Merge branch 'x86-mce-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

* 'x86-mce-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (21 commits)
  x86, mce: Fix compilation with !CONFIG_DEBUG_FS in mce-severity.c
  x86, mce: CE in last bank prevents panic by unknown MCE
  x86, mce: Fake panic support for MCE testing
  x86, mce: Move debugfs mce dir creating to mce.c
  x86, mce: Support specifying raise mode for software MCE injection
  x86, mce: Support specifying context for software mce injection
  x86, mce: fix reporting of Thermal Monitoring mechanism enabled
  x86, mce: remove never executed code
  x86, mce: add missing __cpuinit tags
  x86, mce: fix "mce" boot option handling for CONFIG_X86_NEW_MCE
  x86, mce: don't log boot MCEs on Pentium M (model == 13) CPUs
  x86: mce: Lower maximum number of banks to architecture limit
  x86: mce: macros to compute banks MSRs
  x86: mce: Move per bank data in a single datastructure
  x86: mce: Move code in mce.c
  x86: mce: Rename CONFIG_X86_NEW_MCE to CONFIG_X86_MCE
  x86: mce: Remove old i386 machine check code
  x86: mce: Update X86_MCE description in x86/Kconfig
  x86: mce: Make CONFIG_X86_ANCIENT_MCE dependent on CONFIG_X86_MCE
  x86, mce: use atomic_inc_return() instead of add by 1
  ...

Manually fixed up trivial conflicts:
	Documentation/feature-removal-schedule.txt
	arch/x86/kernel/cpu/mcheck/mce.c

20 files changed, 363 insertions, 777 deletions

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 503d21216d5..fa75220f8d3 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -428,16 +428,6 @@ Who:	Johannes Berg <johannes@sipsolutions.net>
 
 ----------------------------
 
-What:	CONFIG_X86_OLD_MCE
-When:	2.6.32
-Why:	Remove the old legacy 32bit machine check code. This has been
-	superseded by the newer machine check code from the 64bit port,
-	but the old version has been kept around for easier testing. Note this
-	doesn't impact the old P5 and WinChip machine check handlers.
-Who:	Andi Kleen <andi@firstfloor.org>
-
-----------------------------
-
 What:	lock_policy_rwsem_* and unlock_policy_rwsem_* will not be
 	exported interface anymore.
 When:	2.6.33
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index e98e81a0497..e5deee2dfcf 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -783,41 +783,17 @@ config X86_REROUTE_FOR_BROKEN_BOOT_IRQS
 	  increased on these systems.
 
 config X86_MCE
-	bool "Machine Check Exception"
+	bool "Machine Check / overheating reporting"
 	---help---
-	  Machine Check Exception support allows the processor to notify the
-	  kernel if it detects a problem (e.g. overheating, component failure).
+	  Machine Check support allows the processor to notify the
+	  kernel if it detects a problem (e.g. overheating, data corruption).
 	  The action the kernel takes depends on the severity of the problem,
-	  ranging from a warning message on the console, to halting the machine.
-	  Your processor must be a Pentium or newer to support this - check the
-	  flags in /proc/cpuinfo for mce.  Note that some older Pentium systems
-	  have a design flaw which leads to false MCE events - hence MCE is
-	  disabled on all P5 processors, unless explicitly enabled with "mce"
-	  as a boot argument.  Similarly, if MCE is built in and creates a
-	  problem on some new non-standard machine, you can boot with "nomce"
-	  to disable it.  MCE support simply ignores non-MCE processors like
-	  the 386 and 486, so nearly everyone can say Y here.
-
-config X86_OLD_MCE
-	depends on X86_32 && X86_MCE
-	bool "Use legacy machine check code (will go away)"
-	default n
-	select X86_ANCIENT_MCE
-	---help---
-	  Use the old i386 machine check code. This is merely intended for
-	  testing in a transition period. Try this if you run into any machine
-	  check related software problems, but report the problem to
-	  linux-kernel.  When in doubt say no.
-
-config X86_NEW_MCE
-	depends on X86_MCE
-	bool
-	default y if (!X86_OLD_MCE && X86_32) || X86_64
+	  ranging from warning messages to halting the machine.
 
 config X86_MCE_INTEL
 	def_bool y
 	prompt "Intel MCE features"
-	depends on X86_NEW_MCE && X86_LOCAL_APIC
+	depends on X86_MCE && X86_LOCAL_APIC
 	---help---
 	   Additional support for intel specific MCE features such as
 	   the thermal monitor.
@@ -825,14 +801,14 @@ config X86_MCE_INTEL
 config X86_MCE_AMD
 	def_bool y
 	prompt "AMD MCE features"
-	depends on X86_NEW_MCE && X86_LOCAL_APIC
+	depends on X86_MCE && X86_LOCAL_APIC
 	---help---
 	   Additional support for AMD specific MCE features such as
 	   the DRAM Error Threshold.
 
 config X86_ANCIENT_MCE
 	def_bool n
-	depends on X86_32
+	depends on X86_32 && X86_MCE
 	prompt "Support for old Pentium 5 / WinChip machine checks"
 	---help---
 	  Include support for machine check handling on old Pentium 5 or WinChip
@@ -845,36 +821,16 @@ config X86_MCE_THRESHOLD
 	default y
 
 config X86_MCE_INJECT
-	depends on X86_NEW_MCE
+	depends on X86_MCE
 	tristate "Machine check injector support"
 	---help---
 	  Provide support for injecting machine checks for testing purposes.
 	  If you don't know what a machine check is and you don't do kernel
 	  QA it is safe to say n.
 
-config X86_MCE_NONFATAL
-	tristate "Check for non-fatal errors on AMD Athlon/Duron / Intel Pentium 4"
-	depends on X86_OLD_MCE
-	---help---
-	  Enabling this feature starts a timer that triggers every 5 seconds which
-	  will look at the machine check registers to see if anything happened.
-	  Non-fatal problems automatically get corrected (but still logged).
-	  Disable this if you don't want to see these messages.
-	  Seeing the messages this option prints out may be indicative of dying
-	  or out-of-spec (ie, overclocked) hardware.
-	  This option only does something on certain CPUs.
-	  (AMD Athlon/Duron and Intel Pentium 4)
-
-config X86_MCE_P4THERMAL
-	bool "check for P4 thermal throttling interrupt."
-	depends on X86_OLD_MCE && X86_MCE && (X86_UP_APIC || SMP)
-	---help---
-	  Enabling this feature will cause a message to be printed when the P4
-	  enters thermal throttling.
-
 config X86_THERMAL_VECTOR
 	def_bool y
-	depends on X86_MCE_P4THERMAL || X86_MCE_INTEL
+	depends on X86_MCE_INTEL
 
 config VM86
 	bool "Enable VM86 support" if EMBEDDED
diff --git a/arch/x86/include/asm/entry_arch.h b/arch/x86/include/asm/entry_arch.h
index ff8cbfa0785..5e3f2044f0d 100644
--- a/arch/x86/include/asm/entry_arch.h
+++ b/arch/x86/include/asm/entry_arch.h
@@ -61,7 +61,7 @@ BUILD_INTERRUPT(thermal_interrupt,THERMAL_APIC_VECTOR)
 BUILD_INTERRUPT(threshold_interrupt,THRESHOLD_APIC_VECTOR)
 #endif
 
-#ifdef CONFIG_X86_NEW_MCE
+#ifdef CONFIG_X86_MCE
 BUILD_INTERRUPT(mce_self_interrupt,MCE_SELF_VECTOR)
 #endif
 
diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h
index 5cdd8d100ec..b608a64c581 100644
--- a/arch/x86/include/asm/mce.h
+++ b/arch/x86/include/asm/mce.h
@@ -9,7 +9,7 @@
  */
 
 #define MCG_BANKCNT_MASK	0xff         /* Number of Banks */
-#define MCG_CTL_P		(1ULL<<8)    /* MCG_CAP register available */
+#define MCG_CTL_P		(1ULL<<8)    /* MCG_CTL register available */
 #define MCG_EXT_P		(1ULL<<9)    /* Extended registers available */
 #define MCG_CMCI_P		(1ULL<<10)   /* CMCI supported */
 #define MCG_EXT_CNT_MASK	0xff0000     /* Number of Extended registers */
@@ -38,6 +38,14 @@
 #define MCM_ADDR_MEM	 3	/* memory address */
 #define MCM_ADDR_GENERIC 7	/* generic */
 
+#define MCJ_CTX_MASK		3
+#define MCJ_CTX(flags)		((flags) & MCJ_CTX_MASK)
+#define MCJ_CTX_RANDOM		0    /* inject context: random */
+#define MCJ_CTX_PROCESS		1    /* inject context: process */
+#define MCJ_CTX_IRQ		2    /* inject context: IRQ */
+#define MCJ_NMI_BROADCAST	4    /* do NMI broadcasting */
+#define MCJ_EXCEPTION		8    /* raise as exception */
+
 /* Fields are zero when not available */
 struct mce {
 	__u64 status;
@@ -48,8 +56,8 @@ struct mce {
 	__u64 tsc;	/* cpu time stamp counter */
 	__u64 time;	/* wall time_t when error was detected */
 	__u8  cpuvendor;	/* cpu vendor as encoded in system.h */
-	__u8  pad1;
-	__u16 pad2;
+	__u8  inject_flags;	/* software inject flags */
+	__u16  pad;
 	__u32 cpuid;	/* CPUID 1 EAX */
 	__u8  cs;		/* code segment */
 	__u8  bank;	/* machine check bank */
@@ -115,13 +123,6 @@ void mcheck_init(struct cpuinfo_x86 *c);
 static inline void mcheck_init(struct cpuinfo_x86 *c) {}
 #endif
 
-#ifdef CONFIG_X86_OLD_MCE
-extern int nr_mce_banks;
-void amd_mcheck_init(struct cpuinfo_x86 *c);
-void intel_p4_mcheck_init(struct cpuinfo_x86 *c);
-void intel_p6_mcheck_init(struct cpuinfo_x86 *c);
-#endif
-
 #ifdef CONFIG_X86_ANCIENT_MCE
 void intel_p5_mcheck_init(struct cpuinfo_x86 *c);
 void winchip_mcheck_init(struct cpuinfo_x86 *c);
@@ -137,10 +138,11 @@ void mce_log(struct mce *m);
 DECLARE_PER_CPU(struct sys_device, mce_dev);
 
 /*
- * To support more than 128 would need to escape the predefined
- * Linux defined extended banks first.
+ * Maximum banks number.
+ * This is the limit of the current register layout on
+ * Intel CPUs.
  */
-#define MAX_NR_BANKS (MCE_EXTENDED_BANK - 1)
+#define MAX_NR_BANKS 32
 
 #ifdef CONFIG_X86_MCE_INTEL
 extern int mce_cmci_disabled;
@@ -208,11 +210,7 @@ extern void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu);
 
 void intel_init_thermal(struct cpuinfo_x86 *c);
 
-#ifdef CONFIG_X86_NEW_MCE
 void mce_log_therm_throt_event(__u64 status);
-#else
-static inline void mce_log_therm_throt_event(__u64 status) {}
-#endif
 
 #endif /* __KERNEL__ */
 #endif /* _ASM_X86_MCE_H */
diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h
index bd5549034a9..4ffe09b2ad7 100644
--- a/arch/x86/include/asm/msr-index.h
+++ b/arch/x86/include/asm/msr-index.h
@@ -81,8 +81,15 @@
 #define MSR_IA32_MC0_ADDR		0x00000402
 #define MSR_IA32_MC0_MISC		0x00000403
 
+#define MSR_IA32_MCx_CTL(x)		(MSR_IA32_MC0_CTL + 4*(x))
+#define MSR_IA32_MCx_STATUS(x)		(MSR_IA32_MC0_STATUS + 4*(x))
+#define MSR_IA32_MCx_ADDR(x)		(MSR_IA32_MC0_ADDR + 4*(x))
+#define MSR_IA32_MCx_MISC(x)		(MSR_IA32_MC0_MISC + 4*(x))
+
 /* These are consecutive and not in the normal 4er MCE bank block */
 #define MSR_IA32_MC0_CTL2		0x00000280
+#define MSR_IA32_MCx_CTL2(x)		(MSR_IA32_MC0_CTL2 + (x))
+
 #define CMCI_EN			(1ULL << 30)
 #define CMCI_THRESHOLD_MASK		0xffffULL
 
@@ -215,6 +222,10 @@
 
 #define THERM_STATUS_PROCHOT		(1 << 0)
 
+#define MSR_THERM2_CTL			0x0000019d
+
+#define MSR_THERM2_CTL_TM_SELECT	(1ULL << 16)
+
 #define MSR_IA32_MISC_ENABLE		0x000001a0
 
 /* MISC_ENABLE bits: architectural */
diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c
index db7220220d0..cb66a22d98a 100644
--- a/arch/x86/kernel/apic/nmi.c
+++ b/arch/x86/kernel/apic/nmi.c
@@ -66,7 +66,7 @@ static inline unsigned int get_nmi_count(int cpu)
 
 static inline int mce_in_progress(void)
 {
-#if defined(CONFIG_X86_NEW_MCE)
+#if defined(CONFIG_X86_MCE)
 	return atomic_read(&mce_entry) > 0;
 #endif
 	return 0;
diff --git a/arch/x86/kernel/cpu/mcheck/Makefile b/arch/x86/kernel/cpu/mcheck/Makefile
index 188a1ca5ad2..4ac6d48fe11 100644
--- a/arch/x86/kernel/cpu/mcheck/Makefile
+++ b/arch/x86/kernel/cpu/mcheck/Makefile
@@ -1,11 +1,8 @@
-obj-y				=  mce.o
+obj-y				=  mce.o mce-severity.o
 
-obj-$(CONFIG_X86_NEW_MCE)	+= mce-severity.o
-obj-$(CONFIG_X86_OLD_MCE)	+= k7.o p4.o p6.o
 obj-$(CONFIG_X86_ANCIENT_MCE)	+= winchip.o p5.o
 obj-$(CONFIG_X86_MCE_INTEL)	+= mce_intel.o
 obj-$(CONFIG_X86_MCE_AMD)	+= mce_amd.o
-obj-$(CONFIG_X86_MCE_NONFATAL)	+= non-fatal.o
 obj-$(CONFIG_X86_MCE_THRESHOLD) += threshold.o
 obj-$(CONFIG_X86_MCE_INJECT)	+= mce-inject.o
 
diff --git a/arch/x86/kernel/cpu/mcheck/k7.c b/arch/x86/kernel/cpu/mcheck/k7.c
deleted file mode 100644
index b945d5dbc60..00000000000
--- a/arch/x86/kernel/cpu/mcheck/k7.c
+++ /dev/null
@@ -1,116 +0,0 @@
-/*
- * Athlon specific Machine Check Exception Reporting
- * (C) Copyright 2002 Dave Jones <davej@redhat.com>
- */
-#include <linux/interrupt.h>
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/init.h>
-#include <linux/smp.h>
-
-#include <asm/processor.h>
-#include <asm/system.h>
-#include <asm/mce.h>
-#include <asm/msr.h>
-
-/* Machine Check Handler For AMD Athlon/Duron: */
-static void k7_machine_check(struct pt_regs *regs, long error_code)
-{
-	u32 alow, ahigh, high, low;
-	u32 mcgstl, mcgsth;
-	int recover = 1;
-	int i;
-
-	rdmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth);
-	if (mcgstl & (1<<0))	/* Recoverable ? */
-		recover = 0;
-
-	printk(KERN_EMERG "CPU %d: Machine Check Exception: %08x%08x\n",
-		smp_processor_id(), mcgsth, mcgstl);
-
-	for (i = 1; i < nr_mce_banks; i++) {
-		rdmsr(MSR_IA32_MC0_STATUS+i*4, low, high);
-		if (high & (1<<31)) {
-			char misc[20];
-			char addr[24];
-
-			misc[0] = '\0';
-			addr[0] = '\0';
-
-			if (high & (1<<29))
-				recover |= 1;
-			if (high & (1<<25))
-				recover |= 2;
-			high &= ~(1<<31);
-
-			if (high & (1<<27)) {
-				rdmsr(MSR_IA32_MC0_MISC+i*4, alow, ahigh);
-				snprintf(misc, 20, "[%08x%08x]", ahigh, alow);
-			}
-			if (high & (1<<26)) {
-				rdmsr(MSR_IA32_MC0_ADDR+i*4, alow, ahigh);
-				snprintf(addr, 24, " at %08x%08x", ahigh, alow);
-			}
-
-			printk(KERN_EMERG "CPU %d: Bank %d: %08x%08x%s%s\n",
-				smp_processor_id(), i, high, low, misc, addr);
-
-			/* Clear it: */
-			wrmsr(MSR_IA32_MC0_STATUS+i*4, 0UL, 0UL);
-			/* Serialize: */
-			wmb();
-			add_taint(TAINT_MACHINE_CHECK);
-		}
-	}
-
-	if (recover & 2)
-		panic("CPU context corrupt");
-	if (recover & 1)
-		panic("Unable to continue");
-
-	printk(KERN_EMERG "Attempting to continue.\n");
-
-	mcgstl &= ~(1<<2);
-	wrmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth);
-}
-
-
-/* AMD K7 machine check is Intel like: */
-void amd_mcheck_init(struct cpuinfo_x86 *c)
-{
-	u32 l, h;
-	int i;
-
-	if (!cpu_has(c, X86_FEATURE_MCE))
-		return;
-
-	machine_check_vector = k7_machine_check;
-	/* Make sure the vector pointer is visible before we enable MCEs: */
-	wmb();
-
-	printk(KERN_INFO "Intel machine check architecture supported.\n");
-
-	rdmsr(MSR_IA32_MCG_CAP, l, h);
-	if (l & (1<<8))	/* Control register present ? */
-		wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff);
-	nr_mce_banks = l & 0xff;
-
-	/*
-	 * Clear status for MC index 0 separately, we don't touch CTL,
-	 * as some K7 Athlons cause spurious MCEs when its enabled:
-	 */
-	if (boot_cpu_data.x86 == 6) {
-		wrmsr(MSR_IA32_MC0_STATUS, 0x0, 0x0);
-		i = 1;
-	} else
-		i = 0;
-
-	for (; i < nr_mce_banks; i++) {
-		wrmsr(MSR_IA32_MC0_CTL+4*i, 0xffffffff, 0xffffffff);
-		wrmsr(MSR_IA32_MC0_STATUS+4*i, 0x0, 0x0);
-	}
-
-	set_in_cr4(X86_CR4_MCE);
-	printk(KERN_INFO "Intel machine check reporting enabled on CPU#%d.\n",
-		smp_processor_id());
-}
diff --git a/arch/x86/kernel/cpu/mcheck/mce-inject.c b/arch/x86/kernel/cpu/mcheck/mce-inject.c
index a3a235a53f0..7029f0e2aca 100644
--- a/arch/x86/kernel/cpu/mcheck/mce-inject.c
+++ b/arch/x86/kernel/cpu/mcheck/mce-inject.c
@@ -18,7 +18,12 @@
 #include <linux/string.h>
 #include <linux/fs.h>
 #include <linux/smp.h>
+#include <linux/notifier.h>
+#include <linux/kdebug.h>
+#include <linux/cpu.h>
+#include <linux/sched.h>
 #include <asm/mce.h>
+#include <asm/apic.h>
 
 /* Update fake mce registers on current CPU. */
 static void inject_mce(struct mce *m)
@@ -39,44 +44,141 @@ static void inject_mce(struct mce *m)
 	i->finished = 1;
 }
 
-struct delayed_mce {
-	struct timer_list timer;
-	struct mce m;
-};
+static void raise_poll(struct mce *m)
+{
+	unsigned long flags;
+	mce_banks_t b;
 
-/* Inject mce on current CPU */
-static void raise_mce(unsigned long data)
+	memset(&b, 0xff, sizeof(mce_banks_t));
+	local_irq_save(flags);
+	machine_check_poll(0, &b);
+	local_irq_restore(flags);
+	m->finished = 0;
+}
+
+static void raise_exception(struct mce *m, struct pt_regs *pregs)
 {
-	struct delayed_mce *dm = (struct delayed_mce *)data;
-	struct mce *m = &dm->m;
-	int cpu = m->extcpu;
+	struct pt_regs regs;
+	unsigned long flags;
 
-	inject_mce(m);
-	if (m->status & MCI_STATUS_UC) {
-		struct pt_regs regs;
+	if (!pregs) {
 		memset(&regs, 0, sizeof(struct pt_regs));
 		regs.ip = m->ip;
 		regs.cs = m->cs;
+		pregs = &regs;
+	}
+	/* in mcheck exeception handler, irq will be disabled */
+	local_irq_save(flags);
+	do_machine_check(pregs, 0);
+	local_irq_restore(flags);
+	m->finished = 0;
+}
+
+static cpumask_t mce_inject_cpumask;
+
+static int mce_raise_notify(struct notifier_block *self,
+			    unsigned long val, void *data)
+{
+	struct die_args *args = (struct die_args *)data;
+	int cpu = smp_processor_id();
+	struct mce *m = &__get_cpu_var(injectm);
+	if (val != DIE_NMI_IPI || !cpu_isset(cpu, mce_inject_cpumask))
+		return NOTIFY_DONE;
+	cpu_clear(cpu, mce_inject_cpumask);
+	if (m->inject_flags & MCJ_EXCEPTION)
+		raise_exception(m, args->regs);
+	else if (m->status)
+		raise_poll(m);
+	return NOTIFY_STOP;
+}
+
+static struct notifier_block mce_raise_nb = {
+	.notifier_call = mce_raise_notify,
+	.priority = 1000,
+};
+
+/* Inject mce on current CPU */
+static int raise_local(struct mce *m)
+{
+	int context = MCJ_CTX(m->inject_flags);
+	int ret = 0;
+	int cpu = m->extcpu;
+
+	if (m->inject_flags & MCJ_EXCEPTION) {
 		printk(KERN_INFO "Triggering MCE exception on CPU %d\n", cpu);
-		do_machine_check(&regs, 0);
+		switch (context) {
+		case MCJ_CTX_IRQ:
+			/*
+			 * Could do more to fake interrupts like
+			 * calling irq_enter, but the necessary
+			 * machinery isn't exported currently.
+			 */
+			/*FALL THROUGH*/
+		case MCJ_CTX_PROCESS:
+			raise_exception(m, NULL);
+			break;
+		default:
+			printk(KERN_INFO "Invalid MCE context\n");
+			ret = -EINVAL;
+		}
 		printk(KERN_INFO "MCE exception done on CPU %d\n", cpu);
-	} else {
-		mce_banks_t b;
-		memset(&b, 0xff, sizeof(mce_banks_t));
+	} else if (m->status) {
 		printk(KERN_INFO "Starting machine check poll CPU %d\n", cpu);
-		machine_check_poll(0, &b);
+		raise_poll(m);
 		mce_notify_irq();
-		printk(KERN_INFO "Finished machine check poll on CPU %d\n",
-		       cpu);
-	}
-	kfree(dm);
+		printk(KERN_INFO "Machine check poll done on CPU %d\n", cpu);
+	} else
+		m->finished = 0;
+
+	return ret;
+}
+
+static void raise_mce(struct mce *m)
+{
+	int context = MCJ_CTX(m->inject_flags);
+
+	inject_mce(m);
+
+	if (context == MCJ_CTX_RANDOM)
+		return;
+
+#ifdef CONFIG_X86_LOCAL_APIC
+	if (m->inject_flags & MCJ_NMI_BROADCAST) {
+		unsigned long start;
+		int cpu;
+		get_online_cpus();
+		mce_inject_cpumask = cpu_online_map;
+		cpu_clear(get_cpu(), mce_inject_cpumask);
+		for_each_online_cpu(cpu) {
+			struct mce *mcpu = &per_cpu(injectm, cpu);
+			if (!mcpu->finished ||
+			    MCJ_CTX(mcpu->inject_flags) != MCJ_CTX_RANDOM)
+				cpu_clear(cpu, mce_inject_cpumask);
+		}
+		if (!cpus_empty(mce_inject_cpumask))
+			apic->send_IPI_mask(&mce_inject_cpumask, NMI_VECTOR);
+		start = jiffies;
+		while (!cpus_empty(mce_inject_cpumask)) {
+			if (!time_before(jiffies, start + 2*HZ)) {
+				printk(KERN_ERR
+				"Timeout waiting for mce inject NMI %lx\n",
+					*cpus_addr(mce_inject_cpumask));
+				break;
+			}
+			cpu_relax();
+		}
+		raise_local(m);
+		put_cpu();
+		put_online_cpus();
+	} else
+#endif
+		raise_local(m);
 }
 
 /* Error injection interface */
 static ssize_t mce_write(struct file *filp, const char __user *ubuf,
 			 size_t usize, loff_t *off)
 {
-	struct delayed_mce *dm;
 	struct mce m;
 
 	if (!capable(CAP_SYS_ADMIN))
@@ -96,19 +198,12 @@ static ssize_t mce_write(struct file *filp, const char __user *ubuf,
 	if (m.extcpu >= num_possible_cpus() || !cpu_online(m.extcpu))
 		return -EINVAL;
 
-	dm = kmalloc(sizeof(struct delayed_mce), GFP_KERNEL);
-	if (!dm)
-		return -ENOMEM;
-
 	/*
 	 * Need to give user space some time to set everything up,
 	 * so do it a jiffie or two later everywhere.
-	 * Should we use a hrtimer here for better synchronization?
 	 */
-	memcpy(&dm->m, &m, sizeof(struct mce));
-	setup_timer(&dm->timer, raise_mce, (unsigned long)dm);
-	dm->timer.expires = jiffies + 2;
-	add_timer_on(&dm->timer, m.extcpu);
+	schedule_timeout(2);
+	raise_mce(&m);
 	return usize;
 }
 
@@ -116,6 +211,7 @@ static int inject_init(void)
 {
 	printk(KERN_INFO "Machine check injector initialized\n");
 	mce_chrdev_ops.write = mce_write;
+	register_die_notifier(&mce_raise_nb);
 	return 0;
 }
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce-internal.h b/arch/x86/kernel/cpu/mcheck/mce-internal.h
index 54dcb8ff12e..32996f9fab6 100644
--- a/arch/x86/kernel/cpu/mcheck/mce-internal.h
+++ b/arch/x86/kernel/cpu/mcheck/mce-internal.h
@@ -1,3 +1,4 @@
+#include <linux/sysdev.h>
 #include <asm/mce.h>
 
 enum severity_level {
@@ -10,6 +11,20 @@ enum severity_level {
 	MCE_PANIC_SEVERITY,
 };
 
+#define ATTR_LEN		16
+
+/* One object for each MCE bank, shared by all CPUs */
+struct mce_bank {
+	u64			ctl;			/* subevents to enable */
+	unsigned char init;				/* initialise bank? */
+	struct sysdev_attribute attr;			/* sysdev attribute */
+	char			attrname[ATTR_LEN];	/* attribute name */
+};
+
 int mce_severity(struct mce *a, int tolerant, char **msg);
+struct dentry *mce_get_debugfs_dir(void);
 
 extern int mce_ser;
+
+extern struct mce_bank *mce_banks;
+
diff --git a/arch/x86/kernel/cpu/mcheck/mce-severity.c b/arch/x86/kernel/cpu/mcheck/mce-severity.c
index ff0807f9705..8a85dd1b1aa 100644
--- a/arch/x86/kernel/cpu/mcheck/mce-severity.c
+++ b/arch/x86/kernel/cpu/mcheck/mce-severity.c
@@ -139,6 +139,7 @@ int mce_severity(struct mce *a, int tolerant, char **msg)
 	}
 }
 
+#ifdef CONFIG_DEBUG_FS
 static void *s_start(struct seq_file *f, loff_t *pos)
 {
 	if (*pos >= ARRAY_SIZE(severities))
@@ -197,7 +198,7 @@ static int __init severities_debugfs_init(void)
 {
 	struct dentry *dmce = NULL, *fseverities_coverage = NULL;
 
-	dmce = debugfs_create_dir("mce", NULL);
+	dmce = mce_get_debugfs_dir();
 	if (dmce == NULL)
 		goto err_out;
 	fseverities_coverage = debugfs_create_file("severities-coverage",
@@ -209,10 +210,7 @@ static int __init severities_debugfs_init(void)
 	return 0;
 
 err_out:
-	if (fseverities_coverage)
-		debugfs_remove(fseverities_coverage);
-	if (dmce)
-		debugfs_remove(dmce);
 	return -ENOMEM;
 }
 late_initcall(severities_debugfs_init);
+#endif
diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index fdd51b55435..2f5aab26320 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -34,6 +34,7 @@
 #include <linux/smp.h>
 #include <linux/fs.h>
 #include <linux/mm.h>
+#include <linux/debugfs.h>
 
 #include <asm/processor.h>
 #include <asm/hw_irq.h>
@@ -45,21 +46,8 @@
 
 #include "mce-internal.h"
 
-/* Handle unconfigured int18 (should never happen) */
-static void unexpected_machine_check(struct pt_regs *regs, long error_code)
-{
-	printk(KERN_ERR "CPU#%d: Unexpected int18 (Machine Check).\n",
-	       smp_processor_id());
-}
-
-/* Call the installed machine check handler for this CPU setup. */
-void (*machine_check_vector)(struct pt_regs *, long error_code) =
-						unexpected_machine_check;
-
 int mce_disabled __read_mostly;
 
-#ifdef CONFIG_X86_NEW_MCE
-
 #define MISC_MCELOG_MINOR	227
 
 #define SPINUNIT 100	/* 100ns */
@@ -77,7 +65,6 @@ DEFINE_PER_CPU(unsigned, mce_exception_count);
  */
 static int			tolerant		__read_mostly = 1;
 static int			banks			__read_mostly;
-static u64			*bank			__read_mostly;
 static int			rip_msr			__read_mostly;
 static int			mce_bootlog		__read_mostly = -1;
 static int			monarch_timeout		__read_mostly = -1;
@@ -87,13 +74,13 @@ int				mce_cmci_disabled	__read_mostly;
 int				mce_ignore_ce		__read_mostly;
 int				mce_ser			__read_mostly;
 
+struct mce_bank                *mce_banks		__read_mostly;
+
 /* User mode helper program triggered by machine check event */
 static unsigned long		mce_need_notify;
 static char			mce_helper[128];
 static char			*mce_helper_argv[2] = { mce_helper, NULL };
 
-static unsigned long		dont_init_banks;
-
 static DECLARE_WAIT_QUEUE_HEAD(mce_wait);
 static DEFINE_PER_CPU(struct mce, mces_seen);
 static int			cpu_missing;
@@ -104,11 +91,6 @@ DEFINE_PER_CPU(mce_banks_t, mce_poll_banks) = {
 	[0 ... BITS_TO_LONGS(MAX_NR_BANKS)-1] = ~0UL
 };
 
-static inline int skip_bank_init(int i)
-{
-	return i < BITS_PER_LONG && test_bit(i, &dont_init_banks);
-}
-
 static DEFINE_PER_CPU(struct work_struct, mce_work);
 
 /* Do initial initialization of a struct mce */
@@ -232,6 +214,9 @@ static void print_mce_tail(void)
 
 static atomic_t mce_paniced;
 
+static int fake_panic;
+static atomic_t mce_fake_paniced;
+
 /* Panic in progress. Enable interrupts and wait for final IPI */
 static void wait_for_panic(void)
 {
@@ -249,15 +234,21 @@ static void mce_panic(char *msg, struct mce *final, char *exp)
 {
 	int i;
 
-	/*
-	 * Make sure only one CPU runs in machine check panic
-	 */
-	if (atomic_add_return(1, &mce_paniced) > 1)
-		wait_for_panic();
-	barrier();
+	if (!fake_panic) {
+		/*
+		 * Make sure only one CPU runs in machine check panic
+		 */
+		if (atomic_inc_return(&mce_paniced) > 1)
+			wait_for_panic();
+		barrier();
 
-	bust_spinlocks(1);
-	console_verbose();
+		bust_spinlocks(1);
+		console_verbose();
+	} else {
+		/* Don't log too much for fake panic */
+		if (atomic_inc_return(&mce_fake_paniced) > 1)
+			return;
+	}
 	print_mce_head();
 	/* First print corrected ones that are still unlogged */
 	for (i = 0; i < MCE_LOG_LEN; i++) {
@@ -284,9 +275,12 @@ static void mce_panic(char *msg, struct mce *final, char *exp)
 	print_mce_tail();
 	if (exp)
 		printk(KERN_EMERG "Machine check: %s\n", exp);
-	if (panic_timeout == 0)
-		panic_timeout = mce_panic_timeout;
-	panic(msg);
+	if (!fake_panic) {
+		if (panic_timeout == 0)
+			panic_timeout = mce_panic_timeout;
+		panic(msg);
+	} else
+		printk(KERN_EMERG "Fake kernel panic: %s\n", msg);
 }
 
 /* Support code for software error injection */
@@ -296,11 +290,11 @@ static int msr_to_offset(u32 msr)
 	unsigned bank = __get_cpu_var(injectm.bank);
 	if (msr == rip_msr)
 		return offsetof(struct mce, ip);
-	if (msr == MSR_IA32_MC0_STATUS + bank*4)
+	if (msr == MSR_IA32_MCx_STATUS(bank))
 		return offsetof(struct mce, status);
-	if (msr == MSR_IA32_MC0_ADDR + bank*4)
+	if (msr == MSR_IA32_MCx_ADDR(bank))
 		return offsetof(struct mce, addr);
-	if (msr == MSR_IA32_MC0_MISC + bank*4)
+	if (msr == MSR_IA32_MCx_MISC(bank))
 		return offsetof(struct mce, misc);
 	if (msr == MSR_IA32_MCG_STATUS)
 		return offsetof(struct mce, mcgstatus);
@@ -505,7 +499,7 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b)
 
 	m.mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS);
 	for (i = 0; i < banks; i++) {
-		if (!bank[i] || !test_bit(i, *b))
+		if (!mce_banks[i].ctl || !test_bit(i, *b))
 			continue;
 
 		m.misc = 0;
@@ -514,7 +508,7 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b)
 		m.tsc = 0;
 
 		barrier();
-		m.status = mce_rdmsrl(MSR_IA32_MC0_STATUS + i*4);
+		m.status = mce_rdmsrl(MSR_IA32_MCx_STATUS(i));
 		if (!(m.status & MCI_STATUS_VAL))
 			continue;
 
@@ -529,9 +523,9 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b)
 			continue;
 
 		if (m.status & MCI_STATUS_MISCV)
-			m.misc = mce_rdmsrl(MSR_IA32_MC0_MISC + i*4);
+			m.misc = mce_rdmsrl(MSR_IA32_MCx_MISC(i));
 		if (m.status & MCI_STATUS_ADDRV)
-			m.addr = mce_rdmsrl(MSR_IA32_MC0_ADDR + i*4);
+			m.addr = mce_rdmsrl(MSR_IA32_MCx_ADDR(i));
 
 		if (!(flags & MCP_TIMESTAMP))
 			m.tsc = 0;
@@ -547,7 +541,7 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b)
 		/*
 		 * Clear state for this bank.
 		 */
-		mce_wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0);
+		mce_wrmsrl(MSR_IA32_MCx_STATUS(i), 0);
 	}
 
 	/*
@@ -568,7 +562,7 @@ static int mce_no_way_out(struct mce *m, char **msg)
 	int i;
 
 	for (i = 0; i < banks; i++) {
-		m->status = mce_rdmsrl(MSR_IA32_MC0_STATUS + i*4);
+		m->status = mce_rdmsrl(MSR_IA32_MCx_STATUS(i));
 		if (mce_severity(m, tolerant, msg) >= MCE_PANIC_SEVERITY)
 			return 1;
 	}
@@ -628,7 +622,7 @@ out:
  * This way we prevent any potential data corruption in a unrecoverable case
  * and also makes sure always all CPU's errors are examined.
  *
- * Also this detects the case of an machine check event coming from outer
+ * Also this detects the case of a machine check event coming from outer
  * space (not detected by any CPUs) In this case some external agent wants
  * us to shut down, so panic too.
  *
@@ -681,7 +675,7 @@ static void mce_reign(void)
 	 * No machine check event found. Must be some external
 	 * source or one CPU is hung. Panic.
 	 */
-	if (!m && tolerant < 3)
+	if (global_worst <= MCE_KEEP_SEVERITY && tolerant < 3)
 		mce_panic("Machine check from unknown source", NULL, NULL);
 
 	/*
@@ -715,7 +709,7 @@ static int mce_start(int *no_way_out)
 	 * global_nwo should be updated before mce_callin