Merge branch 'master' into x86/memblock

Conflicts & resolutions:

* arch/x86/xen/setup.c

	dc91c728fd "xen: allow extra memory to be in multiple regions"
	24aa07882b "memblock, x86: Replace memblock_x86_reserve/free..."

	conflicted on xen_add_extra_mem() updates.  The resolution is
	trivial as the latter just want to replace
	memblock_x86_reserve_range() with memblock_reserve().

* drivers/pci/intel-iommu.c

	166e9278a3f "x86/ia64: intel-iommu: move to drivers/iommu/"
	5dfe8660a3d "bootmem: Replace work_with_active_regions() with..."

	conflicted as the former moved the file under drivers/iommu/.
	Resolved by applying the chnages from the latter on the moved
	file.

* mm/Kconfig

	6661672053a "memblock: add NO_BOOTMEM config symbol"
	c378ddd53f9 "memblock, x86: Make ARCH_DISCARD_MEMBLOCK a config option"

	conflicted trivially.  Both added config options.  Just
	letting both add their own options resolves the conflict.

* mm/memblock.c

	d1f0ece6cdc "mm/memblock.c: small function definition fixes"
	ed7b56a799c "memblock: Remove memblock_memory_can_coalesce()"

	confliected.  The former updates function removed by the
	latter.  Resolution is trivial.

Signed-off-by: Tejun Heo <tj@kernel.org>

7 files changed, 304 insertions, 256 deletions

diff --git a/arch/x86/kernel/cpu/mcheck/mce-apei.c b/arch/x86/kernel/cpu/mcheck/mce-apei.c
index 83930deec3c..507ea58688e 100644
--- a/arch/x86/kernel/cpu/mcheck/mce-apei.c
+++ b/arch/x86/kernel/cpu/mcheck/mce-apei.c
@@ -28,6 +28,7 @@
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  */
 
+#include <linux/export.h>
 #include <linux/kernel.h>
 #include <linux/acpi.h>
 #include <linux/cper.h>
diff --git a/arch/x86/kernel/cpu/mcheck/mce-inject.c b/arch/x86/kernel/cpu/mcheck/mce-inject.c
index 0ed633c5048..319882ef848 100644
--- a/arch/x86/kernel/cpu/mcheck/mce-inject.c
+++ b/arch/x86/kernel/cpu/mcheck/mce-inject.c
@@ -78,27 +78,20 @@ static void raise_exception(struct mce *m, struct pt_regs *pregs)
 
 static cpumask_var_t mce_inject_cpumask;
 
-static int mce_raise_notify(struct notifier_block *self,
-			    unsigned long val, void *data)
+static int mce_raise_notify(unsigned int cmd, struct pt_regs *regs)
 {
-	struct die_args *args = (struct die_args *)data;
 	int cpu = smp_processor_id();
 	struct mce *m = &__get_cpu_var(injectm);
-	if (val != DIE_NMI || !cpumask_test_cpu(cpu, mce_inject_cpumask))
-		return NOTIFY_DONE;
+	if (!cpumask_test_cpu(cpu, mce_inject_cpumask))
+		return NMI_DONE;
 	cpumask_clear_cpu(cpu, mce_inject_cpumask);
 	if (m->inject_flags & MCJ_EXCEPTION)
-		raise_exception(m, args->regs);
+		raise_exception(m, regs);
 	else if (m->status)
 		raise_poll(m);
-	return NOTIFY_STOP;
+	return NMI_HANDLED;
 }
 
-static struct notifier_block mce_raise_nb = {
-	.notifier_call = mce_raise_notify,
-	.priority = NMI_LOCAL_NORMAL_PRIOR,
-};
-
 /* Inject mce on current CPU */
 static int raise_local(void)
 {
@@ -215,8 +208,9 @@ static int inject_init(void)
 	if (!alloc_cpumask_var(&mce_inject_cpumask, GFP_KERNEL))
 		return -ENOMEM;
 	printk(KERN_INFO "Machine check injector initialized\n");
-	mce_chrdev_ops.write = mce_write;
-	register_die_notifier(&mce_raise_nb);
+	register_mce_write_callback(mce_write);
+	register_nmi_handler(NMI_LOCAL, mce_raise_notify, 0,
+				"mce_notify");
 	return 0;
 }
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce-severity.c b/arch/x86/kernel/cpu/mcheck/mce-severity.c
index 1e8d66c1336..7395d5f4272 100644
--- a/arch/x86/kernel/cpu/mcheck/mce-severity.c
+++ b/arch/x86/kernel/cpu/mcheck/mce-severity.c
@@ -43,61 +43,105 @@ static struct severity {
 	unsigned char covered;
 	char *msg;
 } severities[] = {
-#define KERNEL .context = IN_KERNEL
-#define USER .context = IN_USER
-#define SER .ser = SER_REQUIRED
-#define NOSER .ser = NO_SER
-#define SEV(s) .sev = MCE_ ## s ## _SEVERITY
-#define BITCLR(x, s, m, r...) { .mask = x, .result = 0, SEV(s), .msg = m, ## r }
-#define BITSET(x, s, m, r...) { .mask = x, .result = x, SEV(s), .msg = m, ## r }
-#define MCGMASK(x, res, s, m, r...) \
-	{ .mcgmask = x, .mcgres = res, SEV(s), .msg = m, ## r }
-#define MASK(x, y, s, m, r...) \
-	{ .mask = x, .result = y, SEV(s), .msg = m, ## r }
+#define MCESEV(s, m, c...) { .sev = MCE_ ## s ## _SEVERITY, .msg = m, ## c }
+#define  KERNEL		.context = IN_KERNEL
+#define  USER		.context = IN_USER
+#define  SER		.ser = SER_REQUIRED
+#define  NOSER		.ser = NO_SER
+#define  BITCLR(x)	.mask = x, .result = 0
+#define  BITSET(x)	.mask = x, .result = x
+#define  MCGMASK(x, y)	.mcgmask = x, .mcgres = y
+#define  MASK(x, y)	.mask = x, .result = y
 #define MCI_UC_S (MCI_STATUS_UC|MCI_STATUS_S)
 #define MCI_UC_SAR (MCI_STATUS_UC|MCI_STATUS_S|MCI_STATUS_AR)
 #define MCACOD 0xffff
 
-	BITCLR(MCI_STATUS_VAL, NO, "Invalid"),
-	BITCLR(MCI_STATUS_EN, NO, "Not enabled"),
-	BITSET(MCI_STATUS_PCC, PANIC, "Processor context corrupt"),
+	MCESEV(
+		NO, "Invalid",
+		BITCLR(MCI_STATUS_VAL)
+		),
+	MCESEV(
+		NO, "Not enabled",
+		BITCLR(MCI_STATUS_EN)
+		),
+	MCESEV(
+		PANIC, "Processor context corrupt",
+		BITSET(MCI_STATUS_PCC)
+		),
 	/* When MCIP is not set something is very confused */
-	MCGMASK(MCG_STATUS_MCIP, 0, PANIC, "MCIP not set in MCA handler"),
+	MCESEV(
+		PANIC, "MCIP not set in MCA handler",
+		MCGMASK(MCG_STATUS_MCIP, 0)
+		),
 	/* Neither return not error IP -- no chance to recover -> PANIC */
-	MCGMASK(MCG_STATUS_RIPV|MCG_STATUS_EIPV, 0, PANIC,
-		"Neither restart nor error IP"),
-	MCGMASK(MCG_STATUS_RIPV, 0, PANIC, "In kernel and no restart IP",
-		KERNEL),
-	BITCLR(MCI_STATUS_UC, KEEP, "Corrected error", NOSER),
-	MASK(MCI_STATUS_OVER|MCI_STATUS_UC|MCI_STATUS_EN, MCI_STATUS_UC, SOME,
-	     "Spurious not enabled", SER),
+	MCESEV(
+		PANIC, "Neither restart nor error IP",
+		MCGMASK(MCG_STATUS_RIPV|MCG_STATUS_EIPV, 0)
+		),
+	MCESEV(
+		PANIC, "In kernel and no restart IP",
+		KERNEL, MCGMASK(MCG_STATUS_RIPV, 0)
+		),
+	MCESEV(
+		KEEP, "Corrected error",
+		NOSER, BITCLR(MCI_STATUS_UC)
+		),
 
 	/* ignore OVER for UCNA */
-	MASK(MCI_UC_SAR, MCI_STATUS_UC, KEEP,
-	     "Uncorrected no action required", SER),
-	MASK(MCI_STATUS_OVER|MCI_UC_SAR, MCI_STATUS_UC|MCI_STATUS_AR, PANIC,
-	     "Illegal combination (UCNA with AR=1)", SER),
-	MASK(MCI_STATUS_S, 0, KEEP, "Non signalled machine check", SER),
+	MCESEV(
+		KEEP, "Uncorrected no action required",
+		SER, MASK(MCI_UC_SAR, MCI_STATUS_UC)
+		),
+	MCESEV(
+		PANIC, "Illegal combination (UCNA with AR=1)",
+		SER,
+		MASK(MCI_STATUS_OVER|MCI_UC_SAR, MCI_STATUS_UC|MCI_STATUS_AR)
+		),
+	MCESEV(
+		KEEP, "Non signalled machine check",
+		SER, BITCLR(MCI_STATUS_S)
+		),
 
 	/* AR add known MCACODs here */
-	MASK(MCI_STATUS_OVER|MCI_UC_SAR, MCI_STATUS_OVER|MCI_UC_SAR, PANIC,
-	     "Action required with lost events", SER),
-	MASK(MCI_STATUS_OVER|MCI_UC_SAR|MCACOD, MCI_UC_SAR, PANIC,
-	     "Action required; unknown MCACOD", SER),
+	MCESEV(
+		PANIC, "Action required with lost events",
+		SER, BITSET(MCI_STATUS_OVER|MCI_UC_SAR)
+		),
+	MCESEV(
+		PANIC, "Action required: unknown MCACOD",
+		SER, MASK(MCI_STATUS_OVER|MCI_UC_SAR, MCI_UC_SAR)
+		),
 
 	/* known AO MCACODs: */
-	MASK(MCI_UC_SAR|MCI_STATUS_OVER|0xfff0, MCI_UC_S|0xc0, AO,
-	     "Action optional: memory scrubbing error", SER),
-	MASK(MCI_UC_SAR|MCI_STATUS_OVER|MCACOD, MCI_UC_S|0x17a, AO,
-	     "Action optional: last level cache writeback error", SER),
-
-	MASK(MCI_STATUS_OVER|MCI_UC_SAR, MCI_UC_S, SOME,
-	     "Action optional unknown MCACOD", SER),
-	MASK(MCI_STATUS_OVER|MCI_UC_SAR, MCI_UC_S|MCI_STATUS_OVER, SOME,
-	     "Action optional with lost events", SER),
-	BITSET(MCI_STATUS_UC|MCI_STATUS_OVER, PANIC, "Overflowed uncorrected"),
-	BITSET(MCI_STATUS_UC, UC, "Uncorrected"),
-	BITSET(0, SOME, "No match")	/* always matches. keep at end */
+	MCESEV(
+		AO, "Action optional: memory scrubbing error",
+		SER, MASK(MCI_STATUS_OVER|MCI_UC_SAR|0xfff0, MCI_UC_S|0x00c0)
+		),
+	MCESEV(
+		AO, "Action optional: last level cache writeback error",
+		SER, MASK(MCI_STATUS_OVER|MCI_UC_SAR|MCACOD, MCI_UC_S|0x017a)
+		),
+	MCESEV(
+		SOME, "Action optional: unknown MCACOD",
+		SER, MASK(MCI_STATUS_OVER|MCI_UC_SAR, MCI_UC_S)
+		),
+	MCESEV(
+		SOME, "Action optional with lost events",
+		SER, MASK(MCI_STATUS_OVER|MCI_UC_SAR, MCI_STATUS_OVER|MCI_UC_S)
+		),
+
+	MCESEV(
+		PANIC, "Overflowed uncorrected",
+		BITSET(MCI_STATUS_OVER|MCI_STATUS_UC)
+		),
+	MCESEV(
+		UC, "Uncorrected",
+		BITSET(MCI_STATUS_UC)
+		),
+	MCESEV(
+		SOME, "No match",
+		BITSET(0)
+		)	/* always matches. keep at end */
 };
 
 /*
@@ -112,15 +156,15 @@ static int error_context(struct mce *m)
 	return IN_KERNEL;
 }
 
-int mce_severity(struct mce *a, int tolerant, char **msg)
+int mce_severity(struct mce *m, int tolerant, char **msg)
 {
-	enum context ctx = error_context(a);
+	enum context ctx = error_context(m);
 	struct severity *s;
 
 	for (s = severities;; s++) {
-		if ((a->status & s->mask) != s->result)
+		if ((m->status & s->mask) != s->result)
 			continue;
-		if ((a->mcgstatus & s->mcgmask) != s->mcgres)
+		if ((m->mcgstatus & s->mcgmask) != s->mcgres)
 			continue;
 		if (s->ser == SER_REQUIRED && !mce_ser)
 			continue;
@@ -197,15 +241,15 @@ static const struct file_operations severities_coverage_fops = {
 
 static int __init severities_debugfs_init(void)
 {
-	struct dentry *dmce = NULL, *fseverities_coverage = NULL;
+	struct dentry *dmce, *fsev;
 
 	dmce = mce_get_debugfs_dir();
-	if (dmce == NULL)
+	if (!dmce)
 		goto err_out;
-	fseverities_coverage = debugfs_create_file("severities-coverage",
-						   0444, dmce, NULL,
-						   &severities_coverage_fops);
-	if (fseverities_coverage == NULL)
+
+	fsev = debugfs_create_file("severities-coverage", 0444, dmce, NULL,
+				   &severities_coverage_fops);
+	if (!fsev)
 		goto err_out;
 
 	return 0;
@@ -214,4 +258,4 @@ err_out:
 	return -ENOMEM;
 }
 late_initcall(severities_debugfs_init);
-#endif
+#endif /* CONFIG_DEBUG_FS */
diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index ff1ae9b6464..2af127d4c3d 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -10,7 +10,6 @@
 #include <linux/thread_info.h>
 #include <linux/capability.h>
 #include <linux/miscdevice.h>
-#include <linux/interrupt.h>
 #include <linux/ratelimit.h>
 #include <linux/kallsyms.h>
 #include <linux/rcupdate.h>
@@ -37,24 +36,21 @@
 #include <linux/fs.h>
 #include <linux/mm.h>
 #include <linux/debugfs.h>
-#include <linux/edac_mce.h>
+#include <linux/irq_work.h>
+#include <linux/export.h>
 
 #include <asm/processor.h>
-#include <asm/hw_irq.h>
-#include <asm/apic.h>
-#include <asm/idle.h>
-#include <asm/ipi.h>
 #include <asm/mce.h>
 #include <asm/msr.h>
 
 #include "mce-internal.h"
 
-static DEFINE_MUTEX(mce_read_mutex);
+static DEFINE_MUTEX(mce_chrdev_read_mutex);
 
 #define rcu_dereference_check_mce(p) \
 	rcu_dereference_index_check((p), \
 			      rcu_read_lock_sched_held() || \
-			      lockdep_is_held(&mce_read_mutex))
+			      lockdep_is_held(&mce_chrdev_read_mutex))
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/mce.h>
@@ -94,7 +90,8 @@ static unsigned long		mce_need_notify;
 static char			mce_helper[128];
 static char			*mce_helper_argv[2] = { mce_helper, NULL };
 
-static DECLARE_WAIT_QUEUE_HEAD(mce_wait);
+static DECLARE_WAIT_QUEUE_HEAD(mce_chrdev_wait);
+
 static DEFINE_PER_CPU(struct mce, mces_seen);
 static int			cpu_missing;
 
@@ -147,23 +144,20 @@ static struct mce_log mcelog = {
 void mce_log(struct mce *mce)
 {
 	unsigned next, entry;
+	int ret = 0;
 
 	/* Emit the trace record: */
 	trace_mce_record(mce);
 
+	ret = atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, mce);
+	if (ret == NOTIFY_STOP)
+		return;
+
 	mce->finished = 0;
 	wmb();
 	for (;;) {
 		entry = rcu_dereference_check_mce(mcelog.next);
 		for (;;) {
-			/*
-			 * If edac_mce is enabled, it will check the error type
-			 * and will process it, if it is a known error.
-			 * Otherwise, the error will be sent through mcelog
-			 * interface
-			 */
-			if (edac_mce_parse(mce))
-				return;
 
 			/*
 			 * When the buffer fills up discard new entries.
@@ -220,8 +214,13 @@ static void print_mce(struct mce *m)
 		pr_cont("MISC %llx ", m->misc);
 
 	pr_cont("\n");
-	pr_emerg(HW_ERR "PROCESSOR %u:%x TIME %llu SOCKET %u APIC %x\n",
-		m->cpuvendor, m->cpuid, m->time, m->socketid, m->apicid);
+	/*
+	 * Note this output is parsed by external tools and old fields
+	 * should not be changed.
+	 */
+	pr_emerg(HW_ERR "PROCESSOR %u:%x TIME %llu SOCKET %u APIC %x microcode %x\n",
+		m->cpuvendor, m->cpuid, m->time, m->socketid, m->apicid,
+		cpu_data(m->extcpu).microcode);
 
 	/*
 	 * Print out human-readable details about the MCE error,
@@ -373,6 +372,31 @@ static void mce_wrmsrl(u32 msr, u64 v)
 }
 
 /*
+ * Collect all global (w.r.t. this processor) status about this machine
+ * check into our "mce" struct so that we can use it later to assess
+ * the severity of the problem as we read per-bank specific details.
+ */
+static inline void mce_gather_info(struct mce *m, struct pt_regs *regs)
+{
+	mce_setup(m);
+
+	m->mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS);
+	if (regs) {
+		/*
+		 * Get the address of the instruction at the time of
+		 * the machine check error.
+		 */
+		if (m->mcgstatus & (MCG_STATUS_RIPV|MCG_STATUS_EIPV)) {
+			m->ip = regs->ip;
+			m->cs = regs->cs;
+		}
+		/* Use accurate RIP reporting if available. */
+		if (rip_msr)
+			m->ip = mce_rdmsrl(rip_msr);
+	}
+}
+
+/*
  * Simple lockless ring to communicate PFNs from the exception handler with the
  * process context work function. This is vastly simplified because there's
  * only a single reader and a single writer.
@@ -443,40 +467,13 @@ static void mce_schedule_work(void)
 	}
 }
 
-/*
- * Get the address of the instruction at the time of the machine check
- * error.
- */
-static inline void mce_get_rip(struct mce *m, struct pt_regs *regs)
-{
-
-	if (regs && (m->mcgstatus & (MCG_STATUS_RIPV|MCG_STATUS_EIPV))) {
-		m->ip = regs->ip;
-		m->cs = regs->cs;
-	} else {
-		m->ip = 0;
-		m->cs = 0;
-	}
-	if (rip_msr)
-		m->ip = mce_rdmsrl(rip_msr);
-}
+DEFINE_PER_CPU(struct irq_work, mce_irq_work);
 
-#ifdef CONFIG_X86_LOCAL_APIC
-/*
- * Called after interrupts have been reenabled again
- * when a MCE happened during an interrupts off region
- * in the kernel.
- */
-asmlinkage void smp_mce_self_interrupt(struct pt_regs *regs)
+static void mce_irq_work_cb(struct irq_work *entry)
 {
-	ack_APIC_irq();
-	exit_idle();
-	irq_enter();
 	mce_notify_irq();
 	mce_schedule_work();
-	irq_exit();
 }
-#endif
 
 static void mce_report_event(struct pt_regs *regs)
 {
@@ -492,29 +489,7 @@ static void mce_report_event(struct pt_regs *regs)
 		return;
 	}
 
-#ifdef CONFIG_X86_LOCAL_APIC
-	/*
-	 * Without APIC do not notify. The event will be picked
-	 * up eventually.
-	 */
-	if (!cpu_has_apic)
-		return;
-
-	/*
-	 * When interrupts are disabled we cannot use
-	 * kernel services safely. Trigger an self interrupt
-	 * through the APIC to instead do the notification
-	 * after interrupts are reenabled again.
-	 */
-	apic->send_IPI_self(MCE_SELF_VECTOR);
-
-	/*
-	 * Wait for idle afterwards again so that we don't leave the
-	 * APIC in a non idle state because the normal APIC writes
-	 * cannot exclude us.
-	 */
-	apic_wait_icr_idle();
-#endif
+	irq_work_queue(&__get_cpu_var(mce_irq_work));
 }
 
 DEFINE_PER_CPU(unsigned, mce_poll_count);
@@ -541,9 +516,8 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b)
 
 	percpu_inc(mce_poll_count);
 
-	mce_setup(&m);
+	mce_gather_info(&m, NULL);
 
-	m.mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS);
 	for (i = 0; i < banks; i++) {
 		if (!mce_banks[i].ctl || !test_bit(i, *b))
 			continue;
@@ -579,10 +553,8 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b)
 		 * Don't get the IP here because it's unlikely to
 		 * have anything to do with the actual error location.
 		 */
-		if (!(flags & MCP_DONTLOG) && !mce_dont_log_ce) {
+		if (!(flags & MCP_DONTLOG) && !mce_dont_log_ce)
 			mce_log(&m);
-			atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, &m);
-		}
 
 		/*
 		 * Clear state for this bank.
@@ -879,9 +851,9 @@ static int mce_usable_address(struct mce *m)
 {
 	if (!(m->status & MCI_STATUS_MISCV) || !(m->status & MCI_STATUS_ADDRV))
 		return 0;
-	if ((m->misc & 0x3f) > PAGE_SHIFT)
+	if (MCI_MISC_ADDR_LSB(m->misc) > PAGE_SHIFT)
 		return 0;
-	if (((m->misc >> 6) & 7) != MCM_ADDR_PHYS)
+	if (MCI_MISC_ADDR_MODE(m->misc) != MCI_MISC_ADDR_PHYS)
 		return 0;
 	return 1;
 }
@@ -936,15 +908,11 @@ void do_machine_check(struct pt_regs *regs, long error_code)
 
 	percpu_inc(mce_exception_count);
 
-	if (notify_die(DIE_NMI, "machine check", regs, error_code,
-			   18, SIGKILL) == NOTIFY_STOP)
-		goto out;
 	if (!banks)
 		goto out;
 
-	mce_setup(&m);
+	mce_gather_info(&m, regs);
 
-	m.mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS);
 	final = &__get_cpu_var(mces_seen);
 	*final = m;
 
@@ -1028,7 +996,6 @@ void do_machine_check(struct pt_regs *regs, long error_code)
 		if (severity == MCE_AO_SEVERITY && mce_usable_address(&m))
 			mce_ring_add(m.addr >> PAGE_SHIFT);
 
-		mce_get_rip(&m, regs);
 		mce_log(&m);
 
 		if (severity > worst) {
@@ -1170,6 +1137,15 @@ static void mce_start_timer(unsigned long data)
 	add_timer_on(t, smp_processor_id());
 }
 
+/* Must not be called in IRQ context where del_timer_sync() can deadlock */
+static void mce_timer_delete_all(void)
+{
+	int cpu;
+
+	for_each_online_cpu(cpu)
+		del_timer_sync(&per_cpu(mce_timer, cpu));
+}
+
 static void mce_do_trigger(struct work_struct *work)
 {
 	call_usermodehelper(mce_helper, mce_helper_argv, NULL, UMH_NO_WAIT);
@@ -1190,7 +1166,8 @@ int mce_notify_irq(void)
 	clear_thread_flag(TIF_MCE_NOTIFY);
 
 	if (test_and_clear_bit(0, &mce_need_notify)) {
-		wake_up_interruptible(&mce_wait);
+		/* wake processes polling /dev/mcelog */
+		wake_up_interruptible(&mce_chrdev_wait);
 
 		/*
 		 * There is no risk of missing notifications because
@@ -1363,18 +1340,23 @@ static int __cpuinit __mcheck_cpu_apply_quirks(struct cpuinfo_x86 *c)
 	return 0;
 }
 
-static void __cpuinit __mcheck_cpu_ancient_init(struct cpuinfo_x86 *c)
+static int __cpuinit __mcheck_cpu_ancient_init(struct cpuinfo_x86 *c)
 {
 	if (c->x86 != 5)
-		return;
+		return 0;
+
 	switch (c->x86_vendor) {
 	case X86_VENDOR_INTEL:
 		intel_p5_mcheck_init(c);
+		return 1;
 		break;
 	case X86_VENDOR_CENTAUR:
 		winchip_mcheck_init(c);
+		return 1;
 		break;
 	}
+
+	return 0;
 }
 
 static void __mcheck_cpu_init_vendor(struct cpuinfo_x86 *c)
@@ -1428,7 +1410,8 @@ void __cpuinit mcheck_cpu_init(struct cpuinfo_x86 *c)
 	if (mce_disabled)
 		return;
 
-	__mcheck_cpu_ancient_init(c);
+	if (__mcheck_cpu_ancient_init(c))
+		return;
 
 	if (!mce_available(c))
 		return;
@@ -1444,44 +1427,45 @@ void __cpuinit mcheck_cpu_init(struct cpuinfo_x86 *c)
 	__mcheck_cpu_init_vendor(c);
 	__mcheck_cpu_init_timer();
 	INIT_WORK(&__get_cpu_var(mce_work), mce_process_work);
-
+	init_irq_work(&__get_cpu_var(mce_irq_work), &mce_irq_work_cb);
 }
 
 /*
- * Character device to read and clear the MCE log.
+ * mce_chrdev: Character device /dev/mcelog to read and clear the MCE log.
  */
 
-static DEFINE_SPINLOCK(mce_state_lock);
-static int		open_count;		/* #times opened */
-static int		open_exclu;		/* already open exclusive? */
+static DEFINE_SPINLOCK(mce_chrdev_state_lock);
+static int mce_chrdev_open_count;	/* #times opened */
+static int mce_chrdev_open_exclu;	/* already open exclusive? */
 
-static int mce_open(struct inode *inode, struct file *file)
+static int mce_chrdev_open(struct inode *inode, struct file *file)
 {
-	spin_lock(&mce_state_lock);
+	spin_lock(&mce_chrdev_state_lock);
 
-	if (open_exclu || (open_count && (file->f_flags & O_EXCL))) {
-		spin_unlock(&mce_state_lock);
+	if (mce_chrdev_open_exclu ||
+	    (mce_chrdev_open_count && (file->f_flags & O_EXCL))) {
+		spin_unlock(&mce_chrdev_state_lock);
 
 		return -EBUSY;
 	}
 
 	if (file->f_flags & O_EXCL)
-		open_exclu = 1;
-	open_count++;
+		mce_chrdev_open_exclu = 1;
+	mce_chrdev_open_count++;
 
-	spin_unlock(&mce_state_lock);
+	spin_unlock(&mce_chrdev_state_lock);
 
 	return nonseekable_open(inode, file);
 }
 
-static int mce_release(struct inode *inode, struct file *file)
+static int mce_chrdev_release(struct inode *inode, struct file *file)
 {
-	spin_lock(&mce_state_lock);
+	spin_lock(&mce_chrdev_state_lock);
 
-	open_count--;
-	open_exclu = 0;
+	mce_chrdev_open_count--;
+	mce_chrdev_open_exclu = 0;
 
-	spin_unlock(&mce_state_lock);
+	spin_unlock(&mce_chrdev_state_lock);
 
 	return 0;
 }
@@ -1530,8 +1514,8 @@ static int __mce_read_apei(char __user **ubuf, size_t usize)
 	return 0;
 }
 
-static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize,
-			loff_t *off)
+static ssize_t mce_chrdev_read(struct file *filp, char __user *ubuf,
+				size_t usize, loff_t *off)
 {
 	char __user *buf = ubuf;
 	unsigned long *cpu_tsc;
@@ -1542,7 +1526,7 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize,
 	if (!cpu_tsc)
 		return -ENOMEM;
 
-	mutex_lock(&mce_read_mutex);
+	mutex_lock(&mce_chrdev_read_mutex);
 
 	if (!mce_apei_read_done) {
 		err = __mce_read_apei(&buf, usize);
@@ -1562,19 +1546,18 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize,
 	do {
 		for (i = prev; i < next; i++) {
 			unsigned long start = jiffies;
+			struct mce *m = &mcelog.entry[i];
 
-			while (!mcelog.entry[i].finished) {
+			while (!m->finished) {
 				if (time_after_eq(jiffies, start + 2)) {
-					memset(mcelog.entry + i, 0,
-					       sizeof(struct mce));
+					memset(m, 0, sizeof(*m));
 					goto timeout;
 				}
 				cpu_relax();
 			}
 			smp_rmb();
-			err |= copy_to_user(buf, mcelog.entry + i,
-					    sizeof(struct mce));
-			buf += sizeof(struct mce);
+			err |= copy_to_user(buf, m, sizeof(*m));
+			buf += sizeof(*m);
 timeout:
 			;
 		}
@@ -1594,13 +1577,13 @@ timeout:
 	on_each_cpu(collect_tscs, cpu_tsc, 1);
 
 	for (i = next; i < MCE_LOG_LEN; i++) {
-		if (mcelog.entry[i].finished &&
-		    mcelog.entry[i].tsc < cpu_tsc[mcelog.entry[i].cpu]) {
-			err |= copy_to_user(buf, mcelog.entry+i,
-					    sizeof(struct mce));
+		struct mce *m = &mcelog.entry[i];
+
+		if (m->finished && m->tsc < cpu_tsc[m->cpu]) {
+			err |= copy_to_user(buf, m, sizeof(*m));
 			smp_rmb();
-			buf += sizeof(struct mce);
-			memset(&mcelog.entry[i], 0, sizeof(struct mce));
+			buf += sizeof(*m);
+			memset(m, 0, sizeof(*m));
 		}
 	}
 
@@ -1608,15 +1591,15 @@ timeout:
 		err = -EFAULT;
 
 out:
-	mutex_unlock(&mce_read_mutex);
+	mutex_unlock(&mce_chrdev_read_mutex);
 	kfree(cpu_tsc);
 
 	return err ? err : buf - ubuf;
 }
 
-static unsigned int mce_poll(struct file *file, poll_table *wait)
+static unsigned int mce_chrdev_poll(struct file *file, poll_table *wait)
 {
-	poll_wait(file, &mce_wait, wait);
+	poll_wait(file, &mce_chrdev_wait, wait);
 	if (rcu_access_index(mcelog.next))
 		return POLLIN | POLLRDNORM;
 	if (!mce_apei_read_done && apei_check_mce())
@@ -1624,7 +1607,8 @@ static unsigned int mce_poll(struct file *file, poll_table *wait)
 	return 0;
 }
 
-static long mce_ioctl(struct file *f, unsigned int cmd, unsigned long arg)
+static long mce_chrdev_ioctl(struct file *f, unsigned int cmd,
+				unsigned long arg)
 {
 	int __user *p = (int __user *)arg;
 
@@ -1650,18 +1634,37 @@ static long mce_ioctl(struct file *f, unsigned int cmd, unsigned long arg)
 	}
 }
 
-/* Modified in mce-inject.c, so not static or const */
-struct file_operations mce_chrdev_ops = {
-	.open			= mce_open,
-	.release		= mce_release,
-	.read			= mce_read,
-	.poll			= mce_poll,
-	.unlocked_ioctl		= mce_ioctl,
-	.llseek		= no_llseek,
+static ssize_t (*mce_write)(struct file *filp, const char __user *ubuf,
+			    size_t usize, loff_t *off);
+
+void register_mce_write_callback(ssize_t (*fn)(struct file *filp,
+			     const char __user *ubuf,
+			     size_t usize, loff_t *off))
+{
+	mce_write = fn;
+}
+EXPORT_SYMBOL_GPL(register_mce_write_callback);
+
+ssize_t mce_chrdev_write(struct file *filp, const char __user *ubuf,
+			 size_t usize, loff_t *off)
+{
+	if (mce_write)
+		return mce_write(filp, ubuf, usize, off);
+	else
+		return -EINVAL;
+}
+
+static const struct file_operations mce_chrdev_ops = {
+	.open			= mce_chrdev_open,
+	.release		= mce_chrdev_release,
+	.read			= mce_chrdev_read,
+	.write			= mce_chrdev_write,
+	.poll			= mce_chrdev_poll,
+	.unlocked_ioctl		= mce_chrdev_ioctl,
+	.llseek			= no_llseek,
 };
-EXPORT_SYMBOL_GPL(mce_chrdev_ops);
 
-static struct miscdevice mce_log_device = {
+static struct miscdevice mce_chrdev_device = {
 	MISC_MCELOG_MINOR,
 	"mcelog",
 	&mce_chrdev_ops,
@@ -1719,7 +1722,7 @@ int __init mcheck_init(void)
 }
 
 /*
- * Sysfs support
+ * mce_syscore: PM support
  */
 
 /*
@@ -1739,12 +1742,12 @@ static int mce_disable_error_reporting(void)
 	return 0;
 }
 
-static int mce_suspend(void)
+static int mce_syscore_suspend(void)
 {
 	return mce_disable_error_reporting();
 }
 
-static void mce_shutdown(void)
+static void mce_syscore_shutdown(void)
 {
 	mce_disable_error_reporting();
 }
@@ -1754,21 +1757,24 @@ static void mce_shutdown(void)
  * Only one CPU is active at this time, the others get re-added later using
  * CPU hotplug:
  */
-static void mce_resume(void)
+static void mce_syscore_resume(void)
 {
 	__mcheck_cpu_init_generic();
 	__mcheck_cpu_init_vendor(__this_cpu_ptr(&cpu_info));
 }
 
 static struct syscore_ops mce_syscore_ops = {
-	.suspend	= mce_suspend,
-	.shutdown	= mce_shutdown,
-	.resume		= mce_resume,
+	.suspend	= mce_syscore_suspend,
+	.shutdown	= mce_syscore_shutdown,
+	.resume		= mce_syscore_resume,
 };
 
+/*
+ * mce_sysdev: Sysfs support
+ */
+
 static void mce_cpu_restart(void *data)
 {
-	del_timer_sync(&__get_cpu_var(mce_timer));
 	if (!mce_available(__this_cpu_ptr(&cpu_info)))
 		return;
 	__mcheck_cpu_init_generic();
@@ -1778,16 +1784,15 @@ static void mce_cpu_restart(void *data)
 /* Reinit MCEs after user configuration changes */
 static void mce_restart(void)
 {
+	mce_timer_delete_all();
 	on_each_cpu(mce_cpu_restart, NULL, 1);
 }
 
 /* Toggle features for corrected errors */
-static void mce_disable_ce(void *all)
+static void mce_disable_cmci(void *data)
 {
 	if (!mce_available(__this_cpu_ptr(&cpu_info)))
 		return;
-	if (all)
-		del_timer_sync(&__get_cpu_var(mce_timer));
 	cmci_clear();
 }
 
@@ -1801,11 +1806,11 @@ static void mce_enable_ce(void *all)
 		__mcheck_cpu_init_timer();
 }
 
-static struct sysdev_class mce_sysclass = {
+static struct sysdev_class mce_sysdev_class = {
 	.name		= "machinecheck",
 };
 
-DEFINE_PER_CPU(struct sys_device, mce_dev);
+DEFINE_PER_CPU(struct sys_device, mce_sysdev);
 
 __cpuinitdata
 void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu);
@@ -1870,7 +1875,8 @@ static ssize_t set_ignore_ce(struct sys_device *s,
 	if (mce_ignore_ce ^ !!new) {
 		if (new) {
 			/* disable ce features */
-			on_each_cpu(mce_disable_ce, (void *)1, 1);
+			mce_timer_delete_all();
+			on_each_cpu(mce_disable_cmci, NULL, 1);
 			mce_ignore_ce = 1;
 		} else {
 			/* enable ce features */
@@ -1893,7 +1899,7 @@ static ssize_t set_cmci_disabled(struct sys_device *s,
 	if (mce_cmci_disabled ^ !!new) {
 		if (new) {
 			/* disable cmci */
-			on_each_cpu(mce_disable_ce, NULL, 1);
+			on_each_cpu(mce_disable_cmci, NULL, 1);
 			mce_cmci_disabled = 1;
 		} else {
 			/* enable cmci */
@@ -1934,7 +1940,7 @@ static struct sysdev_ext_attribute attr_cmci_disabled = {
 	&mce_cmci_disabled
 };
 
-static struct sysdev_attribute *mce_attrs[] = {
+static struct sysdev_attribute *mce_sysdev_attrs[] = {
 	&attr_tolerant.attr,
 	&attr_check_interval.attr,
 	&attr_trigger,
@@ -1945,66 +1951,67 @@ static struct sysdev_attribute *mce_attrs[] = {
 	NULL
 };
 
-static cpumask_var_t mce_dev_initialized;
+static cpumask_var_t mce_sysdev_initialized;
 
 /* Per cpu sysdev init. All of the cpus still share the same ctrl bank: */
-static __cpuinit int mce_create_device(unsigned int cpu)
+static __cpuinit int mce_sysdev_create(unsigned int cpu)
 {
+	struct sys_device *sysdev = &per_cpu(mce_sysdev, cpu);
 	int err;
 	int i, j;
 
 	if (!mce_available(&boot_cpu_data))
 		return -EIO;
 
-	memset(&per_cpu(mce_dev, cpu).kobj, 0, sizeof(struct kobject));
-	per_cpu(mce_dev, cpu).id	= cpu;
-	per_cpu(mce_dev, cpu).cls	= &mce_sysclass;
+	memset(&sysdev->kobj, 0, sizeof(struct kobject));
+	sysdev->id  = cpu;
+	sysdev->cls = &mce_sysdev_class;
 
-	err = sysdev_register(&per_cpu(mce_dev, cpu));
+	err = sysdev_register(sysdev);
 	if (err)
 		return err;
 
-	for (i = 0; mce_attrs[i]; i++) {
-		err = sysdev_create_file(&per_cpu(mce_dev, cpu), mce_attrs[i]);
+	for (i = 0; mce_sysdev_attrs[i]; i++) {
+		err = sysdev_create_file(sysdev, mce_sysdev_attrs[i]);
 		if (err)
 			goto error;
 	}
 	for (j = 0; j < banks; j++) {
-		err = sysdev_create_file(&per_cpu(mce_dev, cpu),
-					&mce_banks[j].attr);
+		err = sysdev_create_file(sysdev, &mce_banks[j].attr);
 		if (err)
 			goto error2;
 	}
-	cpumask_set_cpu(cpu, mce_dev_initialized);
+	cpumask_set_cpu(cpu, mce_sysdev_initialized);
 
 	return 0;
 error2:
 	while (--j >= 0)
-		sysdev_remove_file(&per_cpu(mce_dev, cpu), &mce_banks[j].attr);
+		sysdev_remove_file(sysdev, &mce_banks[j].attr);
 error:
 	while (--i >= 0)
-		sysdev_remove_file(&per_cpu(mce_dev, cpu), mce_attrs[i]);
+		sysdev_remove_file(sysdev, mce_sysdev_attrs[i]);
 
-	sysdev_unregister(&per_cpu(mce_dev, cpu));
+	sysdev_unregister(sysdev);
 
 	return err;
 }
 
-static __cpuinit void mce_remove_device(unsigned int cpu)
+static __cpuinit void mce_sysdev_remove(unsigned int cpu)
 {
+	struct sys_device *sysdev = &per_cpu(mce_sysdev, cpu);
 	int i;
 
-	if (!cpumask_test_cpu(cpu, mce_dev_initialized))
+	if (!cpumask_test_cpu(cpu, mce_sysdev_initialized))
 		return;
 
-	for (i = 0; mce_attrs[i]; i++)
-		sysdev_remove_file(&per_cpu(mce_dev, cpu), mce_attrs[i]);
+	for (i = 0; mce_sysdev_attrs[i]; i++)
+		sysdev_remove_file(sysdev, mce_sysdev_attrs[i]);
 
 	for (i = 0; i < banks; i++)
-		sysdev_remove_file(&per_cpu(mce_dev, cpu), &mce_banks[i].attr);
+		sysdev_remove_file(sysdev, &mce_banks[i].attr);
 
-	sysdev_unregister(&per_cpu(mce_dev, cpu));
-	cpumask_clear_cpu(cpu, mce_dev_initialized);
+	sysdev_unregister(sysdev);
+	cpumask_clear_cpu(cpu, mce_sysdev_initialized);
 }
 
 /* Make sure there are no machine checks on offlined CPUs. */
@@ -2054,7 +2061,7 @@ mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
 	switch (action) {
 	case CPU_ONLINE:
 	case CPU_ONLINE_FROZEN:
-		mce_create_device(cpu);
+		mce_sysdev_create(cpu);
 		if (threshold_cpu_callback)
 			threshold_cpu_callback(action, cpu);
 		break;
@@ -2062,7 +2069,7 @@ mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
 	case CPU_DEAD_FROZEN:
 		if (threshold_cpu_callback)
 			threshold_cpu_callback(action, cpu);
-		mce_remove_device(cpu);
+		mce_sysdev_remove(cpu);
 		break;
 	case CPU_DOWN_PREPARE:
 	case CPU_DOWN_PREPARE_FROZEN:
@@ -2116,27 +2123,28 @@ static __init int mcheck_init_device(void)
 	if (!mce_available(&boot_cpu_data))
 		return -EIO;
 
-	zalloc_cpumask_var(&mce_dev_initialized, GFP_KERNEL);
+	zalloc_cpumask_var(&mce_sysdev_initialized, GFP_KERNEL);
 
 	mce_init_banks();
 
-	err = sysdev_class_register(&mce_sysclass);
+	err = sysdev_class_register(&mce_sysdev_class);
 	if (err)
 		return err;
 
 	for_each_online_cpu(i) {
-		err = mce_create_device(i);
+		err = mce_sysdev_create(i);
 		if (err)
 			return err;
 	}
 
 	register_syscore_ops(&mce_syscore_ops);
 	register_hotcpu_notifier(&mce_cpu_notifier);
-	misc_register(&mce_log_device);
+
+	/* register character device /dev/mcelog */
+	misc_register(&mce_chrdev_device);
 
 	return err;
 }
-
 device_initcall(mcheck_init_device);
 
 /*
diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd.c b/arch/x86/kernel/cpu/mcheck/mce_amd.c
index bb0adad3514..f5474218cff 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_amd.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c
@@ -548,7 +548,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
 		if (!b)
 			goto out;
 
-		err = sysfs_create_link(&per_cpu(mce_dev, cpu).kobj,
+		err = sysfs_create_link(&per_cpu(mce_sysdev, cpu).kobj,
 					b->kobj, name);
 		if (err)
 			goto out;
@@ -571,7 +571,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
 		goto out;
 	}
 
-	b->kobj = kobject_create_and_add(name, &per_cpu(mce_dev, cpu).kobj);
+	b->kobj = kobject_create_and_add(name, &per_cpu(mce_sysdev, cpu).kobj);
 	if (!b->kobj)
 		goto out_free;
 
@@ -591,7 +591,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
 		if (i == cpu)
 			continue;
 
-		err = sysfs_create_link(&per_cpu(mce_dev, i).kobj,
+		err = sysfs_create_link(&per_cpu(mce_sysdev, i).kobj,
 					b->kobj, name);
 		if (err)
 			goto out;
@@ -669,7 +669,7 @@ static void threshold_remove_bank(unsigned int cpu, int bank)
 #ifdef CONFIG_SMP
 	/* sibling symlink */
 	if (shared_bank[bank] && b->blocks->cpu != cpu) {
-		sysfs_remove_link(&per_cpu(mce_dev, cpu).kobj, name);
+		sysfs_remove_link(&per_cpu(mce_sysdev, cpu).kobj, name);
 		per_cpu(threshold_banks, cpu)[bank] = NULL;
 
 		return;
@@ -681,7 +681,7 @@ static void threshold_remove_bank(unsigned int cpu, int bank)
 		if (i == cpu)
 			continue;
 
-		sysfs_remove_link(&per_cpu(mce_dev, i).kobj, name);
+		sysfs_remove_link(&per_cpu(mce_sysdev, i).kobj, name);
 		per_cpu(threshold_banks, i)[bank] = NULL;
 	}
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel.c b/arch/x86/kernel/cpu/mcheck/mce_intel.c
index 8694ef56459..38e49bc95ff 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_intel.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_intel.c
@@ -28,7 +28,7 @@ static DEFINE_PER_CPU(mce_banks_t, mce_banks_owned);
  * cmci_discover_lock protects against parallel discovery attempts
  * which could race against each other.
  */
-static DEFINE_SPINLOCK(cmci_discover_lock);
+static DEFINE_RAW_SPINLOCK(cmci_discover_lock);
 
 #define CMCI_THRESHOLD 1
 
@@ -85,7 +85,7 @@ static void cmci_discover(int banks, int boot)
 	int hdr = 0;
 	int i;
 
-	spin_lock_irqsave(&cmci_discover_lock, flags);
+	raw_spin_lock_irqsave(&cmci_discover_lock, flags);
 	for (i = 0; i < banks; i++) {
 		u64 val;
 
@@ -116,7 +116,7 @@ static void cmci_discover(int banks, int boot)
 			WARN_ON(!test_bit(i, __get_cpu_var(mce_poll_banks)));
 		}
 	}
-	spin_unlock_irqrestore(&cmci_discover_lock, flags);
+	raw_spin_unlock_irqrestore(&cmci_discover_lock, flags);
 	if (hdr)
 		printk(KERN_CONT "\n");
 }
@@ -150,7 +150,7 @@ void cmci_clear(void)
 
 	if (!cmci_supported(&banks))
 		return;
-	spin_lock_irqsave(&cmci_discover_lock, flags);
+	raw_spin_lock_irqsave(&cmci_discover_lock, flags);
 	for (i = 0; i < banks; i++) {
 		if (!test_bit(i, __get_cpu_var(mce_banks_owned)))
 			continue;
@@ -160,7 +160,7 @@ void cmci_clear(void)
 		wrmsrl(MSR_IA32_MCx_CTL2(i), val);
 		__clear_bit(i, __get_cpu_var(mce_banks_owned));
 	}
-	spin_unlock_irqrestore(&cmci_discover_lock, flags);
+	raw_spin_unlock_irqrestore(&cmci_discover_lock, flags);
 }
 
 /*
diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c
index 27c625178bf..787e06c84ea 100644
--- a/arch/x86/kernel/cpu/mcheck/therm_throt.c
+++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c
@@ -18,6 +18,7 @@
 #include <linux/jiffies.h>
 #include <linux/kernel.h>
 #include <linux/percpu.h>
+#include <linux/export.h>
 #include <linux/sysdev.h>
 #include <linux/types.h>
 #include <linux/init.h>