Merge branch 'linus' into perfcounters/core

Conflicts:
	fs/exec.c
	include/linux/init_task.h

Simple context conflicts.

80 files changed, 3227 insertions, 2691 deletions

diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 3d4346a73a8..88dd768eab6 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -26,7 +26,7 @@ CFLAGS_tsc.o		:= $(nostackp)
 
 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_$(BITS).o ioport.o ldt.o
+obj-y			+= time_$(BITS).o ioport.o ldt.o dumpstack.o
 obj-y			+= setup.o i8259.o irqinit_$(BITS).o setup_percpu.o
 obj-$(CONFIG_X86_VISWS)	+= visws_quirks.o
 obj-$(CONFIG_X86_32)	+= probe_roms_32.o
@@ -66,6 +66,7 @@ obj-$(CONFIG_X86_LOCAL_APIC)	+= apic.o nmi.o
 obj-$(CONFIG_X86_IO_APIC)	+= io_apic.o
 obj-$(CONFIG_X86_REBOOTFIXUPS)	+= reboot_fixups_32.o
 obj-$(CONFIG_DYNAMIC_FTRACE)	+= ftrace.o
+obj-$(CONFIG_FUNCTION_GRAPH_TRACER)	+= ftrace.o
 obj-$(CONFIG_KEXEC)		+= machine_kexec_$(BITS).o
 obj-$(CONFIG_KEXEC)		+= relocate_kernel_$(BITS).o crash.o
 obj-$(CONFIG_CRASH_DUMP)	+= crash_dump_$(BITS).o
@@ -106,6 +107,8 @@ microcode-$(CONFIG_MICROCODE_INTEL)	+= microcode_intel.o
 microcode-$(CONFIG_MICROCODE_AMD)	+= microcode_amd.o
 obj-$(CONFIG_MICROCODE)			+= microcode.o
 
+obj-$(CONFIG_X86_CHECK_BIOS_CORRUPTION) += check.o
+
 ###
 # 64 bit specific files
 ifeq ($(CONFIG_X86_64),y)
diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 4c51a2f8fd3..65d0b72777e 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -1360,6 +1360,17 @@ static void __init acpi_process_madt(void)
 			disable_acpi();
 		}
 	}
+
+	/*
+	 * ACPI supports both logical (e.g. Hyper-Threading) and physical
+	 * processors, where MPS only supports physical.
+	 */
+	if (acpi_lapic && acpi_ioapic)
+		printk(KERN_INFO "Using ACPI (MADT) for SMP configuration "
+		       "information\n");
+	else if (acpi_lapic)
+		printk(KERN_INFO "Using ACPI for processor (LAPIC) "
+		       "configuration information\n");
 #endif
 	return;
 }
diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index a7b6dec6fc3..2e2da717b35 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 30ae2701b3d..c625800c55c 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 9a32b37ee2e..676debfc170 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 4f859acb156..6c83ac10e6d 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/perf_counter.h>
 #include <asm/atomic.h>
@@ -442,6 +443,7 @@ static void lapic_timer_setup(enum clock_event_mode mode,
 		v = apic_read(APIC_LVTT);
 		v |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR);
 		apic_write(APIC_LVTT, v);
+		apic_write(APIC_TMICT, 0xffffffff);
 		break;
 	case CLOCK_EVT_MODE_RESUME:
 		/* Nothing to do here */
@@ -560,13 +562,13 @@ static int __init calibrate_by_pmtimer(long deltapm, long *delta)
 	} else {
 		res = (((u64)deltapm) *  mult) >> 22;
 		do_div(res, 1000000);
-		printk(KERN_WARNING "APIC calibration not consistent "
+		pr_warning("APIC calibration not consistent "
 			"with PM Timer: %ldms instead of 100ms\n",
 			(long)res);
 		/* Correct the lapic counter value */
 		res = (((u64)(*delta)) * pm_100ms);
 		do_div(res, deltapm);
-		printk(KERN_INFO "APIC delta adjusted to PM-Timer: "
+		pr_info("APIC delta adjusted to PM-Timer: "
 			"%lu (%ld)\n", (unsigned long)res, *delta);
 		*delta = (long)res;
 	}
@@ -646,8 +648,7 @@ static int __init calibrate_APIC_clock(void)
 	 */
 	if (calibration_result < (1000000 / HZ)) {
 		local_irq_enable();
-		printk(KERN_WARNING
-		       "APIC frequency too slow, disabling apic timer\n");
+		pr_warning("APIC frequency too slow, disabling apic timer\n");
 		return -1;
 	}
 
@@ -673,13 +674,9 @@ static int __init calibrate_APIC_clock(void)
 		while (lapic_cal_loops <= LAPIC_CAL_LOOPS)
 			cpu_relax();
 
-		local_irq_disable();
-
 		/* Stop the lapic timer */
 		lapic_timer_setup(CLOCK_EVT_MODE_SHUTDOWN, levt);
 
-		local_irq_enable();
-
 		/* Jiffies delta */
 		deltaj = lapic_cal_j2 - lapic_cal_j1;
 		apic_printk(APIC_VERBOSE, "... jiffies delta = %lu\n", deltaj);
@@ -693,8 +690,7 @@ static int __init calibrate_APIC_clock(void)
 		local_irq_enable();
 
 	if (levt->features & CLOCK_EVT_FEAT_DUMMY) {
-		printk(KERN_WARNING
-		       "APIC timer disabled due to verification failure.\n");
+		pr_warning("APIC timer disabled due to verification failure.\n");
 			return -1;
 	}
 
@@ -715,7 +711,7 @@ void __init setup_boot_APIC_clock(void)
 	 * broadcast mechanism is used. On UP systems simply ignore it.
 	 */
 	if (disable_apic_timer) {
-		printk(KERN_INFO "Disabling APIC timer\n");
+		pr_info("Disabling APIC timer\n");
 		/* No broadcast on UP ! */
 		if (num_possible_cpus() > 1) {
 			lapic_clockevent.mult = 1;
@@ -742,7 +738,7 @@ void __init setup_boot_APIC_clock(void)
 	if (nmi_watchdog != NMI_IO_APIC)
 		lapic_clockevent.features &= ~CLOCK_EVT_FEAT_DUMMY;
 	else
-		printk(KERN_WARNING "APIC timer registered as dummy,"
+		pr_warning("APIC timer registered as dummy,"
 			" due to nmi_watchdog=%d!\n", nmi_watchdog);
 
 	/* Setup the lapic or request the broadcast */
@@ -774,8 +770,7 @@ static void local_apic_timer_interrupt(void)
 	 * spurious.
 	 */
 	if (!evt->event_handler) {
-		printk(KERN_WARNING
-		       "Spurious LAPIC timer interrupt on cpu %d\n", cpu);
+		pr_warning("Spurious LAPIC timer interrupt on cpu %d\n", cpu);
 		/* Switch it off */
 		lapic_timer_setup(CLOCK_EVT_MODE_SHUTDOWN, evt);
 		return;
@@ -797,7 +792,7 @@ 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);
 
@@ -811,9 +806,7 @@ 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();
@@ -1090,7 +1083,7 @@ static void __cpuinit lapic_setup_esr(void)
 	unsigned int oldvalue, value, maxlvt;
 
 	if (!lapic_is_integrated()) {
-		printk(KERN_INFO "No ESR for 82489DX.\n");
+		pr_info("No ESR for 82489DX.\n");
 		return;
 	}
 
@@ -1101,7 +1094,7 @@ static void __cpuinit lapic_setup_esr(void)
 		 * ESR disabled - we can't do anything useful with the
 		 * errors anyway - mbligh
 		 */
-		printk(KERN_INFO "Leaving ESR disabled.\n");
+		pr_info("Leaving ESR disabled.\n");
 		return;
 	}
 
@@ -1296,7 +1289,7 @@ void check_x2apic(void)
 	rdmsr(MSR_IA32_APICBASE, msr, msr2);
 
 	if (msr & X2APIC_ENABLE) {
-		printk("x2apic enabled by BIOS, switching to x2apic ops\n");
+		pr_info("x2apic enabled by BIOS, switching to x2apic ops\n");
 		x2apic_preenabled = x2apic = 1;
 		apic_ops = &x2apic_ops;
 	}
@@ -1308,7 +1301,7 @@ void enable_x2apic(void)
 
 	rdmsr(MSR_IA32_APICBASE, msr, msr2);
 	if (!(msr & X2APIC_ENABLE)) {
-		printk("Enabling x2apic\n");
+		pr_info("Enabling x2apic\n");
 		wrmsr(MSR_IA32_APICBASE, msr | X2APIC_ENABLE, 0);
 	}
 }
@@ -1323,9 +1316,8 @@ void __init enable_IR_x2apic(void)
 		return;
 
 	if (!x2apic_preenabled && disable_x2apic) {
-		printk(KERN_INFO
-		       "Skipped enabling x2apic and Interrupt-remapping "
-		       "because of nox2apic\n");
+		pr_info("Skipped enabling x2apic and Interrupt-remapping "
+			"because of nox2apic\n");
 		return;
 	}
 
@@ -1333,22 +1325,19 @@ void __init enable_IR_x2apic(void)
 		panic("Bios already enabled x2apic, can't enforce nox2apic");
 
 	if (!x2apic_preenabled && skip_ioapic_setup) {
-		printk(KERN_INFO
-		       "Skipped enabling x2apic and Interrupt-remapping "
-		       "because of skipping io-apic setup\n");
+		pr_info("Skipped enabling x2apic and Interrupt-remapping "
+			"because of skipping io-apic setup\n");
 		return;
 	}
 
 	ret = dmar_table_init();
 	if (ret) {
-		printk(KERN_INFO
-		       "dmar_table_init() failed with %d:\n", ret);
+		pr_info("dmar_table_init() failed with %d:\n", ret);
 
 		if (x2apic_preenabled)
 			panic("x2apic enabled by bios. But IR enabling failed");
 		else
-			printk(KERN_INFO
-			       "Not enabling x2apic,Intr-remapping\n");
+			pr_info("Not enabling x2apic,Intr-remapping\n");
 		return;
 	}
 
@@ -1357,7 +1346,7 @@ void __init enable_IR_x2apic(void)
 
 	ret = save_mask_IO_APIC_setup();
 	if (ret) {
-		printk(KERN_INFO "Saving IO-APIC state failed: %d\n", ret);
+		pr_info("Saving IO-APIC state failed: %d\n", ret);
 		goto end;
 	}
 
@@ -1392,14 +1381,11 @@ end:
 
 	if (!ret) {
 		if (!x2apic_preenabled)
-			printk(KERN_INFO
-			       "Enabled x2apic and interrupt-remapping\n");
+			pr_info("Enabled x2apic and interrupt-remapping\n");
 		else
-			printk(KERN_INFO
-			       "Enabled Interrupt-remapping\n");
+			pr_info("Enabled Interrupt-remapping\n");
 	} else
-		printk(KERN_ERR
-		       "Failed to enable Interrupt-remapping and x2apic\n");
+		pr_err("Failed to enable Interrupt-remapping and x2apic\n");
 #else
 	if (!cpu_has_x2apic)
 		return;
@@ -1408,8 +1394,8 @@ end:
 		panic("x2apic enabled prior OS handover,"
 		      " enable CONFIG_INTR_REMAP");
 
-	printk(KERN_INFO "Enable CONFIG_INTR_REMAP for enabling intr-remapping "
-	       " and x2apic\n");
+	pr_info("Enable CONFIG_INTR_REMAP for enabling intr-remapping "
+		" and x2apic\n");
 #endif
 
 	return;
@@ -1426,7 +1412,7 @@ end:
 static int __init detect_init_APIC(void)
 {
 	if (!cpu_has_apic) {
-		printk(KERN_INFO "No local APIC present\n");
+		pr_info("No local APIC present\n");
 		return -1;
 	}
 
@@ -1467,8 +1453,8 @@ static int __init detect_init_APIC(void)
 		 * "lapic" specified.
 		 */
 		if (!force_enable_local_apic) {
-			printk(KERN_INFO "Local APIC disabled by BIOS -- "
-			       "you can enable it with \"lapic\"\n");
+			pr_info("Local APIC disabled by BIOS -- "
+				"you can enable it with \"lapic\"\n");
 			return -1;
 		}
 		/*
@@ -1478,8 +1464,7 @@ static int __init detect_init_APIC(void)
 		 */
 		rdmsr(MSR_IA32_APICBASE, l, h);
 		if (!(l & MSR_IA32_APICBASE_ENABLE)) {
-			printk(KERN_INFO
-			       "Local APIC disabled by BIOS -- reenabling.\n");
+			pr_info("Local APIC disabled by BIOS -- reenabling.\n");
 			l &= ~MSR_IA32_APICBASE_BASE;
 			l |= MSR_IA32_APICBASE_ENABLE | APIC_DEFAULT_PHYS_BASE;
 			wrmsr(MSR_IA32_APICBASE, l, h);
@@ -1492,7 +1477,7 @@ static int __init detect_init_APIC(void)
 	 */
 	features = cpuid_edx(1);
 	if (!(features & (1 << X86_FEATURE_APIC))) {
-		printk(KERN_WARNING "Could not enable APIC!\n");
+		pr_warning("Could not enable APIC!\n");
 		return -1;
 	}
 	set_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC);
@@ -1503,14 +1488,14 @@ static int __init detect_init_APIC(void)
 	if (l & MSR_IA32_APICBASE_ENABLE)
 		mp_lapic_addr = l & MSR_IA32_APICBASE_BASE;
 
-	printk(KERN_INFO "Found and enabled local APIC!\n");
+	pr_info("Found and enabled local APIC!\n");
 
 	apic_pm_activate();
 
 	return 0;
 
 no_apic:
-	printk(KERN_INFO "No local APIC present or hardware disabled\n");
+	pr_info("No local APIC present or hardware disabled\n");
 	return -1;
 }
 #endif
@@ -1586,12 +1571,12 @@ int __init APIC_init_uniprocessor(void)
 {
 #ifdef CONFIG_X86_64
 	if (disable_apic) {
-		printk(KERN_INFO "Apic disabled\n");
+		pr_info("Apic disabled\n");
 		return -1;
 	}
 	if (!cpu_has_apic) {
 		disable_apic = 1;
-		printk(KERN_INFO "Apic disabled by BIOS\n");
+		pr_info("Apic disabled by BIOS\n");
 		return -1;
 	}
 #else
@@ -1603,8 +1588,8 @@ int __init APIC_init_uniprocessor(void)
 	 */
 	if (!cpu_has_apic &&
 	    APIC_INTEGRATED(apic_version[boot_cpu_physical_apicid])) {
-		printk(KERN_ERR "BIOS bug, local APIC 0x%x not detected!...\n",
-		       boot_cpu_physical_apicid);
+		pr_err("BIOS bug, local APIC 0x%x not detected!...\n",
+			boot_cpu_physical_apicid);
 		clear_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC);
 		return -1;
 	}
@@ -1680,9 +1665,7 @@ 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
@@ -1696,8 +1679,8 @@ void smp_spurious_interrupt(struct pt_regs *regs)
 	inc_irq_stat(irq_spurious_count);
 
 	/* see sw-dev-man vol 3, chapter 7.4.13.5 */
-	printk(KERN_INFO "spurious APIC interrupt on CPU#%d, "
-	       "should never happen.\n", smp_processor_id());
+	pr_info("spurious APIC interrupt on CPU#%d, "
+		"should never happen.\n", smp_processor_id());
 	irq_exit();
 }
 
@@ -1708,9 +1691,7 @@ 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);
@@ -1719,17 +1700,18 @@ void smp_error_interrupt(struct pt_regs *regs)
 	ack_APIC_irq();
 	atomic_inc(&irq_err_count);
 
-	/* Here is what the APIC error bits mean:
-	   0: Send CS error
-	   1: Receive CS error
-	   2: Send accept error
-	   3: Receive accept error
-	   4: Reserved
-	   5: Send illegal vector
-	   6: Received illegal vector
-	   7: Illegal register address
-	*/
-	printk(KERN_DEBUG "APIC error on CPU%d: %02x(%02x)\n",
+	/*
+	 * Here is what the APIC error bits mean:
+	 * 0: Send CS error
+	 * 1: Receive CS error
+	 * 2: Send accept error
+	 * 3: Receive accept error
+	 * 4: Reserved
+	 * 5: Send illegal vector
+	 * 6: Received illegal vector
+	 * 7: Illegal register address
+	 */
+	pr_debug("APIC error on CPU%d: %02x(%02x)\n",
 		smp_processor_id(), v , v1);
 	irq_exit();
 }
@@ -1833,15 +1815,15 @@ void __cpuinit generic_processor_info(int apicid, int version)
 	 * Validate version
 	 */
 	if (version == 0x0) {
-		printk(KERN_WARNING "BIOS bug, APIC version is 0 for CPU#%d! "
-				"fixing up to 0x10. (tell your hw vendor)\n",
-				version);
+		pr_warning("BIOS bug, APIC version is 0 for CPU#%d! "
+			"fixing up to 0x10. (tell your hw vendor)\n",
+			version);
 		version = 0x10;
 	}
 	apic_version[apicid] = version;
 
 	if (num_processors >= NR_CPUS) {
-		printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached."
+		pr_warning("WARNING: NR_CPUS limit of %i reached."
 			"  Processor ignored.\n", NR_CPUS);
 		return;
 	}
@@ -2204,7 +2186,7 @@ static int __init apic_set_verbosity(char *arg)
 	else if (strcmp("verbose", arg) == 0)
 		apic_verbosity = APIC_VERBOSE;
 	else {
-		printk(KERN_WARNING "APIC Verbosity level %s not recognised"
+		pr_warning("APIC Verbosity level %s not recognised"
 			" use apic=verbose or apic=debug\n", arg);
 		return -EINVAL;
 	}
diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c
index 5145a6e72bb..3a26525a3f3 100644
--- a/arch/x86/kernel/apm_32.c
+++ b/arch/x86/kernel/apm_32.c
@@ -391,11 +391,7 @@ static int power_off;
 #else
 static int power_off = 1;
 #endif
-#ifdef CONFIG_APM_REAL_MODE_POWER_OFF
-static int realmode_power_off = 1;
-#else
 static int realmode_power_off;
-#endif
 #ifdef CONFIG_APM_ALLOW_INTS
 static int allow_ints = 1;
 #else
diff --git a/arch/x86/kernel/asm-offsets_32.c b/arch/x86/kernel/asm-offsets_32.c
index 6649d09ad88..ee4df08feee 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 7fcf63d22f8..1d41d3f1edb 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 f0dfe6f17e7..2a0a2a3cac2 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 00000000000..2ac0ab71412
--- /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 89e53361fe2..c3813306e0b 100644
--- a/arch/x86/kernel/cpu/Makefile
+++ b/arch/x86/kernel/cpu/Makefile
@@ -2,8 +2,14 @@
 # Makefile for x86-compatible CPU details, features 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 ef8f831af82..2cf23634b6d 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 8f1e31db2ad..7c878f6aa91 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 ad331b4d623..376b9f9d8d2 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -37,6 +37,7 @@
 #include <asm/proto.h>
 #include <asm/sections.h>
 #include <asm/setup.h>
+#include <asm/hypervisor.h>
 
 #include "cpu.h"
 
@@ -704,6 +705,7 @@ 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
@@ -864,7 +866,7 @@ 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)
 {
@@ -905,8 +907,8 @@ 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/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
index 8e48c5d4467..88ea02dcb62 100644
--- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
+++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
@@ -33,6 +33,7 @@
 #include <linux/cpufreq.h>
 #include <linux/compiler.h>
 #include <linux/dmi.h>
+#include <linux/ftrace.h>
 
 #include <linux/acpi.h>
 #include <acpi/processor.h>
@@ -391,6 +392,7 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
 	unsigned int next_perf_state = 0; /* Index into perf table */
 	unsigned int i;
 	int result = 0;
+	struct power_trace it;
 
 	dprintk("acpi_cpufreq_target %d (%d)\n", target_freq, policy->cpu);
 
@@ -427,6 +429,8 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
 		}
 	}
 
+	trace_power_mark(&it, POWER_PSTATE, next_perf_state);
+
 	switch (data->cpu_feature) {
 	case SYSTEM_INTEL_MSR_CAPABLE:
 		cmd.type = SYSTEM_INTEL_MSR_CAPABLE;
diff --git a/arch/x86/kernel/cpu/hypervisor.c b/arch/x86/kernel/cpu/hypervisor.c
new file mode 100644
index 00000000000..fb5b86af0b0
--- /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 cce0b6118d5..8ea6929e974 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);
@@ -307,13 +323,8 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 		set_cpu_cap(c, X86_FEATURE_P4);
 	if (c->x86 == 6)
 		set_cpu_cap(c, X86_FEATURE_P3);
-
-	if (cpu_has_bts)
-		ptrace_bts_init_intel(c);
-
 #endif
 
-	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 3f46afbb1cf..68b5d8681cb 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -644,20 +644,17 @@ 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 4b031a4ac85..1c838032fd3 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 5eb390a4b2e..748c8f9e7a0 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c
@@ -237,7 +237,7 @@ 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 c17eaf5dd6d..4b48f251fd3 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 c78c04821ea..1159e269e59 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 00000000000..284c399e323
--- /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/crash.c b/arch/x86/kernel/crash.c
index 26855381790..d84a852e4cd 100644
--- a/arch/x86/kernel/crash.c
+++ b/arch/x86/kernel/crash.c
@@ -29,34 +29,17 @@
 
 #include <mach_ipi.h>
 
-/* This keeps a track of which one is crashing cpu. */
-static int crashing_cpu;
 
 #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)
-static atomic_t waiting_for_crash_ipi;
 
-static int crash_nmi_callback(struct notifier_block *self,
-			unsigned long val, void *data)
+static void kdump_nmi_callback(int cpu, struct die_args *args)
 {
 	struct pt_regs *regs;
 #ifdef CONFIG_X86_32
 	struct pt_regs fixed_regs;
 #endif
-	int cpu;
 
-	if (val != DIE_NMI_IPI)
-		return NOTIFY_OK;
-
-	regs = ((struct die_args *)data)->regs;
-	cpu = raw_smp_processor_id();
-
-	/* Don't do anything if this handler is invoked on crashing cpu.
-	 * Otherwise, system will completely hang. Crashing cpu can get
-	 * an NMI if system was initially booted with nmi_watchdog parameter.
-	 */
-	if (cpu == crashing_cpu)
-		return NOTIFY_STOP;
-	local_irq_disable();
+	regs = args->regs;
 
 #ifdef CONFIG_X86_32
 	if (!user_mode_vm(regs)) {
@@ -65,54 +48,19 @@ static int crash_nmi_callback(struct notifier_block *self,
 	}
 #endif
 	crash_save_cpu(regs, cpu);
-	disable_local_APIC();
-	atomic_dec(&waiting_for_crash_ipi);
-	/* Assume hlt works */
-	halt();
-	for (;;)
-		cpu_relax();
-
-	return 1;
-}
 
-static void smp_send_nmi_allbutself(void)
-{
-	cpumask_t mask = cpu_online_map;
-	cpu_clear(safe_smp_processor_id(), mask);
-	if (!cpus_empty(mask))
-		send_IPI_mask(mask, NMI_VECTOR);
+	disable_local_APIC();
 }
 
-static struct notifier_block crash_nmi_nb = {
-	.notifier_call = crash_nmi_callback,
-};
-
-static void nmi_shootdown_cpus(void)
+static void kdump_nmi_shootdown_cpus(void)
 {
-	unsigned long msecs;
-
-	atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
-	/* Would it be better to replace the trap vector here? */
-	if (register_die_notifier(&crash_nmi_nb))
-		return;		/* return what? */
-	/* Ensure the new callback function is set before sending
-	 * out the NMI
-	 */
-	wmb();
+	nmi_shootdown_cpus(kdump_nmi_callback);
 
-	smp_send_nmi_allbutself();
-
-	msecs = 1000; /* Wait at most a second for the other cpus to stop */
-	while ((atomic_read(&waiting_for_crash_ipi) > 0) && msecs) {
-		mdelay(1);
-		msecs--;
-	}
-
-	/* Leave the nmi callback set */
 	disable_local_APIC();
 }
+
 #else
-static void nmi_shootdown_cpus(void)
+static void kdump_nmi_shootdown_cpus(void)
 {
 	/* There are no cpus to shootdown */
 }
@@ -131,9 +79,7 @@ void native_machine_crash_shutdown(struct pt_regs *regs)
 	/* The kernel is broken so disable interrupts */
 	local_irq_disable();
 
-	/* Make a note of crashing cpu. Will be used in NMI callback.*/
-	crashing_cpu = safe_smp_processor_id();
-	nmi_shootdown_cpus();
+	kdump_nmi_shootdown_cpus();
 	lapic_shutdown();
 #if defined(CONFIG_X86_IO_APIC)
 	disable_IO_APIC();
diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c
index a2d1176c38e..da91701a234 100644
--- a/arch/x86/kernel/ds.c
+++ b/arch/x86/kernel/ds.c
@@ -6,14 +6,13 @@
  * precise-event based sampling (PEBS).
  *
  * It manages:
- * - per-thread and per-cpu allocation of BTS and PEBS
- * - buffer memory allocation (optional)
- * - buffer overflow handling
+ * - DS and BTS hardware configuration
+ * - buffer overflow handling (to be done)
  * - buffer access
  *
- * It assumes:
- * - get_task_struct on all parameter tasks
- * - current is allowed to trace parameter 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.
@@ -28,22 +27,69 @@
 #include <linux/slab.h>
 #include <linux/sched.h>
 #include <linux/mm.h>
+#include <linux/kernel.h>
 
 
 /*
  * 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.
+ *
+ * This holds the configuration of the tracer and serves as a handle
+ * to identify tracers.
+ */
+struct ds_tracer {
+	/* the DS context (partially) owned by this tracer */
+	struct ds_context *context;
+	/* the buffer provided on ds_request() and its size in bytes */
+	void *buffer;
+	size_t size;
+};
+
+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;
+};
+
+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;
+};
 
 /*
  * Debug Store (DS) save area configuration (see Intel64 and IA32
@@ -109,32 +155,9 @@ static inline void ds_set(unsigned char *base, enum ds_qualifier qual,
 
 
 /*
- * Locking is done only for allocating BTS or PEBS resources and for
- * guarding context and buffer memory allocation.
- *
- * Most functions require the current task to own the ds context part
- * they are going to access. All the locking is done when validating
- * access to the context.
+ * Locking is done only for allocating BTS or PEBS resources.
  */
-static spinlock_t ds_lock = __SPIN_LOCK_UNLOCKED(ds_lock);
-
-/*
- * Validate that the current task is allowed to access the BTS/PEBS
- * buffer of the parameter task.
- *
- * Returns 0, if access is granted; -Eerrno, otherwise.
- */
-static inline int ds_validate_access(struct ds_context *context,
-				     enum ds_qualifier qual)
-{
-	if (!context)
-		return -EPERM;
-
-	if (context->owner[qual] == current)
-		return 0;
-
-	return -EPERM;
-}
+static DEFINE_SPINLOCK(ds_lock);
 
 
 /*
@@ -150,27 +173,32 @@ static inline int ds_validate_access(struct ds_context *context,
  *   >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);
 }
 
 
@@ -183,99 +211,70 @@ 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.
- *
- * We distinguish between an allocating and a non-allocating get of a
- * context:
- * - the allocating get is used for requesting BTS/PEBS resources. It
- *   requires the caller to hold the global ds_lock.
- * - the non-allocating get is used for all other cases. A
- *   non-existing context indicates an error. It acquires and releases
- *   the ds_lock itself for obtaining the context.
- *
- * A context and its DS configuration are allocated and deallocated
- * together. A context always has a DS configuration of the
- * appropriate size.
- */
-static DEFINE_PER_CPU(struct ds_context *, system_context);
-
-#define this_system_context per_cpu(system_context, smp_processor_id())
-
-/*
- * Returns the pointer to the parameter task's context or to the
- * system-wide context, if task is NULL.
- *
- * Increases the use count of the returned context, if not NULL.
  */
-static inline struct ds_context *ds_get_context(struct task_struct *task)
-{
-	struct ds_context *context;
-	unsigned long irq;
+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;
+};
 
-	spin_lock_irqsave(&ds_lock, irq);
+static DEFINE_PER_CPU(struct ds_context *, system_context_array);
 
-	context = (task ? task->thread.ds_ctx : this_system_context);
-	if (context)
-		context->count++;
+#define system_context per_cpu(system_context_array, smp_processor_id())
 
-	spin_unlock_irqrestore(&ds_lock, irq);
-
-	return context;
-}
 
-/*
- * Same as ds_get_context, but allocates the context and it's DS
- * structure, if necessary; returns NULL; if out of memory.
- */
-static inline struct ds_context *ds_alloc_context(struct task_struct *task)
+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;
-
-		context->ds = kzalloc(ds_cfg.sizeof_ds, GFP_KERNEL);
-		if (!context->ds) {
-			kfree(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->ds);
-			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;
 }
 
-/*
- * Decreases the use count of the parameter context, if not NULL.
- * Deallocates the context, if the use count reaches zero.
- */
 static inline void ds_put_context(struct ds_context *context)
 {
 	unsigned long irq;
@@ -285,8 +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;
 
@@ -296,135 +297,263 @@ static inline void ds_put_context(struct ds_context *context)
 	if (!context->task || (context->task == current))
 		wrmsrl(MSR_IA32_DS_AREA, 0);
 
-	put_tracer(context->task);
+	spin_unlock_irqrestore(&ds_lock, irq);
 
-	/* free any leftover buffers from tracers that did not
-	 * deallocate them properly. */
-	kfree(context->buffer[ds_bts]);
-	kfree(context->buffer[ds_pebs]);
-	kfree(context->ds);
 	kfree(context);
- out:
-	spin_unlock_irqrestore(&ds_lock, irq);
 }
 
 
 /*
- * Handle a buffer overflow
+ * Call the tracer's callback on a buffer overflow.
  *
- * task: the task whose buffers are overflowing;
- *       NULL for a buffer overflow on the current cpu
  * context: the ds context
  * qual: the buffer type
  */
-static void ds_overflow(struct task_struct *task, struct ds_context *context,
-			enum ds_qualifier qual)
+static void ds_overflow(struct ds_context *context, enum ds_qualifier qual)
 {
-	if (!context)
-		return;
-
-	if (context->callback[qual])
-		(*context->callback[qual])(task);
-
-	/* todo: do some more overflow handling */
+	switch (qual) {
+	case ds_bts:
+		if (context->bts_master &&
+		    context->bts_master->ovfl)
+			context->bts_master->ovfl(context->bts_master);
+		break;
+	case ds_pebs:
+		if (context->pebs_master &&
+		    context->pebs_master->ovfl)
+			context->pebs_master->ovfl(context->pebs_master);
+		break;
+	}
 }
 
 
 /*
- * Allocate a non-pageable buffer of the parameter size.
- * Checks the memory and the locked memory rlimit.
+ * Write raw data into the BTS or PEBS buffer.
  *
- * Returns the buffer, if successful;
- *         NULL, if out of memory or rlimit exceeded.
+ * The remainder of any partially written record is zeroed out.
  *
- * size: the requested buffer size in bytes
- * pages (out): if not NULL, contains the number of pages reserved
+ * context: the DS context
+ * qual: the buffer type
+ * record: the data to write
+ * size: the size of the data
  */
-static inline void *ds_allocate_buffer(size_t size, unsigned int *pages)
+static int ds_write(struct ds_context *context, enum ds_qualifier qual,
+		    const void *record, size_t size)
 {
-	unsigned long rlim, vm, pgsz;
-	void *buffer;
+	int bytes_written = 0;
 
-	pgsz = PAGE_ALIGN(size) >> PAGE_SHIFT;
+	if (!record)
+		return -EINVAL;
 
-	rlim = current->signal->rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT;
-	vm   = current->mm->total_vm  + pgsz;
-	if (rlim < vm)
-		return NULL;
+	while (size) {
+		unsigned long base, index, end, write_end, int_th;
+		unsigned long write_size, adj_write_size;
 
-	rlim = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur >> PAGE_SHIFT;
-	vm   = current->mm->locked_vm  + pgsz;
-	if (rlim < vm)
-		return NULL;
+		/*
+		 * 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);
 
-	buffer = kzalloc(size, GFP_KERNEL);
-	if (!buffer)
-		return NULL;
+		write_end = min(end, int_th);
 
-	current->mm->total_vm  += pgsz;
-	current->mm->locked_vm += pgsz;
+		/* if we are already beyond the interrupt threshold,
+		 * we fill the entire buffer */
+		if (write_end <= index)
+			write_end = end;
 
-	if (pages)
-		*pages = pgsz;
+		if (write_end <= index)
+			break;
+
+		write_size = min((unsigned long) size, write_end - index);
+		memcpy((void *)index, record, write_size);
 
-	return buffer;
+		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;
 }
 
-static int ds_request(struct task_struct *task, void *base, size_t size,
-		      ds_ovfl_callback_t ovfl, enum ds_qualifier qual)
+
+/*
+ * 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)
 {
-	struct ds_context *context;
-	unsigned long buffer, adj;
-	const unsigned long alignment = (1 << 3);
-	unsigned long irq;
-	int error = 0;
+	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;
+}
 
-	if (!ds_cfg.sizeof_ds)
-		return -EOPNOTSUPP;
 
-	/* we require some space to do alignment adjustments below */
-	if (size < (alignment + ds_cfg.sizeof_rec[qual]))
+/*
+ * 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;
 
-	/* buffer overflow notification is not yet implemented */
-	if (ovfl)
-		return -EOPNOTSUPP;
+	if (at < tracer->trace.ds.begin)
+		return -EINVAL;
 
+	if (tracer->trace.ds.end < (at + tracer->trace.ds.size))
+		return -EINVAL;
 
-	context = ds_alloc_context(task);
-	if (!context)
-		return -ENOMEM;
+	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;
+	}
 
-	spin_lock_irqsave(&ds_lock, irq);
+	return ds_cfg.sizeof_rec[ds_bts];
+}
 
-	error = -EPERM;
-	if (!check_tracer(task))
-		goto out_unlock;
+static int bts_write(struct bts_tracer *tracer, const struct bts_struct *in)
+{
+	unsigned char raw[MAX_SIZEOF_BTS];
 
-	get_tracer(task);
+	if (!tracer)
+		return -EINVAL;
 
-	error = -EALREADY;
-	if (context->owner[qual] == current)
-		goto out_put_tracer;
-	error = -EPERM;
-	if (context->owner[qual] != NULL)
-		goto out_put_tracer;
-	context->owner[qual] = current;
+	if (MAX_SIZEOF_BTS < ds_cfg.sizeof_rec[ds_bts])
+		return -EOVERFLOW;
 
-	spin_unlock_irqrestore(&ds_lock, irq);
+	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]);
+}
 
-	error = -ENOMEM;
-	if (!base) {
-		base = ds_allocate_buffer(size, &context->pages[qual]);
-		if (!base)
-			goto out_release;
 
-		context->buffer[qual]   = base;
-	}
-	error = 0;
+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;
 
-	context->callback[qual] = ovfl;
+	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
 	 * constraints:
@@ -436,410 +565,383 @@ static int ds_request(struct task_struct *task, void *base, size_t size,
 	 */
 	buffer = (unsigned long)base;
 
-	adj = ALIGN(buffer, alignment) - buffer;
+	adj = ALIGN(buffer, DS_ALIGNMENT) - buffer;
 	buffer += adj;
 	size   -= adj;
 
-	size /= ds_cfg.sizeof_rec[qual];
-	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);
+	trace->n = size / ds_cfg.sizeof_rec[qual];
+	trace->size = ds_cfg.sizeof_rec[qual];
 
-	if (ovfl) {
-		/* todo: select a suitable interrupt threshold */
-	} else
-		ds_set(context->ds, qual,
-		       ds_interrupt_threshold, buffer + size + 1);
+	size = (trace->n * trace->size);
 
-	/* we keep the context until ds_release */
-	return error;
-
- out_release:
-	context->owner[qual] = NULL;
-	ds_put_context(context);
-	put_tracer(task);
-	return error;
-
- out_put_tracer:
-	spin_unlock_irqrestore(&ds_lock, irq);
-	ds_put_context(context);
-	put_tracer(task);
-	return error;
+	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.
+	 */
+	trace->ith = (void *)(buffer + size - ith);
 
- out_unlock:
-	spin_unlock_irqrestore(&ds_lock, irq);
-	ds_put_context(context);
-	return error;
+	trace->flags = flags;
 }
 
-int ds_request_bts(struct task_struct *task, void *base, size_t size,
-		   ds_ovfl_callback_t ovfl)
-{
-	return ds_request(task, base, size, ovfl, ds_bts);
-}
 
-int ds_request_pebs(struct task_struct *task, void *base, size_t size,
-		    ds_ovfl_callback_t ovfl)
-{
-	return ds_request(task, base, size, ovfl, ds_pebs);
-}
-
-static int ds_release(struct task_struct *task, enum ds_qualifier qual)
+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;
 	int error;
 
-	context = ds_get_context(task);
-	error = ds_validate_access(context, qual);
-	if (error < 0)
+	error = -EINVAL;
+	if (!base)
 		goto out;
 
-	kfree(context->buffer[qual]);
-	context->buffer[qual] = NULL;
-
-	current->mm->total_vm  -= context->pages[qual];
-	current->mm->locked_vm -= context->pages[qual];
-	context->pages[qual] = 0;
-	context->owner[qual] = NULL;
-
-	/*
-	 * we put the context twice:
-	 *   once for the ds_get_context
-	 *   once for the corresponding ds_request
-	 */
-	ds_put_context(context);
- out:
-	ds_put_context(context);
-	return error;
-}
+	/* we require some space to do alignment adjustments below */
+	error = -EINVAL;
+	if (size < (DS_ALIGNMENT + ds_cfg.sizeof_rec[qual]))
+		goto out;
 
-int ds_release_bts(struct task_struct *task)
-{
-	return ds_release(task, ds_bts);
-}
+	if (th != (size_t)-1) {
+		th *= ds_cfg.sizeof_rec[qual];
 
-int ds_release_pebs(struct task_struct *task)
-{
-	return ds_release(task, ds_pebs);
-}
+		error = -EINVAL;
+		if (size <= th)
+			goto out;
+	}
 
-static int ds_get_index(struct task_struct *task, size_t *pos,
-			enum ds_qualifier qual)
-{
-	struct ds_context *context;
-	unsigned long base, index;
-	int error;
+	tracer->buffer = base;
+	tracer->size = size;
 
+	error = -ENOMEM;
 	context = ds_get_context(task);
-	error = ds_validate_access(context, qual);
-	if (error < 0)
+	if (!context)
 		goto out;
+	tracer->context = context;
 
-	base  = ds_get(context->ds, qual, ds_buffer_base);
-	index = ds_get(context->ds, qual, ds_index);
+	ds_init_ds_trace(trace, qual, base, size, th, flags);
 
-	error = ((index - base) / ds_cfg.sizeof_rec[qual]);
-	if (pos)
-		*pos = error;
+	error = 0;
  out:
-	ds_put_context(context);
 	return error;
 }
 
-int ds_get_bts_index(struct task_struct *task, size_t *pos)
-{
-	return ds_get_index(task, pos, ds_bts);
-}
-
-int ds_get_pebs_index(struct task_struct *task, size_t *pos)
+struct bts_tracer *ds_request_bts(struct task_struct *task,
+				  void *base, size_t size,
+				  bts_ovfl_callback_t ovfl, size_t th,
+				  unsigned int flags)
 {
-	return ds_get_index(task, pos, ds_pebs);
-}
-
-static int ds_get_end(struct task_struct *task, size_t *pos,
-		      enum ds_qualifier qual)
-{
-	struct ds_context *context;
-	unsigned long base, end;
+	struct bts_tracer *tracer;
+	unsigned long irq;
 	int error;
 
-	context = ds_get_context(task);
-	error = ds_validate_access(context, qual);
-	if (error < 0)
+	error = -EOPNOTSUPP;
+	if (!ds_cfg.ctl[dsf_bts])
 		goto out;
 
-	base = ds_get(context->ds, qual, ds_buffer_base);
-	end  = ds_get(context->ds, qual, ds_absolute_maximum);
+	/* buffer overflow notification is not yet implemented */
+	error = -EOPNOTSUPP;
+	if (ovfl)
+		goto out;
 
-	error = ((end - base) / ds_cfg.sizeof_rec[qual]);
-	if (pos)
-		*pos = error;
- out:
-	ds_put_context(context);
-	return error;
-}
+	error = -ENOMEM;
+	tracer = kzalloc(sizeof(*tracer), GFP_KERNEL);
+	if (!tracer)
+		goto out;
+	tracer->ovfl = ovfl;
 
-int ds_get_bts_end(struct task_struct *task, size_t *pos)
-{
-	return ds_get_end(task, pos, ds_bts);
-}
+	error = ds_request(&tracer->ds, &tracer->trace.ds,
+			   ds_bts, task, base, size, th, flags);
+	if (error < 0)
+		goto out_tracer;
 
-int ds_get_pebs_end(struct task_struct *task, size_t *pos)
-{
-	return ds_get_end(task, pos, ds_pebs);
-}
 
-static int ds_access(struct task_struct *task, size_t index,
-		     const void **record, enum ds_qualifier qual)
-{
-	struct ds_context *context;
-	unsigned long base, idx;
-	int error;
+	spin_lock_irqsave(&ds_lock, irq);
 
-	if (!record)
-		return -EINVAL;
+	error = -EPERM;
+	if (!check_tracer(task))
+		goto out_unlock;
+	get_tracer(task);
 
-	context = ds_get_context(task);
-	error = ds_validate_access(context, qual);
-	if (error < 0)
-		goto out;
+	error = -EPERM;
+	if (tracer->ds.context->bts_master)
+		goto out_put_tracer;
+	tracer->ds.context->bts_master = tracer;
 
-	base = ds_get(context->ds, qual, ds_buffer_base);
-	idx = base + (index * ds_cfg.sizeof_rec[qual]);
+	spin_unlock_irqrestore(&ds_lock, irq);
 
-	error = -EINVAL;
-	if (idx > ds_get(context->ds, qual, ds_absolute_maximum))
-		goto out;
 
-	*record = (const void *)idx;
-	error = ds_cfg.sizeof_rec[qual];
- out:
-	ds_put_context(context);
-	return error;
-}
+	tracer->trace.read  = bts_read;
+	tracer->trace.write = bts_write;
 
-int ds_access_bts(struct task_struct *task, size_t index, const void **record)
-{
-	return ds_access(task, index, record, ds_bts);
-}
+	ds_write_config(tracer->ds.context, &tracer->trace.ds, ds_bts);
+	ds_resume_bts(tracer);
 
-int ds_access_pebs(struct task_struct *task, size_t index, const void **record)
-{
-	return ds_access(task, index, record, ds_pebs);
+	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 int ds_write(struct task_struct *task, const void *record, size_t size,
-		    enum ds_qualifier qual, int force)
+struct pebs_tracer *ds_request_pebs(struct task_struct *task,
+				    void *base, size_t size,
+				    pebs_ovfl_callback_t ovfl, size_t th,
+				    unsigned int flags)
 {
-	struct ds_context *context;
+	struct pebs_tracer *tracer;
+	unsigned long irq;
 	int error;
 
-	if (!record)
-		return -EINVAL;
+	/* buffer overflow notification is not yet implemented */
+	error = -EOPNOTSUPP;
+	if (ovfl)
+		goto out;
 
-	error = -EPERM;
-	context = ds_get_context(task);
-	if (!context)
+	error = -ENOMEM;
+	tracer = kzalloc(sizeof(*tracer), GFP_KERNEL);
+	if (!tracer)
 		goto out;
+	tracer->ovfl = ovfl;
 
-	if (!force) {
-		error = ds_validate_access(context, qual);
-		if (error < 0)
-			goto out;
-	}
+	error = ds_request(&tracer->ds, &tracer->trace.ds,
+			   ds_pebs, task, base, size, th, flags);
+	if (error < 0)
+		goto out_tracer;
 
-	error = 0;
-	while (size) {
-		unsigned long base, index, end, write_end, int_th;
-		unsigned long write_size, adj_write_size;
+	spin_lock_irqsave(&ds_lock, irq);
 
-		/*
-		 * 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);
+	error = -EPERM;
+	if (!check_tracer(task))
+		goto out_unlock;
+	get_tracer(task);
 
-		write_end = min(end, int_th);
+	error = -EPERM;
+	if (tracer->ds.context->pebs_master)
+		goto out_put_tracer;
+	tracer->ds.context->pebs_master = tracer;
 
-		/* if we are already beyond the interrupt threshold,
-		 * we fill the entire buffer */
-		if (write_end <= index)
-			write_end = end;
+	spin_unlock_irqrestore(&ds_lock, irq);
 
-		if (write_end <= index)
-			goto out;
+	ds_write_config(tracer->ds.context, &tracer->trace.ds, ds_bts);
+	ds_resume_pebs(tracer);
 
-		write_size = min((unsigned long) size, write_end - index);
-		memcpy((void *)index, record, write_size);
+	return tracer;
 
-		record = (const char *)record + write_size;
-		size  -= write_size;
-		error += write_size;
+ 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);
+}
 
-		adj_write_size = write_size / ds_cfg.sizeof_rec[qual];
-		adj_write_size *= ds_cfg.sizeof_rec[qual];
+void ds_release_bts(struct bts_tracer *tracer)
+{
+	if (!tracer)
+		return;
 
-		/* zero out trailing bytes */
-		memset((char *)index + write_size, 0,
-		       adj_write_size - write_size);
-		index += adj_write_size;
+	ds_suspend_bts(tracer);
 
-		if (index >= end)
-			index = base;
-		ds_set(context->ds, qual, ds_index, index);
+	WARN_ON_ONCE(tracer->ds.context->bts_master != tracer);
+	tracer->ds.context->bts_master = NULL;
 
-		if (index >= int_th)
-			ds_overflow(task, context, qual);
-	}
+	put_tracer(tracer->ds.context->task);
+	ds_put_context(tracer->ds.context);
 
- out:
-	ds_put_context(context);
-	return error;
+	kfree(tracer);
 }
 
-int ds_write_bts(struct task_struct *task, const void *record, size_t size)
+void ds_suspend_bts(struct bts_tracer *tracer)
 {
-	return ds_write(task, record, size, ds_bts, /* force = */ 0);
-}
+	struct task_struct *task;
 
-int ds_write_pebs(struct task_struct *task, const void *record, size_t size)
-{
-	return ds_write(task, record, size, ds_pebs, /* force = */ 0);
-}
+	if (!tracer)
+		return;
 
-int ds_unchecked_write_bts(struct task_struct *task,
-			   const void *record, size_t size)
-{
-	return ds_write(task, record, size, ds_bts, /* force = */ 1);
-}
+	task = tracer->ds.context->task;
 
-int ds_unchecked_write_pebs(struct task_struct *task,
-			    const void *record, size_t size)
-{
-	return ds_write(task, record, size, ds_pebs, /* force = */ 1);
+	if (!task || (task == current))
+		update_debugctlmsr(get_debugctlmsr() & ~BTS_CONTROL);
+
+	if (task) {
+		task->thread.debugctlmsr &= ~BTS_CONTROL;
+
+		if (!task->thread.debugctlmsr)
+			clear_tsk_thread_flag(task, TIF_DEBUGCTLMSR);
+	}
 }
 
-static int ds_reset_or_clear(struct task_struct *task,
-			     enum ds_qualifier qual, int clear)
+void ds_resume_bts(struct bts_tracer *tracer)
 {
-	struct ds_context *context;
-	unsigned long base, end;
-	int error;
+	struct task_struct *task;
+	unsigned long control;
 
-	context = ds_get_context(task);
-	error = ds_validate_access(context, qual);
-	if (error < 0)
-		goto out;
+	if (!tracer)
+		return;
 
-	base = ds_get(context->ds, qual, ds_buffer_base);
-	end  = ds_get(context->ds, qual, ds_absolute_maximum);
+	task = tracer->ds.context->task;
 
-	if (clear)
-		memset((void *)base, 0, end - base);
+	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];
 
-	ds_set(context->ds, qual, ds_index, base);
+	if (task) {
+		task->thread.debugctlmsr |= control;
+		set_tsk_thread_flag(task, TIF_DEBUGCTLMSR);
+	}
 
-	error = 0;
- out:
-	ds_put_context(context);
-	return error;
+	if (!task || (task == current))
+		update_debugctlmsr(get_debugctlmsr() | control);
 }
 
-int ds_reset_bts(struct task_struct *task)
+void ds_release_pebs(struct pebs_tracer *tracer)
 {
-	return ds_reset_or_clear(task, ds_bts, /* clear = */ 0);
+	if (!tracer)
+		return;
+
+	ds_suspend_pebs(tracer);
+
+	WARN_ON_ONCE(tracer->ds.context->pebs_master != tracer);
+	tracer->ds.context->pebs_master = NULL;
+
+	put_tracer(tracer->ds.context->task);
+	ds_put_context(tracer->ds.context);
+
+	kfree(tracer);
 }
 
-int ds_reset_pebs(struct task_struct *task)
+void ds_suspend_pebs(struct pebs_tracer *tracer)
 {
-	return ds_reset_or_clear(task, ds_pebs, /* clear = */ 0);
+
 }
 
-int ds_clear_bts(struct task_struct *task)
+void ds_resume_pebs(struct pebs_tracer *tracer)
 {
-	return ds_reset_or_clear(task, ds_bts, /* clear = */ 1);
+
 }
 
-int ds_clear_pebs(struct task_struct *task)
+const struct bts_trace *ds_read_bts(struct bts_tracer *tracer)
 {
-	return ds_reset_or_clear(task, ds_pebs, /* clear = */ 1);
+	if (!tracer)
+		return NULL;
+
+	ds_read_config(tracer->ds.context, &tracer->trace.ds, ds_bts);
+	return &tracer->trace;
 }
 
-int ds_get_pebs_reset(struct task_struct *task, u64 *value)
+const struct pebs_trace *ds_read_pebs(struct pebs_tracer *tracer)
 {
-	struct ds_context *context;
-	int error;
+	if (!tracer)
+		return NULL;
+
+	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));
 
-	if (!value)
+	return &tracer->trace;
+}
+
+int ds_reset_bts(struct bts_tracer *tracer)
+{
+	if (!tracer)
 		return -EINVAL;
 
-	context = ds_get_context(task);
-	error = ds_validate_access(context, ds_pebs);
-	if (error < 0)
-		goto out;
+	tracer->trace.ds.top = tracer->trace.ds.begin;
 
-	*value = *(u64 *)(context->ds + (ds_cfg.sizeof_field * 8));
+	ds_set(tracer->ds.context->ds, ds_bts, ds_index,
+	       (unsigned long)tracer->trace.ds.top);
 
-	error = 0;
- out:
-	ds_put_context(context);
-	return error;
+	return 0;
 }
 
-int ds_set_pebs_reset(struct task_struct *task, u64 value)
+int ds_reset_pebs(struct pebs_tracer *tracer)
 {
-	struct ds_context *context;
-	int error;
+	if (!tracer)
+		return -EINVAL;
 
-	context = ds_get_context(task);
-	error = ds_validate_access(context, ds_pebs);
-	if (error < 0)
-		goto out;
+	tracer->trace.ds.top = tracer->trace.ds.begin;
 
-	*(u64 *)(context->ds + (ds_cfg.sizeof_field * 8)) = value;
+	ds_set(tracer->ds.context->ds, ds_bts, ds_index,
+	       (unsigned long)tracer->trace.ds.top);
 
-	error = 0;
- out:
-	ds_put_context(context);
-	return error;
+	return 0;
+}
+
+int ds_set_pebs_reset(struct pebs_tracer *tracer, u64 value)
+{
+	if (!tracer)
+		return -EINVAL;
+
+	*(u64 *)(tracer->ds.context->ds + (ds_cfg.sizeof_field * 8)) = 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] 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");
+
+	WARN_ON_ONCE(MAX_SIZEOF_DS < (12 * ds_cfg.sizeof_field));
 }
 
 void __cpuinit ds_init_intel(struct cpuinfo_x86 *c)
@@ -847,16 +949,15 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c)
 	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 */
-			ds_configure(&ds_cfg_var);
+			ds_configure(&ds_cfg_pentium_m);
 			break;
-		case 0xF: /* Core2 */
-		case 0x1C: /* Atom */
-			ds_configure(&ds_cfg_64);
-			break;
-		default:
-			/* sorry, don't know about them */
+		default: /* Core2, Atom, ... */
+			ds_configure(&ds_cfg_core2);
 			break;
 		}
 		break;
@@ -865,7 +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 */
@@ -878,12 +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)
+{
+	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)
 {
-	/* 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--)
-		ds_put_context(context);
+	WARN_ON(tsk->thread.ds_ctx);
 }
diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c
new file mode 100644
index 00000000000..6b1f6f6f866
--- /dev/null
+++ b/arch/x86/kernel/dumpstack.c
@@ -0,0 +1,351 @@
+/*
+ *  Copyright (C) 1991, 1992  Linus Torvalds
+ *  Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs
+ */
+#include <linux/kallsyms.h>
+#include <linux/kprobes.h>
+#include <linux/uaccess.h>
+#include <linux/utsname.h>
+#include <linux/hardirq.h>
+#include <linux/kdebug.h>
+#include <linux/module.h>
+#include <linux/ptrace.h>
+#include <linux/kexec.h>
+#include <linux/bug.h>
+#include <linux/nmi.h>
+#include <linux/sysfs.h>
+
+#include <asm/stacktrace.h>
+
+#include "dumpstack.h"
+
+int panic_on_unrecovered_nmi;
+unsigned int code_bytes = 64;
+int kstack_depth_to_print = 3 * STACKSLOTS_PER_LINE;
+static int die_counter;
+
+void printk_address(unsigned long address, int reliable)
+{
+	printk(" [<%p>] %s%pS\n", (void *) address,
+			reliable ? "" : "? ", (void *) address);
+}
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+static void
+print_ftrace_graph_addr(unsigned long addr, void *data,
+			const struct stacktrace_ops *ops,
+			struct thread_info *tinfo, int *graph)
+{
+	struct task_struct *task = tinfo->task;
+	unsigned long ret_addr;
+	int index = task->curr_ret_stack;
+
+	if (addr != (unsigned long)return_to_handler)
+		return;
+
+	if (!task->ret_stack || index < *graph)
+		return;
+
+	index -= *graph;
+	ret_addr = task->ret_stack[index].ret;
+
+	ops->address(data, ret_addr, 1);
+
+	(*graph)++;
+}
+#else
+static inline void
+print_ftrace_graph_addr(unsigned long addr, void *data,
+			const struct stacktrace_ops *ops,
+			struct thread_info *tinfo, int *graph)
+{ }
+#endif
+
+/*
+ * x86-64 can have up to three kernel stacks:
+ * process stack
+ * interrupt stack
+ * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack
+ */
+
+static inline int valid_stack_ptr(struct thread_info *tinfo,
+			void *p, unsigned int size, void *end)
+{
+	void *t = tinfo;
+	if (end) {
+		if (p < end && p >= (end-THREAD_SIZE))
+			return 1;
+		else
+			return 0;
+	}
+	return p > t && p < t + THREAD_SIZE - size;
+}
+
+unsigned long
+print_context_stack(struct thread_info *tinfo,
+		unsigned long *stack, unsigned long bp,
+		const struct stacktrace_ops *ops, void *data,
+		unsigned long *end, int *graph)
+{
+	struct stack_frame *frame = (struct stack_frame *)bp;
+
+	while (valid_stack_ptr(tinfo, stack, sizeof(*stack), end)) {
+		unsigned long addr;
+
+		addr = *stack;
+		if (__kernel_text_address(addr)) {
+			if ((unsigned long) stack == bp + sizeof(long)) {
+				ops->address(data, addr, 1);
+				frame = frame->next_frame;
+				bp = (unsigned long) frame;
+			} else {
+				ops->address(data, addr, bp == 0);
+			}
+			print_ftrace_graph_addr(addr, data, ops, tinfo, graph);
+		}
+		stack++;
+	}
+	return bp;
+}
+
+
+static void
+print_trace_warning_symbol(void *data, char *msg, unsigned long symbol)
+{
+	printk(data);
+	print_symbol(msg, symbol);
+	printk("\n");
+}
+
+static void print_trace_warning(void *data, char *msg)
+{
+	printk("%s%s\n", (char *)data, msg);
+}
+
+static int print_trace_stack(void *data, char *name)
+{
+	printk("%s <%s> ", (char *)data, name);
+	return 0;
+}
+
+/*
+ * Print one address/symbol entries per line.
+ */
+static void print_trace_address(void *data, unsigned long addr, int reliable)
+{
+	touch_nmi_watchdog();
+	printk(data);
+	printk_address(addr, reliable);
+}
+
+static const struct stacktrace_ops print_trace_ops = {
+	.warning = print_trace_warning,
+	.warning_symbol = print_trace_warning_symbol,
+	.stack = print_trace_stack,
+	.address = print_trace_address,
+};
+
+void
+show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
+		unsigned long *stack, unsigned long bp, char *log_lvl)
+{
+	printk("%sCall Trace:\n", log_lvl);
+	dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl);
+}
+
+void show_trace(struct task_struct *task, struct pt_regs *regs,
+		unsigned long *stack, unsigned long bp)
+{
+	show_trace_log_lvl(task, regs, stack, bp, "");
+}
+
+void show_stack(struct task_struct *task, unsigned long *sp)
+{
+	show_stack_log_lvl(task, NULL, sp, 0, "");
+}
+
+/*
+ * The architecture-independent dump_stack generator
+ */
+void dump_stack(void)
+{
+	unsigned long bp = 0;
+	unsigned long stack;
+
+#ifdef CONFIG_FRAME_POINTER
+	if (!bp)
+		get_bp(bp);
+#endif
+
+	printk("Pid: %d, comm: %.20s %s %s %.*s\n",
+		current->pid, current->comm, print_tainted(),
+		init_utsname()->release,
+		(int)strcspn(init_utsname()->version, " "),
+		init_utsname()->version);
+	show_trace(NULL, NULL, &stack, bp);
+}
+EXPORT_SYMBOL(dump_stack);
+
+static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED;
+static int die_owner = -1;
+static unsigned int die_nest_count;
+
+unsigned __kprobes long oops_begin(void)
+{
+	int cpu;
+	unsigned long flags;
+
+	oops_enter();
+
+	/* racy, but better than risking deadlock. */
+	raw_local_irq_save(flags);
+	cpu = smp_processor_id();
+	if (!__raw_spin_trylock(&die_lock)) {
+		if (cpu == die_owner)
+			/* nested oops. should stop eventually */;
+		else
+			__raw_spin_lock(&die_lock);
+	}
+	die_nest_count++;
+	die_owner = cpu;
+	console_verbose();
+	bust_spinlocks(1);
+	return flags;
+}
+
+void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr)
+{
+	if (regs && kexec_should_crash(current))
+		crash_kexec(regs);
+
+	bust_spinlocks(0);
+	die_owner = -1;
+	add_taint(TAINT_DIE);
+	die_nest_count--;
+	if (!die_nest_count)
+		/* Nest count reaches zero, release the lock. */
+		__raw_spin_unlock(&die_lock);
+	raw_local_irq_restore(flags);
+	oops_exit();
+
+	if (!signr)
+		return;
+	if (in_interrupt())
+		panic("Fatal exception in interrupt");
+	if (panic_on_oops)
+		panic("Fatal exception");
+	do_exit(signr);
+}
+
+int __kprobes __die(const char *str, struct pt_regs *regs, long err)
+{
+#ifdef CONFIG_X86_32
+	unsigned short ss;
+	unsigned long sp;
+#endif
+	printk(KERN_EMERG "%s: %04lx [#%d] ", str, err & 0xffff, ++die_counter);
+#ifdef CONFIG_PREEMPT
+	printk("PREEMPT ");
+#endif
+#ifdef CONFIG_SMP
+	printk("SMP ");
+#endif
+#ifdef CONFIG_DEBUG_PAGEALLOC
+	printk("DEBUG_PAGEALLOC");
+#endif
+	printk("\n");
+	sysfs_printk_last_file();
+	if (notify_die(DIE_OOPS, str, regs, err,
+			current->thread.trap_no, SIGSEGV) == NOTIFY_STOP)
+		return 1;
+
+	show_registers(regs);
+#ifdef CONFIG_X86_32
+	sp = (unsigned long) (&regs->sp);
+	savesegment(ss, ss);
+	if (user_mode(regs)) {
+		sp = regs->sp;
+		ss = regs->ss & 0xffff;
+	}
+	printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip);
+	print_symbol("%s", regs->ip);
+	printk(" SS:ESP %04x:%08lx\n", ss, sp);
+#else
+	/* Executive summary in case the oops scrolled away */
+	printk(KERN_ALERT "RIP ");
+	printk_address(regs->ip, 1);
+	printk(" RSP <%016lx>\n", regs->sp);
+#endif
+	return 0;
+}
+
+/*
+ * This is gone through when something in the kernel has done something bad
+ * and is about to be terminated:
+ */
+void die(const char *str, struct pt_regs *regs, long err)
+{
+	unsigned long flags = oops_begin();
+	int sig = SIGSEGV;
+
+	if (!user_mode_vm(regs))
+		report_bug(regs->ip, regs);
+
+	if (__die(str, regs, err))
+		sig = 0;
+	oops_end(flags, regs, sig);
+}
+
+void notrace __kprobes
+die_nmi(char *str, struct pt_regs *regs, int do_panic)
+{
+	unsigned long flags;
+
+	if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP)
+		return;
+
+	/*
+	 * We are in trouble anyway, lets at least try
+	 * to get a message out.
+	 */
+	flags = oops_begin();
+	printk(KERN_EMERG "%s", str);
+	printk(" on CPU%d, ip %08lx, registers:\n",
+		smp_processor_id(), regs->ip);
+	show_registers(regs);
+	oops_end(flags, regs, 0);
+	if (do_panic || panic_on_oops)
+		panic("Non maskable interrupt");
+	nmi_exit();
+	local_irq_enable();
+	do_exit(SIGBUS);
+}
+
+static int __init oops_setup(char *s)
+{
+	if (!s)
+		return -EINVAL;
+	if (!strcmp(s, "panic"))
+		panic_on_oops = 1;
+	return 0;
+}
+early_param("oops", oops_setup);
+
+static int __init kstack_setup(char *s)
+{
+	if (!s)
+		return -EINVAL;
+	kstack_depth_to_print = simple_strtoul(s, NULL, 0);
+	return 0;
+}
+early_param("kstack", kstack_setup);
+
+static int __init code_bytes_setup(char *s)
+{
+	code_bytes = simple_strtoul(s, NULL, 0);
+	if (code_bytes > 8192)
+		code_bytes = 8192;
+
+	return 1;
+}
+__setup("code_bytes=", code_bytes_setup);
diff --git a/arch/x86/kernel/dumpstack.h b/arch/x86/kernel/dumpstack.h
new file mode 100644
index 00000000000..da87590b869
--- /dev/null
+++ b/arch/x86/kernel/dumpstack.h
@@ -0,0 +1,39 @@
+/*
+ *  Copyright (C) 1991, 1992  Linus Torvalds
+ *  Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs
+ */
+
+#ifndef DUMPSTACK_H
+#define DUMPSTACK_H
+
+#ifdef CONFIG_X86_32
+#define STACKSLOTS_PER_LINE 8
+#define get_bp(bp) asm("movl %%ebp, %0" : "=r" (bp) :)
+#else
+#define STACKSLOTS_PER_LINE 4
+#define get_bp(bp) asm("movq %%rbp, %0" : "=r" (bp) :)
+#endif
+
+extern unsigned long
+print_context_stack(struct thread_info *tinfo,
+		unsigned long *stack, unsigned long bp,
+		const struct stacktrace_ops *ops, void *data,
+		unsigned long *end, int *graph);
+
+extern void
+show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
+		unsigned long *stack, unsigned long bp, char *log_lvl);
+
+extern void
+show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
+		unsigned long *sp, unsigned long bp, char *log_lvl);
+
+extern unsigned int code_bytes;
+extern int kstack_depth_to_print;
+
+/* The form of the top of the frame on the stack */
+struct stack_frame {
+	struct stack_frame *next_frame;
+	unsigned long return_address;
+};
+#endif
diff --git a/arch/x86/kernel/dumpstack_32.c b/arch/x86/kernel/dumpstack_32.c
index b3614752197..d593cd1f58d 100644
--- a/arch/x86/kernel/dumpstack_32.c
+++ b/arch/x86/kernel/dumpstack_32.c
@@ -17,69 +17,14 @@
 
 #include <asm/stacktrace.h>
 
-#define STACKSLOTS_PER_LINE 8
-#define get_bp(bp) asm("movl %%ebp, %0" : "=r" (bp) :)
-
-int panic_on_unrecovered_nmi;
-int kstack_depth_to_print = 3 * STACKSLOTS_PER_LINE;
-static unsigned int code_bytes = 64;
-static int die_counter;
-
-void printk_address(unsigned long address, int reliable)
-{
-	printk(" [<%p>] %s%pS\n", (void *) address,
-			reliable ? "" : "? ", (void *) address);
-}
-
-static inline int valid_stack_ptr(struct thread_info *tinfo,
-			void *p, unsigned int size, void *end)
-{
-	void *t = tinfo;
-	if (end) {
-		if (p < end && p >= (end-THREAD_SIZE))
-			return 1;
-		else
-			return 0;
-	}
-	return p > t && p < t + THREAD_SIZE - size;
-}
-
-/* The form of the top of the frame on the stack */
-struct stack_frame {
-	struct stack_frame *next_frame;
-	unsigned long return_address;
-};
-
-static inline unsigned long
-print_context_stack(struct thread_info *tinfo,
-		unsigned long *stack, unsigned long bp,
-		const struct stacktrace_ops *ops, void *data,
-		unsigned long *end)
-{
-	struct stack_frame *frame = (struct stack_frame *)bp;
-
-	while (valid_stack_ptr(tinfo, stack, sizeof(*stack), end)) {
-		unsigned long addr;
-
-		addr = *stack;
-		if (__kernel_text_address(addr)) {
-			if ((unsigned long) stack == bp + sizeof(long)) {
-				ops->address(data, addr, 1);
-				frame = frame->next_frame;
-				bp = (unsigned long) frame;
-			} else {
-				ops->address(data, addr, bp == 0);
-			}
-		}
-		stack++;
-	}
-	return bp;
-}
+#include "dumpstack.h"
 
 void dump_trace(struct task_struct *task, struct pt_regs *regs,
 		unsigned long *stack, unsigned long bp,
 		const struct stacktrace_ops *ops, void *data)
 {
+	int graph = 0;
+
 	if (!task)
 		task = current;
 
@@ -107,7 +52,8 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 
 		context = (struct thread_info *)
 			((unsigned long)stack & (~(THREAD_SIZE - 1)));
-		bp = print_context_stack(context, stack, bp, ops, data, NULL);
+		bp = print_context_stack(context, stack, bp, ops,
+					 data, NULL, &graph);
 
 		stack = (unsigned long *)context->previous_esp;
 		if (!stack)
@@ -119,57 +65,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 }
 EXPORT_SYMBOL(dump_trace);
 
-static void
-print_trace_warning_symbol(void *data, char *msg, unsigned long symbol)
-{
-	printk(data);
-	print_symbol(msg, symbol);
-	printk("\n");
-}
-
-static void print_trace_warning(void *data, char *msg)
-{
-	printk("%s%s\n", (char *)data, msg);
-}
-
-static int print_trace_stack(void *data, char *name)
-{
-	printk("%s <%s> ", (char *)data, name);
-	return 0;
-}
-
-/*
- * Print one address/symbol entries per line.
- */
-static void print_trace_address(void *data, unsigned long addr, int reliable)
-{
-	touch_nmi_watchdog();
-	printk(data);
-	printk_address(addr, reliable);
-}
-
-static const struct stacktrace_ops print_trace_ops = {
-	.warning = print_trace_warning,
-	.warning_symbol = print_trace_warning_symbol,
-	.stack = print_trace_stack,
-	.address = print_trace_address,
-};
-
-static void
-show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
-		unsigned long *stack, unsigned long bp, char *log_lvl)
-{
-	printk("%sCall Trace:\n", log_lvl);
-	dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl);
-}
-
-void show_trace(struct task_struct *task, struct pt_regs *regs,
-		unsigned long *stack, unsigned long bp)
-{
-	show_trace_log_lvl(task, regs, stack, bp, "");
-}
-
-static void
+void
 show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
 		unsigned long *sp, unsigned long bp, char *log_lvl)
 {
@@ -196,33 +92,6 @@ show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
 	show_trace_log_lvl(task, regs, sp, bp, log_lvl);
 }
 
-void show_stack(struct task_struct *task, unsigned long *sp)
-{
-	show_stack_log_lvl(task, NULL, sp, 0, "");
-}
-
-/*
- * The architecture-independent dump_stack generator
- */
-void dump_stack(void)
-{
-	unsigned long bp = 0;
-	unsigned long stack;
-
-#ifdef CONFIG_FRAME_POINTER
-	if (!bp)
-		get_bp(bp);
-#endif
-
-	printk("Pid: %d, comm: %.20s %s %s %.*s\n",
-		current->pid, current->comm, print_tainted(),
-		init_utsname()->release,
-		(int)strcspn(init_utsname()->version, " "),
-		init_utsname()->version);
-	show_trace(NULL, NULL, &stack, bp);
-}
-
-EXPORT_SYMBOL(dump_stack);
 
 void show_registers(struct pt_regs *regs)
 {
@@ -283,167 +152,3 @@ int is_valid_bugaddr(unsigned long ip)
 	return ud2 == 0x0b0f;
 }
 
-static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED;
-static int die_owner = -1;
-static unsigned int die_nest_count;
-
-unsigned __kprobes long oops_begin(void)
-{
-	unsigned long flags;
-
-	oops_enter();
-
-	if (die_owner != raw_smp_processor_id()) {
-		console_verbose();
-		raw_local_irq_save(flags);
-		__raw_spin_lock(&die_lock);
-		die_owner = smp_processor_id();
-		die_nest_count = 0;
-		bust_spinlocks(1);
-	} else {
-		raw_local_irq_save(flags);
-	}
-	die_nest_count++;
-	return flags;
-}
-
-void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr)
-{
-	bust_spinlocks(0);
-	die_owner = -1;
-	add_taint(TAINT_DIE);
-	__raw_spin_unlock(&die_lock);
-	raw_local_irq_restore(flags);
-
-	if (!regs)
-		return;
-
-	if (kexec_should_crash(current))
-		crash_kexec(regs);
-	if (in_interrupt())
-		panic("Fatal exception in interrupt");
-	if (panic_on_oops)
-		panic("Fatal exception");
-	oops_exit();
-	do_exit(signr);
-}
-
-int __kprobes __die(const char *str, struct pt_regs *regs, long err)
-{
-	unsigned short ss;
-	unsigned long sp;
-
-	printk(KERN_EMERG "%s: %04lx [#%d] ", str, err & 0xffff, ++die_counter);
-#ifdef CONFIG_PREEMPT
-	printk("PREEMPT ");
-#endif
-#ifdef CONFIG_SMP
-	printk("SMP ");
-#endif
-#ifdef CONFIG_DEBUG_PAGEALLOC
-	printk("DEBUG_PAGEALLOC");
-#endif
-	printk("\n");
-	sysfs_printk_last_file();
-	if (notify_die(DIE_OOPS, str, regs, err,
-			current->thread.trap_no, SIGSEGV) == NOTIFY_STOP)
-		return 1;
-
-	show_registers(regs);
-	/* Executive summary in case the oops scrolled away */
-	sp = (unsigned long) (&regs->sp);
-	savesegment(ss, ss);
-	if (user_mode(regs)) {
-		sp = regs->sp;
-		ss = regs->ss & 0xffff;
-	}
-	printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip);
-	print_symbol("%s", regs->ip);
-	printk(" SS:ESP %04x:%08lx\n", ss, sp);
-	return 0;
-}
-
-/*
- * This is gone through when something in the kernel has done something bad
- * and is about to be terminated:
- */
-void die(const char *str, struct pt_regs *regs, long err)
-{
-	unsigned long flags = oops_begin();
-
-	if (die_nest_count < 3) {
-		report_bug(regs->ip, regs);
-
-		if (__die(str, regs, err))
-			regs = NULL;
-	} else {
-		printk(KERN_EMERG "Recursive die() failure, output suppressed\n");
-	}
-
-	oops_end(flags, regs, SIGSEGV);
-}
-
-static DEFINE_SPINLOCK(nmi_print_lock);
-
-void notrace __kprobes
-die_nmi(char *str, struct pt_regs *regs, int do_panic)
-{
-	if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP)
-		return;
-
-	spin_lock(&nmi_print_lock);
-	/*
-	* We are in trouble anyway, lets at least try
-	* to get a message out:
-	*/
-	bust_spinlocks(1);
-	printk(KERN_EMERG "%s", str);
-	printk(" on CPU%d, ip %08lx, registers:\n",
-		smp_processor_id(), regs->ip);
-	show_registers(regs);
-	if (do_panic)
-		panic("Non maskable interrupt");
-	console_silent();
-	spin_unlock(&nmi_print_lock);
-
-	/*
-	 * If we are in kernel we are probably nested up pretty bad
-	 * and might aswell get out now while we still can:
-	 */
-	if (!user_mode_vm(regs)) {
-		current->thread.trap_no = 2;
-		crash_kexec(regs);
-	}
-
-	bust_spinlocks(0);
-	do_exit(SIGSEGV);
-}
-
-static int __init oops_setup(char *s)
-{
-	if (!s)
-		return -EINVAL;
-	if (!strcmp(s, "panic"))
-		panic_on_oops = 1;
-	return 0;
-}
-early_param("oops", oops_setup);
-
-static int __init kstack_setup(char *s)
-{
-	if (!s)
-		return -EINVAL;
-	kstack_depth_to_print = simple_strtoul(s, NULL, 0);
-	return 0;
-}
-early_param("kstack", kstack_setup);
-
-static int __init code_bytes_setup(char *s)
-{
-	code_bytes = simple_strtoul(s, NULL, 0);
-	if (code_bytes > 8192)
-		code_bytes = 8192;
-
-	return 1;
-}
-__setup("code_bytes=", code_bytes_setup);
diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c
index 96a5db7da8a..c302d070704 100644
--- a/arch/x86/kernel/dumpstack_64.c
+++ b/arch/x86/kernel/dumpstack_64.c
@@ -17,19 +17,7 @@
 
 #include <asm/stacktrace.h>
 
-#define STACKSLOTS_PER_LINE 4
-#define get_bp(bp) asm("movq %%rbp, %0" : "=r" (bp) :)
-
-int panic_on_unrecovered_nmi;
-int kstack_depth_to_print = 3 * STACKSLOTS_PER_LINE;
-static unsigned int code_bytes = 64;
-static int die_counter;
-
-void printk_address(unsigned long address, int reliable)
-{
-	printk(" [<%p>] %s%pS\n", (void *) address,
-			reliable ? "" : "? ", (void *) address);
-}
+#include "dumpstack.h"
 
 static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
 					unsigned *usedp, char **idp)
@@ -113,51 +101,6 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
  * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack
  */
 
-static inline int valid_stack_ptr(struct thread_info *tinfo,
-			void *p, unsigned int size, void *end)
-{
-	void *t = tinfo;
-	if (end) {
-		if (p < end && p >= (end-THREAD_SIZE))
-			return 1;
-		else
-			return 0;
-	}
-	return p > t && p < t + THREAD_SIZE - size;
-}
-
-/* The form of the top of the frame on the stack */
-struct stack_frame {
-	struct stack_frame *next_frame;
-	unsigned long return_address;
-};
-
-static inline unsigned long
-print_context_stack(struct thread_info *tinfo,
-		unsigned long *stack, unsigned long bp,
-		const struct stacktrace_ops *ops, void *data,
-		unsigned long *end)
-{
-	struct stack_frame *frame = (struct stack_frame *)bp;
-
-	while (valid_stack_ptr(tinfo, stack, sizeof(*stack), end)) {
-		unsigned long addr;
-
-		addr = *stack;
-		if (__kernel_text_address(addr)) {
-			if ((unsigned long) stack == bp + sizeof(long)) {
-				ops->address(data, addr, 1);
-				frame = frame->next_frame;
-				bp = (unsigned long) frame;
-			} else {
-				ops->address(data, addr, bp == 0);
-			}
-		}
-		stack++;
-	}
-	return bp;
-}
-
 void dump_trace(struct task_struct *task, struct pt_regs *regs,
 		unsigned long *stack, unsigned long bp,
 		const struct stacktrace_ops *ops, void *data)
@@ -166,6 +109,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 	unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr;
 	unsigned used = 0;
 	struct thread_info *tinfo;
+	int graph = 0;
 
 	if (!task)
 		task = current;
@@ -206,7 +150,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 				break;
 
 			bp = print_context_stack(tinfo, stack, bp, ops,
-							data, estack_end);
+						 data, estack_end, &graph);
 			ops->stack(data, "<EOE>");
 			/*
 			 * We link to the next stack via the
@@ -225,7 +169,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 				if (ops->stack(data, "IRQ") < 0)
 					break;
 				bp = print_context_stack(tinfo, stack, bp,
-						ops, data, irqstack_end);
+					ops, data, irqstack_end, &graph);
 				/*
 				 * We link to the next stack (which would be
 				 * the process stack normally) the last
@@ -243,62 +187,12 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 	/*
 	 * This handles the process stack:
 	 */
-	bp = print_context_stack(tinfo, stack, bp, ops, data, NULL);
+	bp = print_context_stack(tinfo, stack, bp, ops, data, NULL, &graph);
 	put_cpu();
 }
 EXPORT_SYMBOL(dump_trace);
 
-static void
-print_trace_warning_symbol(void *data, char *msg, unsigned long symbol)
-{
-	printk(data);
-	print_symbol(msg, symbol);
-	printk("\n");
-}
-
-static void print_trace_warning(void *data, char *msg)
-{
-	printk("%s%s\n", (char *)data, msg);
-}
-
-static int print_trace_stack(void *data, char *name)
-{
-	printk("%s <%s> ", (char *)data, name);
-	return 0;
-}
-
-/*
- * Print one address/symbol entries per line.
- */
-static void print_trace_address(void *data, unsigned long addr, int reliable)
-{
-	touch_nmi_watchdog();
-	printk(data);
-	printk_address(addr, reliable);
-}
-
-static const struct stacktrace_ops print_trace_ops = {
-	.warning = print_trace_warning,
-	.warning_symbol = print_trace_warning_symbol,
-	.stack = print_trace_stack,
-	.address = print_trace_address,
-};
-
-static void
-show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
-		unsigned long *stack, unsigned long bp, char *log_lvl)
-{
-	printk("%sCall Trace:\n", log_lvl);
-	dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl);
-}
-
-void show_trace(struct task_struct *task, struct pt_regs *regs,
-		unsigned long *stack, unsigned long bp)
-{
-	show_trace_log_lvl(task, regs, stack, bp, "");
-}
-
-static void
+void
 show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
 		unsigned long *sp, unsigned long bp, char *log_lvl)
 {
@@ -342,33 +236,6 @@ show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
 	show_trace_log_lvl(task, regs, sp, bp, log_lvl);
 }
 
-void show_stack(struct task_struct *task, unsigned long *sp)
-{
-	show_stack_log_lvl(task, NULL, sp, 0, "");
-}
-
-/*
- * The architecture-independent dump_stack generator
- */
-void dump_stack(void)
-{
-	unsigned long bp = 0;
-	unsigned long stack;
-
-#ifdef CONFIG_FRAME_POINTER
-	if (!bp)
-		get_bp(bp);
-#endif
-
-	printk("Pid: %d, comm: %.20s %s %s %.*s\n",
-		current->pid, current->comm, print_tainted(),
-		init_utsname()->release,
-		(int)strcspn(init_utsname()->version, " "),
-		init_utsname()->version);
-	show_trace(NULL, NULL, &stack, bp);
-}
-EXPORT_SYMBOL(dump_stack);
-
 void show_registers(struct pt_regs *regs)
 {
 	int i;
@@ -429,147 +296,3 @@ int is_valid_bugaddr(unsigned long ip)
 	return ud2 == 0x0b0f;
 }
 
-static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED;
-static int die_owner = -1;
-static unsigned int die_nest_count;
-
-unsigned __kprobes long oops_begin(void)
-{
-	int cpu;
-	unsigned long flags;
-
-	oops_enter();
-
-	/* racy, but better than risking deadlock. */
-	raw_local_irq_save(flags);
-	cpu = smp_processor_id();
-	if (!__raw_spin_trylock(&die_lock)) {
-		if (cpu == die_owner)
-			/* nested oops. should stop eventually */;
-		else
-			__raw_spin_lock(&die_lock);
-	}
-	die_nest_count++;
-	die_owner = cpu;
-	console_verbose();
-	bust_spinlocks(1);
-	return flags;
-}
-
-void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr)
-{
-	die_owner = -1;
-	bust_spinlocks(0);
-	die_nest_count--;
-	if (!die_nest_count)
-		/* Nest count reaches zero, release the lock. */
-		__raw_spin_unlock(&die_lock);
-	raw_local_irq_restore(flags);
-	if (!regs) {
-		oops_exit();
-		return;
-	}
-	if (in_interrupt())
-		panic("Fatal exception in interrupt");
-	if (panic_on_oops)
-		panic("Fatal exception");
-	oops_exit();
-	do_exit(signr);
-}
-
-int __kprobes __die(const char *str, struct pt_regs *regs, long err)
-{
-	printk(KERN_EMERG "%s: %04lx [#%d] ", str, err & 0xffff, ++die_counter);
-#ifdef CONFIG_PREEMPT
-	printk("PREEMPT ");
-#endif
-#ifdef CONFIG_SMP
-	printk("SMP ");
-#endif
-#ifdef CONFIG_DEBUG_PAGEALLOC
-	printk("DEBUG_PAGEALLOC");
-#endif
-	printk("\n");
-	sysfs_printk_last_file();
-	if (notify_die(DIE_OOPS, str, regs, err,
-			current->thread.trap_no, SIGSEGV) == NOTIFY_STOP)
-		return 1;
-
-	show_registers(regs);
-	add_taint(TAINT_DIE);
-	/* Executive summary in case the oops scrolled away */
-	printk(KERN_ALERT "RIP ");
-	printk_address(regs->ip, 1);
-	printk(" RSP <%016lx>\n", regs->sp);
-	if (kexec_should_crash(current))
-		crash_kexec(regs);
-	return 0;
-}
-
-void die(const char *str, struct pt_regs *regs, long err)
-{
-	unsigned long flags = oops_begin();
-
-	if (!user_mode(regs))
-		report_bug(regs->ip, regs);
-
-	if (__die(str, regs, err))
-		regs = NULL;
-	oops_end(flags, regs, SIGSEGV);
-}
-
-notrace __kprobes void
-die_nmi(char *str, struct pt_regs *regs, int do_panic)
-{
-	unsigned long flags;
-
-	if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP)
-		return;
-
-	flags = oops_begin();
-	/*
-	 * We are in trouble anyway, lets at least try
-	 * to get a message out.
-	 */
-	printk(KERN_EMERG "%s", str);
-	printk(" on CPU%d, ip %08lx, registers:\n",
-		smp_processor_id(), regs->ip);
-	show_registers(regs);
-	if (kexec_should_crash(current))
-		crash_kexec(regs);
-	if (do_panic || panic_on_oops)
-		panic("Non maskable interrupt");
-	oops_end(flags, NULL, SIGBUS);
-	nmi_exit();
-	local_irq_enable();
-	do_exit(SIGBUS);
-}
-
-static int __init oops_setup(char *s)
-{
-	if (!s)
-		return -EINVAL;
-	if (!strcmp(s, "panic"))
-		panic_on_oops = 1;
-	return 0;
-}
-early_param("oops", oops_setup);
-
-static int __init kstack_setup(char *s)
-{
-	if (!s)
-		return -EINVAL;
-	kstack_depth_to_print = simple_strtoul(s, NULL, 0);
-	return 0;
-}
-early_param("kstack", kstack_setup);
-
-static int __init code_bytes_setup(char *s)
-{
-	code_bytes = simple_strtoul(s, NULL, 0);
-	if (code_bytes > 8192)
-		code_bytes = 8192;
-
-	return 1;
-}
-__setup("code_bytes=", code_bytes_setup);
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 7aafeb5263e..65a13943e09 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 1b894b72c0f..744aa7fc49d 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 34ad997d383..23b138e31e9 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 fe7014176eb..d6f0490a739 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -954,6 +954,9 @@ ENTRY(mcount)
 END(mcount)
 
 ENTRY(ftrace_caller)
+	cmpl $0, function_trace_stop
+	jne  ftrace_stub
+
 	pushl %eax
 	pushl %ecx
 	pushl %edx
@@ -968,6 +971,11 @@ ftrace_call:
 	popl %edx
 	popl %ecx
 	popl %eax
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+.globl ftrace_graph_call
+ftrace_graph_call:
+	jmp ftrace_stub
+#endif
 
 .globl ftrace_stub
 ftrace_stub:
@@ -977,8 +985,18 @@ END(ftrace_caller)
 #else /* ! CONFIG_DYNAMIC_FTRACE */
 
 ENTRY(mcount)
+	cmpl $0, function_trace_stop
+	jne  ftrace_stub
+
 	cmpl $ftrace_stub, ftrace_trace_function
 	jnz trace
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+	cmpl $ftrace_stub, ftrace_graph_return
+	jnz ftrace_graph_caller
+
+	cmpl $ftrace_graph_entry_stub, ftrace_graph_entry
+	jnz ftrace_graph_caller
+#endif
 .globl ftrace_stub
 ftrace_stub:
 	ret
@@ -997,12 +1015,43 @@ trace:
 	popl %edx
 	popl %ecx
 	popl %eax
-
 	jmp ftrace_stub
 END(mcount)
 #endif /* CONFIG_DYNAMIC_FTRACE */
 #endif /* CONFIG_FUNCTION_TRACER */
 
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+ENTRY(ftrace_graph_caller)
+	cmpl $0, function_trace_stop
+	jne ftrace_stub
+
+	pushl %eax
+	pushl %ecx
+	pushl %edx
+	movl 0xc(%esp), %edx
+	lea 0x4(%ebp), %eax
+	subl $MCOUNT_INSN_SIZE, %edx
+	call prepare_ftrace_return
+	popl %edx
+	popl %ecx
+	popl %eax
+	ret
+END(ftrace_graph_caller)
+
+.globl return_to_handler
+return_to_handler:
+	pushl $0
+	pushl %eax
+	pushl %ecx
+	pushl %edx
+	call ftrace_return_to_handler
+	movl %eax, 0xc(%esp)
+	popl %edx
+	popl %ecx
+	popl %eax
+	ret
+#endif
+
 .section .rodata,"a"
 #include "syscall_table_32.S"
 
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index fc013cfde30..1954a966220 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -67,16 +67,10 @@ ENTRY(mcount)
 END(mcount)
 
 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
@@ -86,14 +80,13 @@ 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
+ftrace_graph_call:
+	jmp ftrace_stub
+#endif
 
 .globl ftrace_stub
 ftrace_stub:
@@ -102,15 +95,63 @@ END(ftrace_caller)
 
 #else /* ! CONFIG_DYNAMIC_FTRACE */
 ENTRY(mcount)
+	cmpl $0, function_trace_stop
+	jne  ftrace_stub
+
 	cmpq $ftrace_stub, ftrace_trace_function
 	jnz trace
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+	cmpq $ftrace_stub, ftrace_graph_return
+	jnz ftrace_graph_caller
+
+	cmpq $ftrace_graph_entry_stub, ftrace_graph_entry
+	jnz ftrace_graph_caller
+#endif
+
 .globl ftrace_stub
 ftrace_stub:
 	retq
 
 trace:
-	/* taken from glibc */
-	subq $0x38, %rsp
+	MCOUNT_SAVE_FRAME
+
+	movq 0x38(%rsp), %rdi
+	movq 8(%rbp), %rsi
+	subq $MCOUNT_INSN_SIZE, %rdi
+
+	call   *ftrace_trace_function
+
+	MCOUNT_RESTORE_FRAME
+
+	jmp ftrace_stub
+END(mcount)
+#endif /* CONFIG_DYNAMIC_FTRACE */
+#endif /* CONFIG_FUNCTION_TRACER */
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+ENTRY(ftrace_graph_caller)
+	cmpl $0, function_trace_stop
+	jne ftrace_stub
+
+	MCOUNT_SAVE_FRAME
+
+	leaq 8(%rbp), %rdi
+	movq 0x38(%rsp), %rsi
+	subq $MCOUNT_INSN_SIZE, %rsi
+
+	call	prepare_ftrace_return
+
+	MCOUNT_RESTORE_FRAME
+
+	retq
+END(ftrace_graph_caller)
+
+
+.globl return_to_handler
+return_to_handler:
+	subq  $80, %rsp
+
 	movq %rax, (%rsp)
 	movq %rcx, 8(%rsp)
 	movq %rdx, 16(%rsp)
@@ -118,13 +159,14 @@ trace:
 	movq %rdi, 32(%rsp)
 	movq %r8, 40(%rsp)
 	movq %r9, 48(%rsp)
+	movq %r10, 56(%rsp)
+	movq %r11, 64(%rsp)
 
-	movq 0x38(%rsp), %rdi
-	movq 8(%rbp), %rsi
-	subq $MCOUNT_INSN_SIZE, %rdi
-
-	call   *ftrace_trace_function
+	call ftrace_return_to_handler
 
+	movq %rax, 72(%rsp)
+	movq 64(%rsp), %r11
+	movq 56(%rsp), %r10
 	movq 48(%rsp), %r9
 	movq 40(%rsp), %r8
 	movq 32(%rsp), %rdi
@@ -132,12 +174,10 @@ trace:
 	movq 16(%rsp), %rdx
 	movq 8(%rsp), %rcx
 	movq (%rsp), %rax
-	addq $0x38, %rsp
+	addq $72, %rsp
+	retq
+#endif
 
-	jmp ftrace_stub
-END(mcount)
-#endif /* CONFIG_DYNAMIC_FTRACE */
-#endif /* CONFIG_FUNCTION_TRACER */
 
 #ifndef CONFIG_PREEMPT
 #define retint_kernel retint_restore_args
diff --git a/arch/x86/kernel/es7000_32.c b/arch/x86/kernel/es7000_32.c
index 0aa2c443d60..53699c931ad 100644
--- a/arch/x86/kernel/es7000_32.c
+++ b/arch/x86/kernel/es7000_32.c
@@ -38,8 +38,11 @@
 #include <asm/io.h>
 #include <asm/nmi.h>
 #include <asm/smp.h>
+#include <asm/atomic.h>
 #include <asm/apicdef.h>
 #include <mach_mpparse.h>
+#include <asm/genapic.h>
+#include <asm/setup.h>
 
 /*
  * ES7000 chipsets
@@ -161,6 +164,43 @@ es7000_rename_gsi(int ioapic, int gsi)
 	return gsi;
 }
 
+static int wakeup_secondary_cpu_via_mip(int cpu, unsigned long eip)
+{
+	unsigned long vect = 0, psaival = 0;
+
+	if (psai == NULL)
+		return -1;
+
+	vect = ((unsigned long)__pa(eip)/0x1000) << 16;
+	psaival = (0x1000000 | vect | cpu);
+
+	while (*psai & 0x1000000)
+		;
+
+	*psai = psaival;
+
+	return 0;
+}
+
+static void noop_wait_for_deassert(atomic_t *deassert_not_used)
+{
+}
+
+static int __init es7000_update_genapic(void)
+{
+	genapic->wakeup_cpu = wakeup_secondary_cpu_via_mip;
+
+	/* MPENTIUMIII */
+	if (boot_cpu_data.x86 == 6 &&
+	    (boot_cpu_data.x86_model >= 7 || boot_cpu_data.x86_model <= 11)) {
+		es7000_update_genapic_to_cluster();
+		genapic->wait_for_init_deassert = noop_wait_for_deassert;
+		genapic->wakeup_cpu = wakeup_secondary_cpu_via_mip;
+	}
+
+	return 0;
+}
+
 void __init
 setup_unisys(void)
 {
@@ -176,6 +216,8 @@ setup_unisys(void)
 	else
 		es7000_plat = ES7000_CLASSIC;
 	ioapic_renumber_irq = es7000_rename_gsi;
+
+	x86_quirks->update_genapic = es7000_update_genapic;
 }
 
 /*
@@ -317,26 +359,6 @@ es7000_mip_write(struct mip_reg *mip_reg)
 	return status;
 }
 
-int
-es7000_start_cpu(int cpu, unsigned long eip)
-{
-	unsigned long vect = 0, psaival = 0;
-
-	if (psai == NULL)
-		return -1;
-
-	vect = ((unsigned long)__pa(eip)/0x1000) << 16;
-	psaival = (0x1000000 | vect | cpu);
-
-	while (*psai & 0x1000000)
-                ;
-
-	*psai = psaival;
-
-	return 0;
-
-}
-
 void __init
 es7000_sw_apic(void)
 {
diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
index 50ea0ac8c9b..1b43086b097 100644
--- a/arch/x86/kernel/ftrace.c
+++ b/arch/x86/kernel/ftrace.c
@@ -14,14 +14,17 @@
 #include <linux/uaccess.h>
 #include <linux/ftrace.h>
 #include <linux/percpu.h>
+#include <linux/sched.h>
 #include <linux/init.h>
 #include <linux/list.h>
 
 #include <asm/ftrace.h>
+#include <linux/ftrace.h>
 #include <asm/nops.h>
+#include <asm/nmi.h>
 
 
-static unsigned char ftrace_nop[MCOUNT_INSN_SIZE];
+#ifdef CONFIG_DYNAMIC_FTRACE
 
 union ftrace_code_union {
 	char code[MCOUNT_INSN_SIZE];
@@ -31,18 +34,12 @@ union ftrace_code_union {
 	} __attribute__((packed));
 };
 
-
 static int ftrace_calc_offset(long ip, long addr)
 {
 	return (int)(addr - ip);
 }
 
-unsigned char *ftrace_nop_replace(void)
-{
-	return ftrace_nop;
-}
-
-unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr)
+static unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr)
 {
 	static union ftrace_code_union calc;
 
@@ -56,7 +53,142 @@ unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr)
 	return calc.code;
 }
 
-int
+/*
+ * Modifying code must take extra care. On an SMP machine, if
+ * the code being modified is also being executed on another CPU
+ * that CPU will have undefined results and possibly take a GPF.
+ * We use kstop_machine to stop other CPUS from exectuing code.
+ * But this does not stop NMIs from happening. We still need
+ * to protect against that. We separate out the modification of
+ * the code to take care of this.
+ *
+ * Two buffers are added: An IP buffer and a "code" buffer.
+ *
+ * 1) Put the instruction pointer into the IP buffer
+ *    and the new code into the "code" buffer.
+ * 2) Set a flag that says we are modifying code
+ * 3) Wait for any running NMIs to finish.
+ * 4) Write the code
+ * 5) clear the flag.
+ * 6) Wait for any running NMIs to finish.
+ *
+ * If an NMI is executed, the first thing it does is to call
+ * "ftrace_nmi_enter". This will check if the flag is set to write
+ * and if it is, it will write what is in the IP and "code" buffers.
+ *
+ * The trick is, it does not matter if everyone is writing the same
+ * content to the code location. Also, if a CPU is executing code
+ * it is OK to write to that code location if the contents being written
+ * are the same as what exists.
+ */
+
+static atomic_t in_nmi = ATOMIC_INIT(0);
+static int mod_code_status;		/* holds return value of text write */
+static int mod_code_write;		/* set when NMI should do the write */
+static void *mod_code_ip;		/* holds the IP to write to */
+static void *mod_code_newcode;		/* holds the text to write to the IP */
+
+static unsigned nmi_wait_count;
+static atomic_t nmi_update_count = ATOMIC_INIT(0);
+
+int ftrace_arch_read_dyn_info(char *buf, int size)
+{
+	int r;
+
+	r = snprintf(buf, size, "%u %u",
+		     nmi_wait_count,
+		     atomic_read(&nmi_update_count));
+	return r;
+}
+
+static void ftrace_mod_code(void)
+{
+	/*
+	 * Yes, more than one CPU process can be writing to mod_code_status.
+	 *    (and the code itself)
+	 * But if one were to fail, then they all should, and if one were
+	 * to succeed, then they all should.
+	 */
+	mod_code_status = probe_kernel_write(mod_code_ip, mod_code_newcode,
+					     MCOUNT_INSN_SIZE);
+}
+
+void ftrace_nmi_enter(void)
+{
+	atomic_inc(&in_nmi);
+	/* Must have in_nmi seen before reading write flag */
+	smp_mb();
+	if (mod_code_write) {
+		ftrace_mod_code();
+		atomic_inc(&nmi_update_count);
+	}
+}
+
+void ftrace_nmi_exit(void)
+{
+	/* Finish all executions before clearing in_nmi */
+	smp_wmb();
+	atomic_dec(&in_nmi);
+}
+
+static void wait_for_nmi(void)
+{
+	int waited = 0;
+
+	while (atomic_read(&in_nmi)) {
+		waited = 1;
+		cpu_relax();
+	}
+
+	if (waited)
+		nmi_wait_count++;
+}
+
+static int
+do_ftrace_mod_code(unsigned long ip, void *new_code)
+{
+	mod_code_ip = (void *)ip;
+	mod_code_newcode = new_code;
+
+	/* The buffers need to be visible before we let NMIs write them */
+	smp_wmb();
+
+	mod_code_write = 1;
+
+	/* Make sure write bit is visible before we wait on NMIs */
+	smp_mb();
+
+	wait_for_nmi();
+
+	/* Make sure all running NMIs have finished before we write the code */
+	smp_mb();
+
+	ftrace_mod_code();
+
+	/* Make sure the write happens before clearing the bit */
+	smp_wmb();
+
+	mod_code_write = 0;
+
+	/* make sure NMIs see the cleared bit */
+	smp_mb();
+
+	wait_for_nmi();
+
+	return mod_code_status;
+}
+
+
+
+
+static unsigned char ftrace_nop[MCOUNT_INSN_SIZE];
+
+static unsigned char *ftrace_nop_replace(void)
+{
+	return ftrace_nop;
+}
+
+static int
 ftrace_modify_code(unsigned long ip, unsigned char *old_code,
 		   unsigned char *new_code)
 {
@@ -81,7 +213,7 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code,
 		return -EINVAL;
 
 	/* replace the text with the new text */
-	if (probe_kernel_write((void *)ip, new_code, MCOUNT_INSN_SIZE))
+	if (do_ftrace_mod_code(ip, new_code))
 		return -EPERM;
 
 	sync_core();
@@ -89,6 +221,29 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code,
 	return 0;
 }
 
+int ftrace_make_nop(struct module *mod,
+		    struct dyn_ftrace *rec, unsigned long addr)
+{
+	unsigned char *new, *old;
+	unsigned long ip = rec->ip;
+
+	old = ftrace_call_replace(ip, addr);
+	new = ftrace_nop_replace();
+
+	return ftrace_modify_code(rec->ip, old, new);
+}
+
+int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr)
+{
+	unsigned char *new, *old;
+	unsigned long ip = rec->ip;
+
+	old = ftrace_nop_replace();
+	new = ftrace_call_replace(ip, addr);
+
+	return ftrace_modify_code(rec->ip, old, new);
+}
+
 int ftrace_update_ftrace_func(ftrace_func_t func)
 {
 	unsigned long ip = (unsigned long)(&ftrace_call);
@@ -165,3 +320,218 @@ int __init ftrace_dyn_arch_init(void *data)
 
 	return 0;
 }
+#endif
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+
+#ifdef CONFIG_DYNAMIC_FTRACE
+extern void ftrace_graph_call(void);
+
+static int ftrace_mod_jmp(unsigned long ip,
+			  int old_offset, int new_offset)
+{
+	unsigned char code[MCOUNT_INSN_SIZE];
+
+	if (probe_kernel_read(code, (void *)ip, MCOUNT_INSN_SIZE))
+		return -EFAULT;
+
+	if (code[0] != 0xe9 || old_offset != *(int *)(&code[1]))
+		return -EINVAL;
+
+	*(int *)(&code[1]) = new_offset;
+
+	if (do_ftrace_mod_code(ip, &code))
+		return -EPERM;
+
+	return 0;
+}
+
+int ftrace_enable_ftrace_graph_caller(void)
+{
+	unsigned long ip = (unsigned long)(&ftrace_graph_call);
+	int old_offset, new_offset;
+
+	old_offset = (unsigned long)(&ftrace_stub) - (ip + MCOUNT_INSN_SIZE);
+	new_offset = (unsigned long)(&ftrace_graph_caller) - (ip + MCOUNT_INSN_SIZE);
+
+	return ftrace_mod_jmp(ip, old_offset, new_offset);
+}
+
+int ftrace_disable_ftrace_graph_caller(void)
+{
+	unsigned long ip = (unsigned long)(&ftrace_graph_call);
+	int old_offset, new_offset;
+
+	old_offset = (unsigned long)(&ftrace_graph_caller) - (ip + MCOUNT_INSN_SIZE);
+	new_offset = (unsigned long)(&ftrace_stub) - (ip + MCOUNT_INSN_SIZE);
+
+	return ftrace_mod_jmp(ip, old_offset, new_offset);
+}
+
+#else /* CONFIG_DYNAMIC_FTRACE */
+
+/*
+ * These functions are picked from those used on
+ * this page for dynamic ftrace. They have been
+ * simplified to ignore all traces in NMI context.
+ */
+static atomic_t in_nmi;
+
+void ftrace_nmi_enter(void)
+{
+	atomic_inc(&in_nmi);
+}
+
+void ftrace_nmi_exit(void)
+{
+	atomic_dec(&in_nmi);
+}
+
+#endif /* !CONFIG_DYNAMIC_FTRACE */
+
+/* Add a function return address to the trace stack on thread info.*/
+static int push_return_trace(unsigned long ret, unsigned long long time,
+				unsigned long func, int *depth)
+{
+	int index;
+
+	if (!current->ret_stack)
+		return -EBUSY;
+
+	/* The return trace stack is full */
+	if (current->curr_ret_stack == FTRACE_RETFUNC_DEPTH - 1) {
+		atomic_inc(&current->trace_overrun);
+		return -EBUSY;
+	}
+
+	index = ++current->curr_ret_stack;
+	barrier();
+	current->ret_stack[index].ret = ret;
+	current->ret_stack[index].func = func;
+	current->ret_stack[index].calltime = time;
+	*depth = index;
+
+	return 0;
+}
+
+/* Retrieve a function return address to the trace stack on thread info.*/
+static void pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret)
+{
+	int index;
+
+	index = current->curr_ret_stack;
+
+	if (unlikely(index < 0)) {
+		ftrace_graph_stop();
+		WARN_ON(1);
+		/* Might as well panic, otherwise we have no where to go */
+		*ret = (unsigned long)panic;
+		return;
+	}
+
+	*ret = current->ret_stack[index].ret;
+	trace->func = current->ret_stack[index].func;
+	trace->calltime = current->ret_stack[index].calltime;
+	trace->overrun = atomic_read(&current->trace_overrun);
+	trace->depth = index;
+	barrier();
+	current->curr_ret_stack--;
+
+}
+
+/*
+ * Send the trace to the ring-buffer.
+ * @return the original return address.
+ */
+unsigned long ftrace_return_to_handler(void)
+{
+	struct ftrace_graph_ret trace;
+	unsigned long ret;
+
+	pop_return_trace(&trace, &ret);
+	trace.rettime = cpu_clock(raw_smp_processor_id());
+	ftrace_graph_return(&trace);
+
+	if (unlikely(!ret)) {
+		ftrace_graph_stop();
+		WARN_ON(1);
+		/* Might as well panic. What else to do? */
+		ret = (unsigned long)panic;
+	}
+
+	return ret;
+}
+
+/*
+ * Hook the return address and push it in the stack of return addrs
+ * in current thread info.
+ */
+void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr)
+{
+	unsigned long old;
+	unsigned long long calltime;
+	int faulted;
+	struct ftrace_graph_ent trace;
+	unsigned long return_hooker = (unsigned long)
+				&return_to_handler;
+
+	/* Nmi's are currently unsupported */
+	if (unlikely(atomic_read(&in_nmi)))
+		return;
+
+	if (unlikely(atomic_read(&current->tracing_graph_pause)))
+		return;
+
+	/*
+	 * Protect against fault, even if it shouldn't
+	 * happen. This tool is too much intrusive to
+	 * ignore such a protection.
+	 */
+	asm volatile(
+		"1: " _ASM_MOV " (%[parent_old]), %[old]\n"
+		"2: " _ASM_MOV " %[return_hooker], (%[parent_replaced])\n"
+		"   movl $0, %[faulted]\n"
+
+		".section .fixup, \"ax\"\n"
+		"3: movl $1, %[faulted]\n"
+		".previous\n"
+
+		_ASM_EXTABLE(1b, 3b)
+		_ASM_EXTABLE(2b, 3b)
+
+		: [parent_replaced] "=r" (parent), [old] "=r" (old),
+		  [faulted] "=r" (faulted)
+		: [parent_old] "0" (parent), [return_hooker] "r" (return_hooker)
+		: "memory"
+	);
+
+	if (unlikely(faulted)) {
+		ftrace_graph_stop();
+		WARN_ON(1);
+		return;
+	}
+
+	if (unlikely(!__kernel_text_address(old))) {
+		ftrace_graph_stop();
+		*parent = old;
+		WARN_ON(1);
+		return;
+	}
+
+	calltime = cpu_clock(raw_smp_processor_id());
+
+	if (push_return_trace(old, calltime,
+				self_addr, &trace.depth) == -EBUSY) {
+		*parent = old;
+		return;
+	}
+
+	trace.func = self_addr;
+
+	/* Only trace if the calling function expects to */
+	if (!ftrace_graph_entry(&trace)) {
+		current->curr_ret_stack--;
+		*parent = old;
+	}
+}
+#endif /* CONFIG_FUNCTION_GRAPH_TRACER */
diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c
index 6c9bfc9e1e9..2bced78b0b8 100644
--- a/arch/x86/kernel/genapic_64.c
+++ b/arch/x86/kernel/genapic_64.c
@@ -21,6 +21,7 @@
 #include <asm/smp.h>
 #include <asm/ipi.h>
 #include <asm/genapic.h>
+#include <asm/setup.h>
 
 extern struct genapic apic_flat;
 extern struct genapic apic_physflat;
@@ -53,6 +54,9 @@ void __init setup_apic_routing(void)
 			genapic = &apic_physflat;
 		printk(KERN_INFO "Setting APIC routing to %s\n", genapic->name);
 	}
+
+	if (x86_quirks->update_genapic)
+		x86_quirks->update_genapic();
 }
 
 /* Same for both flat and physical. */
diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c
index 2c7dbdb9827..dece1728973 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>
@@ -356,6 +360,103 @@ 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
  */
@@ -428,7 +529,7 @@ 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) {
@@ -439,8 +540,7 @@ 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;
@@ -450,7 +550,8 @@ 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);
@@ -467,4 +568,6 @@ 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 1dcb0f13897..3e66bd364a9 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 fa1d25dd83e..ac108d1fe18 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 d16084f9064..388e05a5fc1 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 067d8de913f..3f0a3edf0a5 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 {
diff --git a/arch/x86/kernel/init_task.c b/arch/x86/kernel/init_task.c
index a4f93b4120c..d39918076bb 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 9043251210f..679e7bbbbcd 100644
--- a/arch/x86/kernel/io_apic.c
+++ b/arch/x86/kernel/io_apic.c
@@ -2216,10 +2216,9 @@ static void set_ir_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
 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();
diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c
index 1d3d0e71b04..1df869e5bd0 100644
--- a/arch/x86/kernel/irq_64.c
+++ b/arch/x86/kernel/irq_64.c
@@ -13,6 +13,7 @@
 #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>
@@ -45,7 +46,7 @@ static inline void stack_overflow_check(struct pt_regs *regs)
  * 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;
diff --git a/arch/x86/kernel/machine_kexec_32.c b/arch/x86/kernel/machine_kexec_32.c
index 7a385746509..37f420018a4 100644
--- a/arch/x86/kernel/machine_kexec_32.c
+++ b/arch/x86/kernel/machine_kexec_32.c
@@ -13,6 +13,7 @@
 #include <linux/numa.h>
 #include <linux/ftrace.h>
 #include <linux/suspend.h>
+#include <linux/gfp.h>
 
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
@@ -25,15 +26,6 @@
 #include <asm/system.h>
 #include <asm/cacheflush.h>
 
-#define PAGE_ALIGNED __attribute__ ((__aligned__(PAGE_SIZE)))
-static u32 kexec_pgd[1024] PAGE_ALIGNED;
-#ifdef CONFIG_X86_PAE
-static u32 kexec_pmd0[1024] PAGE_ALIGNED;
-static u32 kexec_pmd1[1024] PAGE_ALIGNED;
-#endif
-static u32 kexec_pte0[1024] PAGE_ALIGNED;
-static u32 kexec_pte1[1024] PAGE_ALIGNED;
-
 static void set_idt(void *newidt, __u16 limit)
 {
 	struct desc_ptr curidt;
@@ -76,6 +68,76 @@ static void load_segments(void)
 #undef __STR
 }
 
+static void machine_kexec_free_page_tables(struct kimage *image)
+{
+	free_page((unsigned long)image->arch.pgd);
+#ifdef CONFIG_X86_PAE
+	free_page((unsigned long)image->arch.pmd0);
+	free_page((unsigned long)image->arch.pmd1);
+#endif
+	free_page((unsigned long)image->arch.pte0);
+	free_page((unsigned long)image->arch.pte1);
+}
+
+static int machine_kexec_alloc_page_tables(struct kimage *image)
+{
+	image->arch.pgd = (pgd_t *)get_zeroed_page(GFP_KERNEL);
+#ifdef CONFIG_X86_PAE
+	image->arch.pmd0 = (pmd_t *)get_zeroed_page(GFP_KERNEL);
+	image->arch.pmd1 = (pmd_t *)get_zeroed_page(GFP_KERNEL);
+#endif
+	image->arch.pte0 = (pte_t *)get_zeroed_page(GFP_KERNEL);
+	image->arch.pte1 = (pte_t *)get_zeroed_page(GFP_KERNEL);
+	if (!image->arch.pgd ||
+#ifdef CONFIG_X86_PAE
+	    !image->arch.pmd0 || !image->arch.pmd1 ||
+#endif
+	    !image->arch.pte0 || !image->arch.pte1) {
+		machine_kexec_free_page_tables(image);
+		return -ENOMEM;
+	}
+	return 0;
+}
+
+static void machine_kexec_page_table_set_one(
+	pgd_t *pgd, pmd_t *pmd, pte_t *pte,
+	unsigned long vaddr, unsigned long paddr)
+{
+	pud_t *pud;
+
+	pgd += pgd_index(vaddr);
+#ifdef CONFIG_X86_PAE
+	if (!(pgd_val(*pgd) & _PAGE_PRESENT))
+		set_pgd(pgd, __pgd(__pa(pmd) | _PAGE_PRESENT));
+#endif
+	pud = pud_offset(pgd, vaddr);
+	pmd = pmd_offset(pud, vaddr);
+	if (!(pmd_val(*pmd) & _PAGE_PRESENT))
+		set_pmd(pmd, __pmd(__pa(pte) | _PAGE_TABLE));
+	pte = pte_offset_kernel(pmd, vaddr);
+	set_pte(pte, pfn_pte(paddr >> PAGE_SHIFT, PAGE_KERNEL_EXEC));
+}
+
+static void machine_kexec_prepare_page_tables(struct kimage *image)
+{
+	void *control_page;
+	pmd_t *pmd = 0;
+
+	control_page = page_address(image->control_code_page);
+#ifdef CONFIG_X86_PAE
+	pmd = image->arch.pmd0;
+#endif
+	machine_kexec_page_table_set_one(
+		image->arch.pgd, pmd, image->arch.pte0,
+		(unsigned long)control_page, __pa(control_page));
+#ifdef CONFIG_X86_PAE
+	pmd = image->arch.pmd1;
+#endif
+	machine_kexec_page_table_set_one(
+		image->arch.pgd, pmd, image->arch.pte1,
+		__pa(control_page), __pa(control_page));
+}
+
 /*
  * A architecture hook called to validate the
  * proposed image and prepare the control pages
@@ -87,12 +149,20 @@ static void load_segments(void)
  * reboot code buffer to allow us to avoid allocations
  * later.
  *
- * Make control page executable.
+ * - Make control page executable.
+ * - Allocate page tables
+ * - Setup page tables
  */
 int machine_kexec_prepare(struct kimage *image)
 {
+	int error;
+
 	if (nx_enabled)
 		set_pages_x(image->control_code_page, 1);
+	error = machine_kexec_alloc_page_tables(image);
+	if (error)
+		return error;
+	machine_kexec_prepare_page_tables(image);
 	return 0;
 }
 
@@ -104,6 +174,7 @@ void machine_kexec_cleanup(struct kimage *image)
 {
 	if (nx_enabled)
 		set_pages_nx(image->control_code_page, 1);
+	machine_kexec_free_page_tables(image);
 }
 
 /*
@@ -150,18 +221,7 @@ void machine_kexec(struct kimage *image)
 	relocate_kernel_ptr = control_page;
 	page_list[PA_CONTROL_PAGE] = __pa(control_page);
 	page_list[VA_CONTROL_PAGE] = (unsigned long)control_page;
-	page_list[PA_PGD] = __pa(kexec_pgd);
-	page_list[VA_PGD] = (unsigned long)kexec_pgd;
-#ifdef CONFIG_X86_PAE
-	page_list[PA_PMD_0] = __pa(kexec_pmd0);
-	page_list[VA_PMD_0] = (unsigned long)kexec_pmd0;
-	page_list[PA_PMD_1] = __pa(kexec_pmd1);
-	page_list[VA_PMD_1] = (unsigned long)kexec_pmd1;
-#endif
-	page_list[PA_PTE_0] = __pa(kexec_pte0);
-	page_list[VA_PTE_0] = (unsigned long)kexec_pte0;
-	page_list[PA_PTE_1] = __pa(kexec_pte1);
-	page_list[VA_PTE_1] = (unsigned long)kexec_pte1;
+	page_list[PA_PGD] = __pa(image->arch.pgd);
 
 	if (image->type == KEXEC_TYPE_DEFAULT)
 		page_list[PA_SWAP_PAGE] = (page_to_pfn(image->swap_page)
diff --git a/arch/x86/kernel/microcode_amd.c b/arch/x86/kernel/microcode_amd.c
index 5f8e5d75a25..c25fdb38229 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 82fb2809ce3..c9b721ba968 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 622dc4a2178..b7f4c929e61 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/mpparse.c b/arch/x86/kernel/mpparse.c
index 0f4c1fd5a1f..45e3b69808b 100644
--- a/arch/x86/kernel/mpparse.c
+++ b/arch/x86/kernel/mpparse.c
@@ -586,26 +586,23 @@ static void __init __get_smp_config(unsigned int early)
 {
 	struct intel_mp_floating *mpf = mpf_found;
 
-	if (x86_quirks->mach_get_smp_config) {
-		if (x86_quirks->mach_get_smp_config(early))
-			return;
-	}
+	if (!mpf)
+		return;
+
 	if (acpi_lapic && early)
 		return;
+
 	/*
-	 * ACPI supports both logical (e.g. Hyper-Threading) and physical
-	 * processors, where MPS only supports physical.
+	 * MPS doesn't support hyperthreading, aka only have
+	 * thread 0 apic id in MPS table
 	 */
-	if (acpi_lapic && acpi_ioapic) {
-		printk(KERN_INFO "Using ACPI (MADT) for SMP configuration "
-		       "information\n");
+	if (acpi_lapic && acpi_ioapic)
 		return;
-	} else if (acpi_lapic)
-		printk(KERN_INFO "Using ACPI for processor (LAPIC) "
-		       "configuration information\n");
 
-	if (!mpf)
-		return;
+	if (x86_quirks->mach_get_smp_config) {
+		if (x86_quirks->mach_get_smp_config(early))
+			return;
+	}
 
 	printk(KERN_INFO "Intel MultiProcessor Specification v1.%d\n",
 	       mpf->mpf_specification);
diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
index 2c97f07f1c2..8bd1bf9622a 100644
--- a/arch/x86/kernel/nmi.c
+++ b/arch/x86/kernel/nmi.c
@@ -131,6 +131,11 @@ static void report_broken_nmi(int cpu, int *prev_nmi_count)
 	atomic_dec(&nmi_active);
 }
 
+static void __acpi_nmi_disable(void *__unused)
+{
+	apic_write(APIC_LVT0, APIC_DM_NMI | APIC_LVT_MASKED);
+}
+
 int __init check_nmi_watchdog(void)
 {
 	unsigned int *prev_nmi_count;
@@ -179,8 +184,12 @@ int __init check_nmi_watchdog(void)
 	kfree(prev_nmi_count);
 	return 0;
 error:
-	if (nmi_watchdog == NMI_IO_APIC && !timer_through_8259)
-		disable_8259A_irq(0);
+	if (nmi_watchdog == NMI_IO_APIC) {
+		if (!timer_through_8259)
+			disable_8259A_irq(0);
+		on_each_cpu(__acpi_nmi_disable, NULL, 1);
+	}
+
 #ifdef CONFIG_X86_32
 	timer_ack = 0;
 #endif
@@ -199,12 +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);
@@ -285,11 +299,6 @@ void acpi_nmi_enable(void)
 		on_each_cpu(__acpi_nmi_enable, NULL, 1);
 }
 
-static void __acpi_nmi_disable(void *__unused)
-{
-	apic_write(APIC_LVT0, APIC_DM_NMI | APIC_LVT_MASKED);
-}
-
 /*
  * Disable timer based NMIs on all CPUs:
  */
@@ -340,6 +349,8 @@ void stop_apic_nmi_watchdog(void *unused)
 		return;
 	if (nmi_watchdog == NMI_LOCAL_APIC)
 		lapic_watchdog_stop();
+	else
+		__acpi_nmi_disable(NULL);
 	__get_cpu_var(wd_enabled) = 0;
 	atomic_dec(&nmi_active);
 }
@@ -465,6 +476,24 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
 
 #ifdef CONFIG_SYSCTL
 
+static void enable_ioapic_nmi_watchdog_single(void *unused)
+{
+	__get_cpu_var(wd_enabled) = 1;
+	atomic_inc(&nmi_active);
+	__acpi_nmi_enable(NULL);
+}
+
+static void enable_ioapic_nmi_watchdog(void)
+{
+	on_each_cpu(enable_ioapic_nmi_watchdog_single, NULL, 1);
+	touch_nmi_watchdog();
+}
+
+static void disable_ioapic_nmi_watchdog(void)
+{
+	on_each_cpu(stop_apic_nmi_watchdog, NULL, 1);
+}
+
 static int __init setup_unknown_nmi_panic(char *str)
 {
 	unknown_nmi_panic = 1;
@@ -507,6 +536,11 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file,
 			enable_lapic_nmi_watchdog();
 		else
 			disable_lapic_nmi_watchdog();
+	} else if (nmi_watchdog == NMI_IO_APIC) {
+		if (nmi_watchdog_enabled)
+			enable_ioapic_nmi_watchdog();
+		else
+			disable_ioapic_nmi_watchdog();
 	} else {
 		printk(KERN_WARNING
 			"NMI watchdog doesn't know what hardware to touch\n");
diff --git a/arch/x86/kernel/numaq_32.c b/arch/x86/kernel/numaq_32.c
index 4caff39078e..0deea37a53c 100644
--- a/arch/x86/kernel/numaq_32.c
+++ b/arch/x86/kernel/numaq_32.c
@@ -31,7 +31,7 @@
 #include <asm/numaq.h>
 #include <asm/topology.h>
 #include <asm/processor.h>
-#include <asm/mpspec.h>
+#include <asm/genapic.h>
 #include <asm/e820.h>
 #include <asm/setup.h>
 
@@ -235,6 +235,13 @@ static int __init numaq_setup_ioapic_ids(void)
 	return 1;
 }
 
+static int __init numaq_update_genapic(void)
+{
+	genapic->wakeup_cpu = wakeup_secondary_cpu_via_nmi;
+
+	return 0;
+}
+
 static struct x86_quirks numaq_x86_quirks __initdata = {
 	.arch_pre_time_init	= numaq_pre_time_init,
 	.arch_time_init		= NULL,
@@ -250,6 +257,7 @@ static struct x86_quirks numaq_x86_quirks __initdata = {
 	.mpc_oem_pci_bus	= mpc_oem_pci_bus,
 	.smp_read_mpc_oem	= smp_read_mpc_oem,
 	.setup_ioapic_ids	= numaq_setup_ioapic_ids,
+	.update_genapic		= numaq_update_genapic,
 };
 
 void numaq_mps_oem_check(struct mp_config_table *mpc, char *oem,
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 19262482021..7a3dfceb90e 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);
 
@@ -188,7 +184,6 @@ static __init int iommu_setup(char *p)
 		}
 
 		if (!strncmp(p, "biomerge", 8)) {
-			iommu_bio_merge = 4096;
 			iommu_merge = 1;
 			force_iommu = 1;
 		}
@@ -300,8 +295,8 @@ 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 ba7ad83e20a..a35eaa379ff 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/process.c b/arch/x86/kernel/process.c
index c622772744d..e68bb9e3086 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -1,13 +1,16 @@
 #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>
 #include <linux/module.h>
 #include <linux/pm.h>
 #include <linux/clockchips.h>
+#include <linux/ftrace.h>
 #include <asm/system.h>
+#include <asm/apic.h>
 
 unsigned long idle_halt;
 EXPORT_SYMBOL(idle_halt);
@@ -100,6 +103,9 @@ static inline int hlt_use_halt(void)
 void default_idle(void)
 {
 	if (hlt_use_halt()) {
+		struct power_trace it;
+
+		trace_power_start(&it, POWER_CSTATE, 1);
 		current_thread_info()->status &= ~TS_POLLING;
 		/*
 		 * TS_POLLING-cleared state must be visible before we
@@ -112,6 +118,7 @@ void default_idle(void)
 		else
 			local_irq_enable();
 		current_thread_info()->status |= TS_POLLING;
+		trace_power_end(&it);
 	} else {
 		local_irq_enable();
 		/* loop is done by the caller */
@@ -122,6 +129,21 @@ void default_idle(void)
 EXPORT_SYMBOL(default_idle);
 #endif
 
+void stop_this_cpu(void *dummy)
+{
+	local_irq_disable();
+	/*
+	 * Remove this CPU:
+	 */
+	cpu_clear(smp_processor_id(), cpu_online_map);
+	disable_local_APIC();
+
+	for (;;) {
+		if (hlt_works(smp_processor_id()))
+			halt();
+	}
+}
+
 static void do_nothing(void *unused)
 {
 }
@@ -154,24 +176,31 @@ EXPORT_SYMBOL_GPL(cpu_idle_wait);
  */
 void mwait_idle_with_hints(unsigned long ax, unsigned long cx)
 {
+	struct power_trace it;
+
+	trace_power_start(&it, POWER_CSTATE, (ax>>4)+1);
 	if (!need_resched()) {
 		__monitor((void *)&current_thread_info()->flags, 0, 0);
 		smp_mb();
 		if (!need_resched())
 			__mwait(ax, cx);
 	}
+	trace_power_end(&it);
 }
 
 /* Default MONITOR/MWAIT with no hints, used for default C1 state */
 static void mwait_idle(void)
 {
+	struct power_trace it;
 	if (!need_resched()) {
+		trace_power_start(&it, POWER_CSTATE, 1);
 		__monitor((void *)&current_thread_info()->flags, 0, 0);
 		smp_mb();
 		if (!need_resched())
 			__sti_mwait(0, 0);
 		else
 			local_irq_enable();
+		trace_power_end(&it);
 	} else
 		local_irq_enable();
 }
@@ -183,9 +212,13 @@ static void mwait_idle(void)
  */
 static void poll_idle(void)
 {
+	struct power_trace it;
+
+	trace_power_start(&it, POWER_CSTATE, 0);
 	local_irq_enable();
 	while (!need_resched())
 		cpu_relax();
+	trace_power_end(&it);
 }
 
 /*
@@ -270,7 +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 0a1302fe6d4..3ba155d2488 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -38,6 +38,7 @@
 #include <linux/percpu.h>
 #include <linux/prctl.h>
 #include <linux/dmi.h>
+#include <linux/ftrace.h>
 
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -59,6 +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");
 
@@ -250,14 +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)
@@ -339,6 +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;
 }
 
@@ -419,48 +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)) {
@@ -482,15 +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
@@ -548,7 +512,8 @@ __switch_to_xtra(struct task_struct *prev_p, struct task_struct *next_p,
  * the task-switch, and shows up in ret_from_fork in entry.S,
  * for example.
  */
-struct task_struct * __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
+__notrace_funcgraph struct task_struct *
+__switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 {
 	struct thread_struct *prev = &prev_p->thread,
 				 *next = &next_p->thread;
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index c958120fb1b..416fb9282f4 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -39,6 +39,7 @@
 #include <linux/prctl.h>
 #include <linux/uaccess.h>
 #include <linux/io.h>
+#include <linux/ftrace.h>
 
 #include <asm/pgtable.h>
 #include <asm/system.h>
@@ -52,6 +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);
 
@@ -235,14 +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)
@@ -372,6 +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) {
@@ -470,35 +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)) {
@@ -533,14 +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 */
 }
 
 /*
@@ -551,8 +524,9 @@ static inline void __switch_to_xtra(struct task_struct *prev_p,
  * - could test fs/gs bitsliced
  *
  * Kprobes not supported here. Set the probe on schedule instead.
+ * Function graph tracer not supported too.
  */
-struct task_struct *
+__notrace_funcgraph struct task_struct *
 __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 {
 	struct thread_struct *prev = &prev_p->thread;
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 0a6d8c12e10..0a5df5f82fb 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -581,158 +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_end);
-	if (error < 0)
-		return error;
-
-	if (bts_end <= index)
-		return -EINVAL;
+	trace = ds_read_bts(child->bts);
+	if (!trace)
+		return -EPERM;
 
-	error = ds_get_bts_index(child, &bts_index);
-	if (error < 0)
-		return error;
+	at = trace->ds.top - ((index + 1) * trace->ds.size);
+	if ((void *)at < trace->ds.begin)
+		at += (trace->ds.n * trace->ds.size);
 
-	/* 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_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, &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, 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);
+	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 void ptrace_bts_ovfl(struct task_struct *child)
+static int ptrace_bts_allocate_buffer(struct task_struct *child, size_t size)
 {
-	send_sig(child->thread.bts_ovfl_signal, child, 0);
+	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,
@@ -740,114 +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;
+		return -EFAULT;
 
-	error = -EINVAL;
-	if ((cfg.flags & PTRACE_BTS_O_SIGNAL) &&
-	    !(cfg.flags & PTRACE_BTS_O_ALLOC))
-		goto errout;
+	if (child->bts) {
+		ds_release_bts(child->bts);
+		child->bts = NULL;
+	}
 
-	if (cfg.flags & PTRACE_BTS_O_ALLOC) {
-		ds_ovfl_callback_t ovfl = NULL;
-		unsigned int sig = 0;
+	if (cfg.flags & PTRACE_BTS_O_SIGNAL) {
+		if (!cfg.signal)
+			return -EINVAL;
 
-		/* we ignore the error in case we were not tracing child */
-		(void)ds_release_bts(child);
+		return -EOPNOTSUPP;
 
-		if (cfg.flags & PTRACE_BTS_O_SIGNAL) {
-			if (!cfg.signal)
-				goto errout;
+		child->thread.bts_ovfl_signal = cfg.signal;
+	}
 
-			sig  = cfg.signal;
-			ovfl = ptrace_bts_ovfl;
-		}
+	if ((cfg.flags & PTRACE_BTS_O_ALLOC) &&
+	    (cfg.size != child->bts_size)) {
+		int error;
 
-		error = ds_request_bts(child, /* base = */ NULL, cfg.size, ovfl);
-		if (error < 0)
-			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, &end);
-	if (error < 0)
-		return error;
-
-	error = ds_access_bts(child, /* index = */ 0, &base);
-	if (error < 0)
-		return error;
-
-	error = ds_access_bts(child, /* 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)))
@@ -856,110 +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;
 
-	/* The writing task will be the switched-to task on a context
-	 * switch. It needs to write into the switched-from task's BTS
-	 * buffer. */
-	return ds_unchecked_write_bts(child, 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;
+
+		/* 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_pentium_m = {
-	.sizeof_bts    = sizeof(long) * 3,
-	.sizeof_field  = sizeof(long),
-	.debugctl_mask = (1<<6)|(1<<7)
-};
+static void ptrace_bts_detach(struct task_struct *child)
+{
+	if (unlikely(child->bts)) {
+		ds_release_bts(child->bts);
+		child->bts = NULL;
 
-static const struct bts_configuration bts_cfg_core2 = {
-	.sizeof_bts    = 8 * 3,
-	.sizeof_field  = 8,
-	.debugctl_mask = (1<<6)|(1<<7)|(1<<9)
-};
+		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 */
 
-static inline void bts_configure(const struct bts_configuration *cfg)
+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 0xD:
-		case 0xE: /* Pentium M */
-			bts_configure(&bts_cfg_pentium_m);
-			break;
-		case 0xF: /* Core2 */
-        case 0x1C: /* Atom */
-			bts_configure(&bts_cfg_core2);
-			break;
-		default:
-			/* sorry, don't know about them */
-			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..
@@ -972,15 +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
-	(void)ds_release_bts(child);
-
-	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
@@ -1112,7 +976,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 		break;
 
 	case PTRACE_BTS_SIZE:
-		ret = ds_get_bts_index(child, /* pos = */ NULL);
+		ret = ptrace_bts_size(child);
 		break;
 
 	case PTRACE_BTS_GET:
@@ -1121,7 +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);
+		ret = ptrace_bts_clear(child);
 		break;
 
 	case PTRACE_BTS_DRAIN:
@@ -1384,6 +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/reboot.c b/arch/x86/kernel/reboot.c
index cc5a2545dd4..61f718df6ee 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -21,6 +21,9 @@
 # include <asm/iommu.h>
 #endif
 
+#include <mach_ipi.h>
+
+
 /*
  * Power off function, if any
  */
@@ -36,7 +39,10 @@ int reboot_force;
 static int reboot_cpu = -1;
 #endif
 
-/* reboot=b[ios] | s[mp] | t[riple] | k[bd] | e[fi] [, [w]arm | [c]old]
+/* This is set by the PCI code if either type 1 or type 2 PCI is detected */
+bool port_cf9_safe = false;
+
+/* reboot=b[ios] | s[mp] | t[riple] | k[bd] | e[fi] [, [w]arm | [c]old] | p[ci]
    warm   Don't set the cold reboot flag
    cold   Set the cold reboot flag
    bios   Reboot by jumping through the BIOS (only for X86_32)
@@ -45,6 +51,7 @@ static int reboot_cpu = -1;
    kbd    Use the keyboard controller. cold reset (default)
    acpi   Use the RESET_REG in the FADT
    efi    Use efi reset_system runtime service
+   pci    Use the so-called "PCI reset register", CF9
    force  Avoid anything that could hang.
  */
 static int __init reboot_setup(char *str)
@@ -79,6 +86,7 @@ static int __init reboot_setup(char *str)
 		case 'k':
 		case 't':
 		case 'e':
+		case 'p':
 			reboot_type = *str;
 			break;
 
@@ -404,12 +412,27 @@ static void native_machine_emergency_restart(void)
 			reboot_type = BOOT_KBD;
 			break;
 
-
 		case BOOT_EFI:
 			if (efi_enabled)
-				efi.reset_system(reboot_mode ? EFI_RESET_WARM : EFI_RESET_COLD,
+				efi.reset_system(reboot_mode ?
+						 EFI_RESET_WARM :
+						 EFI_RESET_COLD,
 						 EFI_SUCCESS, 0, NULL);
+			reboot_type = BOOT_KBD;
+			break;
 
+		case BOOT_CF9:
+			port_cf9_safe = true;
+			/* fall through */
+
+		case BOOT_CF9_COND:
+			if (port_cf9_safe) {
+				u8 cf9 = inb(0xcf9) & ~6;
+				outb(cf9|2, 0xcf9); /* Request hard reset */
+				udelay(50);
+				outb(cf9|6, 0xcf9); /* Actually do the reset */
+				udelay(50);
+			}
 			reboot_type = BOOT_KBD;
 			break;
 		}
@@ -470,6 +493,11 @@ static void native_machine_restart(char *__unused)
 
 static void native_machine_halt(void)
 {
+	/* stop other cpus and apics */
+	machine_shutdown();
+
+	/* stop this cpu */
+	stop_this_cpu(NULL);
 }
 
 static void native_machine_power_off(void)
@@ -523,3 +551,95 @@ void machine_crash_shutdown(struct pt_regs *regs)
 	machine_ops.crash_shutdown(regs);
 }
 #endif
+
+
+#if defined(CONFIG_SMP)
+
+/* This keeps a track of which one is crashing cpu. */
+static int crashing_cpu;
+static nmi_shootdown_cb shootdown_callback;
+
+static atomic_t waiting_for_crash_ipi;
+
+static int crash_nmi_callback(struct notifier_block *self,
+			unsigned long val, void *data)
+{
+	int cpu;
+
+	if (val != DIE_NMI_IPI)
+		return NOTIFY_OK;
+
+	cpu = raw_smp_processor_id();
+
+	/* Don't do anything if this handler is invoked on crashing cpu.
+	 * Otherwise, system will completely hang. Crashing cpu can get
+	 * an NMI if system was initially booted with nmi_watchdog parameter.
+	 */
+	if (cpu == crashing_cpu)
+		return NOTIFY_STOP;
+	local_irq_disable();
+
+	shootdown_callback(cpu, (struct die_args *)data);
+
+	atomic_dec(&waiting_for_crash_ipi);
+	/* Assume hlt works */
+	halt();
+	for (;;)
+		cpu_relax();
+
+	return 1;
+}
+
+static void smp_send_nmi_allbutself(void)
+{
+	cpumask_t mask = cpu_online_map;
+	cpu_clear(safe_smp_processor_id(), mask);
+	if (!cpus_empty(mask))
+		send_IPI_mask(mask, NMI_VECTOR);
+}
+
+static struct notifier_block crash_nmi_nb = {
+	.notifier_call = crash_nmi_callback,
+};
+
+/* Halt all other CPUs, calling the specified function on each of them
+ *
+ * This function can be used to halt all other CPUs on crash
+ * or emergency reboot time. The function passed as parameter
+ * will be called inside a NMI handler on all CPUs.
+ */
+void nmi_shootdown_cpus(nmi_shootdown_cb callback)
+{
+	unsigned long msecs;
+	local_irq_disable();
+
+	/* Make a note of crashing cpu. Will be used in NMI callback.*/
+	crashing_cpu = safe_smp_processor_id();
+
+	shootdown_callback = callback;
+
+	atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
+	/* Would it be better to replace the trap vector here? */
+	if (register_die_notifier(&crash_nmi_nb))
+		return;		/* return what? */
+	/* Ensure the new callback function is set before sending
+	 * out the NMI
+	 */
+	wmb();
+
+	smp_send_nmi_allbutself();
+
+	msecs = 1000; /* Wait at most a second for the other cpus to stop */
+	while ((atomic_read(&waiting_for_crash_ipi) > 0) && msecs) {
+		mdelay(1);
+		msecs--;
+	}
+
+	/* Leave the nmi callback set */
+}
+#else /* !CONFIG_SMP */
+void nmi_shootdown_cpus(nmi_shootdown_cb callback)
+{
+	/* No other CPUs to shoot down */
+}
+#endif
diff --git a/arch/x86/kernel/relocate_kernel_32.S b/arch/x86/kernel/relocate_kernel_32.S
index 6f50664b2ba..a160f311972 100644
--- a/arch/x86/kernel/relocate_kernel_32.S
+++ b/arch/x86/kernel/relocate_kernel_32.S
@@ -10,15 +10,12 @@
 #include <asm/page.h>
 #include <asm/kexec.h>
 #include <asm/processor-flags.h>
-#include <asm/pgtable.h>
 
 /*
  * Must be relocatable PIC code callable as a C function
  */
 
 #define PTR(x) (x << 2)
-#define PAGE_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
-#define PAE_PGD_ATTR (_PAGE_PRESENT)
 
 /* control_page + KEXEC_CONTROL_CODE_MAX_SIZE
  * ~ control_page + PAGE_SIZE are used as data storage and stack for
@@ -39,7 +36,6 @@
 #define CP_PA_BACKUP_PAGES_MAP	DATA(0x1c)
 
 	.text
-	.align PAGE_SIZE
 	.globl relocate_kernel
 relocate_kernel:
 	/* Save the CPU context, used for jumping back */
@@ -60,117 +56,6 @@ relocate_kernel:
 	movl	%cr4, %eax
 	movl	%eax, CR4(%edi)
 
-#ifdef CONFIG_X86_PAE
-	/* map the control page at its virtual address */
-
-	movl	PTR(VA_PGD)(%ebp), %edi
-	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
-	andl	$0xc0000000, %eax
-	shrl	$27, %eax
-	addl	%edi, %eax
-
-	movl	PTR(PA_PMD_0)(%ebp), %edx
-	orl	$PAE_PGD_ATTR, %edx
-	movl	%edx, (%eax)
-
-	movl	PTR(VA_PMD_0)(%ebp), %edi
-	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
-	andl	$0x3fe00000, %eax
-	shrl	$18, %eax
-	addl	%edi, %eax
-
-	movl	PTR(PA_PTE_0)(%ebp), %edx
-	orl	$PAGE_ATTR, %edx
-	movl	%edx, (%eax)
-
-	movl	PTR(VA_PTE_0)(%ebp), %edi
-	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
-	andl	$0x001ff000, %eax
-	shrl	$9, %eax
-	addl	%edi, %eax
-
-	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edx
-	orl	$PAGE_ATTR, %edx
-	movl	%edx, (%eax)
-
-	/* identity map the control page at its physical address */
-
-	movl	PTR(VA_PGD)(%ebp), %edi
-	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
-	andl	$0xc0000000, %eax
-	shrl	$27, %eax
-	addl	%edi, %eax
-
-	movl	PTR(PA_PMD_1)(%ebp), %edx
-	orl	$PAE_PGD_ATTR, %edx
-	movl	%edx, (%eax)
-
-	movl	PTR(VA_PMD_1)(%ebp), %edi
-	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
-	andl	$0x3fe00000, %eax
-	shrl	$18, %eax
-	addl	%edi, %eax
-
-	movl	PTR(PA_PTE_1)(%ebp), %edx
-	orl	$PAGE_ATTR, %edx
-	movl	%edx, (%eax)
-
-	movl	PTR(VA_PTE_1)(%ebp), %edi
-	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
-	andl	$0x001ff000, %eax
-	shrl	$9, %eax
-	addl	%edi, %eax
-
-	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edx
-	orl	$PAGE_ATTR, %edx
-	movl	%edx, (%eax)
-#else
-	/* map the control page at its virtual address */
-
-	movl	PTR(VA_PGD)(%ebp), %edi
-	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
-	andl	$0xffc00000, %eax
-	shrl	$20, %eax
-	addl	%edi, %eax
-
-	movl	PTR(PA_PTE_0)(%ebp), %edx
-	orl	$PAGE_ATTR, %edx
-	movl	%edx, (%eax)
-
-	movl	PTR(VA_PTE_0)(%ebp), %edi
-	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
-	andl	$0x003ff000, %eax
-	shrl	$10, %eax
-	addl	%edi, %eax
-
-	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edx
-	orl	$PAGE_ATTR, %edx
-	movl	%edx, (%eax)
-
-	/* identity map the control page at its physical address */
-
-	movl	PTR(VA_PGD)(%ebp), %edi
-	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
-	andl	$0xffc00000, %eax
-	shrl	$20, %eax
-	addl	%edi, %eax
-
-	movl	PTR(PA_PTE_1)(%ebp), %edx
-	orl	$PAGE_ATTR, %edx
-	movl	%edx, (%eax)
-
-	movl	PTR(VA_PTE_1)(%ebp), %edi
-	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
-	andl	$0x003ff000, %eax
-	shrl	$10, %eax
-	addl	%edi, %eax
-
-	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edx
-	orl	$PAGE_ATTR, %edx
-	movl	%edx, (%eax)
-#endif
-
-relocate_new_kernel:
 	/* read the arguments and say goodbye to the stack */
 	movl  20+4(%esp), %ebx /* page_list */
 	movl  20+8(%esp), %ebp /* list of pages */
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 9d5674f7b6c..08e02e8453c 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 */
@@ -583,161 +586,24 @@ static int __init setup_elfcorehdr(char *arg)
 early_param("elfcorehdr", setup_elfcorehdr);
 #endif
 
-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)
+static int __init default_update_genapic(void)
 {
-	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
+#ifdef CONFIG_X86_SMP
+# if defined(CONFIG_X86_GENERICARCH) || defined(CONFIG_X86_64)
+	genapic->wakeup_cpu = wakeup_secondary_cpu_via_init;
+# endif
 #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));
+	return 0;
 }
 
-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);
+static struct x86_quirks default_x86_quirks __initdata = {
+	.update_genapic         = default_update_genapic,
+};
 
-	init_timer(&periodic_check_timer);
-	periodic_check_timer.function = &periodic_check_for_corruption;
-	periodic_check_for_corruption(0);
-}
-#endif
+struct x86_quirks *x86_quirks __initdata = &default_x86_quirks;
 
+#ifdef CONFIG_X86_RESERVE_LOW_64K
 static int __init dmi_low_memory_corruption(const struct dmi_system_id *d)
 {
 	printk(KERN_NOTICE
@@ -749,6 +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[] = {
@@ -794,6 +661,9 @@ void __init setup_arch(char **cmdline_p)
 	printk(KERN_INFO "Command line: %s\n", boot_command_line);
 #endif
 
+	/* VMI may relocate the fixmap; do this before touching ioremap area */
+	vmi_init();
+
 	early_cpu_init();
 	early_ioremap_init();
 
@@ -880,13 +750,8 @@ void __init setup_arch(char **cmdline_p)
 	check_efer();
 #endif
 
-#if defined(CONFIG_VMI) && defined(CONFIG_X86_32)
-	/*
-	 * Must be before kernel pagetables are setup
-	 * or fixmap area is touched.
-	 */
-	vmi_init();
-#endif
+	/* Must be before kernel pagetables are setup */
+	vmi_activate();
 
 	/* after early param, so could get panic from serial */
 	reserve_early_setup_data();
@@ -909,6 +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/sigframe.h b/arch/x86/kernel/sigframe.h
deleted file mode 100644
index cc673aa55ce..00000000000
--- 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.c b/arch/x86/kernel/signal.c
index dee553c503d..4fa5243c206 100644
--- a/arch/x86/kernel/signal.c
+++ b/arch/x86/kernel/signal.c
@@ -35,7 +35,7 @@
 #include <asm/syscall.h>
 #include <asm/syscalls.h>
 
-#include "sigframe.h"
+#include <asm/sigframe.h>
 
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
@@ -594,17 +594,7 @@ asmlinkage unsigned long sys_sigreturn(unsigned long __unused)
 	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);
+	signal_fault(regs, frame, "sigreturn");
 
 	return 0;
 }
@@ -681,6 +671,11 @@ static int signr_convert(int sig)
 #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
@@ -906,8 +901,9 @@ 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/smp.c b/arch/x86/kernel/smp.c
index d18537ce2c7..7e558db362c 100644
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
@@ -140,19 +140,6 @@ void native_send_call_func_ipi(cpumask_t mask)
 		send_IPI_mask(mask, CALL_FUNCTION_VECTOR);
 }
 
-static void stop_this_cpu(void *dummy)
-{
-	local_irq_disable();
-	/*
-	 * Remove this CPU:
-	 */
-	cpu_clear(smp_processor_id(), cpu_online_map);
-	disable_local_APIC();
-	if (hlt_works(smp_processor_id()))
-		for (;;) halt();
-	for (;;);
-}
-
 /*
  * this function calls the 'stop' function on all other CPUs in the system.
  */
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 7b109339731..f8500c96944 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -62,6 +62,7 @@
 #include <asm/mtrr.h>
 #include <asm/vmi.h>
 #include <asm/genapic.h>
+#include <asm/setup.h>
 #include <linux/mc146818rtc.h>
 
 #include <mach_apic.h>
@@ -287,16 +288,14 @@ 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
 	 * fragile that we want to limit the things done here to the
 	 * most necessary things.
 	 */
-#ifdef CONFIG_VMI
 	vmi_bringup();
-#endif
 	cpu_init();
 	preempt_disable();
 	smp_callin();
@@ -536,7 +535,7 @@ static void impress_friends(void)
 	pr_debug("Before bogocount - setting activated=1.\n");
 }
 
-static inline void __inquire_remote_apic(int apicid)
+void __inquire_remote_apic(int apicid)
 {
 	unsigned i, regs[] = { APIC_ID >> 4, APIC_LVR >> 4, APIC_SPIV >> 4 };
 	char *names[] = { "ID", "VERSION", "SPIV" };
@@ -575,14 +574,13 @@ static inline void __inquire_remote_apic(int apicid)
 	}
 }
 
-#ifdef WAKE_SECONDARY_VIA_NMI
 /*
  * Poke the other CPU in the eye via NMI to wake it up. Remember that the normal
  * INIT, INIT, STARTUP sequence will reset the chip hard for us, and this
  * won't ... remember to clear down the APIC, etc later.
  */
-static int __devinit
-wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip)
+int __devinit
+wakeup_secondary_cpu_via_nmi(int logical_apicid, unsigned long start_eip)
 {
 	unsigned long send_status, accept_status = 0;
 	int maxlvt;
@@ -599,7 +597,7 @@ wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip)
 	 * Give the other CPU some time to accept the IPI.
 	 */
 	udelay(200);
-	if (APIC_INTEGRATED(apic_version[phys_apicid])) {
+	if (APIC_INTEGRATED(apic_version[boot_cpu_physical_apicid])) {
 		maxlvt = lapic_get_maxlvt();
 		if (maxlvt > 3)			/* Due to the Pentium erratum 3AP.  */
 			apic_write(APIC_ESR, 0);
@@ -614,11 +612,9 @@ wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip)
 
 	return (send_status | accept_status);
 }
-#endif	/* WAKE_SECONDARY_VIA_NMI */
 
-#ifdef WAKE_SECONDARY_VIA_INIT
-static int __devinit
-wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip)
+int __devinit
+wakeup_secondary_cpu_via_init(int phys_apicid, unsigned long start_eip)
 {
 	unsigned long send_status, accept_status = 0;
 	int maxlvt, num_starts, j;
@@ -737,7 +733,6 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip)
 
 	return (send_status | accept_status);
 }
-#endif	/* WAKE_SECONDARY_VIA_INIT */
 
 struct create_idle {
 	struct work_struct work;
@@ -1086,8 +1081,10 @@ 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/stacktrace.c b/arch/x86/kernel/stacktrace.c
index a03e7f6d90c..10786af9554 100644
--- a/arch/x86/kernel/stacktrace.c
+++ b/arch/x86/kernel/stacktrace.c
@@ -6,6 +6,7 @@
 #include <linux/sched.h>
 #include <linux/stacktrace.h>
 #include <linux/module.h>
+#include <linux/uaccess.h>
 #include <asm/stacktrace.h>
 
 static void save_stack_warning(void *data, char *msg)
@@ -83,3 +84,66 @@ void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
 }
 EXPORT_SYMBOL_GPL(save_stack_trace_tsk);
+
+/* Userspace stacktrace - based on kernel/trace/trace_sysprof.c */
+
+struct stack_frame {
+	const void __user	*next_fp;
+	unsigned long		ret_addr;
+};
+
+static int copy_stack_frame(const void __user *fp, struct stack_frame *frame)
+{
+	int ret;
+
+	if (!access_ok(VERIFY_READ, fp, sizeof(*frame)))
+		return 0;
+
+	ret = 1;
+	pagefault_disable();
+	if (__copy_from_user_inatomic(frame, fp, sizeof(*frame)))
+		ret = 0;
+	pagefault_enable();
+
+	return ret;
+}
+
+static inline void __save_stack_trace_user(struct stack_trace *trace)
+{
+	const struct pt_regs *regs = task_pt_regs(current);
+	const void __user *fp = (const void __user *)regs->bp;
+
+	if (trace->nr_entries < trace->max_entries)
+		trace->entries[trace->nr_entries++] = regs->ip;
+
+	while (trace->nr_entries < trace->max_entries) {
+		struct stack_frame frame;
+
+		frame.next_fp = NULL;
+		frame.ret_addr = 0;
+		if (!copy_stack_frame(fp, &frame))
+			break;
+		if ((unsigned long)fp < regs->sp)
+			break;
+		if (frame.ret_addr) {
+			trace->entries[trace->nr_entries++] =
+				frame.ret_addr;
+		}
+		if (fp == frame.next_fp)
+			break;
+		fp = frame.next_fp;
+	}
+}
+
+void save_stack_trace_user(struct stack_trace *trace)
+{
+	/*
+	 * Trace user stack if we are not a kernel thread
+	 */
+	if (current->mm) {
+		__save_stack_trace_user(trace);
+	}
+	if (trace->nr_entries < trace->max_entries)
+		trace->entries[trace->nr_entries++] = ULONG_MAX;
+}
+
diff --git a/arch/x86/kernel/time_32.c b/arch/x86/kernel/time_32.c
index 77b400f06ea..65309e4cb1c 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 418a095c579..891e7a7c433 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);
 
diff --git a/arch/x86/kernel/tlb_32.c b/arch/x86/kernel/tlb_32.c
index f4049f3513b..8da059f949b 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 8f919ca6949..29887d7081a 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 04431f34fd1..6a00e5faaa7 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 1106fac6024..808031a5ba1 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 d815293e6d9..141907ab6e2 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -660,7 +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.
@@ -671,7 +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
@@ -685,34 +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 424093b157d..599e5816863 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 1c0dfbca87c..bf36328f6ef 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 8b6c393ab9f..23206ba1687 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
@@ -960,8 +841,6 @@ static inline int __init activate_vmi(void)
 
 void __init vmi_init(void)
 {
-	unsigned long flags;
-
 	if (!vmi_rom)
 		probe_vmi_rom();
 	else
@@ -973,13 +852,21 @@ void __init vmi_init(void)
 
 	reserve_top_address(-vmi_rom->virtual_top);
 
-	local_irq_save(flags);
-	activate_vmi();
-
 #ifdef CONFIG_X86_IO_APIC
 	/* This is virtual hardware; timer routing is wired correctly */
 	no_timer_check = 1;
 #endif
+}
+
+void vmi_activate(void)
+{
+	unsigned long flags;
+
+	if (!vmi_rom)
+		return;
+
+	local_irq_save(flags);
+	activate_vmi();
 	local_irq_restore(flags & X86_EFLAGS_IF);
 }
 
diff --git a/arch/x86/kernel/vmlinux_32.lds.S b/arch/x86/kernel/vmlinux_32.lds.S
index a9b8560adbc..82c67559dde 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 46e05447405..1a614c0e6be 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 ebf2f12900f..44153afc906 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -17,6 +17,9 @@
  *  want per guest time just set the kernel.vsyscall64 sysctl to 0.
  */
 
+/* Disable profiling for userspace code: */
+#define DISABLE_BRANCH_PROFILING
+
 #include <linux/time.h>
 #include <linux/init.h>
 #include <linux/kernel.h>