFIG_X86_32 */
+
+#ifdef CONFIG_IA32_EMULATION
+#define is_ia32	test_thread_flag(TIF_IA32)
+#else /* !CONFIG_IA32_EMULATION */
+#define is_ia32	0
+#endif /* CONFIG_IA32_EMULATION */
+
+int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+		sigset_t *set, struct pt_regs *regs);
+int ia32_setup_frame(int sig, struct k_sigaction *ka,
+		sigset_t *set, struct pt_regs *regs);
+
+#endif /* CONFIG_X86_32 */
+
 static int
 setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	       sigset_t *set, struct pt_regs *regs)
@@ -592,7 +778,13 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	return 0;
 }
 
+#ifdef CONFIG_X86_32
 #define NR_restart_syscall	__NR_restart_syscall
+#else /* !CONFIG_X86_32 */
+#define NR_restart_syscall	\
+	test_thread_flag(TIF_IA32) ? __NR_ia32_restart_syscall : __NR_restart_syscall
+#endif /* CONFIG_X86_32 */
+
 /*
  * Note that 'init' is a special process: it doesn't get signals it doesn't
  * want to handle. Thus you cannot kill init even with a SIGKILL even by
@@ -704,8 +896,9 @@ void signal_fault(struct pt_regs *regs, void __user *frame, char *where)
 	struct task_struct *me = current;
 
 	if (show_unhandled_signals && printk_ratelimit()) {
-		printk(KERN_INFO
+		printk("%s"
 		       "%s[%d] bad frame in %s frame:%p ip:%lx sp:%lx orax:%lx",
+		       task_pid_nr(current) > 1 ? KERN_INFO : KERN_EMERG,
 		       me->comm, me->pid, where, frame,
 		       regs->ip, regs->sp, regs->orig_ax);
 		print_vma_addr(" in ", regs->ip);
diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c
deleted file mode 100644
index a5c9627f4db9..000000000000
--- a/arch/x86/kernel/signal_64.c
+++ /dev/null
@@ -1,516 +0,0 @@
-/*
- *  Copyright (C) 1991, 1992  Linus Torvalds
- *  Copyright (C) 2000, 2001, 2002 Andi Kleen SuSE Labs
- *
- *  1997-11-28  Modified for POSIX.1b signals by Richard Henderson
- *  2000-06-20  Pentium III FXSR, SSE support by Gareth Hughes
- *  2000-2002   x86-64 support by Andi Kleen
- */
-
-#include <linux/sched.h>
-#include <linux/mm.h>
-#include <linux/smp.h>
-#include <linux/kernel.h>
-#include <linux/signal.h>
-#include <linux/errno.h>
-#include <linux/wait.h>
-#include <linux/ptrace.h>
-#include <linux/tracehook.h>
-#include <linux/unistd.h>
-#include <linux/stddef.h>
-#include <linux/personality.h>
-#include <linux/compiler.h>
-#include <linux/uaccess.h>
-
-#include <asm/processor.h>
-#include <asm/ucontext.h>
-#include <asm/i387.h>
-#include <asm/proto.h>
-#include <asm/ia32_unistd.h>
-#include <asm/mce.h>
-#include <asm/syscall.h>
-#include <asm/syscalls.h>
-#include "sigframe.h"
-
-#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
-
-#define __FIX_EFLAGS	(X86_EFLAGS_AC | X86_EFLAGS_OF | \
-			 X86_EFLAGS_DF | X86_EFLAGS_TF | X86_EFLAGS_SF | \
-			 X86_EFLAGS_ZF | X86_EFLAGS_AF | X86_EFLAGS_PF | \
-			 X86_EFLAGS_CF)
-
-#ifdef CONFIG_X86_32
-# define FIX_EFLAGS	(__FIX_EFLAGS | X86_EFLAGS_RF)
-#else
-# define FIX_EFLAGS	__FIX_EFLAGS
-#endif
-
-asmlinkage long
-sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
-		struct pt_regs *regs)
-{
-	return do_sigaltstack(uss, uoss, regs->sp);
-}
-
-#define COPY(x)			{		\
-	err |= __get_user(regs->x, &sc->x);	\
-}
-
-#define COPY_SEG_STRICT(seg)	{			\
-		unsigned short tmp;			\
-		err |= __get_user(tmp, &sc->seg);	\
-		regs->seg = tmp | 3;			\
-}
-
-/*
- * Do a signal return; undo the signal stack.
- */
-static int
-restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc,
-		   unsigned long *pax)
-{
-	void __user *buf;
-	unsigned int tmpflags;
-	unsigned int err = 0;
-
-	/* Always make any pending restarted system calls return -EINTR */
-	current_thread_info()->restart_block.fn = do_no_restart_syscall;
-
-	COPY(di); COPY(si); COPY(bp); COPY(sp); COPY(bx);
-	COPY(dx); COPY(cx); COPY(ip);
-	COPY(r8);
-	COPY(r9);
-	COPY(r10);
-	COPY(r11);
-	COPY(r12);
-	COPY(r13);
-	COPY(r14);
-	COPY(r15);
-
-	/* Kernel saves and restores only the CS segment register on signals,
-	 * which is the bare minimum needed to allow mixed 32/64-bit code.
-	 * App's signal handler can save/restore other segments if needed. */
-	COPY_SEG_STRICT(cs);
-
-	err |= __get_user(tmpflags, &sc->flags);
-	regs->flags = (regs->flags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS);
-	regs->orig_ax = -1;		/* disable syscall checks */
-
-	err |= __get_user(buf, &sc->fpstate);
-	err |= restore_i387_xstate(buf);
-
-	err |= __get_user(*pax, &sc->ax);
-	return err;
-}
-
-static long do_rt_sigreturn(struct pt_regs *regs)
-{
-	struct rt_sigframe __user *frame;
-	unsigned long ax;
-	sigset_t set;
-
-	frame = (struct rt_sigframe __user *)(regs->sp - sizeof(long));
-	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
-		goto badframe;
-	if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
-		goto badframe;
-
-	sigdelsetmask(&set, ~_BLOCKABLE);
-	spin_lock_irq(&current->sighand->siglock);
-	current->blocked = set;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ax))
-		goto badframe;
-
-	if (do_sigaltstack(&frame->uc.uc_stack, NULL, regs->sp) == -EFAULT)
-		goto badframe;
-
-	return ax;
-
-badframe:
-	signal_fault(regs, frame, "rt_sigreturn");
-	return 0;
-}
-
-asmlinkage long sys_rt_sigreturn(struct pt_regs *regs)
-{
-	return do_rt_sigreturn(regs);
-}
-
-/*
- * Set up a signal frame.
- */
-
-static inline int
-setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs,
-		unsigned long mask, struct task_struct *me)
-{
-	int err = 0;
-
-	err |= __put_user(regs->cs, &sc->cs);
-	err |= __put_user(0, &sc->gs);
-	err |= __put_user(0, &sc->fs);
-
-	err |= __put_user(regs->di, &sc->di);
-	err |= __put_user(regs->si, &sc->si);
-	err |= __put_user(regs->bp, &sc->bp);
-	err |= __put_user(regs->sp, &sc->sp);
-	err |= __put_user(regs->bx, &sc->bx);
-	err |= __put_user(regs->dx, &sc->dx);
-	err |= __put_user(regs->cx, &sc->cx);
-	err |= __put_user(regs->ax, &sc->ax);
-	err |= __put_user(regs->r8, &sc->r8);
-	err |= __put_user(regs->r9, &sc->r9);
-	err |= __put_user(regs->r10, &sc->r10);
-	err |= __put_user(regs->r11, &sc->r11);
-	err |= __put_user(regs->r12, &sc->r12);
-	err |= __put_user(regs->r13, &sc->r13);
-	err |= __put_user(regs->r14, &sc->r14);
-	err |= __put_user(regs->r15, &sc->r15);
-	err |= __put_user(me->thread.trap_no, &sc->trapno);
-	err |= __put_user(me->thread.error_code, &sc->err);
-	err |= __put_user(regs->ip, &sc->ip);
-	err |= __put_user(regs->flags, &sc->flags);
-	err |= __put_user(mask, &sc->oldmask);
-	err |= __put_user(me->thread.cr2, &sc->cr2);
-
-	return err;
-}
-
-/*
- * Determine which stack to use..
- */
-
-static void __user *
-get_stack(struct k_sigaction *ka, struct pt_regs *regs, unsigned long size)
-{
-	unsigned long sp;
-
-	/* Default to using normal stack - redzone*/
-	sp = regs->sp - 128;
-
-	/* This is the X/Open sanctioned signal stack switching.  */
-	if (ka->sa.sa_flags & SA_ONSTACK) {
-		if (sas_ss_flags(sp) == 0)
-			sp = current->sas_ss_sp + current->sas_ss_size;
-	}
-
-	return (void __user *)round_down(sp - size, 64);
-}
-
-static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
-			    sigset_t *set, struct pt_regs *regs)
-{
-	struct rt_sigframe __user *frame;
-	void __user *fp = NULL;
-	int err = 0;
-	struct task_struct *me = current;
-
-	if (used_math()) {
-		fp = get_stack(ka, regs, sig_xstate_size);
-		frame = (void __user *)round_down(
-			(unsigned long)fp - sizeof(struct rt_sigframe), 16) - 8;
-
-		if (save_i387_xstate(fp) < 0)
-			return -EFAULT;
-	} else
-		frame = get_stack(ka, regs, sizeof(struct rt_sigframe)) - 8;
-
-	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
-		return -EFAULT;
-
-	if (ka->sa.sa_flags & SA_SIGINFO) {
-		if (copy_siginfo_to_user(&frame->info, info))
-			return -EFAULT;
-	}
-
-	/* Create the ucontext.  */
-	if (cpu_has_xsave)
-		err |= __put_user(UC_FP_XSTATE, &frame->uc.uc_flags);
-	else
-		err |= __put_user(0, &frame->uc.uc_flags);
-	err |= __put_user(0, &frame->uc.uc_link);
-	err |= __put_user(me->sas_ss_sp, &frame->uc.uc_stack.ss_sp);
-	err |= __put_user(sas_ss_flags(regs->sp),
-			  &frame->uc.uc_stack.ss_flags);
-	err |= __put_user(me->sas_ss_size, &frame->uc.uc_stack.ss_size);
-	err |= setup_sigcontext(&frame->uc.uc_mcontext, regs, set->sig[0], me);
-	err |= __put_user(fp, &frame->uc.uc_mcontext.fpstate);
-	if (sizeof(*set) == 16) {
-		__put_user(set->sig[0], &frame->uc.uc_sigmask.sig[0]);
-		__put_user(set->sig[1], &frame->uc.uc_sigmask.sig[1]);
-	} else
-		err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
-
-	/* Set up to return from userspace.  If provided, use a stub
-	   already in userspace.  */
-	/* x86-64 should always use SA_RESTORER. */
-	if (ka->sa.sa_flags & SA_RESTORER) {
-		err |= __put_user(ka->sa.sa_restorer, &frame->pretcode);
-	} else {
-		/* could use a vstub here */
-		return -EFAULT;
-	}
-
-	if (err)
-		return -EFAULT;
-
-	/* Set up registers for signal handler */
-	regs->di = sig;
-	/* In case the signal handler was declared without prototypes */
-	regs->ax = 0;
-
-	/* This also works for non SA_SIGINFO handlers because they expect the
-	   next argument after the signal number on the stack. */
-	regs->si = (unsigned long)&frame->info;
-	regs->dx = (unsigned long)&frame->uc;
-	regs->ip = (unsigned long) ka->sa.sa_handler;
-
-	regs->sp = (unsigned long)frame;
-
-	/* Set up the CS register to run signal handlers in 64-bit mode,
-	   even if the handler happens to be interrupting 32-bit code. */
-	regs->cs = __USER_CS;
-
-	return 0;
-}
-
-/*
- * OK, we're invoking a handler
- */
-static int signr_convert(int sig)
-{
-	return sig;
-}
-
-#ifdef CONFIG_IA32_EMULATION
-#define is_ia32	test_thread_flag(TIF_IA32)
-#else
-#define is_ia32	0
-#endif
-
-static int
-setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
-	       sigset_t *set, struct pt_regs *regs)
-{
-	int usig = signr_convert(sig);
-	int ret;
-
-	/* Set up the stack frame */
-	if (is_ia32) {
-		if (ka->sa.sa_flags & SA_SIGINFO)
-			ret = ia32_setup_rt_frame(usig, ka, info, set, regs);
-		else
-			ret = ia32_setup_frame(usig, ka, set, regs);
-	} else
-		ret = __setup_rt_frame(sig, ka, info, set, regs);
-
-	if (ret) {
-		force_sigsegv(sig, current);
-		return -EFAULT;
-	}
-
-	return ret;
-}
-
-static int
-handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
-	      sigset_t *oldset, struct pt_regs *regs)
-{
-	int ret;
-
-	/* Are we from a system call? */
-	if (syscall_get_nr(current, regs) >= 0) {
-		/* If so, check system call restarting.. */
-		switch (syscall_get_error(current, regs)) {
-		case -ERESTART_RESTARTBLOCK:
-		case -ERESTARTNOHAND:
-			regs->ax = -EINTR;
-			break;
-
-		case -ERESTARTSYS:
-			if (!(ka->sa.sa_flags & SA_RESTART)) {
-				regs->ax = -EINTR;
-				break;
-			}
-		/* fallthrough */
-		case -ERESTARTNOINTR:
-			regs->ax = regs->orig_ax;
-			regs->ip -= 2;
-			break;
-		}
-	}
-
-	/*
-	 * If TF is set due to a debugger (TIF_FORCED_TF), clear the TF
-	 * flag so that register information in the sigcontext is correct.
-	 */
-	if (unlikely(regs->flags & X86_EFLAGS_TF) &&
-	    likely(test_and_clear_thread_flag(TIF_FORCED_TF)))
-		regs->flags &= ~X86_EFLAGS_TF;
-
-	ret = setup_rt_frame(sig, ka, info, oldset, regs);
-
-	if (ret)
-		return ret;
-
-#ifdef CONFIG_X86_64
-	/*
-	 * This has nothing to do with segment registers,
-	 * despite the name.  This magic affects uaccess.h
-	 * macros' behavior.  Reset it to the normal setting.
-	 */
-	set_fs(USER_DS);
-#endif
-
-	/*
-	 * Clear the direction flag as per the ABI for function entry.
-	 */
-	regs->flags &= ~X86_EFLAGS_DF;
-
-	/*
-	 * Clear TF when entering the signal handler, but
-	 * notify any tracer that was single-stepping it.
-	 * The tracer may want to single-step inside the
-	 * handler too.
-	 */
-	regs->flags &= ~X86_EFLAGS_TF;
-
-	spin_lock_irq(&current->sighand->siglock);
-	sigorsets(&current->blocked, &current->blocked, &ka->sa.sa_mask);
-	if (!(ka->sa.sa_flags & SA_NODEFER))
-		sigaddset(&current->blocked, sig);
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	tracehook_signal_handler(sig, info, ka, regs,
-				 test_thread_flag(TIF_SINGLESTEP));
-
-	return 0;
-}
-
-#define NR_restart_syscall	\
-	test_thread_flag(TIF_IA32) ? __NR_ia32_restart_syscall : __NR_restart_syscall
-/*
- * Note that 'init' is a special process: it doesn't get signals it doesn't
- * want to handle. Thus you cannot kill init even with a SIGKILL even by
- * mistake.
- */
-static void do_signal(struct pt_regs *regs)
-{
-	struct k_sigaction ka;
-	siginfo_t info;
-	int signr;
-	sigset_t *oldset;
-
-	/*
-	 * We want the common case to go fast, which is why we may in certain
-	 * cases get here from kernel mode. Just return without doing anything
-	 * if so.
-	 * X86_32: vm86 regs switched out by assembly code before reaching
-	 * here, so testing against kernel CS suffices.
-	 */
-	if (!user_mode(regs))
-		return;
-
-	if (current_thread_info()->status & TS_RESTORE_SIGMASK)
-		oldset = &current->saved_sigmask;
-	else
-		oldset = &current->blocked;
-
-	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
-	if (signr > 0) {
-		/*
-		 * Re-enable any watchpoints before delivering the
-		 * signal to user space. The processor register will
-		 * have been cleared if the watchpoint triggered
-		 * inside the kernel.
-		 */
-		if (current->thread.debugreg7)
-			set_debugreg(current->thread.debugreg7, 7);
-
-		/* Whee! Actually deliver the signal.  */
-		if (handle_signal(signr, &info, &ka, oldset, regs) == 0) {
-			/*
-			 * A signal was successfully delivered; the saved
-			 * sigmask will have been stored in the signal frame,
-			 * and will be restored by sigreturn, so we can simply
-			 * clear the TS_RESTORE_SIGMASK flag.
-			 */
-			current_thread_info()->status &= ~TS_RESTORE_SIGMASK;
-		}
-		return;
-	}
-
-	/* Did we come from a system call? */
-	if (syscall_get_nr(current, regs) >= 0) {
-		/* Restart the system call - no handlers present */
-		switch (syscall_get_error(current, regs)) {
-		case -ERESTARTNOHAND:
-		case -ERESTARTSYS:
-		case -ERESTARTNOINTR:
-			regs->ax = regs->orig_ax;
-			regs->ip -= 2;
-			break;
-
-		case -ERESTART_RESTARTBLOCK:
-			regs->ax = NR_restart_syscall;
-			regs->ip -= 2;
-			break;
-		}
-	}
-
-	/*
-	 * If there's no signal to deliver, we just put the saved sigmask
-	 * back.
-	 */
-	if (current_thread_info()->status & TS_RESTORE_SIGMASK) {
-		current_thread_info()->status &= ~TS_RESTORE_SIGMASK;
-		sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
-	}
-}
-
-/*
- * notification of userspace execution resumption
- * - triggered by the TIF_WORK_MASK flags
- */
-void
-do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags)
-{
-#if defined(CONFIG_X86_64) && defined(CONFIG_X86_MCE)
-	/* notify userspace of pending MCEs */
-	if (thread_info_flags & _TIF_MCE_NOTIFY)
-		mce_notify_user();
-#endif /* CONFIG_X86_64 && CONFIG_X86_MCE */
-
-	/* deal with pending signal delivery */
-	if (thread_info_flags & _TIF_SIGPENDING)
-		do_signal(regs);
-
-	if (thread_info_flags & _TIF_NOTIFY_RESUME) {
-		clear_thread_flag(TIF_NOTIFY_RESUME);
-		tracehook_notify_resume(regs);
-	}
-
-#ifdef CONFIG_X86_32
-	clear_thread_flag(TIF_IRET);
-#endif /* CONFIG_X86_32 */
-}
-
-void signal_fault(struct pt_regs *regs, void __user *frame, char *where)
-{
-	struct task_struct *me = current;
-
-	if (show_unhandled_signals && printk_ratelimit()) {
-		printk(KERN_INFO
-		       "%s[%d] bad frame in %s frame:%p ip:%lx sp:%lx orax:%lx",
-		       me->comm, me->pid, where, frame,
-		       regs->ip, regs->sp, regs->orig_ax);
-		print_vma_addr(" in ", regs->ip);
-		printk(KERN_CONT "\n");
-	}
-
-	force_sig(SIGSEGV, me);
-}
diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c
index 49ed667b06f3..beea2649a240 100644
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
@@ -165,11 +165,7 @@ static void native_smp_send_stop(void)
 void smp_reschedule_interrupt(struct pt_regs *regs)
 {
 	ack_APIC_irq();
-#ifdef CONFIG_X86_32
-	__get_cpu_var(irq_stat).irq_resched_count++;
-#else
-	add_pda(irq_resched_count, 1);
-#endif
+	inc_irq_stat(irq_resched_count);
 }
 
 void smp_call_function_interrupt(struct pt_regs *regs)
@@ -177,11 +173,7 @@ void smp_call_function_interrupt(struct pt_regs *regs)
 	ack_APIC_irq();
 	irq_enter();
 	generic_smp_call_function_interrupt();
-#ifdef CONFIG_X86_32
-	__get_cpu_var(irq_stat).irq_call_count++;
-#else
-	add_pda(irq_call_count, 1);
-#endif
+	inc_irq_stat(irq_call_count);
 	irq_exit();
 }
 
@@ -190,11 +182,7 @@ void smp_call_function_single_interrupt(struct pt_regs *regs)
 	ack_APIC_irq();
 	irq_enter();
 	generic_smp_call_function_single_interrupt();
-#ifdef CONFIG_X86_32
-	__get_cpu_var(irq_stat).irq_call_count++;
-#else
-	add_pda(irq_call_count, 1);
-#endif
+	inc_irq_stat(irq_call_count);
 	irq_exit();
 }
 
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 1a9941b11150..9e177a4077ee 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -282,7 +282,7 @@ static int __cpuinitdata unsafe_smp;
 /*
  * Activate a secondary processor.
  */
-static void __cpuinit start_secondary(void *unused)
+notrace static void __cpuinit start_secondary(void *unused)
 {
 	/*
 	 * Don't put *anything* before cpu_init(), SMP booting is too
@@ -496,7 +496,7 @@ void __cpuinit set_cpu_sibling_map(int cpu)
 }
 
 /* maps the cpu to the sched domain representing multi-core */
-cpumask_t cpu_coregroup_map(int cpu)
+const struct cpumask *cpu_coregroup_mask(int cpu)
 {
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
 	/*
@@ -504,9 +504,14 @@ cpumask_t cpu_coregroup_map(int cpu)
 	 * And for power savings, we return cpu_core_map
 	 */
 	if (sched_mc_power_savings || sched_smt_power_savings)
-		return per_cpu(cpu_core_map, cpu);
+		return &per_cpu(cpu_core_map, cpu);
 	else
-		return c->llc_shared_map;
+		return &c->llc_shared_map;
+}
+
+cpumask_t cpu_coregroup_map(int cpu)
+{
+	return *cpu_coregroup_mask(cpu);
 }
 
 static void impress_friends(void)
@@ -1075,8 +1080,10 @@ static int __init smp_sanity_check(unsigned max_cpus)
 #endif
 
 	if (!physid_isset(hard_smp_processor_id(), phys_cpu_present_map)) {
-		printk(KERN_WARNING "weird, boot CPU (#%d) not listed"
-				    "by the BIOS.\n", hard_smp_processor_id());
+		printk(KERN_WARNING
+			"weird, boot CPU (#%d) not listed by the BIOS.\n",
+			hard_smp_processor_id());
+
 		physid_set(hard_smp_processor_id(), phys_cpu_present_map);
 	}
 
diff --git a/arch/x86/kernel/time_32.c b/arch/x86/kernel/time_32.c
index 77b400f06ea2..65309e4cb1c0 100644
--- a/arch/x86/kernel/time_32.c
+++ b/arch/x86/kernel/time_32.c
@@ -75,7 +75,7 @@ EXPORT_SYMBOL(profile_pc);
 irqreturn_t timer_interrupt(int irq, void *dev_id)
 {
 	/* Keep nmi watchdog up to date */
-	per_cpu(irq_stat, smp_processor_id()).irq0_irqs++;
+	inc_irq_stat(irq0_irqs);
 
 #ifdef CONFIG_X86_IO_APIC
 	if (timer_ack) {
diff --git a/arch/x86/kernel/time_64.c b/arch/x86/kernel/time_64.c
index cb19d650c216..891e7a7c4334 100644
--- a/arch/x86/kernel/time_64.c
+++ b/arch/x86/kernel/time_64.c
@@ -49,9 +49,9 @@ unsigned long profile_pc(struct pt_regs *regs)
 }
 EXPORT_SYMBOL(profile_pc);
 
-irqreturn_t timer_interrupt(int irq, void *dev_id)
+static irqreturn_t timer_interrupt(int irq, void *dev_id)
 {
-	add_pda(irq0_irqs, 1);
+	inc_irq_stat(irq0_irqs);
 
 	global_clock_event->event_handler(global_clock_event);
 
@@ -80,6 +80,8 @@ unsigned long __init calibrate_cpu(void)
 			break;
 	no_ctr_free = (i == 4);
 	if (no_ctr_free) {
+		WARN(1, KERN_WARNING "Warning: AMD perfctrs busy ... "
+		     "cpu_khz value may be incorrect.\n");
 		i = 3;
 		rdmsrl(MSR_K7_EVNTSEL3, evntsel3);
 		wrmsrl(MSR_K7_EVNTSEL3, 0);
diff --git a/arch/x86/kernel/tlb_32.c b/arch/x86/kernel/tlb_32.c
index 174ea90d1cbd..ce5054642247 100644
--- a/arch/x86/kernel/tlb_32.c
+++ b/arch/x86/kernel/tlb_32.c
@@ -34,9 +34,8 @@ static DEFINE_SPINLOCK(tlbstate_lock);
  */
 void leave_mm(int cpu)
 {
-	if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK)
-		BUG();
-	cpu_clear(cpu, per_cpu(cpu_tlbstate, cpu).active_mm->cpu_vm_mask);
+	BUG_ON(x86_read_percpu(cpu_tlbstate.state) == TLBSTATE_OK);
+	cpu_clear(cpu, x86_read_percpu(cpu_tlbstate.active_mm)->cpu_vm_mask);
 	load_cr3(swapper_pg_dir);
 }
 EXPORT_SYMBOL_GPL(leave_mm);
@@ -104,8 +103,8 @@ void smp_invalidate_interrupt(struct pt_regs *regs)
 		 * BUG();
 		 */
 
-	if (flush_mm == per_cpu(cpu_tlbstate, cpu).active_mm) {
-		if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK) {
+	if (flush_mm == x86_read_percpu(cpu_tlbstate.active_mm)) {
+		if (x86_read_percpu(cpu_tlbstate.state) == TLBSTATE_OK) {
 			if (flush_va == TLB_FLUSH_ALL)
 				local_flush_tlb();
 			else
@@ -119,7 +118,7 @@ void smp_invalidate_interrupt(struct pt_regs *regs)
 	smp_mb__after_clear_bit();
 out:
 	put_cpu_no_resched();
-	__get_cpu_var(irq_stat).irq_tlb_count++;
+	inc_irq_stat(irq_tlb_count);
 }
 
 void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm,
@@ -238,7 +237,7 @@ static void do_flush_tlb_all(void *info)
 	unsigned long cpu = smp_processor_id();
 
 	__flush_tlb_all();
-	if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_LAZY)
+	if (x86_read_percpu(cpu_tlbstate.state) == TLBSTATE_LAZY)
 		leave_mm(cpu);
 }
 
diff --git a/arch/x86/kernel/tlb_64.c b/arch/x86/kernel/tlb_64.c
index de6f1bda0c50..f8be6f1d2e48 100644
--- a/arch/x86/kernel/tlb_64.c
+++ b/arch/x86/kernel/tlb_64.c
@@ -154,7 +154,7 @@ asmlinkage void smp_invalidate_interrupt(struct pt_regs *regs)
 out:
 	ack_APIC_irq();
 	cpu_clear(cpu, f->flush_cpumask);
-	add_pda(irq_tlb_count, 1);
+	inc_irq_stat(irq_tlb_count);
 }
 
 void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm,
diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c
index 04431f34fd16..6a00e5faaa74 100644
--- a/arch/x86/kernel/tlb_uv.c
+++ b/arch/x86/kernel/tlb_uv.c
@@ -566,14 +566,10 @@ static int __init uv_ptc_init(void)
 	if (!is_uv_system())
 		return 0;
 
-	if (!proc_mkdir("sgi_uv", NULL))
-		return -EINVAL;
-
 	proc_uv_ptc = create_proc_entry(UV_PTC_BASENAME, 0444, NULL);
 	if (!proc_uv_ptc) {
 		printk(KERN_ERR "unable to create %s proc entry\n",
 		       UV_PTC_BASENAME);
-		remove_proc_entry("sgi_uv", NULL);
 		return -EINVAL;
 	}
 	proc_uv_ptc->proc_fops = &proc_uv_ptc_operations;
diff --git a/arch/x86/kernel/trampoline.c b/arch/x86/kernel/trampoline.c
index 1106fac6024d..808031a5ba19 100644
--- a/arch/x86/kernel/trampoline.c
+++ b/arch/x86/kernel/trampoline.c
@@ -1,10 +1,26 @@
 #include <linux/io.h>
 
 #include <asm/trampoline.h>
+#include <asm/e820.h>
 
 /* ready for x86_64 and x86 */
 unsigned char *trampoline_base = __va(TRAMPOLINE_BASE);
 
+void __init reserve_trampoline_memory(void)
+{
+#ifdef CONFIG_X86_32
+	/*
+	 * But first pinch a few for the stack/trampoline stuff
+	 * FIXME: Don't need the extra page at 4K, but need to fix
+	 * trampoline before removing it. (see the GDT stuff)
+	 */
+	reserve_early(PAGE_SIZE, PAGE_SIZE + PAGE_SIZE, "EX TRAMPOLINE");
+#endif
+	/* Has to be in very low memory so we can execute real-mode AP code. */
+	reserve_early(TRAMPOLINE_BASE, TRAMPOLINE_BASE + TRAMPOLINE_SIZE,
+			"TRAMPOLINE");
+}
+
 /*
  * Currently trivial. Write the real->protected mode
  * bootstrap into the page concerned. The caller
@@ -12,7 +28,6 @@ unsigned char *trampoline_base = __va(TRAMPOLINE_BASE);
  */
 unsigned long setup_trampoline(void)
 {
-	memcpy(trampoline_base, trampoline_data,
-	       trampoline_end - trampoline_data);
+	memcpy(trampoline_base, trampoline_data, TRAMPOLINE_SIZE);
 	return virt_to_phys(trampoline_base);
 }
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 4a6dff39a470..2d1f4c7e4052 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -481,11 +481,7 @@ do_nmi(struct pt_regs *regs, long error_code)
 {
 	nmi_enter();
 
-#ifdef CONFIG_X86_32
-	{ int cpu; cpu = smp_processor_id(); ++nmi_count(cpu); }
-#else
-	add_pda(__nmi_count, 1);
-#endif
+	inc_irq_stat(__nmi_count);
 
 	if (!ignore_nmis)
 		default_do_nmi(regs);
@@ -664,7 +660,7 @@ void math_error(void __user *ip)
 {
 	struct task_struct *task;
 	siginfo_t info;
-	unsigned short cwd, swd;
+	unsigned short cwd, swd, err;
 
 	/*
 	 * Save the info for the exception handler and clear the error.
@@ -675,7 +671,6 @@ void math_error(void __user *ip)
 	task->thread.error_code = 0;
 	info.si_signo = SIGFPE;
 	info.si_errno = 0;
-	info.si_code = __SI_FAULT;
 	info.si_addr = ip;
 	/*
 	 * (~cwd & swd) will mask out exceptions that are not set to unmasked
@@ -689,34 +684,31 @@ void math_error(void __user *ip)
 	 */
 	cwd = get_fpu_cwd(task);
 	swd = get_fpu_swd(task);
-	switch (swd & ~cwd & 0x3f) {
-	case 0x000: /* No unmasked exception */
+
+	err = swd & ~cwd & 0x3f;
+
 #ifdef CONFIG_X86_32
+	if (!err)
 		return;
 #endif
-	default: /* Multiple exceptions */
-		break;
-	case 0x001: /* Invalid Op */
+
+	if (err & 0x001) {	/* Invalid op */
 		/*
 		 * swd & 0x240 == 0x040: Stack Underflow
 		 * swd & 0x240 == 0x240: Stack Overflow
 		 * User must clear the SF bit (0x40) if set
 		 */
 		info.si_code = FPE_FLTINV;
-		break;
-	case 0x002: /* Denormalize */
-	case 0x010: /* Underflow */
-		info.si_code = FPE_FLTUND;
-		break;
-	case 0x004: /* Zero Divide */
+	} else if (err & 0x004) { /* Divide by Zero */
 		info.si_code = FPE_FLTDIV;
-		break;
-	case 0x008: /* Overflow */
+	} else if (err & 0x008) { /* Overflow */
 		info.si_code = FPE_FLTOVF;
-		break;
-	case 0x020: /* Precision */
+	} else if (err & 0x012) { /* Denormal, Underflow */
+		info.si_code = FPE_FLTUND;
+	} else if (err & 0x020) { /* Precision */
 		info.si_code = FPE_FLTRES;
-		break;
+	} else {
+		info.si_code = __SI_FAULT|SI_KERNEL; /* WTF? */
 	}
 	force_sig_info(SIGFPE, &info, task);
 }
diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index 424093b157d3..599e58168631 100644
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -15,6 +15,7 @@
 #include <asm/vgtod.h>
 #include <asm/time.h>
 #include <asm/delay.h>
+#include <asm/hypervisor.h>
 
 unsigned int cpu_khz;           /* TSC clocks / usec, not used here */
 EXPORT_SYMBOL(cpu_khz);
@@ -31,6 +32,7 @@ static int tsc_unstable;
    erroneous rdtsc usage on !cpu_has_tsc processors */
 static int tsc_disabled = -1;
 
+static int tsc_clocksource_reliable;
 /*
  * Scheduler clock - returns current time in nanosec units.
  */
@@ -98,6 +100,15 @@ int __init notsc_setup(char *str)
 
 __setup("notsc", notsc_setup);
 
+static int __init tsc_setup(char *str)
+{
+	if (!strcmp(str, "reliable"))
+		tsc_clocksource_reliable = 1;
+	return 1;
+}
+
+__setup("tsc=", tsc_setup);
+
 #define MAX_RETRIES     5
 #define SMI_TRESHOLD    50000
 
@@ -352,9 +363,15 @@ unsigned long native_calibrate_tsc(void)
 {
 	u64 tsc1, tsc2, delta, ref1, ref2;
 	unsigned long tsc_pit_min = ULONG_MAX, tsc_ref_min = ULONG_MAX;
-	unsigned long flags, latch, ms, fast_calibrate;
+	unsigned long flags, latch, ms, fast_calibrate, tsc_khz;
 	int hpet = is_hpet_enabled(), i, loopmin;
 
+	tsc_khz = get_hypervisor_tsc_freq();
+	if (tsc_khz) {
+		printk(KERN_INFO "TSC: Frequency read from the hypervisor\n");
+		return tsc_khz;
+	}
+
 	local_irq_save(flags);
 	fast_calibrate = quick_pit_calibrate();
 	local_irq_restore(flags);
@@ -731,24 +748,21 @@ static struct dmi_system_id __initdata bad_tsc_dmi_table[] = {
 	{}
 };
 
-/*
- * Geode_LX - the OLPC CPU has a possibly a very reliable TSC
- */
+static void __init check_system_tsc_reliable(void)
+{
 #ifdef CONFIG_MGEODE_LX
-/* RTSC counts during suspend */
+	/* RTSC counts during suspend */
 #define RTSC_SUSP 0x100
-
-static void __init check_geode_tsc_reliable(void)
-{
 	unsigned long res_low, res_high;
 
 	rdmsr_safe(MSR_GEODE_BUSCONT_CONF0, &res_low, &res_high);
+	/* Geode_LX - the OLPC CPU has a possibly a very reliable TSC */
 	if (res_low & RTSC_SUSP)
-		clocksource_tsc.flags &= ~CLOCK_SOURCE_MUST_VERIFY;
-}
-#else
-static inline void check_geode_tsc_reliable(void) { }
+		tsc_clocksource_reliable = 1;
 #endif
+	if (boot_cpu_has(X86_FEATURE_TSC_RELIABLE))
+		tsc_clocksource_reliable = 1;
+}
 
 /*
  * Make an educated guess if the TSC is trustworthy and synchronized
@@ -783,6 +797,8 @@ static void __init init_tsc_clocksource(void)
 {
 	clocksource_tsc.mult = clocksource_khz2mult(tsc_khz,
 			clocksource_tsc.shift);
+	if (tsc_clocksource_reliable)
+		clocksource_tsc.flags &= ~CLOCK_SOURCE_MUST_VERIFY;
 	/* lower the rating if we already know its unstable: */
 	if (check_tsc_unstable()) {
 		clocksource_tsc.rating = 0;
@@ -843,7 +859,7 @@ void __init tsc_init(void)
 	if (unsynchronized_tsc())
 		mark_tsc_unstable("TSCs unsynchronized");
 
-	check_geode_tsc_reliable();
+	check_system_tsc_reliable();
 	init_tsc_clocksource();
 }
 
diff --git a/arch/x86/kernel/tsc_sync.c b/arch/x86/kernel/tsc_sync.c
index 1c0dfbca87c1..bf36328f6ef9 100644
--- a/arch/x86/kernel/tsc_sync.c
+++ b/arch/x86/kernel/tsc_sync.c
@@ -112,6 +112,12 @@ void __cpuinit check_tsc_sync_source(int cpu)
 	if (unsynchronized_tsc())
 		return;
 
+	if (boot_cpu_has(X86_FEATURE_TSC_RELIABLE)) {
+		printk(KERN_INFO
+		       "Skipping synchronization checks as TSC is reliable.\n");
+		return;
+	}
+
 	printk(KERN_INFO "checking TSC synchronization [CPU#%d -> CPU#%d]:",
 			  smp_processor_id(), cpu);
 
@@ -165,7 +171,7 @@ void __cpuinit check_tsc_sync_target(void)
 {
 	int cpus = 2;
 
-	if (unsynchronized_tsc())
+	if (unsynchronized_tsc() || boot_cpu_has(X86_FEATURE_TSC_RELIABLE))
 		return;
 
 	/*
diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c
index 22fd6577156a..23206ba16874 100644
--- a/arch/x86/kernel/vmi_32.c
+++ b/arch/x86/kernel/vmi_32.c
@@ -266,109 +266,6 @@ static void vmi_nop(void)
 {
 }
 
-#ifdef CONFIG_DEBUG_PAGE_TYPE
-
-#ifdef CONFIG_X86_PAE
-#define MAX_BOOT_PTS (2048+4+1)
-#else
-#define MAX_BOOT_PTS (1024+1)
-#endif
-
-/*
- * During boot, mem_map is not yet available in paging_init, so stash
- * all the boot page allocations here.
- */
-static struct {
-	u32 pfn;
-	int type;
-} boot_page_allocations[MAX_BOOT_PTS];
-static int num_boot_page_allocations;
-static int boot_allocations_applied;
-
-void vmi_apply_boot_page_allocations(void)
-{
-	int i;
-	BUG_ON(!mem_map);
-	for (i = 0; i < num_boot_page_allocations; i++) {
-		struct page *page = pfn_to_page(boot_page_allocations[i].pfn);
-		page->type = boot_page_allocations[i].type;
-		page->type = boot_page_allocations[i].type &
-				~(VMI_PAGE_ZEROED | VMI_PAGE_CLONE);
-	}
-	boot_allocations_applied = 1;
-}
-
-static void record_page_type(u32 pfn, int type)
-{
-	BUG_ON(num_boot_page_allocations >= MAX_BOOT_PTS);
-	boot_page_allocations[num_boot_page_allocations].pfn = pfn;
-	boot_page_allocations[num_boot_page_allocations].type = type;
-	num_boot_page_allocations++;
-}
-
-static void check_zeroed_page(u32 pfn, int type, struct page *page)
-{
-	u32 *ptr;
-	int i;
-	int limit = PAGE_SIZE / sizeof(int);
-
-	if (page_address(page))
-		ptr = (u32 *)page_address(page);
-	else
-		ptr = (u32 *)__va(pfn << PAGE_SHIFT);
-	/*
-	 * When cloning the root in non-PAE mode, only the userspace
-	 * pdes need to be zeroed.
-	 */
-	if (type & VMI_PAGE_CLONE)
-		limit = KERNEL_PGD_BOUNDARY;
-	for (i = 0; i < limit; i++)
-		BUG_ON(ptr[i]);
-}
-
-/*
- * We stash the page type into struct page so we can verify the page
- * types are used properly.
- */
-static void vmi_set_page_type(u32 pfn, int type)
-{
-	/* PAE can have multiple roots per page - don't track */
-	if (PTRS_PER_PMD > 1 && (type & VMI_PAGE_PDP))
-		return;
-
-	if (boot_allocations_applied) {
-		struct page *page = pfn_to_page(pfn);
-		if (type != VMI_PAGE_NORMAL)
-			BUG_ON(page->type);
-		else
-			BUG_ON(page->type == VMI_PAGE_NORMAL);
-		page->type = type & ~(VMI_PAGE_ZEROED | VMI_PAGE_CLONE);
-		if (type & VMI_PAGE_ZEROED)
-			check_zeroed_page(pfn, type, page);
-	} else {
-		record_page_type(pfn, type);
-	}
-}
-
-static void vmi_check_page_type(u32 pfn, int type)
-{
-	/* PAE can have multiple roots per page - skip checks */
-	if (PTRS_PER_PMD > 1 && (type & VMI_PAGE_PDP))
-		return;
-
-	type &= ~(VMI_PAGE_ZEROED | VMI_PAGE_CLONE);
-	if (boot_allocations_applied) {
-		struct page *page = pfn_to_page(pfn);
-		BUG_ON((page->type ^ type) & VMI_PAGE_PAE);
-		BUG_ON(type == VMI_PAGE_NORMAL && page->type);
-		BUG_ON((type & page->type) == 0);
-	}
-}
-#else
-#define vmi_set_page_type(p,t) do { } while (0)
-#define vmi_check_page_type(p,t) do { } while (0)
-#endif
-
 #ifdef CONFIG_HIGHPTE
 static void *vmi_kmap_atomic_pte(struct page *page, enum km_type type)
 {
@@ -395,7 +292,6 @@ static void *vmi_kmap_atomic_pte(struct page *page, enum km_type type)
 
 static void vmi_allocate_pte(struct mm_struct *mm, unsigned long pfn)
 {
-	vmi_set_page_type(pfn, VMI_PAGE_L1);
 	vmi_ops.allocate_page(pfn, VMI_PAGE_L1, 0, 0, 0);
 }
 
@@ -406,27 +302,22 @@ static void vmi_allocate_pmd(struct mm_struct *mm, unsigned long pfn)
 	 * It is called only for swapper_pg_dir, which already has
 	 * data on it.
 	 */
- 	vmi_set_page_type(pfn, VMI_PAGE_L2);
 	vmi_ops.allocate_page(pfn, VMI_PAGE_L2, 0, 0, 0);
 }
 
 static void vmi_allocate_pmd_clone(unsigned long pfn, unsigned long clonepfn, unsigned long start, unsigned long count)
 {
- 	vmi_set_page_type(pfn, VMI_PAGE_L2 | VMI_PAGE_CLONE);
-	vmi_check_page_type(clonepfn, VMI_PAGE_L2);
 	vmi_ops.allocate_page(pfn, VMI_PAGE_L2 | VMI_PAGE_CLONE, clonepfn, start, count);
 }
 
 static void vmi_release_pte(unsigned long pfn)
 {
 	vmi_ops.release_page(pfn, VMI_PAGE_L1);
-	vmi_set_page_type(pfn, VMI_PAGE_NORMAL);
 }
 
 static void vmi_release_pmd(unsigned long pfn)
 {
 	vmi_ops.release_page(pfn, VMI_PAGE_L2);
-	vmi_set_page_type(pfn, VMI_PAGE_NORMAL);
 }
 
 /*
@@ -450,26 +341,22 @@ static void vmi_release_pmd(unsigned long pfn)
 
 static void vmi_update_pte(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
-	vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE);
 	vmi_ops.update_pte(ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0));
 }
 
 static void vmi_update_pte_defer(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
-	vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE);
 	vmi_ops.update_pte(ptep, vmi_flags_addr_defer(mm, addr, VMI_PAGE_PT, 0));
 }
 
 static void vmi_set_pte(pte_t *ptep, pte_t pte)
 {
 	/* XXX because of set_pmd_pte, this can be called on PT or PD layers */
-	vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE | VMI_PAGE_PD);
 	vmi_ops.set_pte(pte, ptep, VMI_PAGE_PT);
 }
 
 static void vmi_set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte)
 {
-	vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE);
 	vmi_ops.set_pte(pte, ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0));
 }
 
@@ -477,10 +364,8 @@ static void vmi_set_pmd(pmd_t *pmdp, pmd_t pmdval)
 {
 #ifdef CONFIG_X86_PAE
 	const pte_t pte = { .pte = pmdval.pmd };
-	vmi_check_page_type(__pa(pmdp) >> PAGE_SHIFT, VMI_PAGE_PMD);
 #else
 	const pte_t pte = { pmdval.pud.pgd.pgd };
-	vmi_check_page_type(__pa(pmdp) >> PAGE_SHIFT, VMI_PAGE_PGD);
 #endif
 	vmi_ops.set_pte(pte, (pte_t *)pmdp, VMI_PAGE_PD);
 }
@@ -502,7 +387,6 @@ static void vmi_set_pte_atomic(pte_t *ptep, pte_t pteval)
 
 static void vmi_set_pte_present(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte)
 {
-	vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE);
 	vmi_ops.set_pte(pte, ptep, vmi_flags_addr_defer(mm, addr, VMI_PAGE_PT, 1));
 }
 
@@ -510,21 +394,18 @@ static void vmi_set_pud(pud_t *pudp, pud_t pudval)
 {
 	/* Um, eww */
 	const pte_t pte = { .pte = pudval.pgd.pgd };
-	vmi_check_page_type(__pa(pudp) >> PAGE_SHIFT, VMI_PAGE_PGD);
 	vmi_ops.set_pte(pte, (pte_t *)pudp, VMI_PAGE_PDP);
 }
 
 static void vmi_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
 	const pte_t pte = { .pte = 0 };
-	vmi_check_page_type(__pa(ptep) >> PAGE_SHIFT, VMI_PAGE_PTE);
 	vmi_ops.set_pte(pte, ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0));
 }
 
 static void vmi_pmd_clear(pmd_t *pmd)
 {
 	const pte_t pte = { .pte = 0 };
-	vmi_check_page_type(__pa(pmd) >> PAGE_SHIFT, VMI_PAGE_PMD);
 	vmi_ops.set_pte(pte, (pte_t *)pmd, VMI_PAGE_PD);
 }
 #endif
diff --git a/arch/x86/kernel/vmlinux_32.lds.S b/arch/x86/kernel/vmlinux_32.lds.S
index a9b8560adbc2..82c67559dde7 100644
--- a/arch/x86/kernel/vmlinux_32.lds.S
+++ b/arch/x86/kernel/vmlinux_32.lds.S
@@ -44,6 +44,7 @@ SECTIONS
 	SCHED_TEXT
 	LOCK_TEXT
 	KPROBES_TEXT
+	IRQENTRY_TEXT
 	*(.fixup)
 	*(.gnu.warning)
   	_etext = .;			/* End of text section */
diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S
index 46e05447405b..1a614c0e6bef 100644
--- a/arch/x86/kernel/vmlinux_64.lds.S
+++ b/arch/x86/kernel/vmlinux_64.lds.S
@@ -35,6 +35,7 @@ SECTIONS
 	SCHED_TEXT
 	LOCK_TEXT
 	KPROBES_TEXT
+	IRQENTRY_TEXT
 	*(.fixup)
 	*(.gnu.warning)
 	_etext = .;		/* End of text section */
diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c
index 6f3d3d4cd973..44153afc9067 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -131,7 +131,16 @@ static __always_inline void do_vgettimeofday(struct timeval * tv)
 			gettimeofday(tv,NULL);
 			return;
 		}
+
+		/*
+		 * Surround the RDTSC by barriers, to make sure it's not
+		 * speculated to outside the seqlock critical section and
+		 * does not cause time warps:
+		 */
+		rdtsc_barrier();
 		now = vread();
+		rdtsc_barrier();
+
 		base = __vsyscall_gtod_data.clock.cycle_last;
 		mask = __vsyscall_gtod_data.clock.mask;
 		mult = __vsyscall_gtod_data.clock.mult;