20 files changed, 748 insertions, 102 deletions

diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile
index d30a6a9a0121..a0e067d3d96c 100644
--- a/arch/x86/kernel/cpu/Makefile
+++ b/arch/x86/kernel/cpu/Makefile
@@ -32,7 +32,7 @@ obj-$(CONFIG_PERF_EVENTS)		+= perf_event.o
 
 ifdef CONFIG_PERF_EVENTS
 obj-$(CONFIG_CPU_SUP_AMD)		+= perf_event_amd.o
-obj-$(CONFIG_CPU_SUP_INTEL)		+= perf_event_p6.o perf_event_p4.o
+obj-$(CONFIG_CPU_SUP_INTEL)		+= perf_event_p6.o perf_event_knc.o perf_event_p4.o
 obj-$(CONFIG_CPU_SUP_INTEL)		+= perf_event_intel_lbr.o perf_event_intel_ds.o perf_event_intel.o
 obj-$(CONFIG_CPU_SUP_INTEL)		+= perf_event_intel_uncore.o
 endif
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 9d92e19039f0..f7e98a2c0d12 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -737,6 +737,72 @@ static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c,
 }
 #endif
 
+static void __cpuinit cpu_set_tlb_flushall_shift(struct cpuinfo_x86 *c)
+{
+	if (!cpu_has_invlpg)
+		return;
+
+	tlb_flushall_shift = 5;
+
+	if (c->x86 <= 0x11)
+		tlb_flushall_shift = 4;
+}
+
+static void __cpuinit cpu_detect_tlb_amd(struct cpuinfo_x86 *c)
+{
+	u32 ebx, eax, ecx, edx;
+	u16 mask = 0xfff;
+
+	if (c->x86 < 0xf)
+		return;
+
+	if (c->extended_cpuid_level < 0x80000006)
+		return;
+
+	cpuid(0x80000006, &eax, &ebx, &ecx, &edx);
+
+	tlb_lld_4k[ENTRIES] = (ebx >> 16) & mask;
+	tlb_lli_4k[ENTRIES] = ebx & mask;
+
+	/*
+	 * K8 doesn't have 2M/4M entries in the L2 TLB so read out the L1 TLB
+	 * characteristics from the CPUID function 0x80000005 instead.
+	 */
+	if (c->x86 == 0xf) {
+		cpuid(0x80000005, &eax, &ebx, &ecx, &edx);
+		mask = 0xff;
+	}
+
+	/* Handle DTLB 2M and 4M sizes, fall back to L1 if L2 is disabled */
+	if (!((eax >> 16) & mask)) {
+		u32 a, b, c, d;
+
+		cpuid(0x80000005, &a, &b, &c, &d);
+		tlb_lld_2m[ENTRIES] = (a >> 16) & 0xff;
+	} else {
+		tlb_lld_2m[ENTRIES] = (eax >> 16) & mask;
+	}
+
+	/* a 4M entry uses two 2M entries */
+	tlb_lld_4m[ENTRIES] = tlb_lld_2m[ENTRIES] >> 1;
+
+	/* Handle ITLB 2M and 4M sizes, fall back to L1 if L2 is disabled */
+	if (!(eax & mask)) {
+		/* Erratum 658 */
+		if (c->x86 == 0x15 && c->x86_model <= 0x1f) {
+			tlb_lli_2m[ENTRIES] = 1024;
+		} else {
+			cpuid(0x80000005, &eax, &ebx, &ecx, &edx);
+			tlb_lli_2m[ENTRIES] = eax & 0xff;
+		}
+	} else
+		tlb_lli_2m[ENTRIES] = eax & mask;
+
+	tlb_lli_4m[ENTRIES] = tlb_lli_2m[ENTRIES] >> 1;
+
+	cpu_set_tlb_flushall_shift(c);
+}
+
 static const struct cpu_dev __cpuinitconst amd_cpu_dev = {
 	.c_vendor	= "AMD",
 	.c_ident	= { "AuthenticAMD" },
@@ -756,6 +822,7 @@ static const struct cpu_dev __cpuinitconst amd_cpu_dev = {
 	.c_size_cache	= amd_size_cache,
 #endif
 	.c_early_init   = early_init_amd,
+	.c_detect_tlb	= cpu_detect_tlb_amd,
 	.c_bsp_init	= bsp_init_amd,
 	.c_init		= init_amd,
 	.c_x86_vendor	= X86_VENDOR_AMD,
diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c
index c97bb7b5a9f8..d0e910da16c5 100644
--- a/arch/x86/kernel/cpu/bugs.c
+++ b/arch/x86/kernel/cpu/bugs.c
@@ -165,10 +165,15 @@ void __init check_bugs(void)
 	print_cpu_info(&boot_cpu_data);
 #endif
 	check_config();
-	check_fpu();
 	check_hlt();
 	check_popad();
 	init_utsname()->machine[1] =
 		'0' + (boot_cpu_data.x86 > 6 ? 6 : boot_cpu_data.x86);
 	alternative_instructions();
+
+	/*
+	 * kernel_fpu_begin/end() in check_fpu() relies on the patched
+	 * alternative instructions.
+	 */
+	check_fpu();
 }
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index a5fbc3c5fccc..7505f7b13e71 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -259,23 +259,36 @@ static inline void squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
 }
 #endif
 
-static int disable_smep __cpuinitdata;
 static __init int setup_disable_smep(char *arg)
 {
-	disable_smep = 1;
+	setup_clear_cpu_cap(X86_FEATURE_SMEP);
 	return 1;
 }
 __setup("nosmep", setup_disable_smep);
 
-static __cpuinit void setup_smep(struct cpuinfo_x86 *c)
+static __always_inline void setup_smep(struct cpuinfo_x86 *c)
 {
-	if (cpu_has(c, X86_FEATURE_SMEP)) {
-		if (unlikely(disable_smep)) {
-			setup_clear_cpu_cap(X86_FEATURE_SMEP);
-			clear_in_cr4(X86_CR4_SMEP);
-		} else
-			set_in_cr4(X86_CR4_SMEP);
-	}
+	if (cpu_has(c, X86_FEATURE_SMEP))
+		set_in_cr4(X86_CR4_SMEP);
+}
+
+static __init int setup_disable_smap(char *arg)
+{
+	setup_clear_cpu_cap(X86_FEATURE_SMAP);
+	return 1;
+}
+__setup("nosmap", setup_disable_smap);
+
+static __always_inline void setup_smap(struct cpuinfo_x86 *c)
+{
+	unsigned long eflags;
+
+	/* This should have been cleared long ago */
+	raw_local_save_flags(eflags);
+	BUG_ON(eflags & X86_EFLAGS_AC);
+
+	if (cpu_has(c, X86_FEATURE_SMAP))
+		set_in_cr4(X86_CR4_SMAP);
 }
 
 /*
@@ -476,7 +489,7 @@ void __cpuinit cpu_detect_tlb(struct cpuinfo_x86 *c)
 
 	printk(KERN_INFO "Last level iTLB entries: 4KB %d, 2MB %d, 4MB %d\n" \
 		"Last level dTLB entries: 4KB %d, 2MB %d, 4MB %d\n"	     \
-		"tlb_flushall_shift is 0x%x\n",
+		"tlb_flushall_shift: %d\n",
 		tlb_lli_4k[ENTRIES], tlb_lli_2m[ENTRIES],
 		tlb_lli_4m[ENTRIES], tlb_lld_4k[ENTRIES],
 		tlb_lld_2m[ENTRIES], tlb_lld_4m[ENTRIES],
@@ -712,8 +725,6 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c)
 	c->cpu_index = 0;
 	filter_cpuid_features(c, false);
 
-	setup_smep(c);
-
 	if (this_cpu->c_bsp_init)
 		this_cpu->c_bsp_init(c);
 }
@@ -798,8 +809,6 @@ static void __cpuinit generic_identify(struct cpuinfo_x86 *c)
 		c->phys_proc_id = c->initial_apicid;
 	}
 
-	setup_smep(c);
-
 	get_model_name(c); /* Default name */
 
 	detect_nopl(c);
@@ -864,6 +873,10 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 	/* Disable the PN if appropriate */
 	squash_the_stupid_serial_number(c);
 
+	/* Set up SMEP/SMAP */
+	setup_smep(c);
+	setup_smap(c);
+
 	/*
 	 * The vendor-specific functions might have changed features.
 	 * Now we do "generic changes."
@@ -942,8 +955,7 @@ void __init identify_boot_cpu(void)
 #else
 	vgetcpu_set_mode();
 #endif
-	if (boot_cpu_data.cpuid_level >= 2)
-		cpu_detect_tlb(&boot_cpu_data);
+	cpu_detect_tlb(&boot_cpu_data);
 }
 
 void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c)
@@ -1023,14 +1035,16 @@ void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
 		printk(KERN_CONT "%s ", vendor);
 
 	if (c->x86_model_id[0])
-		printk(KERN_CONT "%s", c->x86_model_id);
+		printk(KERN_CONT "%s", strim(c->x86_model_id));
 	else
 		printk(KERN_CONT "%d86", c->x86);
 
+	printk(KERN_CONT " (fam: %02x, model: %02x", c->x86, c->x86_model);
+
 	if (c->x86_mask || c->cpuid_level >= 0)
-		printk(KERN_CONT " stepping %02x\n", c->x86_mask);
+		printk(KERN_CONT ", stepping: %02x)\n", c->x86_mask);
 	else
-		printk(KERN_CONT "\n");
+		printk(KERN_CONT ")\n");
 
 	print_cpu_msr(c);
 }
@@ -1113,11 +1127,10 @@ void syscall_init(void)
 
 	/* Flags to clear on syscall */
 	wrmsrl(MSR_SYSCALL_MASK,
-	       X86_EFLAGS_TF|X86_EFLAGS_DF|X86_EFLAGS_IF|X86_EFLAGS_IOPL);
+	       X86_EFLAGS_TF|X86_EFLAGS_DF|X86_EFLAGS_IF|
+	       X86_EFLAGS_IOPL|X86_EFLAGS_AC);
 }
 
-unsigned long kernel_eflags;
-
 /*
  * Copies of the original ist values from the tss are only accessed during
  * debugging, no special alignment required.
@@ -1297,9 +1310,6 @@ void __cpuinit cpu_init(void)
 	dbg_restore_debug_regs();
 
 	fpu_init();
-	xsave_init();
-
-	raw_local_save_flags(kernel_eflags);
 
 	if (is_uv_system())
 		uv_cpu_init();
@@ -1352,6 +1362,5 @@ void __cpuinit cpu_init(void)
 	dbg_restore_debug_regs();
 
 	fpu_init();
-	xsave_init();
 }
 #endif
diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index 0a4ce2980a5a..198e019a531a 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -648,6 +648,10 @@ static void __cpuinit intel_detect_tlb(struct cpuinfo_x86 *c)
 	int i, j, n;
 	unsigned int regs[4];
 	unsigned char *desc = (unsigned char *)regs;
+
+	if (c->cpuid_level < 2)
+		return;
+
 	/* Number of times to iterate */
 	n = cpuid_eax(2) & 0xFF;
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce-inject.c b/arch/x86/kernel/cpu/mcheck/mce-inject.c
index fc4beb393577..ddc72f839332 100644
--- a/arch/x86/kernel/cpu/mcheck/mce-inject.c
+++ b/arch/x86/kernel/cpu/mcheck/mce-inject.c
@@ -78,6 +78,7 @@ static void raise_exception(struct mce *m, struct pt_regs *pregs)
 }
 
 static cpumask_var_t mce_inject_cpumask;
+static DEFINE_MUTEX(mce_inject_mutex);
 
 static int mce_raise_notify(unsigned int cmd, struct pt_regs *regs)
 {
@@ -194,7 +195,11 @@ static void raise_mce(struct mce *m)
 		put_online_cpus();
 	} else
 #endif
+	{
+		preempt_disable();
 		raise_local();
+		preempt_enable();
+	}
 }
 
 /* Error injection interface */
@@ -225,7 +230,10 @@ static ssize_t mce_write(struct file *filp, const char __user *ubuf,
 	 * so do it a jiffie or two later everywhere.
 	 */
 	schedule_timeout(2);
+
+	mutex_lock(&mce_inject_mutex);
 	raise_mce(&m);
+	mutex_unlock(&mce_inject_mutex);
 	return usize;
 }
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce-internal.h b/arch/x86/kernel/cpu/mcheck/mce-internal.h
index ed44c8a65858..6a05c1d327a9 100644
--- a/arch/x86/kernel/cpu/mcheck/mce-internal.h
+++ b/arch/x86/kernel/cpu/mcheck/mce-internal.h
@@ -28,6 +28,18 @@ extern int mce_ser;
 
 extern struct mce_bank *mce_banks;
 
+#ifdef CONFIG_X86_MCE_INTEL
+unsigned long mce_intel_adjust_timer(unsigned long interval);
+void mce_intel_cmci_poll(void);
+void mce_intel_hcpu_update(unsigned long cpu);
+#else
+# define mce_intel_adjust_timer mce_adjust_timer_default
+static inline void mce_intel_cmci_poll(void) { }
+static inline void mce_intel_hcpu_update(unsigned long cpu) { }
+#endif
+
+void mce_timer_kick(unsigned long interval);
+
 #ifdef CONFIG_ACPI_APEI
 int apei_write_mce(struct mce *m);
 ssize_t apei_read_mce(struct mce *m, u64 *record_id);
diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index 292d0258311c..29e87d3b2843 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -83,6 +83,7 @@ static int			mce_dont_log_ce		__read_mostly;
 int				mce_cmci_disabled	__read_mostly;
 int				mce_ignore_ce		__read_mostly;
 int				mce_ser			__read_mostly;
+int				mce_bios_cmci_threshold	__read_mostly;
 
 struct mce_bank                *mce_banks		__read_mostly;
 
@@ -1266,6 +1267,14 @@ static unsigned long check_interval = 5 * 60; /* 5 minutes */
 static DEFINE_PER_CPU(unsigned long, mce_next_interval); /* in jiffies */
 static DEFINE_PER_CPU(struct timer_list, mce_timer);
 
+static unsigned long mce_adjust_timer_default(unsigned long interval)
+{
+	return interval;
+}
+
+static unsigned long (*mce_adjust_timer)(unsigned long interval) =
+	mce_adjust_timer_default;
+
 static void mce_timer_fn(unsigned long data)
 {
 	struct timer_list *t = &__get_cpu_var(mce_timer);
@@ -1276,6 +1285,7 @@ static void mce_timer_fn(unsigned long data)
 	if (mce_available(__this_cpu_ptr(&cpu_info))) {
 		machine_check_poll(MCP_TIMESTAMP,
 				&__get_cpu_var(mce_poll_banks));
+		mce_intel_cmci_poll();
 	}
 
 	/*
@@ -1283,14 +1293,38 @@ static void mce_timer_fn(unsigned long data)
 	 * polling interval, otherwise increase the polling interval.
 	 */
 	iv = __this_cpu_read(mce_next_interval);
-	if (mce_notify_irq())
+	if (mce_notify_irq()) {
 		iv = max(iv / 2, (unsigned long) HZ/100);
-	else
+	} else {
 		iv = min(iv * 2, round_jiffies_relative(check_interval * HZ));
+		iv = mce_adjust_timer(iv);
+	}
 	__this_cpu_write(mce_next_interval, iv);
+	/* Might have become 0 after CMCI storm subsided */
+	if (iv) {
+		t->expires = jiffies + iv;
+		add_timer_on(t, smp_processor_id());
+	}
+}
 
-	t->expires = jiffies + iv;
-	add_timer_on(t, smp_processor_id());
+/*
+ * Ensure that the timer is firing in @interval from now.
+ */
+void mce_timer_kick(unsigned long interval)
+{
+	struct timer_list *t = &__get_cpu_var(mce_timer);
+	unsigned long when = jiffies + interval;
+	unsigned long iv = __this_cpu_read(mce_next_interval);
+
+	if (timer_pending(t)) {
+		if (time_before(when, t->expires))
+			mod_timer_pinned(t, when);
+	} else {
+		t->expires = round_jiffies(when);
+		add_timer_on(t, smp_processor_id());
+	}
+	if (interval < iv)
+		__this_cpu_write(mce_next_interval, interval);
 }
 
 /* Must not be called in IRQ context where del_timer_sync() can deadlock */
@@ -1585,6 +1619,7 @@ static void __mcheck_cpu_init_vendor(struct cpuinfo_x86 *c)
 	switch (c->x86_vendor) {
 	case X86_VENDOR_INTEL:
 		mce_intel_feature_init(c);
+		mce_adjust_timer = mce_intel_adjust_timer;
 		break;
 	case X86_VENDOR_AMD:
 		mce_amd_feature_init(c);
@@ -1594,23 +1629,28 @@ static void __mcheck_cpu_init_vendor(struct cpuinfo_x86 *c)
 	}
 }
 
-static void __mcheck_cpu_init_timer(void)
+static void mce_start_timer(unsigned int cpu, struct timer_list *t)
 {
-	struct timer_list *t = &__get_cpu_var(mce_timer);
-	unsigned long iv = check_interval * HZ;
+	unsigned long iv = mce_adjust_timer(check_interval * HZ);
 
-	setup_timer(t, mce_timer_fn, smp_processor_id());
+	__this_cpu_write(mce_next_interval, iv);
 
-	if (mce_ignore_ce)
+	if (mce_ignore_ce || !iv)
 		return;
 
-	__this_cpu_write(mce_next_interval, iv);
-	if (!iv)
-		return;
 	t->expires = round_jiffies(jiffies + iv);
 	add_timer_on(t, smp_processor_id());
 }
 
+static void __mcheck_cpu_init_timer(void)
+{
+	struct timer_list *t = &__get_cpu_var(mce_timer);
+	unsigned int cpu = smp_processor_id();
+
+	setup_timer(t, mce_timer_fn, cpu);
+	mce_start_timer(cpu, t);
+}
+
 /* Handle unconfigured int18 (should never happen) */
 static void unexpected_machine_check(struct pt_regs *regs, long error_code)
 {
@@ -1907,6 +1947,7 @@ static struct miscdevice mce_chrdev_device = {
  *	check, or 0 to not wait
  * mce=bootlog Log MCEs from before booting. Disabled by default on AMD.
  * mce=nobootlog Don't log MCEs from before booting.
+ * mce=bios_cmci_threshold Don't program the CMCI threshold
  */
 static int __init mcheck_enable(char *str)
 {
@@ -1926,6 +1967,8 @@ static int __init mcheck_enable(char *str)
 		mce_ignore_ce = 1;
 	else if (!strcmp(str, "bootlog") || !strcmp(str, "nobootlog"))
 		mce_bootlog = (str[0] == 'b');
+	else if (!strcmp(str, "bios_cmci_threshold"))
+		mce_bios_cmci_threshold = 1;
 	else if (isdigit(str[0])) {
 		get_option(&str, &tolerant);
 		if (*str == ',') {
@@ -2166,6 +2209,11 @@ static struct dev_ext_attribute dev_attr_cmci_disabled = {
 	&mce_cmci_disabled
 };
 
+static struct dev_ext_attribute dev_attr_bios_cmci_threshold = {
+	__ATTR(bios_cmci_threshold, 0444, device_show_int, NULL),
+	&mce_bios_cmci_threshold
+};
+
 static struct device_attribute *mce_device_attrs[] = {
 	&dev_attr_tolerant.attr,
 	&dev_attr_check_interval.attr,
@@ -2174,6 +2222,7 @@ static struct device_attribute *mce_device_attrs[] = {
 	&dev_attr_dont_log_ce.attr,
 	&dev_attr_ignore_ce.attr,
 	&dev_attr_cmci_disabled.attr,
+	&dev_attr_bios_cmci_threshold.attr,
 	NULL
 };
 
@@ -2294,38 +2343,33 @@ mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
 	unsigned int cpu = (unsigned long)hcpu;
 	struct timer_list *t = &per_cpu(mce_timer, cpu);
 
-	switch (action) {
+	switch (action & ~CPU_TASKS_FROZEN) {
 	case CPU_ONLINE:
-	case CPU_ONLINE_FROZEN:
 		mce_device_create(cpu);
 		if (threshold_cpu_callback)
 			threshold_cpu_callback(action, cpu);
 		break;
 	case CPU_DEAD:
-	case CPU_DEAD_FROZEN:
 		if (threshold_cpu_callback)
 			threshold_cpu_callback(action, cpu);
 		mce_device_remove(cpu);
+		mce_intel_hcpu_update(cpu);
 		break;
 	case CPU_DOWN_PREPARE:
-	case CPU_DOWN_PREPARE_FROZEN:
-		del_timer_sync(t);
 		smp_call_function_single(cpu, mce_disable_cpu, &action, 1);
+		del_timer_sync(t);
 		break;
 	case CPU_DOWN_FAILED:
-	case CPU_DOWN_FAILED_FROZEN:
-		if (!mce_ignore_ce && check_interval) {
-			t->expires = round_jiffies(jiffies +
-					per_cpu(mce_next_interval, cpu));
-			add_timer_on(t, cpu);
-		}
 		smp_call_function_single(cpu, mce_reenable_cpu, &action, 1);
+		mce_start_timer(cpu, t);
 		break;
-	case CPU_POST_DEAD:
+	}
+
+	if (action == CPU_POST_DEAD) {
 		/* intentionally ignoring frozen here */
 		cmci_rediscover(cpu);
-		break;
 	}
+
 	return NOTIFY_OK;
 }
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel.c b/arch/x86/kernel/cpu/mcheck/mce_intel.c
index 38e49bc95ffc..5f88abf07e9c 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_intel.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_intel.c
@@ -15,6 +15,8 @@
 #include <asm/msr.h>
 #include <asm/mce.h>
 
+#include "mce-internal.h"
+
 /*
  * Support for Intel Correct Machine Check Interrupts. This allows
  * the CPU to raise an interrupt when a corrected machine check happened.
@@ -30,7 +32,22 @@ static DEFINE_PER_CPU(mce_banks_t, mce_banks_owned);
  */
 static DEFINE_RAW_SPINLOCK(cmci_discover_lock);
 
-#define CMCI_THRESHOLD 1
+#define CMCI_THRESHOLD		1
+#define CMCI_POLL_INTERVAL	(30 * HZ)
+#define CMCI_STORM_INTERVAL	(1 * HZ)
+#define CMCI_STORM_THRESHOLD	15
+
+static DEFINE_PER_CPU(unsigned long, cmci_time_stamp);
+static DEFINE_PER_CPU(unsigned int, cmci_storm_cnt);
+static DEFINE_PER_CPU(unsigned int, cmci_storm_state);
+
+enum {
+	CMCI_STORM_NONE,
+	CMCI_STORM_ACTIVE,
+	CMCI_STORM_SUBSIDED,
+};
+
+static atomic_t cmci_storm_on_cpus;
 
 static int cmci_supported(int *banks)
 {
@@ -53,6 +70,93 @@ static int cmci_supported(int *banks)
 	return !!(cap & MCG_CMCI_P);
 }
 
+void mce_intel_cmci_poll(void)
+{
+	if (__this_cpu_read(cmci_storm_state) == CMCI_STORM_NONE)
+		return;
+	machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned));
+}
+
+void mce_intel_hcpu_update(unsigned long cpu)
+{
+	if (per_cpu(cmci_storm_state, cpu) == CMCI_STORM_ACTIVE)
+		atomic_dec(&cmci_storm_on_cpus);
+
+	per_cpu(cmci_storm_state, cpu) = CMCI_STORM_NONE;
+}
+
+unsigned long mce_intel_adjust_timer(unsigned long interval)
+{
+	int r;
+
+	if (interval < CMCI_POLL_INTERVAL)
+		return interval;
+
+	switch (__this_cpu_read(cmci_storm_state)) {
+	case CMCI_STORM_ACTIVE:
+		/*
+		 * We switch back to interrupt mode once the poll timer has
+		 * silenced itself. That means no events recorded and the
+		 * timer interval is back to our poll interval.
+		 */
+		__this_cpu_write(cmci_storm_state, CMCI_STORM_SUBSIDED);
+		r = atomic_sub_return(1, &cmci_storm_on_cpus);
+		if (r == 0)
+			pr_notice("CMCI storm subsided: switching to interrupt mode\n");
+		/* FALLTHROUGH */
+
+	case CMCI_STORM_SUBSIDED:
+		/*
+		 * We wait for all cpus to go back to SUBSIDED
+		 * state. When that happens we switch back to
+		 * interrupt mode.
+		 */
+		if (!atomic_read(&cmci_storm_on_cpus)) {
+			__this_cpu_write(cmci_storm_state, CMCI_STORM_NONE);
+			cmci_reenable();
+			cmci_recheck();
+		}
+		return CMCI_POLL_INTERVAL;
+	default:
+		/*
+		 * We have shiny weather. Let the poll do whatever it
+		 * thinks.
+		 */
+		return interval;
+	}
+}
+
+static bool cmci_storm_detect(void)
+{
+	unsigned int cnt = __this_cpu_read(cmci_storm_cnt);
+	unsigned long ts = __this_cpu_read(cmci_time_stamp);
+	unsigned long now = jiffies;
+	int r;
+
+	if (__this_cpu_read(cmci_storm_state) != CMCI_STORM_NONE)
+		return true;
+
+	if (time_before_eq(now, ts + CMCI_STORM_INTERVAL)) {
+		cnt++;
+	} else {
+		cnt = 1;
+		__this_cpu_write(cmci_time_stamp, now);
+	}
+	__this_cpu_write(cmci_storm_cnt, cnt);
+
+	if (cnt <= CMCI_STORM_THRESHOLD)
+		return false;
+
+	cmci_clear();
+	__this_cpu_write(cmci_storm_state, CMCI_STORM_ACTIVE);
+	r = atomic_add_return(1, &cmci_storm_on_cpus);
+	mce_timer_kick(CMCI_POLL_INTERVAL);
+
+	if (r == 1)
+		pr_notice("CMCI storm detected: switching to poll mode\n");
+	return true;
+}
+
 /*
  * The interrupt handler. This is called on every event.
  * Just call the poller directly to log any events.
@@ -61,33 +165,28 @@ static int cmci_supported(int *banks)
  */
 static void intel_threshold_interrupt(void)
 {
+	if (cmci_storm_detect())
+		return;
 	machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned));
 	mce_notify_irq();
 }
 
-static void print_update(char *type, int *hdr, int num)
-{
-	if (*hdr == 0)
-		printk(KERN_INFO "CPU %d MCA banks", smp_processor_id());
-	*hdr = 1;
-	printk(KERN_CONT " %s:%d", type, num);
-}
-
 /*
  * Enable CMCI (Corrected Machine Check Interrupt) for available MCE banks
  * on this CPU. Use the algorithm recommended in the SDM to discover shared
  * banks.
  */
-static void cmci_discover(int banks, int boot)
+static void cmci_discover(int banks)
 {
 	unsigned long *owned = (void *)&__get_cpu_var(mce_banks_owned);
 	unsigned long flags;
-	int hdr = 0;
 	int i;
+	int bios_wrong_thresh = 0;
 
 	raw_spin_lock_irqsave(&cmci_discover_lock, flags);
 	for (i = 0; i < banks; i++) {
 		u64 val;
+		int bios_zero_thresh = 0;
 
 		if (test_bit(i, owned))
 			continue;
@@ -96,29 +195,52 @@ static void cmci_discover(int banks, int boot)
 
 		/* Already owned by someone else? */
 		if (val & MCI_CTL2_CMCI_EN) {
-			if (test_and_clear_bit(i, owned) && !boot)
-				print_update("SHD", &hdr, i);
+			clear_bit(i, owned);
 			__clear_bit(i, __get_cpu_var(mce_poll_banks));
 			continue;
 		}
 
-		val &= ~MCI_CTL2_CMCI_THRESHOLD_MASK;
-		val |= MCI_CTL2_CMCI_EN | CMCI_THRESHOLD;
+		if (!mce_bios_cmci_threshold) {
+			val &= ~MCI_CTL2_CMCI_THRESHOLD_MASK;
+			val |= CMCI_THRESHOLD;
+		} else if (!(val & MCI_CTL2_CMCI_THRESHOLD_MASK)) {
+			/*
+			 * If bios_cmci_threshold boot option was specified
+			 * but the threshold is zero, we'll try to initialize
+			 * it to 1.
+			 */
+			bios_zero_thresh = 1;
+			val |= CMCI_THRESHOLD;
+		}
+
+		val |= MCI_CTL2_CMCI_EN;
 		wrmsrl(MSR_IA32_MCx_CTL2(i), val);
 		rdmsrl(MSR_IA32_MCx_CTL2(i), val);
 
 		/* Did the enable bit stick? -- the bank supports CMCI */
 		if (val & MCI_CTL2_CMCI_EN) {
-			if (!test_and_set_bit(i, owned) && !boot)
-				print_update("CMCI", &hdr, i);
+			set_bit(i, owned);
 			__clear_bit(i, __get_cpu_var(mce_poll_banks));
+			/*
+			 * We are able to set thresholds for some banks that
+			 * had a threshold of 0. This means the BIOS has not
+			 * set the thresholds properly or does not work with
+			 * this boot option. Note down now and report later.
+			 */
+			if (mce_bios_cmci_threshold && bios_zero_thresh &&
+					(val & MCI_CTL2_CMCI_THRESHOLD_MASK))
+				bios_wrong_thresh = 1;
 		} else {
 			WARN_ON(!test_bit(i, __get_cpu_var(mce_poll_banks)));
 		}
 	}
 	raw_spin_unlock_irqrestore(&cmci_discover_lock, flags);
-	if (hdr)
-		printk(KERN_CONT "\n");
+	if (mce_bios_cmci_threshold && bios_wrong_thresh) {
+		pr_info_once(
+			"bios_cmci_threshold: Some banks do not have valid thresholds set\n");
+		pr_info_once(
+			"bios_cmci_threshold: Make sure your BIOS supports this boot option\n");
+	}
 }
 
 /*
@@ -156,7 +278,7 @@ void cmci_clear(void)
 			continue;
 		/* Disable CMCI */
 		rdmsrl(MSR_IA32_MCx_CTL2(i), val);
-		val &= ~(MCI_CTL2_CMCI_EN|MCI_CTL2_CMCI_THRESHOLD_MASK);
+		val &= ~MCI_CTL2_CMCI_EN;
 		wrmsrl(MSR_IA32_MCx_CTL2(i), val);
 		__clear_bit(i, __get_cpu_var(mce_banks_owned));
 	}
@@ -186,7 +308,7 @@ void cmci_rediscover(int dying)
 			continue;
 		/* Recheck banks in case CPUs don't all have the same */
 		if (cmci_supported(&banks))
-			cmci_discover(banks, 0);
+			cmci_discover(banks);
 	}
 
 	set_cpus_allowed_ptr(current, old);
@@ -200,7 +322,7 @@ void cmci_reenable(void)
 {
 	int banks;
 	if (cmci_supported(&banks))
-		cmci_discover(banks, 0);
+		cmci_discover(banks);
 }
 
 static void intel_init_cmci(void)
@@ -211,7 +333,7 @@ static void intel_init_cmci(void)
 		return;
 
 	mce_threshold_vector = intel_threshold_interrupt;
-	cmci_discover(banks, 1);
+	cmci_discover(banks);
 	/*
 	 * For CPU #0 this runs with still disabled APIC, but that's
 	 * ok because only the vector is set up. We still do another
diff --git a/arch/x86/kernel/cpu/mkcapflags.pl b/arch/x86/kernel/cpu/mkcapflags.pl
index c7b3fe2d72e0..091972ef49de 100644
--- a/arch/x86/kernel/cpu/mkcapflags.pl
+++ b/arch/x86/kernel/cpu/mkcapflags.pl
@@ -8,7 +8,10 @@
 open(IN, "< $in\0")   or die "$0: cannot open: $in: $!\n";
 open(OUT, "> $out\0") or die "$0: cannot create: $out: $!\n";
 
-print OUT "#include <asm/cpufeature.h>\n\n";
+print OUT "#ifndef _ASM_X86_CPUFEATURE_H\n";
+print OUT "#include <asm/cpufeature.h>\n";
+print OUT "#endif\n";
+print OUT "\n";
 print OUT "const char * const x86_cap_flags[NCAPINTS*32] = {\n";
 
 %features = ();
diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h
index 6605a81ba339..271d25700297 100644
--- a/arch/x86/kernel/cpu/perf_event.h
+++ b/arch/x86/kernel/cpu/perf_event.h
@@ -586,6 +586,8 @@ extern struct event_constraint intel_westmere_pebs_event_constraints[];
 
 extern struct event_constraint intel_snb_pebs_event_constraints[];
 
+extern struct event_constraint intel_ivb_pebs_event_constraints[];
+
 struct event_constraint *intel_pebs_constraints(struct perf_event *event);
 
 void intel_pmu_pebs_enable(struct perf_event *event);
@@ -624,6 +626,8 @@ int p4_pmu_init(void);
 
 int p6_pmu_init(void);
 
+int knc_pmu_init(void);
+
 #else /* CONFIG_CPU_SUP_INTEL */
 
 static inline void reserve_ds_buffers(void)
diff --git a/arch/x86/kernel/cpu/perf_event_amd_ibs.c b/arch/x86/kernel/cpu/perf_event_amd_ibs.c
index 7bfb5bec8630..6336bcbd0618 100644
--- a/arch/x86/kernel/cpu/perf_event_amd_ibs.c
+++ b/arch/x86/kernel/cpu/perf_event_amd_ibs.c
@@ -41,17 +41,22 @@ struct cpu_perf_ibs {
 };
 
 struct perf_ibs {
-	struct pmu	pmu;
-	unsigned int	msr;
-	u64		config_mask;
-	u64		cnt_mask;
-	u64		enable_mask;
-	u64		valid_mask;
-	u64		max_period;
-	unsigned long	offset_mask[1];
-	int		offset_max;
-	struct cpu_perf_ibs __percpu *pcpu;
-	u64		(*get_count)(u64 config);
+	struct pmu			pmu;
+	unsigned int			msr;
+	u64				config_mask;
+	u64				cnt_mask;
+	u64				enable_mask;
+	u64				valid_mask;
+	u64				max_period;
+	unsigned long			offset_mask[1];
+	int				offset_max;
+	struct cpu_perf_ibs __percpu	*pcpu;
+
+	struct attribute		**format_attrs;
+	struct attribute_group		format_group;
+	const struct attribute_group	*attr_groups[2];
+
+	u64				(*get_count)(u64 config);
 };
 
 struct perf_ibs_data {
@@ -209,6 +214,15 @@ static int perf_ibs_precise_event(struct perf_event *event, u64 *config)
 	return -EOPNOTSUPP;
 }
 
+static const struct perf_event_attr ibs_notsupp = {
+	.exclude_user	= 1,
+	.exclude_kernel	= 1,
+	.exclude_hv	= 1,
+	.exclude_idle	= 1,
+	.exclude_host	= 1,
+	.exclude_guest	= 1,
+};
+
 static int perf_ibs_init(struct perf_event *event)
 {
 	struct hw_perf_event *hwc = &event->hw;
@@ -229,6 +243,9 @@ static int perf_ibs_init(struct perf_event *event)
 	if (event->pmu != &perf_ibs->pmu)
 		return -ENOENT;
 
+	if (perf_flags(&event->attr) & perf_flags(&ibs_notsupp))
+		return -EINVAL;
+
 	if (config & ~perf_ibs->config_mask)
 		return -EINVAL;
 
@@ -434,6 +451,19 @@ static void perf_ibs_del(struct perf_event *event, int flags)
 
 static void perf_ibs_read(struct perf_event *event) { }
 
+PMU_FORMAT_ATTR(rand_en,	"config:57");
+PMU_FORMAT_ATTR(cnt_ctl,	"config:19");
+
+static struct attribute *ibs_fetch_format_attrs[] = {
+	&format_attr_rand_en.attr,
+	NULL,
+};
+
+static struct attribute *ibs_op_format_attrs[] = {
+	NULL,	/* &format_attr_cnt_ctl.attr if IBS_CAPS_OPCNT */
+	NULL,
+};
+
 static struct perf_ibs perf_ibs_fetch = {
 	.pmu = {
 		.task_ctx_nr	= perf_invalid_context,
@@ -453,6 +483,7 @@ static struct perf_ibs perf_ibs_fetch = {
 	.max_period		= IBS_FETCH_MAX_CNT << 4,
 	.offset_mask		= { MSR_AMD64_IBSFETCH_REG_MASK },
 	.offset_max		= MSR_AMD64_IBSFETCH_REG_COUNT,
+	.format_attrs		= ibs_fetch_format_attrs,
 
 	.get_count		= get_ibs_fetch_count,
 };
@@ -476,6 +507,7 @@ static struct perf_ibs perf_ibs_op = {
 	.max_period		= IBS_OP_MAX_CNT << 4,
 	.offset_mask		= { MSR_AMD64_IBSOP_REG_MASK },
 	.offset_max		= MSR_AMD64_IBSOP_REG_COUNT,
+	.format_attrs		= ibs_op_format_attrs,
 
 	.get_count		= get_ibs_op_count,
 };
@@ -585,6 +617,17 @@ static __init int perf_ibs_pmu_init(struct perf_ibs *perf_ibs, char *name)
 
 	perf_ibs->pcpu = pcpu;
 
+	/* register attributes */
+	if (perf_ibs->format_attrs[0]) {
+		memset(&perf_ibs->format_group, 0, sizeof(perf_ibs->format_group));
+		perf_ibs->format_group.name	= "format";
+		perf_ibs->format_group.attrs	= perf_ibs->format_attrs;
+
+		memset(&perf_ibs->attr_groups, 0, sizeof(perf_ibs->attr_groups));
+		perf_ibs->attr_groups[0]	= &perf_ibs->format_group;
+		perf_ibs->pmu.attr_groups	= perf_ibs->attr_groups;
+	}
+
 	ret = perf_pmu_register(&perf_ibs->pmu, name, -1);
 	if (ret) {
 		perf_ibs->pcpu = NULL;
@@ -596,13 +639,19 @@ static __init int perf_ibs_pmu_init(struct perf_ibs *perf_ibs, char *name)
 
 static __init int perf_event_ibs_init(void)
 {
+	struct attribute **attr = ibs_op_format_attrs;
+
 	if (!ibs_caps)
 		return -ENODEV;	/* ibs not supported by the cpu */
 
 	perf_ibs_pmu_init(&perf_ibs_fetch, "ibs_fetch");
-	if (ibs_caps & IBS_CAPS_OPCNT)
+
+	if (ibs_caps & IBS_CAPS_OPCNT) {
 		perf_ibs_op.config_mask |= IBS_OP_CNT_CTL;
+		*attr++ = &format_attr_cnt_ctl.attr;
+	}
 	perf_ibs_pmu_init(&perf_ibs_op, "ibs_op");
+
 	register_nmi_handler(NMI_LOCAL, perf_ibs_nmi_handler, 0, "perf_ibs");
 	printk(KERN_INFO "perf: AMD IBS detected (0x%08x)\n", ibs_caps);
 
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index 7f2739e03e79..324bb523d9d9 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -1906,6 +1906,8 @@ __init int intel_pmu_init(void)
 		switch (boot_cpu_data.x86) {
 		case 0x6:
 			return p6_pmu_init();
+		case 0xb:
+			return knc_pmu_init();
 		case 0xf:
 			return p4_pmu_init();
 		}
@@ -2008,6 +2010,7 @@ __init int intel_pmu_init(void)
 		break;
 
 	case 28: /* Atom */
+	case 54: /* Cedariew */
 		memcpy(hw_cache_event_ids, atom_hw_cache_event_ids,
 		       sizeof(hw_cache_event_ids));
 
@@ -2047,7 +2050,6 @@ __init int intel_pmu_init(void)
 	case 42: /* SandyBridge */
 	case 45: /* SandyBridge, "Romely-EP" */
 		x86_add_quirk(intel_sandybridge_quirk);
-	case 58: /* IvyBridge */
 		memcpy(hw_cache_event_ids, snb_hw_cache_event_ids,
 		       sizeof(hw_cache_event_ids));
 		memcpy(hw_cache_extra_regs, snb_hw_cache_extra_regs,
@@ -2072,6 +2074,29 @@ __init int intel_pmu_init(void)
 
 		pr_cont("SandyBridge events, ");
 		break;
+	case 58: /* IvyBridge */
+		memcpy(hw_cache_event_ids, snb_hw_cache_event_ids,
+		       sizeof(hw_cache_event_ids));
+		memcpy(hw_cache_extra_regs, snb_hw_cache_extra_regs,
+		       sizeof(hw_cache_extra_regs));
+
+		intel_pmu_lbr_init_snb();
+
+		x86_pmu.event_constraints = intel_snb_event_constraints;
+		x86_pmu.pebs_constraints = intel_ivb_pebs_event_constraints;
+		x86_pmu.pebs_aliases = intel_pebs_aliases_snb;
+		x86_pmu.extra_regs = intel_snb_extra_regs;
+		/* all extra regs are per-cpu when HT is on */
+		x86_pmu.er_flags |= ERF_HAS_RSP_1;
+		x86_pmu.er_flags |= ERF_NO_HT_SHARING;
+
+		/* UOPS_ISSUED.ANY,c=1,i=1 to count stall cycles */
+		intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] =
+			X86_CONFIG(.event=0x0e, .umask=0x01, .inv=1, .cmask=1);
+
+		pr_cont("IvyBridge events, ");
+		break;
+
 
 	default:
 		switch (x86_pmu.version) {
diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c
index e38d97bf4259..826054a4f2ee 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_ds.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c
@@ -407,6 +407,20 @@ struct event_constraint intel_snb_pebs_event_constraints[] = {
 	EVENT_CONSTRAINT_END
 };
 
+struct event_constraint intel_ivb_pebs_event_constraints[] = {
+        INTEL_UEVENT_CONSTRAINT(0x01c0, 0x2), /* INST_RETIRED.PRECDIST */
+        INTEL_UEVENT_CONSTRAINT(0x01c2, 0xf), /* UOPS_RETIRED.ALL */
+        INTEL_UEVENT_CONSTRAINT(0x02c2, 0xf), /* UOPS_RETIRED.RETIRE_SLOTS */
+        INTEL_EVENT_CONSTRAINT(0xc4, 0xf),    /* BR_INST_RETIRED.* */
+        INTEL_EVENT_CONSTRAINT(0xc5, 0xf),    /* BR_MISP_RETIRED.* */
+        INTEL_EVENT_CONSTRAINT(0xcd, 0x8),    /* MEM_TRANS_RETIRED.* */
+        INTEL_EVENT_CONSTRAINT(0xd0, 0xf),    /* MEM_UOP_RETIRED.* */
+        INTEL_EVENT_CONSTRAINT(0xd1, 0xf),    /* MEM_LOAD_UOPS_RETIRED.* */
+        INTEL_EVENT_CONSTRAINT(0xd2, 0xf),    /* MEM_LOAD_UOPS_LLC_HIT_RETIRED.* */
+        INTEL_EVENT_CONSTRAINT(0xd3, 0xf),    /* MEM_LOAD_UOPS_LLC_MISS_RETIRED.* */
+        EVENT_CONSTRAINT_END
+};
+
 struct event_constraint *intel_pebs_constraints(struct perf_event *event)
 {
 	struct event_constraint *c;
diff --git a/arch/x86/kernel/cpu/perf_event_intel_lbr.c b/arch/x86/kernel/cpu/perf_event_intel_lbr.c
index 520b4265fcd2..da02e9cc3754 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_lbr.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_lbr.c
@@ -686,7 +686,8 @@ void intel_pmu_lbr_init_atom(void)
 	 * to have an operational LBR which can freeze
 	 * on PMU interrupt
 	 */
-	if (boot_cpu_data.x86_mask < 10) {
+	if (boot_cpu_data.x86_model == 28
+	    && boot_cpu_data.x86_mask < 10) {
 		pr_cont("LBR disabled due to erratum");
 		return;
 	}
diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore.c b/arch/x86/kernel/cpu/perf_event_intel_uncore.c
index 0a5571080e74..99d96a4978b5 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_uncore.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_uncore.c
@@ -661,6 +661,11 @@ static void snb_uncore_msr_init_box(struct intel_uncore_box *box)
 	}
 }
 
+static struct uncore_event_desc snb_uncore_events[] = {
+	INTEL_UNCORE_EVENT_DESC(clockticks, "event=0xff,umask=0x00"),
+	{ /* end: all zeroes */ },
+};
+
 static struct attribute *snb_uncore_formats_attr[] = {
 	&format_attr_event.attr,
 	&format_attr_umask.attr,
@@ -704,6 +709,7 @@ static struct intel_uncore_type snb_uncore_cbox = {
 	.constraints	= snb_uncore_cbox_constraints,
 	.ops		= &snb_uncore_msr_ops,
 	.format_group	= &snb_uncore_format_group,
+	.event_descs	= snb_uncore_events,
 };
 
 static struct intel_uncore_type *snb_msr_uncores[] = {
@@ -1944,7 +1950,7 @@ struct intel_uncore_box *uncore_alloc_box(struct intel_uncore_type *type, int cp
 static struct intel_uncore_box *
 uncore_pmu_to_box(struct intel_uncore_pmu *pmu, int cpu)
 {
-	static struct intel_uncore_box *box;
+	struct intel_uncore_box *box;
 
 	box = *per_cpu_ptr(pmu->box, cpu);
 	if (box)
@@ -2341,6 +2347,27 @@ int uncore_pmu_event_init(struct perf_event *event)
 	return ret;
 }
 
+static ssize_t uncore_get_attr_cpumask(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	int n = cpulist_scnprintf(buf, PAGE_SIZE - 2, &uncore_cpu_mask);
+
+	buf[n++] = '\n';
+	buf[n] = '\0';
+	return n;
+}
+
+static DEVICE_ATTR(cpumask, S_IRUGO, uncore_get_attr_cpumask, NULL);
+
+static struct attribute *uncore_pmu_attrs[] = {
+	&dev_attr_cpumask.attr,
+	NULL,
+};
+
+static struct attribute_group uncore_pmu_attr_group = {
+	.attrs = uncore_pmu_attrs,
+};
+
 static int __init uncore_pmu_register(struct intel_uncore_pmu *pmu)
 {
 	int ret;
@@ -2378,8 +2405,8 @@ static void __init uncore_type_exit(struct intel_uncore_type *type)
 		free_percpu(type->pmus[i].box);
 	kfree(type->pmus);
 	type->pmus = NULL;
-	kfree(type->attr_groups[1]);
-	type->attr_groups[1] = NULL;
+	kfree(type->events_group);
+	type->events_group = NULL;
 }
 
 static void __init uncore_types_exit(struct intel_uncore_type **types)
@@ -2431,9 +2458,10 @@ static int __init uncore_type_init(struct intel_uncore_type *type)
 		for (j = 0; j < i; j++)
 			attrs[j] = &type->event_descs[j].attr.attr;
 
-		type->attr_groups[1] = events_group;
+		type->events_group = events_group;
 	}
 
+	type->pmu_group = &uncore_pmu_attr_group;
 	type->pmus = pmus;
 	return 0;
 fail:
diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore.h b/arch/x86/kernel/cpu/perf_event_intel_uncore.h
index 5b81c1856aac..e68a4550e952 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_uncore.h
+++ b/arch/x86/kernel/cpu/perf_event_intel_uncore.h
@@ -369,10 +369,12 @@ struct intel_uncore_type {
 	struct intel_uncore_pmu *pmus;
 	struct intel_uncore_ops *ops;
 	struct uncore_event_desc *event_descs;
-	const struct attribute_group *attr_groups[3];
+	const struct attribute_group *attr_groups[4];
 };
 
-#define format_group attr_groups[0]
+#define pmu_group attr_groups[0]
+#define format_group attr_groups[1]
+#define events_group attr_groups[2]
 
 struct intel_uncore_ops {
 	void (*init_box)(struct intel_uncore_box *);
diff --git a/arch/x86/kernel/cpu/perf_event_knc.c b/arch/x86/kernel/cpu/perf_event_knc.c
new file mode 100644
index 000000000000..7c46bfdbc373
--- /dev/null
+++ b/arch/x86/kernel/cpu/perf_event_knc.c
@@ -0,0 +1,248 @@
+/* Driver for Intel Xeon Phi "Knights Corner" PMU */
+
+#include <linux/perf_event.h>
+#include <linux/types.h>
+
+#include "perf_event.h"
+
+static const u64 knc_perfmon_event_map[] =
+{
+  [PERF_COUNT_HW_CPU_CYCLES]		= 0x002a,
+  [PERF_COUNT_HW_INSTRUCTIONS]		= 0x0016,
+  [PERF_COUNT_HW_CACHE_REFERENCES]	= 0x0028,
+  [PERF_COUNT_HW_CACHE_MISSES]		= 0x0029,
+  [PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= 0x0012,
+  [PERF_COUNT_HW_BRANCH_MISSES]		= 0x002b,
+};
+
+static __initconst u64 knc_hw_cache_event_ids
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX] =
+{
+ [ C(L1D) ] = {
+	[ C(OP_READ) ] = {
+		/* On Xeon Phi event "0" is a valid DATA_READ          */
+		/*   (L1 Data Cache Reads) Instruction.                */
+		/* We code this as ARCH_PERFMON_EVENTSEL_INT as this   */
+		/* bit will always be set in x86_pmu_hw_config().      */
+		[ C(RESULT_ACCESS) ] = ARCH_PERFMON_EVENTSEL_INT,
+						/* DATA_READ           */
+		[ C(RESULT_MISS)   ] = 0x0003,	/* DATA_READ_MISS      */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0001,	/* DATA_WRITE          */
+		[ C(RESULT_MISS)   ] = 0x0004,	/* DATA_WRITE_MISS     */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0011,	/* L1_DATA_PF1         */
+		[ C(RESULT_MISS)   ] = 0x001c,	/* L1_DATA_PF1_MISS    */
+	},
+ },
+ [ C(L1I ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x000c,	/* CODE_READ          */
+		[ C(RESULT_MISS)   ] = 0x000e,	/* CODE_CACHE_MISS    */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(LL  ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0,
+		[ C(RESULT_MISS)   ] = 0x10cb,	/* L2_READ_MISS */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x10cc,	/* L2_WRITE_HIT */
+		[ C(RESULT_MISS)   ] = 0,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x10fc,	/* L2_DATA_PF2      */
+		[ C(RESULT_MISS)   ] = 0x10fe,	/* L2_DATA_PF2_MISS */
+	},
+ },
+ [ C(DTLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = ARCH_PERFMON_EVENTSEL_INT,
+						/* DATA_READ */
+						/* see note on L1 OP_READ */
+		[ C(RESULT_MISS)   ] = 0x0002,	/* DATA_PAGE_WALK */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0001,	/* DATA_WRITE */
+		[ C(RESULT_MISS)   ] = 0x0002,	/* DATA_PAGE_WALK */
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0,
+		[ C(RESULT_MISS)   ] = 0x0,
+	},
+ },
+ [ C(ITLB) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x000c,	/* CODE_READ */
+		[ C(RESULT_MISS)   ] = 0x000d,	/* CODE_PAGE_WALK */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+ [ C(BPU ) ] = {
+	[ C(OP_READ) ] = {
+		[ C(RESULT_ACCESS) ] = 0x0012,	/* BRANCHES */
+		[ C(RESULT_MISS)   ] = 0x002b,	/* BRANCHES_MISPREDICTED */
+	},
+	[ C(OP_WRITE) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+	[ C(OP_PREFETCH) ] = {
+		[ C(RESULT_ACCESS) ] = -1,
+		[ C(RESULT_MISS)   ] = -1,
+	},
+ },
+};
+
+
+static u64 knc_pmu_event_map(int hw_event)
+{
+	return knc_perfmon_event_map[hw_event];
+}
+
+static struct event_constraint knc_event_constraints[] =
+{
+	INTEL_EVENT_CONSTRAINT(0xc3, 0x1),	/* HWP_L2HIT */
+	INTEL_EVENT_CONSTRAINT(0xc4, 0x1),	/* HWP_L2MISS */
+	INTEL_EVENT_CONSTRAINT(0xc8, 0x1),	/* L2_READ_HIT_E */
+	INTEL_EVENT_CONSTRAINT(0xc9, 0x1),	/* L2_READ_HIT_M */
+	INTEL_EVENT_CONSTRAINT(0xca, 0x1),	/* L2_READ_HIT_S */
+	INTEL_EVENT_CONSTRAINT(0xcb, 0x1),	/* L2_READ_MISS */
+	INTEL_EVENT_CONSTRAINT(0xcc, 0x1),	/* L2_WRITE_HIT */
+	INTEL_EVENT_CONSTRAINT(0xce, 0x1),	/* L2_STRONGLY_ORDERED_STREAMING_VSTORES_MISS */
+	INTEL_EVENT_CONSTRAINT(0xcf, 0x1),	/* L2_WEAKLY_ORDERED_STREAMING_VSTORE_MISS */
+	INTEL_EVENT_CONSTRAINT(0xd7, 0x1),	/* L2_VICTIM_REQ_WITH_DATA */
+	INTEL_EVENT_CONSTRAINT(0xe3, 0x1),	/* SNP_HITM_BUNIT */
+	INTEL_EVENT_CONSTRAINT(0xe6, 0x1),	/* SNP_HIT_L2 */
+	INTEL_EVENT_CONSTRAINT(0xe7, 0x1),	/* SNP_HITM_L2 */
+	INTEL_EVENT_CONSTRAINT(0xf1, 0x1),	/* L2_DATA_READ_MISS_CACHE_FILL */
+	INTEL_EVENT_CONSTRAINT(0xf2, 0x1),	/* L2_DATA_WRITE_MISS_CACHE_FILL */
+	INTEL_EVENT_CONSTRAINT(0xf6, 0x1),	/* L2_DATA_READ_MISS_MEM_FILL */
+	INTEL_EVENT_CONSTRAINT(0xf7, 0x1),	/* L2_DATA_WRITE_MISS_MEM_FILL */
+	INTEL_EVENT_CONSTRAINT(0xfc, 0x1),	/* L2_DATA_PF2 */
+	INTEL_EVENT_CONSTRAINT(0xfd, 0x1),	/* L2_DATA_PF2_DROP */
+	INTEL_EVENT_CONSTRAINT(0xfe, 0x1),	/* L2_DATA_PF2_MISS */
+	INTEL_EVENT_CONSTRAINT(0xff, 0x1),	/* L2_DATA_HIT_INFLIGHT_PF2 */
+	EVENT_CONSTRAINT_END
+};
+
+#define MSR_KNC_IA32_PERF_GLOBAL_STATUS		0x0000002d
+#define MSR_KNC_IA32_PERF_GLOBAL_OVF_CONTROL	0x0000002e
+#define MSR_KNC_IA32_PERF_GLOBAL_CTRL		0x0000002f
+
+#define KNC_ENABLE_COUNTER0			0x00000001
+#define KNC_ENABLE_COUNTER1			0x00000002
+
+static void knc_pmu_disable_all(void)
+{
+	u64 val;
+
+	rdmsrl(MSR_KNC_IA32_PERF_GLOBAL_CTRL, val);
+	val &= ~(KNC_ENABLE_COUNTER0|KNC_ENABLE_COUNTER1);
+	wrmsrl(MSR_KNC_IA32_PERF_GLOBAL_CTRL, val);
+}
+
+static void knc_pmu_enable_all(int added)
+{
+	u64 val;
+
+	rdmsrl(MSR_KNC_IA32_PERF_GLOBAL_CTRL, val);
+	val |= (KNC_ENABLE_COUNTER0|KNC_ENABLE_COUNTER1);
+	wrmsrl(MSR_KNC_IA32_PERF_GLOBAL_CTRL, val);
+}
+
+static inline void
+knc_pmu_disable_event(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	u64 val;
+
+	val = hwc->config;
+	if (cpuc->enabled)
+		val &= ~ARCH_PERFMON_EVENTSEL_ENABLE;
+
+	(void)wrmsrl_safe(hwc->config_base + hwc->idx, val);
+}
+
+static void knc_pmu_enable_event(struct perf_event *event)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	u64 val;
+
+	val = hwc->config;
+	if (cpuc->enabled)
+		val |= ARCH_PERFMON_EVENTSEL_ENABLE;
+
+	(void)wrmsrl_safe(hwc->config_base + hwc->idx, val);
+}
+
+PMU_FORMAT_ATTR(event,	"config:0-7"	);
+PMU_FORMAT_ATTR(umask,	"config:8-15"	);
+PMU_FORMAT_ATTR(edge,	"config:18"	);
+PMU_FORMAT_ATTR(inv,	"config:23"	);
+PMU_FORMAT_ATTR(cmask,	"config:24-31"	);
+
+static struct attribute *intel_knc_formats_attr[] = {
+	&format_attr_event.attr,
+	&format_attr_umask.attr,
+	&format_attr_edge.attr,
+	&format_attr_inv.attr,
+	&format_attr_cmask.attr,
+	NULL,
+};
+
+static __initconst struct x86_pmu knc_pmu = {
+	.name			= "knc",
+	.handle_irq		= x86_pmu_handle_irq,
+	.disable_all		= knc_pmu_disable_all,
+	.enable_all		= knc_pmu_enable_all,
+	.enable			= knc_pmu_enable_event,
+	.disable		= knc_pmu_disable_event,
+	.hw_config		= x86_pmu_hw_config,
+	.schedule_events	= x86_schedule_events,
+	.eventsel		= MSR_KNC_EVNTSEL0,
+	.perfctr		= MSR_KNC_PERFCTR0,
+	.event_map		= knc_pmu_event_map,
+	.max_events             = ARRAY_SIZE(knc_perfmon_event_map),
+	.apic			= 1,
+	.max_period		= (1ULL << 31) - 1,
+	.version		= 0,
+	.num_counters		= 2,
+	/* in theory 40 bits, early silicon is buggy though */
+	.cntval_bits		= 32,
+	.cntval_mask		= (1ULL << 32) - 1,
+	.get_event_constraints	= x86_get_event_constraints,
+	.event_constraints	= knc_event_constraints,
+	.format_attrs		= intel_knc_formats_attr,
+};
+
+__init int knc_pmu_init(void)
+{
+	x86_pmu = knc_pmu;
+
+	memcpy(hw_cache_event_ids, knc_hw_cache_event_ids, 
+		sizeof(hw_cache_event_ids));
+
+	return 0;
+}
diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c
index 966512b2cacf..2e8caf03f593 100644
--- a/arch/x86/kernel/cpu/perfctr-watchdog.c
+++ b/arch/x86/kernel/cpu/perfctr-watchdog.c
@@ -56,6 +56,8 @@ static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
 		switch (boot_cpu_data.x86) {
 		case 6:
 			return msr - MSR_P6_PERFCTR0;
+		case 11:
+			return msr - MSR_KNC_PERFCTR0;
 		case 15:
 			return msr - MSR_P4_BPU_PERFCTR0;
 		}
@@ -82,6 +84,8 @@ static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr)
 		switch (boot_cpu_data.x86) {
 		case 6:
 			return msr - MSR_P6_EVNTSEL0;
+		case 11:
+			return msr - MSR_KNC_EVNTSEL0;
 		case 15:
 			return msr - MSR_P4_BSU_ESCR0;
 		}
diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c
index 8022c6681485..fbd895562292 100644
--- a/arch/x86/kernel/cpu/proc.c
+++ b/arch/x86/kernel/cpu/proc.c
@@ -140,10 +140,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 
 static void *c_start(struct seq_file *m, loff_t *pos)
 {
-	if (*pos == 0)	/* just in case, cpu 0 is not the first */
-		*pos = cpumask_first(cpu_online_mask);
-	else
-		*pos = cpumask_next(*pos - 1, cpu_online_mask);
+	*pos = cpumask_next(*pos - 1, cpu_online_mask);
 	if ((*pos) < nr_cpu_ids)
 		return &cpu_data(*pos);
 	return NULL;