Merge branch 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull perf events changes from Ingo Molnar:

 "- kernel side:

   - Intel uncore PMU support for Nehalem and Sandy Bridge CPUs, we
     support both the events available via the MSR and via the PCI
     access space.

   - various uprobes cleanups and restructurings

   - PMU driver quirks by microcode version and required x86 microcode
     loader cleanups/robustization

   - various tracing robustness updates

   - static keys: remove obsolete static_branch()

  - tooling side:

   - GTK browser improvements

   - perf report browser: support screenshots to file

   - more automated tests

   - perf kvm improvements

   - perf bench refinements

   - build environment improvements

   - pipe mode improvements

   - libtraceevent updates, we have now hopefully merged most bits with
     the out of tree forked code base

  ... and many other goodies."

* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (138 commits)
  tracing: Check for allocation failure in __tracing_open()
  perf/x86: Fix intel_perfmon_event_mapformatting
  jump label: Remove static_branch()
  tracepoint: Use static_key_false(), since static_branch() is deprecated
  perf/x86: Uncore filter support for SandyBridge-EP
  perf/x86: Detect number of instances of uncore CBox
  perf/x86: Fix event constraint for SandyBridge-EP C-Box
  perf/x86: Use 0xff as pseudo code for fixed uncore event
  perf/x86: Save a few bytes in 'struct x86_pmu'
  perf/x86: Add a microcode revision check for SNB-PEBS
  perf/x86: Improve debug output in check_hw_exists()
  perf/x86/amd: Unify AMD's generic and family 15h pmus
  perf/x86: Move Intel specific code to intel_pmu_init()
  perf/x86: Rename Intel specific macros
  perf/x86: Fix USER/KERNEL tagging of samples
  perf tools: Split event symbols arrays to hw and sw parts
  perf tools: Split out PE_VALUE_SYM parsing token to SW and HW tokens
  perf tools: Add empty rule for new line in event syntax parsing
  perf test: Use ARRAY_SIZE in parse events tests
  tools lib traceevent: Cleanup realloc use
  ...

arch/x86/include/asm/kvm_host.h

@@ -313,8 +313,8 @@ struct kvm_pmu {
 	u64 counter_bitmask[2];
 	u64 global_ctrl_mask;
 	u8 version;
-	struct kvm_pmc gp_counters[X86_PMC_MAX_GENERIC];
-	struct kvm_pmc fixed_counters[X86_PMC_MAX_FIXED];
+	struct kvm_pmc gp_counters[INTEL_PMC_MAX_GENERIC];
+	struct kvm_pmc fixed_counters[INTEL_PMC_MAX_FIXED];
 	struct irq_work irq_work;
 	u64 reprogram_pmi;
 };

arch/x86/include/asm/msr.h

@@ -115,8 +115,8 @@ notrace static inline int native_write_msr_safe(unsigned int msr,
 
 extern unsigned long long native_read_tsc(void);
 
-extern int native_rdmsr_safe_regs(u32 regs[8]);
-extern int native_wrmsr_safe_regs(u32 regs[8]);
+extern int rdmsr_safe_regs(u32 regs[8]);
+extern int wrmsr_safe_regs(u32 regs[8]);
 
 static __always_inline unsigned long long __native_read_tsc(void)
 {
@@ -187,43 +187,6 @@ static inline int rdmsrl_safe(unsigned msr, unsigned long long *p)
 	return err;
 }
 
-static inline int rdmsrl_amd_safe(unsigned msr, unsigned long long *p)
-{
-	u32 gprs[8] = { 0 };
-	int err;
-
-	gprs[1] = msr;
-	gprs[7] = 0x9c5a203a;
-
-	err = native_rdmsr_safe_regs(gprs);
-
-	*p = gprs[0] | ((u64)gprs[2] << 32);
-
-	return err;
-}
-
-static inline int wrmsrl_amd_safe(unsigned msr, unsigned long long val)
-{
-	u32 gprs[8] = { 0 };
-
-	gprs[0] = (u32)val;
-	gprs[1] = msr;
-	gprs[2] = val >> 32;
-	gprs[7] = 0x9c5a203a;
-
-	return native_wrmsr_safe_regs(gprs);
-}
-
-static inline int rdmsr_safe_regs(u32 regs[8])
-{
-	return native_rdmsr_safe_regs(regs);
-}
-
-static inline int wrmsr_safe_regs(u32 regs[8])
-{
-	return native_wrmsr_safe_regs(regs);
-}
-
 #define rdtscl(low)						\
 	((low) = (u32)__native_read_tsc())
 
@@ -237,6 +200,8 @@ do {							\
 	(high) = (u32)(_l >> 32);			\
 } while (0)
 
+#define rdpmcl(counter, val) ((val) = native_read_pmc(counter))
+
 #define rdtscp(low, high, aux)					\
 do {                                                            \
 	unsigned long long _val = native_read_tscp(&(aux));     \
@@ -248,8 +213,7 @@ do {                                                            \
 
 #endif	/* !CONFIG_PARAVIRT */
 
-
-#define checking_wrmsrl(msr, val) wrmsr_safe((msr), (u32)(val),		\
+#define wrmsrl_safe(msr, val) wrmsr_safe((msr), (u32)(val),		\
 					     (u32)((val) >> 32))
 
 #define write_tsc(val1, val2) wrmsr(MSR_IA32_TSC, (val1), (val2))

arch/x86/include/asm/paravirt.h

@@ -128,21 +128,11 @@ static inline u64 paravirt_read_msr(unsigned msr, int *err)
 	return PVOP_CALL2(u64, pv_cpu_ops.read_msr, msr, err);
 }
 
-static inline int paravirt_rdmsr_regs(u32 *regs)
-{
-	return PVOP_CALL1(int, pv_cpu_ops.rdmsr_regs, regs);
-}
-
 static inline int paravirt_write_msr(unsigned msr, unsigned low, unsigned high)
 {
 	return PVOP_CALL3(int, pv_cpu_ops.write_msr, msr, low, high);
 }
 
-static inline int paravirt_wrmsr_regs(u32 *regs)
-{
-	return PVOP_CALL1(int, pv_cpu_ops.wrmsr_regs, regs);
-}
-
 /* These should all do BUG_ON(_err), but our headers are too tangled. */
 #define rdmsr(msr, val1, val2)			\
 do {						\
@@ -176,9 +166,6 @@ do {						\
 	_err;					\
 })
 
-#define rdmsr_safe_regs(regs)	paravirt_rdmsr_regs(regs)
-#define wrmsr_safe_regs(regs)	paravirt_wrmsr_regs(regs)
-
 static inline int rdmsrl_safe(unsigned msr, unsigned long long *p)
 {
 	int err;
@@ -186,32 +173,6 @@ static inline int rdmsrl_safe(unsigned msr, unsigned long long *p)
 	*p = paravirt_read_msr(msr, &err);
 	return err;
 }
-static inline int rdmsrl_amd_safe(unsigned msr, unsigned long long *p)
-{
-	u32 gprs[8] = { 0 };
-	int err;
-
-	gprs[1] = msr;
-	gprs[7] = 0x9c5a203a;
-
-	err = paravirt_rdmsr_regs(gprs);
-
-	*p = gprs[0] | ((u64)gprs[2] << 32);
-
-	return err;
-}
-
-static inline int wrmsrl_amd_safe(unsigned msr, unsigned long long val)
-{
-	u32 gprs[8] = { 0 };
-
-	gprs[0] = (u32)val;
-	gprs[1] = msr;
-	gprs[2] = val >> 32;
-	gprs[7] = 0x9c5a203a;
-
-	return paravirt_wrmsr_regs(gprs);
-}
 
 static inline u64 paravirt_read_tsc(void)
 {
@@ -252,6 +213,8 @@ do {						\
 	high = _l >> 32;			\
 } while (0)
 
+#define rdpmcl(counter, val) ((val) = paravirt_read_pmc(counter))
+
 static inline unsigned long long paravirt_rdtscp(unsigned int *aux)
 {
 	return PVOP_CALL1(u64, pv_cpu_ops.read_tscp, aux);

arch/x86/include/asm/paravirt_types.h

@@ -153,9 +153,7 @@ struct pv_cpu_ops {
 	/* MSR, PMC and TSR operations.
 	   err = 0/-EFAULT.  wrmsr returns 0/-EFAULT. */
 	u64 (*read_msr)(unsigned int msr, int *err);
-	int (*rdmsr_regs)(u32 *regs);
 	int (*write_msr)(unsigned int msr, unsigned low, unsigned high);
-	int (*wrmsr_regs)(u32 *regs);
 
 	u64 (*read_tsc)(void);
 	u64 (*read_pmc)(int counter);

arch/x86/include/asm/perf_event.h

@@ -5,11 +5,10 @@
  * Performance event hw details:
  */
 
-#define X86_PMC_MAX_GENERIC				       32
-#define X86_PMC_MAX_FIXED					3
+#define INTEL_PMC_MAX_GENERIC				       32
+#define INTEL_PMC_MAX_FIXED					3
+#define INTEL_PMC_IDX_FIXED				       32
 
-#define X86_PMC_IDX_GENERIC				        0
-#define X86_PMC_IDX_FIXED				       32
 #define X86_PMC_IDX_MAX					       64
 
 #define MSR_ARCH_PERFMON_PERFCTR0			      0xc1
@@ -48,8 +47,7 @@
 	(X86_RAW_EVENT_MASK          |  \
 	 AMD64_EVENTSEL_EVENT)
 #define AMD64_NUM_COUNTERS				4
-#define AMD64_NUM_COUNTERS_F15H				6
-#define AMD64_NUM_COUNTERS_MAX				AMD64_NUM_COUNTERS_F15H
+#define AMD64_NUM_COUNTERS_CORE				6
 
 #define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL		0x3c
 #define ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK		(0x00 << 8)
@@ -121,16 +119,16 @@ struct x86_pmu_capability {
 
 /* Instr_Retired.Any: */
 #define MSR_ARCH_PERFMON_FIXED_CTR0	0x309
-#define X86_PMC_IDX_FIXED_INSTRUCTIONS	(X86_PMC_IDX_FIXED + 0)
+#define INTEL_PMC_IDX_FIXED_INSTRUCTIONS	(INTEL_PMC_IDX_FIXED + 0)
 
 /* CPU_CLK_Unhalted.Core: */
 #define MSR_ARCH_PERFMON_FIXED_CTR1	0x30a
-#define X86_PMC_IDX_FIXED_CPU_CYCLES	(X86_PMC_IDX_FIXED + 1)
+#define INTEL_PMC_IDX_FIXED_CPU_CYCLES	(INTEL_PMC_IDX_FIXED + 1)
 
 /* CPU_CLK_Unhalted.Ref: */
 #define MSR_ARCH_PERFMON_FIXED_CTR2	0x30b
-#define X86_PMC_IDX_FIXED_REF_CYCLES	(X86_PMC_IDX_FIXED + 2)
-#define X86_PMC_MSK_FIXED_REF_CYCLES	(1ULL << X86_PMC_IDX_FIXED_REF_CYCLES)
+#define INTEL_PMC_IDX_FIXED_REF_CYCLES	(INTEL_PMC_IDX_FIXED + 2)
+#define INTEL_PMC_MSK_FIXED_REF_CYCLES	(1ULL << INTEL_PMC_IDX_FIXED_REF_CYCLES)
 
 /*
  * We model BTS tracing as another fixed-mode PMC.
@@ -139,7 +137,7 @@ struct x86_pmu_capability {
  * values are used by actual fixed events and higher values are used
  * to indicate other overflow conditions in the PERF_GLOBAL_STATUS msr.
  */
-#define X86_PMC_IDX_FIXED_BTS				(X86_PMC_IDX_FIXED + 16)
+#define INTEL_PMC_IDX_FIXED_BTS				(INTEL_PMC_IDX_FIXED + 16)
 
 /*
  * IBS cpuid feature detection
@@ -234,6 +232,7 @@ struct perf_guest_switch_msr {
 
 extern struct perf_guest_switch_msr *perf_guest_get_msrs(int *nr);
 extern void perf_get_x86_pmu_capability(struct x86_pmu_capability *cap);
+extern void perf_check_microcode(void);
 #else
 static inline perf_guest_switch_msr *perf_guest_get_msrs(int *nr)
 {
@@ -247,6 +246,7 @@ static inline void perf_get_x86_pmu_capability(struct x86_pmu_capability *cap)
 }
 
 static inline void perf_events_lapic_init(void)	{ }
+static inline void perf_check_microcode(void) { }
 #endif
 
 #if defined(CONFIG_PERF_EVENTS) && defined(CONFIG_CPU_SUP_AMD)

arch/x86/include/asm/uprobes.h

@@ -48,7 +48,7 @@ struct arch_uprobe_task {
 #endif
 };
 
-extern int  arch_uprobe_analyze_insn(struct arch_uprobe *aup, struct mm_struct *mm);
+extern int  arch_uprobe_analyze_insn(struct arch_uprobe *aup, struct mm_struct *mm, unsigned long addr);
 extern int  arch_uprobe_pre_xol(struct arch_uprobe *aup, struct pt_regs *regs);
 extern int  arch_uprobe_post_xol(struct arch_uprobe *aup, struct pt_regs *regs);
 extern bool arch_uprobe_xol_was_trapped(struct task_struct *tsk);

arch/x86/kernel/cpu/Makefile

@@ -32,7 +32,9 @@ 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 perf_event_intel_lbr.o perf_event_intel_ds.o perf_event_intel.o
+obj-$(CONFIG_CPU_SUP_INTEL)		+= perf_event_p6.o perf_event_p4.o
+obj-$(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
 
 obj-$(CONFIG_X86_MCE)			+= mcheck/

arch/x86/kernel/cpu/amd.c

@@ -19,6 +19,39 @@
 
 #include "cpu.h"
 
+static inline int rdmsrl_amd_safe(unsigned msr, unsigned long long *p)
+{
+	struct cpuinfo_x86 *c = &cpu_data(smp_processor_id());
+	u32 gprs[8] = { 0 };
+	int err;
+
+	WARN_ONCE((c->x86 != 0xf), "%s should only be used on K8!\n", __func__);
+
+	gprs[1] = msr;
+	gprs[7] = 0x9c5a203a;
+
+	err = rdmsr_safe_regs(gprs);
+
+	*p = gprs[0] | ((u64)gprs[2] << 32);
+
+	return err;
+}
+
+static inline int wrmsrl_amd_safe(unsigned msr, unsigned long long val)
+{
+	struct cpuinfo_x86 *c = &cpu_data(smp_processor_id());
+	u32 gprs[8] = { 0 };
+
+	WARN_ONCE((c->x86 != 0xf), "%s should only be used on K8!\n", __func__);
+
+	gprs[0] = (u32)val;
+	gprs[1] = msr;
+	gprs[2] = val >> 32;
+	gprs[7] = 0x9c5a203a;
+
+	return wrmsr_safe_regs(gprs);
+}
+
 #ifdef CONFIG_X86_32
 /*
  *	B step AMD K6 before B 9730xxxx have hardware bugs that can cause
@@ -586,9 +619,9 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 	    !cpu_has(c, X86_FEATURE_TOPOEXT)) {
 		u64 val;
 
-		if (!rdmsrl_amd_safe(0xc0011005, &val)) {
+		if (!rdmsrl_safe(0xc0011005, &val)) {
 			val |= 1ULL << 54;
-			wrmsrl_amd_safe(0xc0011005, val);
+			wrmsrl_safe(0xc0011005, val);
 			rdmsrl(0xc0011005, val);
 			if (val & (1ULL << 54)) {
 				set_cpu_cap(c, X86_FEATURE_TOPOEXT);
@@ -679,7 +712,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 		err = rdmsrl_safe(MSR_AMD64_MCx_MASK(4), &mask);
 		if (err == 0) {
 			mask |= (1 << 10);
-			checking_wrmsrl(MSR_AMD64_MCx_MASK(4), mask);
+			wrmsrl_safe(MSR_AMD64_MCx_MASK(4), mask);
 		}
 	}
 

arch/x86/kernel/cpu/common.c

@@ -947,7 +947,7 @@ static void __cpuinit __print_cpu_msr(void)
 		index_max = msr_range_array[i].max;
 
 		for (index = index_min; index < index_max; index++) {
-			if (rdmsrl_amd_safe(index, &val))
+			if (rdmsrl_safe(index, &val))
 				continue;
 			printk(KERN_INFO " MSR%08x: %016llx\n", index, val);
 		}

arch/x86/kernel/cpu/perf_event.c

@@ -35,17 +35,6 @@
 
 #include "perf_event.h"
 
-#if 0
-#undef wrmsrl
-#define wrmsrl(msr, val) 					\
-do {								\
-	trace_printk("wrmsrl(%lx, %lx)\n", (unsigned long)(msr),\
-			(unsigned long)(val));			\
-	native_write_msr((msr), (u32)((u64)(val)), 		\
-			(u32)((u64)(val) >> 32));		\
-} while (0)
-#endif
-
 struct x86_pmu x86_pmu __read_mostly;
 
 DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
@@ -74,7 +63,7 @@ u64 x86_perf_event_update(struct perf_event *event)
 	int idx = hwc->idx;
 	s64 delta;
 
-	if (idx == X86_PMC_IDX_FIXED_BTS)
+	if (idx == INTEL_PMC_IDX_FIXED_BTS)
 		return 0;
 
 	/*
@@ -86,7 +75,7 @@ u64 x86_perf_event_update(struct perf_event *event)
 	 */
 again:
 	prev_raw_count = local64_read(&hwc->prev_count);
-	rdmsrl(hwc->event_base, new_raw_count);
+	rdpmcl(hwc->event_base_rdpmc, new_raw_count);
 
 	if (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
 					new_raw_count) != prev_raw_count)
@@ -189,7 +178,7 @@ static void release_pmc_hardware(void) {}
 
 static bool check_hw_exists(void)
 {
-	u64 val, val_new = 0;
+	u64 val, val_new = ~0;
 	int i, reg, ret = 0;
 
 	/*
@@ -222,8 +211,9 @@ static bool check_hw_exists(void)
 	 * that don't trap on the MSR access and always return 0s.
 	 */
 	val = 0xabcdUL;
-	ret = checking_wrmsrl(x86_pmu_event_addr(0), val);
-	ret |= rdmsrl_safe(x86_pmu_event_addr(0), &val_new);
+	reg = x86_pmu_event_addr(0);
+	ret = wrmsrl_safe(reg, val);
+	ret |= rdmsrl_safe(reg, &val_new);
 	if (ret || val != val_new)
 		goto msr_fail;
 
@@ -240,6 +230,7 @@ static bool check_hw_exists(void)
 
 msr_fail:
 	printk(KERN_CONT "Broken PMU hardware detected, using software events only.\n");
+	printk(KERN_ERR "Failed to access perfctr msr (MSR %x is %Lx)\n", reg, val_new);
 
 	return false;
 }
@@ -388,7 +379,7 @@ int x86_pmu_hw_config(struct perf_event *event)
 		int precise = 0;
 
 		/* Support for constant skid */
-		if (x86_pmu.pebs_active) {
+		if (x86_pmu.pebs_active && !x86_pmu.pebs_broken) {
 			precise++;
 
 			/* Support for IP fixup */
@@ -637,8 +628,8 @@ static bool __perf_sched_find_counter(struct perf_sched *sched)
 	c = sched->constraints[sched->state.event];
 
 	/* Prefer fixed purpose counters */
-	if (x86_pmu.num_counters_fixed) {
-		idx = X86_PMC_IDX_FIXED;
+	if (c->idxmsk64 & (~0ULL << INTEL_PMC_IDX_FIXED)) {
+		idx = INTEL_PMC_IDX_FIXED;
 		for_each_set_bit_from(idx, c->idxmsk, X86_PMC_IDX_MAX) {
 			if (!__test_and_set_bit(idx, sched->state.used))
 				goto done;
@@ -646,7 +637,7 @@ static bool __perf_sched_find_counter(struct perf_sched *sched)
 	}
 	/* Grab the first unused counter starting with idx */
 	idx = sched->state.counter;
-	for_each_set_bit_from(idx, c->idxmsk, X86_PMC_IDX_FIXED) {
+	for_each_set_bit_from(idx, c->idxmsk, INTEL_PMC_IDX_FIXED) {
 		if (!__test_and_set_bit(idx, sched->state.used))
 			goto done;
 	}
@@ -704,8 +695,8 @@ static bool perf_sched_next_event(struct perf_sched *sched)
 /*
  * Assign a counter for each event.
  */
-static int perf_assign_events(struct event_constraint **constraints, int n,
-			      int wmin, int wmax, int *assign)
+int perf_assign_events(struct event_constraint **constraints, int n,
+			int wmin, int wmax, int *assign)
 {
 	struct perf_sched sched;
 
@@ -824,15 +815,17 @@ static inline void x86_assign_hw_event(struct perf_event *event,
 	hwc->last_cpu = smp_processor_id();
 	hwc->last_tag = ++cpuc->tags[i];
 
-	if (hwc->idx == X86_PMC_IDX_FIXED_BTS) {
+	if (hwc->idx == INTEL_PMC_IDX_FIXED_BTS) {
 		hwc->config_base = 0;
 		hwc->event_base	= 0;
-	} else if (hwc->idx >= X86_PMC_IDX_FIXED) {
+	} else if (hwc->idx >= INTEL_PMC_IDX_FIXED) {
 		hwc->config_base = MSR_ARCH_PERFMON_FIXED_CTR_CTRL;
-		hwc->event_base = MSR_ARCH_PERFMON_FIXED_CTR0 + (hwc->idx - X86_PMC_IDX_FIXED);
+		hwc->event_base = MSR_ARCH_PERFMON_FIXED_CTR0 + (hwc->idx - INTEL_PMC_IDX_FIXED);
+		hwc->event_base_rdpmc = (hwc->idx - INTEL_PMC_IDX_FIXED) | 1<<30;
 	} else {
 		hwc->config_base = x86_pmu_config_addr(hwc->idx);
 		hwc->event_base  = x86_pmu_event_addr(hwc->idx);
+		hwc->event_base_rdpmc = hwc->idx;
 	}
 }
 
@@ -930,7 +923,7 @@ int x86_perf_event_set_period(struct perf_event *event)
 	s64 period = hwc->sample_period;
 	int ret = 0, idx = hwc->idx;
 
-	if (idx == X86_PMC_IDX_FIXED_BTS)
+	if (idx == INTEL_PMC_IDX_FIXED_BTS)
 		return 0;
 
 	/*
@@ -1316,7 +1309,6 @@ static struct attribute_group x86_pmu_format_group = {
 static int __init init_hw_perf_events(void)
 {
 	struct x86_pmu_quirk *quirk;
-	struct event_constraint *c;
 	int err;
 
 	pr_info("Performance Events: ");
@@ -1347,21 +1339,8 @@ static int __init init_hw_perf_events(void)
 	for (quirk = x86_pmu.quirks; quirk; quirk = quirk->next)
 		quirk->func();
 
-	if (x86_pmu.num_counters > X86_PMC_MAX_GENERIC) {
-		WARN(1, KERN_ERR "hw perf events %d > max(%d), clipping!",
-		     x86_pmu.num_counters, X86_PMC_MAX_GENERIC);
-		x86_pmu.num_counters = X86_PMC_MAX_GENERIC;
-	}
-	x86_pmu.intel_ctrl = (1 << x86_pmu.num_counters) - 1;
-
-	if (x86_pmu.num_counters_fixed > X86_PMC_MAX_FIXED) {
-		WARN(1, KERN_ERR "hw perf events fixed %d > max(%d), clipping!",
-		     x86_pmu.num_counters_fixed, X86_PMC_MAX_FIXED);
-		x86_pmu.num_counters_fixed = X86_PMC_MAX_FIXED;
-	}
-
-	x86_pmu.intel_ctrl |=
-		((1LL << x86_pmu.num_counters_fixed)-1) << X86_PMC_IDX_FIXED;
+	if (!x86_pmu.intel_ctrl)
+		x86_pmu.intel_ctrl = (1 << x86_pmu.num_counters) - 1;
 
 	perf_events_lapic_init();
 	register_nmi_handler(NMI_LOCAL, perf_event_nmi_handler, 0, "PMI");
@@ -1370,22 +1349,6 @@ static int __init init_hw_perf_events(void)
 		__EVENT_CONSTRAINT(0, (1ULL << x86_pmu.num_counters) - 1,
 				   0, x86_pmu.num_counters, 0);
 
-	if (x86_pmu.event_constraints) {
-		/*
-		 * event on fixed counter2 (REF_CYCLES) only works on this
-		 * counter, so do not extend mask to generic counters
-		 */
-		for_each_event_constraint(c, x86_pmu.event_constraints) {
-			if (c->cmask != X86_RAW_EVENT_MASK
-			    || c->idxmsk64 == X86_PMC_MSK_FIXED_REF_CYCLES) {
-				continue;
-			}
-
-			c->idxmsk64 |= (1ULL << x86_pmu.num_counters) - 1;
-			c->weight += x86_pmu.num_counters;
-		}
-	}
-
 	x86_pmu.attr_rdpmc = 1; /* enable userspace RDPMC usage by default */
 	x86_pmu_format_group.attrs = x86_pmu.format_attrs;
 
@@ -1620,8 +1583,8 @@ static int x86_pmu_event_idx(struct perf_event *event)
 	if (!x86_pmu.attr_rdpmc)
 		return 0;
 
-	if (x86_pmu.num_counters_fixed && idx >= X86_PMC_IDX_FIXED) {
-		idx -= X86_PMC_IDX_FIXED;
+	if (x86_pmu.num_counters_fixed && idx >= INTEL_PMC_IDX_FIXED) {
+		idx -= INTEL_PMC_IDX_FIXED;
 		idx |= 1 << 30;
 	}
 
@@ -1649,7 +1612,12 @@ static ssize_t set_attr_rdpmc(struct device *cdev,
 			      struct device_attribute *attr,
 			      const char *buf, size_t count)
 {
-	unsigned long val = simple_strtoul(buf, NULL, 0);
+	unsigned long val;
+	ssize_t ret;
+
+	ret = kstrtoul(buf, 0, &val);
+	if (ret)
+		return ret;
 
 	if (!!val != !!x86_pmu.attr_rdpmc) {
 		x86_pmu.attr_rdpmc = !!val;
@@ -1682,13 +1650,20 @@ static void x86_pmu_flush_branch_stack(void)
 		x86_pmu.flush_branch_stack();
 }
 
+void perf_check_microcode(void)
+{
+	if (x86_pmu.check_microcode)
+		x86_pmu.check_microcode();
+}
+EXPORT_SYMBOL_GPL(perf_check_microcode);
+
 static struct pmu pmu = {
 	.pmu_enable		= x86_pmu_enable,
 	.pmu_disable		= x86_pmu_disable,
 
-	.attr_groups	= x86_pmu_attr_groups,
+	.attr_groups		= x86_pmu_attr_groups,
 
-	.event_init	= x86_pmu_event_init,
+	.event_init		= x86_pmu_event_init,
 
 	.add			= x86_pmu_add,
 	.del			= x86_pmu_del,
@@ -1696,11 +1671,11 @@ static struct pmu pmu = {
 	.stop			= x86_pmu_stop,
 	.read			= x86_pmu_read,
 
-	.start_txn	= x86_pmu_start_txn,
-	.cancel_txn	= x86_pmu_cancel_txn,
-	.commit_txn	= x86_pmu_commit_txn,
+	.start_txn		= x86_pmu_start_txn,
+	.cancel_txn		= x86_pmu_cancel_txn,
+	.commit_txn		= x86_pmu_commit_txn,
 
-	.event_idx	= x86_pmu_event_idx,
+	.event_idx		= x86_pmu_event_idx,
 	.flush_branch_stack	= x86_pmu_flush_branch_stack,
 };
 
@@ -1863,7 +1838,7 @@ unsigned long perf_misc_flags(struct pt_regs *regs)
 		else
 			misc |= PERF_RECORD_MISC_GUEST_KERNEL;
 	} else {
-		if (user_mode(regs))
+		if (!kernel_ip(regs->ip))
 			misc |= PERF_RECORD_MISC_USER;
 		else
 			misc |= PERF_RECORD_MISC_KERNEL;

arch/x86/kernel/cpu/perf_event.h

@@ -14,6 +14,18 @@
 
 #include <linux/perf_event.h>
 
+#if 0
+#undef wrmsrl
+#define wrmsrl(msr, val) 						\
+do {									\
+	unsigned int _msr = (msr);					\
+	u64 _val = (val);						\
+	trace_printk("wrmsrl(%x, %Lx)\n", (unsigned int)(_msr),		\
+			(unsigned long long)(_val));			\
+	native_write_msr((_msr), (u32)(_val), (u32)(_val >> 32));	\
+} while (0)
+#endif
+
 /*
  *          |   NHM/WSM    |      SNB     |
  * register -------------------------------
@@ -57,7 +69,7 @@ struct amd_nb {
 };
 
 /* The maximal number of PEBS events: */
-#define MAX_PEBS_EVENTS		4
+#define MAX_PEBS_EVENTS		8
 
 /*
  * A debug store configuration.
@@ -349,6 +361,8 @@ struct x86_pmu {
 	void		(*cpu_starting)(int cpu);
 	void		(*cpu_dying)(int cpu);
 	void		(*cpu_dead)(int cpu);
+
+	void		(*check_microcode)(void);
 	void		(*flush_branch_stack)(void);
 
 	/*
@@ -360,12 +374,16 @@ struct x86_pmu {
 	/*
 	 * Intel DebugStore bits
 	 */
-	int		bts, pebs;
-	int		bts_active, pebs_active;
+	int		bts		:1,
+			bts_active	:1,
+			pebs		:1,
+			pebs_active	:1,
+			pebs_broken	:1;
 	int		pebs_record_size;
 	void		(*drain_pebs)(struct pt_regs *regs);
 	struct event_constraint *pebs_constraints;
 	void		(*pebs_aliases)(struct perf_event *event);
+	int 		max_pebs_events;
 
 	/*
 	 * Intel LBR
@@ -468,6 +486,8 @@ static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc,
 
 void x86_pmu_enable_all(int added);
 
+int perf_assign_events(struct event_constraint **constraints, int n,
+			int wmin, int wmax, int *assign);
 int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign);
 
 void x86_pmu_stop(struct perf_event *event, int flags);

arch/x86/kernel/cpu/perf_event_amd.c

@@ -366,7 +366,7 @@ static void amd_pmu_cpu_starting(int cpu)
 
 	cpuc->perf_ctr_virt_mask = AMD_PERFMON_EVENTSEL_HOSTONLY;
 
-	if (boot_cpu_data.x86_max_cores < 2 || boot_cpu_data.x86 == 0x15)
+	if (boot_cpu_data.x86_max_cores < 2)
 		return;
 
 	nb_id = amd_get_nb_id(cpu);
@@ -422,35 +422,6 @@ static struct attribute *amd_format_attr[] = {
 	NULL,
 };
 
-static __initconst const struct x86_pmu amd_pmu = {
-	.name			= "AMD",
-	.handle_irq		= x86_pmu_handle_irq,
-	.disable_all		= x86_pmu_disable_all,
-	.enable_all		= x86_pmu_enable_all,
-	.enable			= x86_pmu_enable_event,
-	.disable		= x86_pmu_disable_event,
-	.hw_config		= amd_pmu_hw_config,
-	.schedule_events	= x86_schedule_events,
-	.eventsel		= MSR_K7_EVNTSEL0,
-	.perfctr		= MSR_K7_PERFCTR0,
-	.event_map		= amd_pmu_event_map,
-	.max_events		= ARRAY_SIZE(amd_perfmon_event_map),
-	.num_counters		= AMD64_NUM_COUNTERS,
-	.cntval_bits		= 48,
-	.cntval_mask		= (1ULL << 48) - 1,
-	.apic			= 1,
-	/* use highest bit to detect overflow */
-	.max_period		= (1ULL << 47) - 1,
-	.get_event_constraints	= amd_get_event_constraints,
-	.put_event_constraints	= amd_put_event_constraints,
-
-	.format_attrs		= amd_format_attr,
-
-	.cpu_prepare		= amd_pmu_cpu_prepare,
-	.cpu_starting		= amd_pmu_cpu_starting,
-	.cpu_dead		= amd_pmu_cpu_dead,
-};
-
 /* AMD Family 15h */
 
 #define AMD_EVENT_TYPE_MASK	0x000000F0ULL
@@ -597,8 +568,8 @@ amd_get_event_constraints_f15h(struct cpu_hw_events *cpuc, struct perf_event *ev
 	}
 }
 
-static __initconst const struct x86_pmu amd_pmu_f15h = {
-	.name			= "AMD Family 15h",
+static __initconst const struct x86_pmu amd_pmu = {
+	.name			= "AMD",
 	.handle_irq		= x86_pmu_handle_irq,
 	.disable_all		= x86_pmu_disable_all,
 	.enable_all		= x86_pmu_enable_all,
@@ -606,50 +577,68 @@ static __initconst const struct x86_pmu amd_pmu_f15h = {
 	.disable		= x86_pmu_disable_event,
 	.hw_config		= amd_pmu_hw_config,
 	.schedule_events	= x86_schedule_events,
-	.eventsel		= MSR_F15H_PERF_CTL,
-	.perfctr		= MSR_F15H_PERF_CTR,
+	.eventsel		= MSR_K7_EVNTSEL0,
+	.perfctr		= MSR_K7_PERFCTR0,
 	.event_map		= amd_pmu_event_map,
 	.max_events		= ARRAY_SIZE(amd_perfmon_event_map),
-	.num_counters		= AMD64_NUM_COUNTERS_F15H,
+	.num_counters		= AMD64_NUM_COUNTERS,
 	.cntval_bits		= 48,
 	.cntval_mask		= (1ULL << 48) - 1,
 	.apic			= 1,
 	/* use highest bit to detect overflow */
 	.max_period		= (1ULL << 47) - 1,
-	.get_event_constraints	= amd_get_event_constraints_f15h,
-	/* nortbridge counters not yet implemented: */
-#if 0
+	.get_event_constraints	= amd_get_event_constraints,
 	.put_event_constraints	= amd_put_event_constraints,
 
+	.format_attrs		= amd_format_attr,
+
 	.cpu_prepare		= amd_pmu_cpu_prepare,
-	.cpu_dead		= amd_pmu_cpu_dead,
-#endif
 	.cpu_starting		= amd_pmu_cpu_starting,
-	.format_attrs		= amd_format_attr,
+	.cpu_dead		= amd_pmu_cpu_dead,
 };
 
+static int setup_event_constraints(void)
+{
+	if (boot_cpu_data.x86 >= 0x15)
+		x86_pmu.get_event_constraints = amd_get_event_constraints_f15h;
+	return 0;
+}
+
+static int setup_perfctr_core(void)
+{
+	if (!cpu_has_perfctr_core) {
+		WARN(x86_pmu.get_event_constraints == amd_get_event_constraints_f15h,
+		     KERN_ERR "Odd, counter constraints enabled but no core perfctrs detected!");
+		return -ENODEV;
+	}
+
+	WARN(x86_pmu.get_event_constraints == amd_get_event_constraints,
+	     KERN_ERR "hw perf events core counters need constraints handler!");
+
+	/*
+	 * If core performance counter extensions exists, we must use
+	 * MSR_F15H_PERF_CTL/MSR_F15H_PERF_CTR msrs. See also
+	 * x86_pmu_addr_offset().
+	 */
+	x86_pmu.eventsel	= MSR_F15H_PERF_CTL;
+	x86_pmu.perfctr		= MSR_F15H_PERF_CTR;
+	x86_pmu.num_counters	= AMD64_NUM_COUNTERS_CORE;
+
+	printk(KERN_INFO "perf: AMD core performance counters detected\n");
+
+	return 0;
+}
+
 __init int amd_pmu_init(void)
 {
 	/* Performance-monitoring supported from K7 and later: */
 	if (boot_cpu_data.x86 < 6)
 		return -ENODEV;
 
-	/*
-	 * If core performance counter extensions exists, it must be
-	 * family 15h, otherwise fail. See x86_pmu_addr_offset().
-	 */
-	switch (boot_cpu_data.x86) {
-	case 0x15:
-		if (!cpu_has_perfctr_core)
-			return -ENODEV;
-		x86_pmu = amd_pmu_f15h;
-		break;
-	default:
-		if (cpu_has_perfctr_core)
-			return -ENODEV;
-		x86_pmu = amd_pmu;
-		break;
-	}
+	x86_pmu = amd_pmu;
+
+	setup_event_constraints();
+	setup_perfctr_core();
 
 	/* Events are common for all AMDs */
 	memcpy(hw_cache_event_ids, amd_hw_cache_event_ids,

arch/x86/kernel/cpu/perf_event_intel.c

@@ -21,14 +21,14 @@
  */
 static u64 intel_perfmon_event_map[PERF_COUNT_HW_MAX] __read_mostly =
 {
-  [PERF_COUNT_HW_CPU_CYCLES]		= 0x003c,
-  [PERF_COUNT_HW_INSTRUCTIONS]		= 0x00c0,
-  [PERF_COUNT_HW_CACHE_REFERENCES]	= 0x4f2e,
-  [PERF_COUNT_HW_CACHE_MISSES]		= 0x412e,
-  [PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= 0x00c4,
-  [PERF_COUNT_HW_BRANCH_MISSES]		= 0x00c5,
-  [PERF_COUNT_HW_BUS_CYCLES]		= 0x013c,
-  [PERF_COUNT_HW_REF_CPU_CYCLES]	= 0x0300, /* pseudo-encoding */
+	[PERF_COUNT_HW_CPU_CYCLES]		= 0x003c,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= 0x00c0,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= 0x4f2e,
+	[PERF_COUNT_HW_CACHE_MISSES]		= 0x412e,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= 0x00c4,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= 0x00c5,
+	[PERF_COUNT_HW_BUS_CYCLES]		= 0x013c,
+	[PERF_COUNT_HW_REF_CPU_CYCLES]		= 0x0300, /* pseudo-encoding */
 };
 
 static struct event_constraint intel_core_event_constraints[] __read_mostly =
@@ -747,7 +747,7 @@ static void intel_pmu_disable_all(void)
 
 	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0);
 
-	if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask))
+	if (test_bit(INTEL_PMC_IDX_FIXED_BTS, cpuc->active_mask))
 		intel_pmu_disable_bts();
 
 	intel_pmu_pebs_disable_all();
@@ -763,9 +763,9 @@ static void intel_pmu_enable_all(int added)
 	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL,
 			x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_guest_mask);
 
-	if (test_bit(X86_PMC_IDX_FIXED_BTS, cpuc->active_mask)) {
+	if (test_bit(INTEL_PMC_IDX_FIXED_BTS, cpuc->active_mask)) {
 		struct perf_event *event =
-			cpuc->events[X86_PMC_IDX_FIXED_BTS];
+			cpuc->events[INTEL_PMC_IDX_FIXED_BTS];
 
 		if (WARN_ON_ONCE(!event))
 			return;
@@ -871,7 +871,7 @@ static inline void intel_pmu_ack_status(u64 ack)
 
 static void intel_pmu_disable_fixed(struct hw_perf_event *hwc)
 {
-	int idx = hwc->idx - X86_PMC_IDX_FIXED;
+	int idx = hwc->idx - INTEL_PMC_IDX_FIXED;
 	u64 ctrl_val, mask;
 
 	mask = 0xfULL << (idx * 4);
@@ -886,7 +886,7 @@ static void intel_pmu_disable_event(struct perf_event *event)
 	struct hw_perf_event *hwc = &event->hw;
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
-	if (unlikely(hwc->idx == X86_PMC_IDX_FIXED_BTS)) {
+	if (unlikely(hwc->idx == INTEL_PMC_IDX_FIXED_BTS)) {
 		intel_pmu_disable_bts();
 		intel_pmu_drain_bts_buffer();
 		return;
@@ -915,7 +915,7 @@ static void intel_pmu_disable_event(struct perf_event *event)
 
 static void intel_pmu_enable_fixed(struct hw_perf_event *hwc)
 {
-	int idx = hwc->idx - X86_PMC_IDX_FIXED;
+	int idx = hwc->idx - INTEL_PMC_IDX_FIXED;
 	u64 ctrl_val, bits, mask;
 
 	/*
@@ -949,7 +949,7 @@ static void intel_pmu_enable_event(struct perf_event *event)
 	struct hw_perf_event *hwc = &event->hw;
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
-	if (unlikely(hwc->idx == X86_PMC_IDX_FIXED_BTS)) {
+	if (unlikely(hwc->idx == INTEL_PMC_IDX_FIXED_BTS)) {
 		if (!__this_cpu_read(cpu_hw_events.enabled))
 			return;
 
@@ -1003,11 +1003,11 @@ static void intel_pmu_reset(void)
 	printk("clearing PMU state on CPU#%d\n", smp_processor_id());
 
 	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
-		checking_wrmsrl(x86_pmu_config_addr(idx), 0ull);
-		checking_wrmsrl(x86_pmu_event_addr(idx),  0ull);
+		wrmsrl_safe(x86_pmu_config_addr(idx), 0ull);
+		wrmsrl_safe(x86_pmu_event_addr(idx),  0ull);
 	}
 	for (idx = 0; idx < x86_pmu.num_counters_fixed; idx++)
-		checking_wrmsrl(MSR_ARCH_PERFMON_FIXED_CTR0 + idx, 0ull);
+		wrmsrl_safe(MSR_ARCH_PERFMON_FIXED_CTR0 + idx, 0ull);
 
 	if (ds)
 		ds->bts_index = ds->bts_buffer_base;
@@ -1712,11 +1712,56 @@ static __init void intel_clovertown_quirk(void)
 	x86_pmu.pebs_constraints = NULL;
 }
 
+static int intel_snb_pebs_broken(int cpu)
+{
+	u32 rev = UINT_MAX; /* default to broken for unknown models */
+
+	switch (cpu_data(cpu).x86_model) {
+	case 42: /* SNB */
+		rev = 0x28;
+		break;
+
+	case 45: /* SNB-EP */
+		switch (cpu_data(cpu).x86_mask) {
+		case 6: rev = 0x618; break;
+		case 7: rev = 0x70c; break;
+		}
+	}
+
+	return (cpu_data(cpu).microcode < rev);
+}
+
+static void intel_snb_check_microcode(void)
+{
+	int pebs_broken = 0;
+	int cpu;
+
+	get_online_cpus();
+	for_each_online_cpu(cpu) {
+		if ((pebs_broken = intel_snb_pebs_broken(cpu)))
+			break;
+	}
+	put_online_cpus();
+
+	if (pebs_broken == x86_pmu.pebs_broken)
+		return;
+
+	/*
+	 * Serialized by the microcode lock..
+	 */
+	if (x86_pmu.pebs_broken) {
+		pr_info("PEBS enabled due to microcode update\n");
+		x86_pmu.pebs_broken = 0;
+	} else {
+		pr_info("PEBS disabled due to CPU errata, please upgrade microcode\n");
+		x86_pmu.pebs_broken = 1;
+	}
+}
+
 static __init void intel_sandybridge_quirk(void)
 {
-	printk(KERN_WARNING "PEBS disabled due to CPU errata.\n");
-	x86_pmu.pebs = 0;
-	x86_pmu.pebs_constraints = NULL;
+	x86_pmu.check_microcode = intel_snb_check_microcode;
+	intel_snb_check_microcode();
 }
 
 static const struct { int id; char *name; } intel_arch_events_map[] __initconst = {
@@ -1765,6 +1810,7 @@ __init int intel_pmu_init(void)
 	union cpuid10_edx edx;
 	union cpuid10_eax eax;
 	union cpuid10_ebx ebx;
+	struct event_constraint *c;
 	unsigned int unused;
 	int version;
 
@@ -1800,6 +1846,8 @@ __init int intel_pmu_init(void)
 	x86_pmu.events_maskl		= ebx.full;
 	x86_pmu.events_mask_len		= eax.split.mask_length;
 
+	x86_pmu.max_pebs_events		= min_t(unsigned, MAX_PEBS_EVENTS, x86_pmu.num_counters);
+
 	/*
 	 * Quirk: v2 perfmon does not report fixed-purpose events, so
 	 * assume at least 3 events:
@@ -1951,5 +1999,37 @@ __init int intel_pmu_init(void)
 		}
 	}
 
+	if (x86_pmu.num_counters > INTEL_PMC_MAX_GENERIC) {
+		WARN(1, KERN_ERR "hw perf events %d > max(%d), clipping!",
+		     x86_pmu.num_counters, INTEL_PMC_MAX_GENERIC);
+		x86_pmu.num_counters = INTEL_PMC_MAX_GENERIC;
+	}
+	x86_pmu.intel_ctrl = (1 << x86_pmu.num_counters) - 1;
+
+	if (x86_pmu.num_counters_fixed > INTEL_PMC_MAX_FIXED) {
+		WARN(1, KERN_ERR "hw perf events fixed %d > max(%d), clipping!",
+		     x86_pmu.num_counters_fixed, INTEL_PMC_MAX_FIXED);
+		x86_pmu.num_counters_fixed = INTEL_PMC_MAX_FIXED;
+	}
+
+	x86_pmu.intel_ctrl |=
+		((1LL << x86_pmu.num_counters_fixed)-1) << INTEL_PMC_IDX_FIXED;
+
+	if (x86_pmu.event_constraints) {
+		/*
+		 * event on fixed counter2 (REF_CYCLES) only works on this
+		 * counter, so do not extend mask to generic counters
+		 */
+		for_each_event_constraint(c, x86_pmu.event_constraints) {
+			if (c->cmask != X86_RAW_EVENT_MASK
+			    || c->idxmsk64 == INTEL_PMC_MSK_FIXED_REF_CYCLES) {
+				continue;
+			}
+
+			c->idxmsk64 |= (1ULL << x86_pmu.num_counters) - 1;
+			c->weight += x86_pmu.num_counters;
+		}
+	}
+
 	return 0;
 }

arch/x86/kernel/cpu/perf_event_intel_ds.c

@@ -248,7 +248,7 @@ void reserve_ds_buffers(void)
  */
 
 struct event_constraint bts_constraint =
-	EVENT_CONSTRAINT(0, 1ULL << X86_PMC_IDX_FIXED_BTS, 0);
+	EVENT_CONSTRAINT(0, 1ULL << INTEL_PMC_IDX_FIXED_BTS, 0);
 
 void intel_pmu_enable_bts(u64 config)
 {
@@ -295,7 +295,7 @@ int intel_pmu_drain_bts_buffer(void)
 		u64	to;
 		u64	flags;
 	};
-	struct perf_event *event = cpuc->events[X86_PMC_IDX_FIXED_BTS];
+	struct perf_event *event = cpuc->events[INTEL_PMC_IDX_FIXED_BTS];
 	struct bts_record *at, *top;
 	struct perf_output_handle handle;
 	struct perf_event_header header;
@@ -620,7 +620,7 @@ static void intel_pmu_drain_pebs_core(struct pt_regs *iregs)
 	 * Should not happen, we program the threshold at 1 and do not
 	 * set a reset value.
 	 */
-	WARN_ON_ONCE(n > 1);
+	WARN_ONCE(n > 1, "bad leftover pebs %d\n", n);
 	at += n - 1;
 
 	__intel_pmu_pebs_event(event, iregs, at);
@@ -651,10 +651,10 @@ static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs)
 	 * Should not happen, we program the threshold at 1 and do not
 	 * set a reset value.
 	 */
-	WARN_ON_ONCE(n > MAX_PEBS_EVENTS);
+	WARN_ONCE(n > x86_pmu.max_pebs_events, "Unexpected number of pebs records %d\n", n);
 
 	for ( ; at < top; at++) {
-		for_each_set_bit(bit, (unsigned long *)&at->status, MAX_PEBS_EVENTS) {
+		for_each_set_bit(bit, (unsigned long *)&at->status, x86_pmu.max_pebs_events) {
 			event = cpuc->events[bit];
 			if (!test_bit(bit, cpuc->active_mask))
 				continue;
@@ -670,7 +670,7 @@ static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs)
 			break;
 		}
 
-		if (!event || bit >= MAX_PEBS_EVENTS)
+		if (!event || bit >= x86_pmu.max_pebs_events)
 			continue;
 
 		__intel_pmu_pebs_event(event, iregs, at);

arch/x86/kernel/cpu/perf_event_p4.c

@@ -895,8 +895,8 @@ static void p4_pmu_disable_pebs(void)
 	 * So at moment let leave metrics turned on forever -- it's
 	 * ok for now but need to be revisited!
 	 *
-	 * (void)checking_wrmsrl(MSR_IA32_PEBS_ENABLE, (u64)0);
-	 * (void)checking_wrmsrl(MSR_P4_PEBS_MATRIX_VERT, (u64)0);
+	 * (void)wrmsrl_safe(MSR_IA32_PEBS_ENABLE, (u64)0);
+	 * (void)wrmsrl_safe(MSR_P4_PEBS_MATRIX_VERT, (u64)0);
 	 */
 }
 
@@ -909,7 +909,7 @@ static inline void p4_pmu_disable_event(struct perf_event *event)
 	 * state we need to clear P4_CCCR_OVF, otherwise interrupt get
 	 * asserted again and again
 	 */
-	(void)checking_wrmsrl(hwc->config_base,
+	(void)wrmsrl_safe(hwc->config_base,
 		(u64)(p4_config_unpack_cccr(hwc->config)) &
 			~P4_CCCR_ENABLE & ~P4_CCCR_OVF & ~P4_CCCR_RESERVED);
 }
@@ -943,8 +943,8 @@ static void p4_pmu_enable_pebs(u64 config)
 
 	bind = &p4_pebs_bind_map[idx];
 
-	(void)checking_wrmsrl(MSR_IA32_PEBS_ENABLE,	(u64)bind->metric_pebs);
-	(void)checking_wrmsrl(MSR_P4_PEBS_MATRIX_VERT,	(u64)bind->metric_vert);
+	(void)wrmsrl_safe(MSR_IA32_PEBS_ENABLE,	(u64)bind->metric_pebs);
+	(void)wrmsrl_safe(MSR_P4_PEBS_MATRIX_VERT,	(u64)bind->metric_vert);
 }
 
 static void p4_pmu_enable_event(struct perf_event *event)
@@ -978,8 +978,8 @@ static void p4_pmu_enable_event(struct perf_event *event)
 	 */
 	p4_pmu_enable_pebs(hwc->config);
 
-	(void)checking_wrmsrl(escr_addr, escr_conf);
-	(void)checking_wrmsrl(hwc->config_base,
+	(void)wrmsrl_safe(escr_addr, escr_conf);
+	(void)wrmsrl_safe(hwc->config_base,
 				(cccr & ~P4_CCCR_RESERVED) | P4_CCCR_ENABLE);
 }
 
@@ -1325,7 +1325,7 @@ __init int p4_pmu_init(void)
 	unsigned int low, high;
 
 	/* If we get stripped -- indexing fails */
-	BUILD_BUG_ON(ARCH_P4_MAX_CCCR > X86_PMC_MAX_GENERIC);
+	BUILD_BUG_ON(ARCH_P4_MAX_CCCR > INTEL_PMC_MAX_GENERIC);
 
 	rdmsr(MSR_IA32_MISC_ENABLE, low, high);
 	if (!(low & (1 << 7))) {

arch/x86/kernel/cpu/perf_event_p6.c

@@ -71,7 +71,7 @@ p6_pmu_disable_event(struct perf_event *event)
 	if (cpuc->enabled)
 		val |= ARCH_PERFMON_EVENTSEL_ENABLE;
 
-	(void)checking_wrmsrl(hwc->config_base, val);
+	(void)wrmsrl_safe(hwc->config_base, val);
 }
 
 static void p6_pmu_enable_event(struct perf_event *event)
@@ -84,7 +84,7 @@ static void p6_pmu_enable_event(struct perf_event *event)
 	if (cpuc->enabled)
 		val |= ARCH_PERFMON_EVENTSEL_ENABLE;
 
-	(void)checking_wrmsrl(hwc->config_base, val);
+	(void)wrmsrl_safe(hwc->config_base, val);
 }
 
 PMU_FORMAT_ATTR(event,	"config:0-7"	);

arch/x86/kernel/microcode_core.c

@@ -87,6 +87,7 @@
 #include <asm/microcode.h>
 #include <asm/processor.h>
 #include <asm/cpu_device_id.h>
+#include <asm/perf_event.h>
 
 MODULE_DESCRIPTION("Microcode Update Driver");
 MODULE_AUTHOR("Tigran Aivazian <tigran@aivazian.fsnet.co.uk>");
@@ -277,7 +278,6 @@ static int reload_for_cpu(int cpu)
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
 	int err = 0;
 
-	mutex_lock(&microcode_mutex);
 	if (uci->valid) {
 		enum ucode_state ustate;
 
@@ -288,7 +288,6 @@ static int reload_for_cpu(int cpu)
 			if (ustate == UCODE_ERROR)
 				err = -EINVAL;
 	}
-	mutex_unlock(&microcode_mutex);
 
 	return err;
 }
@@ -298,19 +297,31 @@ static ssize_t reload_store(struct device *dev,
 			    const char *buf, size_t size)
 {
 	unsigned long val;
-	int cpu = dev->id;
-	ssize_t ret = 0;
+	int cpu;
+	ssize_t ret = 0, tmp_ret;
 
 	ret = kstrtoul(buf, 0, &val);
 	if (ret)
 		return ret;
 
-	if (val == 1) {
-		get_online_cpus();
-		if (cpu_online(cpu))
-			ret = reload_for_cpu(cpu);
-		put_online_cpus();
+	if (val != 1)
+		return size;
+
+	get_online_cpus();
+	mutex_lock(&microcode_mutex);
+	for_each_online_cpu(cpu) {
+		tmp_ret = reload_for_cpu(cpu);
+		if (tmp_ret != 0)
+			pr_warn("Error reloading microcode on CPU %d\n", cpu);
+
+		/* save retval of the first encountered reload error */
+		if (!ret)
+			ret = tmp_ret;
 	}
+	if (!ret)
+		perf_check_microcode();
+	mutex_unlock(&microcode_mutex);
+	put_online_cpus();
 
 	if (!ret)
 		ret = size;
@@ -339,7 +350,6 @@ static DEVICE_ATTR(version, 0400, version_show, NULL);
 static DEVICE_ATTR(processor_flags, 0400, pf_show, NULL);
 
 static struct attribute *mc_default_attrs[] = {
-	&dev_attr_reload.attr,
 	&dev_attr_version.attr,
 	&dev_attr_processor_flags.attr,
 	NULL
@@ -516,6 +526,16 @@ static const struct x86_cpu_id microcode_id[] = {
 MODULE_DEVICE_TABLE(x86cpu, microcode_id);
 #endif
 
+static struct attribute *cpu_root_microcode_attrs[] = {
+	&dev_attr_reload.attr,
+	NULL
+};
+
+static struct attribute_group cpu_root_microcode_group = {
+	.name  = "microcode",
+	.attrs = cpu_root_microcode_attrs,
+};
+
 static int __init microcode_init(void)
 {
 	struct cpuinfo_x86 *c = &cpu_data(0);
@@ -540,16 +560,25 @@ static int __init microcode_init(void)
 	mutex_lock(&microcode_mutex);
 
 	error = subsys_interface_register(&mc_cpu_interface);
-
+	if (!error)
+		perf_check_microcode();
 	mutex_unlock(&microcode_mutex);
 	put_online_cpus();
 
 	if (error)
 		goto out_pdev;
 
+	error = sysfs_create_group(&cpu_subsys.dev_root->kobj,
+				   &cpu_root_microcode_group);
+
+	if (error) {
+		pr_err("Error creating microcode group!\n");
+		goto out_driver;
+	}
+
 	error = microcode_dev_init();
 	if (error)
-		goto out_driver;
+		goto out_ucode_group;
 
 	register_syscore_ops(&mc_syscore_ops);
 	register_hotcpu_notifier(&mc_cpu_notifier);
@@ -559,7 +588,11 @@ static int __init microcode_init(void)
 
 	return 0;
 
-out_driver:
+ out_ucode_group:
+	sysfs_remove_group(&cpu_subsys.dev_root->kobj,
+			   &cpu_root_microcode_group);
+
+ out_driver:
 	get_online_cpus();
 	mutex_lock(&microcode_mutex);
 
@@ -568,7 +601,7 @@ static int __init microcode_init(void)
 	mutex_unlock(&microcode_mutex);
 	put_online_cpus();
 
-out_pdev:
+ out_pdev:
 	platform_device_unregister(microcode_pdev);
 	return error;
 
@@ -584,6 +617,9 @@ static void __exit microcode_exit(void)
 	unregister_hotcpu_notifier(&mc_cpu_notifier);
 	unregister_syscore_ops(&mc_syscore_ops);
 
+	sysfs_remove_group(&cpu_subsys.dev_root->kobj,
+			   &cpu_root_microcode_group);
+
 	get_online_cpus();
 	mutex_lock(&microcode_mutex);
 

arch/x86/kernel/paravirt.c

@@ -352,9 +352,7 @@ struct pv_cpu_ops pv_cpu_ops = {
 #endif
 	.wbinvd = native_wbinvd,
 	.read_msr = native_read_msr_safe,
-	.rdmsr_regs = native_rdmsr_safe_regs,
 	.write_msr = native_write_msr_safe,
-	.wrmsr_regs = native_wrmsr_safe_regs,
 	.read_tsc = native_read_tsc,
 	.read_pmc = native_read_pmc,
 	.read_tscp = native_read_tscp,

arch/x86/kernel/process_64.c

@@ -466,7 +466,7 @@ long do_arch_prctl(struct task_struct *task, int code, unsigned long addr)
 			task->thread.gs = addr;
 			if (doit) {
 				load_gs_index(0);
-				ret = checking_wrmsrl(MSR_KERNEL_GS_BASE, addr);
+				ret = wrmsrl_safe(MSR_KERNEL_GS_BASE, addr);
 			}
 		}
 		put_cpu();
@@ -494,7 +494,7 @@ long do_arch_prctl(struct task_struct *task, int code, unsigned long addr)
 				/* set the selector to 0 to not confuse
 				   __switch_to */
 				loadsegment(fs, 0);
-				ret = checking_wrmsrl(MSR_FS_BASE, addr);
+				ret = wrmsrl_safe(MSR_FS_BASE, addr);
 			}
 		}
 		put_cpu();

arch/x86/kernel/uprobes.c

@@ -409,9 +409,10 @@ static int validate_insn_bits(struct arch_uprobe *auprobe, struct mm_struct *mm,
  * arch_uprobe_analyze_insn - instruction analysis including validity and fixups.
  * @mm: the probed address space.
  * @arch_uprobe: the probepoint information.
+ * @addr: virtual address at which to install the probepoint
  * Return 0 on success or a -ve number on error.
  */
-int arch_uprobe_analyze_insn(struct arch_uprobe *auprobe, struct mm_struct *mm)
+int arch_uprobe_analyze_insn(struct arch_uprobe *auprobe, struct mm_struct *mm, unsigned long addr)
 {
 	int ret;
 	struct insn insn;

arch/x86/kvm/pmu.c

@@ -80,10 +80,10 @@ static inline struct kvm_pmc *get_fixed_pmc_idx(struct kvm_pmu *pmu, int idx)
 
 static struct kvm_pmc *global_idx_to_pmc(struct kvm_pmu *pmu, int idx)
 {
-	if (idx < X86_PMC_IDX_FIXED)
+	if (idx < INTEL_PMC_IDX_FIXED)
 		return get_gp_pmc(pmu, MSR_P6_EVNTSEL0 + idx, MSR_P6_EVNTSEL0);
 	else
-		return get_fixed_pmc_idx(pmu, idx - X86_PMC_IDX_FIXED);
+		return get_fixed_pmc_idx(pmu, idx - INTEL_PMC_IDX_FIXED);
 }
 
 void kvm_deliver_pmi(struct kvm_vcpu *vcpu)
@@ -291,7 +291,7 @@ static void reprogram_idx(struct kvm_pmu *pmu, int idx)
 	if (pmc_is_gp(pmc))
 		reprogram_gp_counter(pmc, pmc->eventsel);
 	else {
-		int fidx = idx - X86_PMC_IDX_FIXED;
+		int fidx = idx - INTEL_PMC_IDX_FIXED;
 		reprogram_fixed_counter(pmc,
 				fixed_en_pmi(pmu->fixed_ctr_ctrl, fidx), fidx);
 	}
@@ -452,7 +452,7 @@ void kvm_pmu_cpuid_update(struct kvm_vcpu *vcpu)
 		return;
 
 	pmu->nr_arch_gp_counters = min((int)(entry->eax >> 8) & 0xff,
-			X86_PMC_MAX_GENERIC);
+			INTEL_PMC_MAX_GENERIC);
 	pmu->counter_bitmask[KVM_PMC_GP] =
 		((u64)1 << ((entry->eax >> 16) & 0xff)) - 1;
 	bitmap_len = (entry->eax >> 24) & 0xff;
@@ -462,13 +462,13 @@ void kvm_pmu_cpuid_update(struct kvm_vcpu *vcpu)
 		pmu->nr_arch_fixed_counters = 0;
 	} else {
 		pmu->nr_arch_fixed_counters = min((int)(entry->edx & 0x1f),
-				X86_PMC_MAX_FIXED);
+				INTEL_PMC_MAX_FIXED);
 		pmu->counter_bitmask[KVM_PMC_FIXED] =
 			((u64)1 << ((entry->edx >> 5) & 0xff)) - 1;
 	}
 
 	pmu->global_ctrl = ((1 << pmu->nr_arch_gp_counters) - 1) |
-		(((1ull << pmu->nr_arch_fixed_counters) - 1) << X86_PMC_IDX_FIXED);
+		(((1ull << pmu->nr_arch_fixed_counters) - 1) << INTEL_PMC_IDX_FIXED);
 	pmu->global_ctrl_mask = ~pmu->global_ctrl;
 }
 
@@ -478,15 +478,15 @@ void kvm_pmu_init(struct kvm_vcpu *vcpu)
 	struct kvm_pmu *pmu = &vcpu->arch.pmu;
 
 	memset(pmu, 0, sizeof(*pmu));
-	for (i = 0; i < X86_PMC_MAX_GENERIC; i++) {
+	for (i = 0; i < INTEL_PMC_MAX_GENERIC; i++) {
 		pmu->gp_counters[i].type = KVM_PMC_GP;
 		pmu->gp_counters[i].vcpu = vcpu;
 		pmu->gp_counters[i].idx = i;
 	}
-	for (i = 0; i < X86_PMC_MAX_FIXED; i++) {
+	for (i = 0; i < INTEL_PMC_MAX_FIXED; i++) {
 		pmu->fixed_counters[i].type = KVM_PMC_FIXED;
 		pmu->fixed_counters[i].vcpu = vcpu;
-		pmu->fixed_counters[i].idx = i + X86_PMC_IDX_FIXED;
+		pmu->fixed_counters[i].idx = i + INTEL_PMC_IDX_FIXED;
 	}
 	init_irq_work(&pmu->irq_work, trigger_pmi);
 	kvm_pmu_cpuid_update(vcpu);
@@ -498,13 +498,13 @@ void kvm_pmu_reset(struct kvm_vcpu *vcpu)
 	int i;
 
 	irq_work_sync(&pmu->irq_work);
-	for (i = 0; i < X86_PMC_MAX_GENERIC; i++) {
+	for (i = 0; i < INTEL_PMC_MAX_GENERIC; i++) {
 		struct kvm_pmc *pmc = &pmu->gp_counters[i];
 		stop_counter(pmc);
 		pmc->counter = pmc->eventsel = 0;
 	}
 
-	for (i = 0; i < X86_PMC_MAX_FIXED; i++)
+	for (i = 0; i < INTEL_PMC_MAX_FIXED; i++)
 		stop_counter(&pmu->fixed_counters[i]);
 
 	pmu->fixed_ctr_ctrl = pmu->global_ctrl = pmu->global_status =

arch/x86/kvm/trace.h

@@ -710,16 +710,6 @@ TRACE_EVENT(kvm_skinit,
 		  __entry->rip, __entry->slb)
 );
 
-#define __print_insn(insn, ilen) ({		                 \
-	int i;							 \
-	const char *ret = p->buffer + p->len;			 \
-								 \
-	for (i = 0; i < ilen; ++i)				 \
-		trace_seq_printf(p, " %02x", insn[i]);		 \
-	trace_seq_printf(p, "%c", 0);				 \
-	ret;							 \
-	})
-
 #define KVM_EMUL_INSN_F_CR0_PE (1 << 0)
 #define KVM_EMUL_INSN_F_EFL_VM (1 << 1)
 #define KVM_EMUL_INSN_F_CS_D   (1 << 2)
@@ -786,7 +776,7 @@ TRACE_EVENT(kvm_emulate_insn,
 
 	TP_printk("%x:%llx:%s (%s)%s",
 		  __entry->csbase, __entry->rip,
-		  __print_insn(__entry->insn, __entry->len),
+		  __print_hex(__entry->insn, __entry->len),
 		  __print_symbolic(__entry->flags,
 				   kvm_trace_symbol_emul_flags),
 		  __entry->failed ? " failed" : ""

arch/x86/lib/msr-reg-export.c

 #include <linux/module.h>
 #include <asm/msr.h>
 
-EXPORT_SYMBOL(native_rdmsr_safe_regs);
-EXPORT_SYMBOL(native_wrmsr_safe_regs);
+EXPORT_SYMBOL(rdmsr_safe_regs);
+EXPORT_SYMBOL(wrmsr_safe_regs);

arch/x86/lib/msr-reg.S

@@ -6,13 +6,13 @@
 
 #ifdef CONFIG_X86_64
 /*
- * int native_{rdmsr,wrmsr}_safe_regs(u32 gprs[8]);
+ * int {rdmsr,wrmsr}_safe_regs(u32 gprs[8]);
  *
  * reg layout: u32 gprs[eax, ecx, edx, ebx, esp, ebp, esi, edi]
  *
  */
 .macro op_safe_regs op
-ENTRY(native_\op\()_safe_regs)
+ENTRY(\op\()_safe_regs)
 	CFI_STARTPROC
 	pushq_cfi %rbx
 	pushq_cfi %rbp
@@ -45,13 +45,13 @@ ENTRY(native_\op\()_safe_regs)
 
 	_ASM_EXTABLE(1b, 3b)
 	CFI_ENDPROC
-ENDPROC(native_\op\()_safe_regs)
+ENDPROC(\op\()_safe_regs)
 .endm
 
 #else /* X86_32 */
 
 .macro op_safe_regs op
-ENTRY(native_\op\()_safe_regs)
+ENTRY(\op\()_safe_regs)
 	CFI_STARTPROC
 	pushl_cfi %ebx
 	pushl_cfi %ebp
@@ -92,7 +92,7 @@ ENTRY(native_\op\()_safe_regs)
 
 	_ASM_EXTABLE(1b, 3b)
 	CFI_ENDPROC
-ENDPROC(native_\op\()_safe_regs)
+ENDPROC(\op\()_safe_regs)
 .endm
 
 #endif

arch/x86/oprofile/op_model_amd.c

@@ -312,7 +312,7 @@ static int op_amd_fill_in_addresses(struct op_msrs * const msrs)
 			goto fail;
 		}
 		/* both registers must be reserved */
-		if (num_counters == AMD64_NUM_COUNTERS_F15H) {
+		if (num_counters == AMD64_NUM_COUNTERS_CORE) {
 			msrs->counters[i].addr = MSR_F15H_PERF_CTR + (i << 1);
 			msrs->controls[i].addr = MSR_F15H_PERF_CTL + (i << 1);
 		} else {
@@ -514,7 +514,7 @@ static int op_amd_init(struct oprofile_operations *ops)
 	ops->create_files = setup_ibs_files;
 
 	if (boot_cpu_data.x86 == 0x15) {
-		num_counters = AMD64_NUM_COUNTERS_F15H;
+		num_counters = AMD64_NUM_COUNTERS_CORE;
 	} else {
 		num_counters = AMD64_NUM_COUNTERS;
 	}

arch/x86/vdso/vdso32-setup.c

@@ -205,9 +205,9 @@ void syscall32_cpu_init(void)
 {
 	/* Load these always in case some future AMD CPU supports
 	   SYSENTER from compat mode too. */
-	checking_wrmsrl(MSR_IA32_SYSENTER_CS, (u64)__KERNEL_CS);
-	checking_wrmsrl(MSR_IA32_SYSENTER_ESP, 0ULL);
-	checking_wrmsrl(MSR_IA32_SYSENTER_EIP, (u64)ia32_sysenter_target);
+	wrmsrl_safe(MSR_IA32_SYSENTER_CS, (u64)__KERNEL_CS);
+	wrmsrl_safe(MSR_IA32_SYSENTER_ESP, 0ULL);
+	wrmsrl_safe(MSR_IA32_SYSENTER_EIP, (u64)ia32_sysenter_target);
 
 	wrmsrl(MSR_CSTAR, ia32_cstar_target);
 }

arch/x86/xen/enlighten.c

@@ -1124,9 +1124,7 @@ static const struct pv_cpu_ops xen_cpu_ops __initconst = {
 	.wbinvd = native_wbinvd,
 
 	.read_msr = native_read_msr_safe,
-	.rdmsr_regs = native_rdmsr_safe_regs,
 	.write_msr = xen_write_msr_safe,
-	.wrmsr_regs = native_wrmsr_safe_regs,
 
 	.read_tsc = native_read_tsc,
 	.read_pmc = native_read_pmc,

drivers/oprofile/oprofile_perf.c

 /*
  * Copyright 2010 ARM Ltd.
+ * Copyright 2012 Advanced Micro Devices, Inc., Robert Richter
  *
  * Perf-events backend for OProfile.
  */
@@ -25,7 +26,7 @@ static int oprofile_perf_enabled;
 static DEFINE_MUTEX(oprofile_perf_mutex);
 
 static struct op_counter_config *counter_config;
-static struct perf_event **perf_events[NR_CPUS];
+static DEFINE_PER_CPU(struct perf_event **, perf_events);
 static int num_counters;
 
 /*
@@ -38,7 +39,7 @@ static void op_overflow_handler(struct perf_event *event,
 	u32 cpu = smp_processor_id();
 
 	for (id = 0; id < num_counters; ++id)
-		if (perf_events[cpu][id] == event)
+		if (per_cpu(perf_events, cpu)[id] == event)
 			break;
 
 	if (id != num_counters)
@@ -74,7 +75,7 @@ static int op_create_counter(int cpu, int event)
 {
 	struct perf_event *pevent;
 
-	if (!counter_config[event].enabled || perf_events[cpu][event])
+	if (!counter_config[event].enabled || per_cpu(perf_events, cpu)[event])
 		return 0;
 
 	pevent = perf_event_create_kernel_counter(&counter_config[event].attr,
@@ -91,18 +92,18 @@ static int op_create_counter(int cpu, int event)
 		return -EBUSY;
 	}
 
-	perf_events[cpu][event] = pevent;
+	per_cpu(perf_events, cpu)[event] = pevent;
 
 	return 0;
 }
 
 static void op_destroy_counter(int cpu, int event)
 {
-	struct perf_event *pevent = perf_events[cpu][event];
+	struct perf_event *pevent = per_cpu(perf_events, cpu)[event];
 
 	if (pevent) {
 		perf_event_release_kernel(pevent);
-		perf_events[cpu][event] = NULL;
+		per_cpu(perf_events, cpu)[event] = NULL;
 	}
 }
 
@@ -257,12 +258,12 @@ void oprofile_perf_exit(void)
 
 	for_each_possible_cpu(cpu) {
 		for (id = 0; id < num_counters; ++id) {
-			event = perf_events[cpu][id];
+			event = per_cpu(perf_events, cpu)[id];
 			if (event)
 				perf_event_release_kernel(event);
 		}
 
-		kfree(perf_events[cpu]);
+		kfree(per_cpu(perf_events, cpu));
 	}
 
 	kfree(counter_config);
@@ -277,8 +278,6 @@ int __init oprofile_perf_init(struct oprofile_operations *ops)
 	if (ret)
 		return ret;
 
-	memset(&perf_events, 0, sizeof(perf_events));
-
 	num_counters = perf_num_counters();
 	if (num_counters <= 0) {
 		pr_info("oprofile: no performance counters\n");
@@ -298,9 +297,9 @@ int __init oprofile_perf_init(struct oprofile_operations *ops)
 	}
 
 	for_each_possible_cpu(cpu) {
-		perf_events[cpu] = kcalloc(num_counters,
+		per_cpu(perf_events, cpu) = kcalloc(num_counters,
 				sizeof(struct perf_event *), GFP_KERNEL);
-		if (!perf_events[cpu]) {
+		if (!per_cpu(perf_events, cpu)) {
 			pr_info("oprofile: failed to allocate %d perf events "
 					"for cpu %d\n", num_counters, cpu);
 			ret = -ENOMEM;

include/linux/ftrace_event.h

@@ -65,7 +65,7 @@ struct trace_iterator {
 	void			*private;
 	int			cpu_file;
 	struct mutex		mutex;
-	struct ring_buffer_iter	*buffer_iter[NR_CPUS];
+	struct ring_buffer_iter	**buffer_iter;
 	unsigned long		iter_flags;
 
 	/* trace_seq for __print_flags() and __print_symbolic() etc. */
@@ -207,6 +207,9 @@ struct ftrace_event_call {
 	 *   bit 1:		enabled
 	 *   bit 2:		filter_active
 	 *   bit 3:		enabled cmd record
+	 *   bit 4:		allow trace by non root (cap any)
+	 *   bit 5:		failed to apply filter
+	 *   bit 6:		ftrace internal event (do not enable)
 	 *
 	 * Changes to flags must hold the event_mutex.
 	 *

include/linux/jump_label.h

@@ -42,8 +42,7 @@
  * allowed.
  *
  * Not initializing the key (static data is initialized to 0s anyway) is the
- * same as using STATIC_KEY_INIT_FALSE and static_key_false() is
- * equivalent with static_branch().
+ * same as using STATIC_KEY_INIT_FALSE.
  *
 */
 
@@ -107,12 +106,6 @@ static __always_inline bool static_key_true(struct static_key *key)
 	return !static_key_false(key);
 }
 
-/* Deprecated. Please use 'static_key_false() instead. */
-static __always_inline bool static_branch(struct static_key *key)
-{
-	return arch_static_branch(key);
-}
-
 extern struct jump_entry __start___jump_table[];
 extern struct jump_entry __stop___jump_table[];
 
@@ -166,14 +159,6 @@ static __always_inline bool static_key_true(struct static_key *key)
 	return false;
 }
 
-/* Deprecated. Please use 'static_key_false() instead. */
-static __always_inline bool static_branch(struct static_key *key)
-{
-	if (unlikely(atomic_read(&key->enabled)) > 0)
-		return true;
-	return false;
-}
-
 static inline void static_key_slow_inc(struct static_key *key)
 {
 	atomic_inc(&key->enabled);

include/linux/pci_ids.h

@@ -2755,6 +2755,17 @@
 #define PCI_DEVICE_ID_INTEL_IOAT_SNB7	0x3c27
 #define PCI_DEVICE_ID_INTEL_IOAT_SNB8	0x3c2e
 #define PCI_DEVICE_ID_INTEL_IOAT_SNB9	0x3c2f
+#define PCI_DEVICE_ID_INTEL_UNC_HA	0x3c46
+#define PCI_DEVICE_ID_INTEL_UNC_IMC0	0x3cb0
+#define PCI_DEVICE_ID_INTEL_UNC_IMC1	0x3cb1
+#define PCI_DEVICE_ID_INTEL_UNC_IMC2	0x3cb4
+#define PCI_DEVICE_ID_INTEL_UNC_IMC3	0x3cb5
+#define PCI_DEVICE_ID_INTEL_UNC_QPI0	0x3c41
+#define PCI_DEVICE_ID_INTEL_UNC_QPI1	0x3c42
+#define PCI_DEVICE_ID_INTEL_UNC_R2PCIE	0x3c43
+#define PCI_DEVICE_ID_INTEL_UNC_R3QPI0	0x3c44
+#define PCI_DEVICE_ID_INTEL_UNC_R3QPI1	0x3c45
+#define PCI_DEVICE_ID_INTEL_JAKETOWN_UBOX	0x3ce0
 #define PCI_DEVICE_ID_INTEL_IOAT_SNB	0x402f
 #define PCI_DEVICE_ID_INTEL_5100_16	0x65f0
 #define PCI_DEVICE_ID_INTEL_5100_21	0x65f5

include/linux/perf_event.h

@@ -677,6 +677,7 @@ struct hw_perf_event {
 			u64		last_tag;
 			unsigned long	config_base;
 			unsigned long	event_base;
+			int		event_base_rdpmc;
 			int		idx;
 			int		last_cpu;
 
@@ -1106,6 +1107,8 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr,
 				struct task_struct *task,
 				perf_overflow_handler_t callback,
 				void *context);
+extern void perf_pmu_migrate_context(struct pmu *pmu,
+				int src_cpu, int dst_cpu);
 extern u64 perf_event_read_value(struct perf_event *event,
 				 u64 *enabled, u64 *running);
 

include/linux/sched.h

@@ -1581,7 +1581,6 @@ struct task_struct {
 #endif
 #ifdef CONFIG_UPROBES
 	struct uprobe_task *utask;
-	int uprobe_srcu_id;
 #endif
 };
 

include/linux/tracepoint.h

@@ -153,7 +153,7 @@ static inline void tracepoint_synchronize_unregister(void)
 	}								\
 	static inline void trace_##name##_rcuidle(proto)		\
 	{								\
-		if (static_branch(&__tracepoint_##name.key))		\
+		if (static_key_false(&__tracepoint_##name.key))		\
 			__DO_TRACE(&__tracepoint_##name,		\
 				TP_PROTO(data_proto),			\
 				TP_ARGS(data_args),			\

include/trace/ftrace.h

@@ -571,6 +571,7 @@ static inline void ftrace_test_probe_##call(void)			\
 
 #undef __print_flags
 #undef __print_symbolic
+#undef __print_hex
 #undef __get_dynamic_array
 #undef __get_str
 

kernel/events/core.c

@@ -1645,6 +1645,8 @@ perf_install_in_context(struct perf_event_context *ctx,
 	lockdep_assert_held(&ctx->mutex);
 
 	event->ctx = ctx;
+	if (event->cpu != -1)
+		event->cpu = cpu;
 
 	if (!task) {
 		/*
@@ -6252,6 +6254,8 @@ SYSCALL_DEFINE5(perf_event_open,
 		}
 	}
 
+	get_online_cpus();
+
 	event = perf_event_alloc(&attr, cpu, task, group_leader, NULL,
 				 NULL, NULL);
 	if (IS_ERR(event)) {
@@ -6304,7 +6308,7 @@ SYSCALL_DEFINE5(perf_event_open,
 	/*
 	 * Get the target context (task or percpu):
 	 */
-	ctx = find_get_context(pmu, task, cpu);
+	ctx = find_get_context(pmu, task, event->cpu);
 	if (IS_ERR(ctx)) {
 		err = PTR_ERR(ctx);
 		goto err_alloc;
@@ -6377,20 +6381,23 @@ SYSCALL_DEFINE5(perf_event_open,
 	mutex_lock(&ctx->mutex);
 
 	if (move_group) {
-		perf_install_in_context(ctx, group_leader, cpu);
+		synchronize_rcu();
+		perf_install_in_context(ctx, group_leader, event->cpu);
 		get_ctx(ctx);
 		list_for_each_entry(sibling, &group_leader->sibling_list,
 				    group_entry) {
-			perf_install_in_context(ctx, sibling, cpu);
+			perf_install_in_context(ctx, sibling, event->cpu);
 			get_ctx(ctx);
 		}
 	}
 
-	perf_install_in_context(ctx, event, cpu);
+	perf_install_in_context(ctx, event, event->cpu);
 	++ctx->generation;
 	perf_unpin_context(ctx);
 	mutex_unlock(&ctx->mutex);
 
+	put_online_cpus();
+
 	event->owner = current;
 
 	mutex_lock(&current->perf_event_mutex);
@@ -6419,6 +6426,7 @@ SYSCALL_DEFINE5(perf_event_open,
 err_alloc:
 	free_event(event);
 err_task:
+	put_online_cpus();
 	if (task)
 		put_task_struct(task);
 err_group_fd:
@@ -6479,6 +6487,39 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu,
 }
 EXPORT_SYMBOL_GPL(perf_event_create_kernel_counter);
 
+void perf_pmu_migrate_context(struct pmu *pmu, int src_cpu, int dst_cpu)
+{
+	struct perf_event_context *src_ctx;
+	struct perf_event_context *dst_ctx;
+	struct perf_event *event, *tmp;
+	LIST_HEAD(events);
+
+	src_ctx = &per_cpu_ptr(pmu->pmu_cpu_context, src_cpu)->ctx;
+	dst_ctx = &per_cpu_ptr(pmu->pmu_cpu_context, dst_cpu)->ctx;
+
+	mutex_lock(&src_ctx->mutex);
+	list_for_each_entry_safe(event, tmp, &src_ctx->event_list,
+				 event_entry) {
+		perf_remove_from_context(event);
+		put_ctx(src_ctx);
+		list_add(&event->event_entry, &events);
+	}
+	mutex_unlock(&src_ctx->mutex);
+
+	synchronize_rcu();
+
+	mutex_lock(&dst_ctx->mutex);
+	list_for_each_entry_safe(event, tmp, &events, event_entry) {
+		list_del(&event->event_entry);
+		if (event->state >= PERF_EVENT_STATE_OFF)
+			event->state = PERF_EVENT_STATE_INACTIVE;
+		perf_install_in_context(dst_ctx, event, dst_cpu);
+		get_ctx(dst_ctx);
+	}
+	mutex_unlock(&dst_ctx->mutex);
+}
+EXPORT_SYMBOL_GPL(perf_pmu_migrate_context);
+
 static void sync_child_event(struct perf_event *child_event,
 			       struct task_struct *child)
 {

kernel/trace/ftrace.c

@@ -312,7 +312,7 @@ static int remove_ftrace_list_ops(struct ftrace_ops **list,
 
 static int __register_ftrace_function(struct ftrace_ops *ops)
 {
-	if (ftrace_disabled)
+	if (unlikely(ftrace_disabled))
 		return -ENODEV;
 
 	if (FTRACE_WARN_ON(ops == &global_ops))
@@ -4299,16 +4299,12 @@ int register_ftrace_function(struct ftrace_ops *ops)
 
 	mutex_lock(&ftrace_lock);
 
-	if (unlikely(ftrace_disabled))
-		goto out_unlock;
-
 	ret = __register_ftrace_function(ops);
 	if (!ret)
 		ret = ftrace_startup(ops, 0);
 
-
- out_unlock:
 	mutex_unlock(&ftrace_lock);
+
 	return ret;
 }
 EXPORT_SYMBOL_GPL(register_ftrace_function);

kernel/trace/ring_buffer.c

@@ -3239,6 +3239,10 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer)
 	if (cpu_buffer->commit_page == cpu_buffer->reader_page)
 		goto out;
 
+	/* Don't bother swapping if the ring buffer is empty */
+	if (rb_num_of_entries(cpu_buffer) == 0)
+		goto out;
+
 	/*
 	 * Reset the reader page to size zero.
 	 */

kernel/trace/trace.c

@@ -830,6 +830,8 @@ int register_tracer(struct tracer *type)
 		current_trace = saved_tracer;
 		if (ret) {
 			printk(KERN_CONT "FAILED!\n");
+			/* Add the warning after printing 'FAILED' */
+			WARN_ON(1);
 			goto out;
 		}
 		/* Only reset on passing, to avoid touching corrupted buffers */
@@ -1708,9 +1710,11 @@ EXPORT_SYMBOL_GPL(trace_vprintk);
 
 static void trace_iterator_increment(struct trace_iterator *iter)
 {
+	struct ring_buffer_iter *buf_iter = trace_buffer_iter(iter, iter->cpu);
+
 	iter->idx++;
-	if (iter->buffer_iter[iter->cpu])
-		ring_buffer_read(iter->buffer_iter[iter->cpu], NULL);
+	if (buf_iter)
+		ring_buffer_read(buf_iter, NULL);
 }
 
 static struct trace_entry *
@@ -1718,7 +1722,7 @@ peek_next_entry(struct trace_iterator *iter, int cpu, u64 *ts,
 		unsigned long *lost_events)
 {
 	struct ring_buffer_event *event;
-	struct ring_buffer_iter *buf_iter = iter->buffer_iter[cpu];
+	struct ring_buffer_iter *buf_iter = trace_buffer_iter(iter, cpu);
 
 	if (buf_iter)
 		event = ring_buffer_iter_peek(buf_iter, ts);
@@ -1856,10 +1860,10 @@ void tracing_iter_reset(struct trace_iterator *iter, int cpu)
 
 	tr->data[cpu]->skipped_entries = 0;
 
-	if (!iter->buffer_iter[cpu])
+	buf_iter = trace_buffer_iter(iter, cpu);
+	if (!buf_iter)
 		return;
 
-	buf_iter = iter->buffer_iter[cpu];
 	ring_buffer_iter_reset(buf_iter);
 
 	/*
@@ -2205,13 +2209,15 @@ static enum print_line_t print_bin_fmt(struct trace_iterator *iter)
 
 int trace_empty(struct trace_iterator *iter)
 {
+	struct ring_buffer_iter *buf_iter;
 	int cpu;
 
 	/* If we are looking at one CPU buffer, only check that one */
 	if (iter->cpu_file != TRACE_PIPE_ALL_CPU) {
 		cpu = iter->cpu_file;
-		if (iter->buffer_iter[cpu]) {
-			if (!ring_buffer_iter_empty(iter->buffer_iter[cpu]))
+		buf_iter = trace_buffer_iter(iter, cpu);
+		if (buf_iter) {
+			if (!ring_buffer_iter_empty(buf_iter))
 				return 0;
 		} else {
 			if (!ring_buffer_empty_cpu(iter->tr->buffer, cpu))
@@ -2221,8 +2227,9 @@ int trace_empty(struct trace_iterator *iter)
 	}
 
 	for_each_tracing_cpu(cpu) {
-		if (iter->buffer_iter[cpu]) {
-			if (!ring_buffer_iter_empty(iter->buffer_iter[cpu]))
+		buf_iter = trace_buffer_iter(iter, cpu);
+		if (buf_iter) {
+			if (!ring_buffer_iter_empty(buf_iter))
 				return 0;
 		} else {
 			if (!ring_buffer_empty_cpu(iter->tr->buffer, cpu))
@@ -2381,6 +2388,11 @@ __tracing_open(struct inode *inode, struct file *file)
 	if (!iter)
 		return ERR_PTR(-ENOMEM);
 
+	iter->buffer_iter = kzalloc(sizeof(*iter->buffer_iter) * num_possible_cpus(),
+				    GFP_KERNEL);
+	if (!iter->buffer_iter)
+		goto release;
+
 	/*
 	 * We make a copy of the current tracer to avoid concurrent
 	 * changes on it while we are reading.
@@ -2441,6 +2453,8 @@ __tracing_open(struct inode *inode, struct file *file)
  fail:
 	mutex_unlock(&trace_types_lock);
 	kfree(iter->trace);
+	kfree(iter->buffer_iter);
+release:
 	seq_release_private(inode, file);
 	return ERR_PTR(-ENOMEM);
 }
@@ -2481,6 +2495,7 @@ static int tracing_release(struct inode *inode, struct file *file)
 	mutex_destroy(&iter->mutex);
 	free_cpumask_var(iter->started);
 	kfree(iter->trace);
+	kfree(iter->buffer_iter);
 	seq_release_private(inode, file);
 	return 0;
 }

kernel/trace/trace.h

@@ -317,6 +317,14 @@ struct tracer {
 
 #define TRACE_PIPE_ALL_CPU	-1
 
+static inline struct ring_buffer_iter *
+trace_buffer_iter(struct trace_iterator *iter, int cpu)
+{
+	if (iter->buffer_iter && iter->buffer_iter[cpu])
+		return iter->buffer_iter[cpu];
+	return NULL;
+}
+
 int tracer_init(struct tracer *t, struct trace_array *tr);
 int tracing_is_enabled(void);
 void trace_wake_up(void);

kernel/trace/trace_functions_graph.c

@@ -538,7 +538,7 @@ get_return_for_leaf(struct trace_iterator *iter,
 		next = &data->ret;
 	} else {
 
-		ring_iter = iter->buffer_iter[iter->cpu];
+		ring_iter = trace_buffer_iter(iter, iter->cpu);
 
 		/* First peek to compare current entry and the next one */
 		if (ring_iter)

kernel/trace/trace_output.c

@@ -1325,4 +1325,4 @@ __init static int init_events(void)
 
 	return 0;
 }
-device_initcall(init_events);
+early_initcall(init_events);

tools/lib/traceevent/Makefile

@@ -250,8 +250,12 @@ endef
 all_objs := $(sort $(ALL_OBJS))
 all_deps := $(all_objs:%.o=.%.d)
 
+# let .d file also depends on the source and header files
 define check_deps
-		$(CC) -M $(CFLAGS) $< > $@;
+		@set -e; $(RM) $@; \
+		$(CC) -M $(CFLAGS) $< > $@.$$$$; \
+		sed 's,\($*\)\.o[ :]*,\1.o $@ : ,g' < $@.$$$$ > $@; \
+		$(RM) $@.$$$$
 endef
 
 $(gui_deps): ks_version.h
@@ -270,11 +274,13 @@ endif
 
 tags:	force
 	$(RM) tags
-	find . -name '*.[ch]' | xargs ctags --extra=+f --c-kinds=+px
+	find . -name '*.[ch]' | xargs ctags --extra=+f --c-kinds=+px \
+	--regex-c++='/_PE\(([^,)]*).*/PEVENT_ERRNO__\1/'
 
 TAGS:	force
 	$(RM) TAGS
-	find . -name '*.[ch]' | xargs etags
+	find . -name '*.[ch]' | xargs etags \
+	--regex='/_PE(\([^,)]*\).*/PEVENT_ERRNO__\1/'
 
 define do_install
 	$(print_install)				\
@@ -290,7 +296,7 @@ install_lib: all_cmd install_plugins install_python
 install: install_lib
 
 clean:
-	$(RM) *.o *~ $(TARGETS) *.a *.so $(VERSION_FILES).*.d
+	$(RM) *.o *~ $(TARGETS) *.a *.so $(VERSION_FILES) .*.d
 	$(RM) tags TAGS
 
 endif # skip-makefile

tools/lib/traceevent/event-parse.h

@@ -226,6 +226,11 @@ struct print_arg_symbol {
 	struct print_flag_sym	*symbols;
 };
 
+struct print_arg_hex {
+	struct print_arg	*field;
+	struct print_arg	*size;
+};
+
 struct print_arg_dynarray {
 	struct format_field	*field;
 	struct print_arg	*index;
@@ -253,6 +258,7 @@ enum print_arg_type {
 	PRINT_FIELD,
 	PRINT_FLAGS,
 	PRINT_SYMBOL,
+	PRINT_HEX,
 	PRINT_TYPE,
 	PRINT_STRING,
 	PRINT_BSTRING,
@@ -270,6 +276,7 @@ struct print_arg {
 		struct print_arg_typecast	typecast;
 		struct print_arg_flags		flags;
 		struct print_arg_symbol		symbol;
+		struct print_arg_hex		hex;
 		struct print_arg_func		func;
 		struct print_arg_string		string;
 		struct print_arg_op		op;

tools/perf/Documentation/perf-report.txt

@@ -57,7 +57,7 @@ OPTIONS
 
 -s::
 --sort=::
-	Sort by key(s): pid, comm, dso, symbol, parent.
+	Sort by key(s): pid, comm, dso, symbol, parent, srcline.
 
 -p::
 --parent=<regex>::

tools/perf/Documentation/perf-top.txt

@@ -112,7 +112,7 @@ Default is to monitor all CPUS.
 
 -s::
 --sort::
-	Sort by key(s): pid, comm, dso, symbol, parent
+	Sort by key(s): pid, comm, dso, symbol, parent, srcline.
 
 -n::
 --show-nr-samples::

tools/perf/Makefile

@@ -155,7 +155,7 @@ endif
 
 ### --- END CONFIGURATION SECTION ---
 
-BASIC_CFLAGS = -Iutil/include -Iarch/$(ARCH)/include -I$(OUTPUT)/util -I$(TRACE_EVENT_DIR) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -D_GNU_SOURCE
+BASIC_CFLAGS = -Iutil/include -Iarch/$(ARCH)/include -I$(OUTPUT)util -I$(TRACE_EVENT_DIR) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -D_GNU_SOURCE
 BASIC_LDFLAGS =
 
 # Guard against environment variables
@@ -503,6 +503,7 @@ else
 		LIB_OBJS += $(OUTPUT)ui/progress.o
 		LIB_OBJS += $(OUTPUT)ui/util.o
 		LIB_OBJS += $(OUTPUT)ui/tui/setup.o
+		LIB_OBJS += $(OUTPUT)ui/tui/util.o
 		LIB_H += ui/browser.h
 		LIB_H += ui/browsers/map.h
 		LIB_H += ui/helpline.h
@@ -522,13 +523,18 @@ else
 		msg := $(warning GTK2 not found, disables GTK2 support. Please install gtk2-devel or libgtk2.0-dev);
 		BASIC_CFLAGS += -DNO_GTK2_SUPPORT
 	else
+		ifeq ($(call try-cc,$(SOURCE_GTK2_INFOBAR),$(FLAGS_GTK2)),y)
+			BASIC_CFLAGS += -DHAVE_GTK_INFO_BAR
+		endif
 		BASIC_CFLAGS += $(shell pkg-config --cflags gtk+-2.0)
 		EXTLIBS += $(shell pkg-config --libs gtk+-2.0)
 		LIB_OBJS += $(OUTPUT)ui/gtk/browser.o
 		LIB_OBJS += $(OUTPUT)ui/gtk/setup.o
+		LIB_OBJS += $(OUTPUT)ui/gtk/util.o
 		# Make sure that it'd be included only once.
 		ifneq ($(findstring -DNO_NEWT_SUPPORT,$(BASIC_CFLAGS)),)
 			LIB_OBJS += $(OUTPUT)ui/setup.o
+			LIB_OBJS += $(OUTPUT)ui/util.o
 		endif
 	endif
 endif