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

* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (121 commits)
  perf symbols: Increase symbol KSYM_NAME_LEN size
  perf hists browser: Refuse 'a' hotkey on non symbolic views
  perf ui browser: Use libslang to read keys
  perf tools: Fix tracing info recording
  perf hists browser: Elide DSO column when it is set to just one DSO, ditto for threads
  perf hists: Don't consider filtered entries when calculating column widths
  perf hists: Don't decay total_period for filtered entries
  perf hists browser: Honour symbol_conf.show_{nr_samples,total_period}
  perf hists browser: Do not exit on tab key with single event
  perf annotate browser: Don't change selection line when returning from callq
  perf tools: handle endianness of feature bitmap
  perf tools: Add prelink suggestion to dso update message
  perf script: Fix unknown feature comment
  perf hists browser: Apply the dso and thread filters when merging new batches
  perf hists: Move the dso and thread filters from hist_browser
  perf ui browser: Honour the xterm colors
  perf top tui: Give color hints just on the percentage, like on --stdio
  perf ui browser: Make the colors configurable and change the defaults
  perf tui: Remove unneeded call to newtCls on startup
  perf hists: Don't format the percentage on hist_entry__snprintf
  ...

Fix up conflicts in arch/x86/kernel/kprobes.c manually.

Ingo's tree did the insane "add volatile to const array", which just
doesn't make sense ("volatile const"?).  But we could remove the const
*and* make the array volatile to make doubly sure that gcc doesn't
optimize it away..

Also fix up kernel/trace/ring_buffer.c non-data-conflicts manually: the
reader_lock has been turned into a raw lock by the core locking merge,
and there was a new user of it introduced in this perf core merge.  Make
sure that new use also uses the raw accessor functions.

12 files changed, 1048 insertions, 553 deletions

diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile
index 6042981d030..fe6eb197f84 100644
--- a/arch/x86/kernel/cpu/Makefile
+++ b/arch/x86/kernel/cpu/Makefile
@@ -28,10 +28,15 @@ obj-$(CONFIG_CPU_SUP_UMC_32)		+= umc.o
 
 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
+endif
+
 obj-$(CONFIG_X86_MCE)			+= mcheck/
 obj-$(CONFIG_MTRR)			+= mtrr/
 
-obj-$(CONFIG_X86_LOCAL_APIC)		+= perfctr-watchdog.o
+obj-$(CONFIG_X86_LOCAL_APIC)		+= perfctr-watchdog.o perf_event_amd_ibs.o
 
 quiet_cmd_mkcapflags = MKCAP   $@
       cmd_mkcapflags = $(PERL) $(srctree)/$(src)/mkcapflags.pl $< $@
diff --git a/arch/x86/kernel/cpu/mcheck/mce-inject.c b/arch/x86/kernel/cpu/mcheck/mce-inject.c
index 0ed633c5048..6199232161c 100644
--- a/arch/x86/kernel/cpu/mcheck/mce-inject.c
+++ b/arch/x86/kernel/cpu/mcheck/mce-inject.c
@@ -78,27 +78,20 @@ static void raise_exception(struct mce *m, struct pt_regs *pregs)
 
 static cpumask_var_t mce_inject_cpumask;
 
-static int mce_raise_notify(struct notifier_block *self,
-			    unsigned long val, void *data)
+static int mce_raise_notify(unsigned int cmd, struct pt_regs *regs)
 {
-	struct die_args *args = (struct die_args *)data;
 	int cpu = smp_processor_id();
 	struct mce *m = &__get_cpu_var(injectm);
-	if (val != DIE_NMI || !cpumask_test_cpu(cpu, mce_inject_cpumask))
-		return NOTIFY_DONE;
+	if (!cpumask_test_cpu(cpu, mce_inject_cpumask))
+		return NMI_DONE;
 	cpumask_clear_cpu(cpu, mce_inject_cpumask);
 	if (m->inject_flags & MCJ_EXCEPTION)
-		raise_exception(m, args->regs);
+		raise_exception(m, regs);
 	else if (m->status)
 		raise_poll(m);
-	return NOTIFY_STOP;
+	return NMI_HANDLED;
 }
 
-static struct notifier_block mce_raise_nb = {
-	.notifier_call = mce_raise_notify,
-	.priority = NMI_LOCAL_NORMAL_PRIOR,
-};
-
 /* Inject mce on current CPU */
 static int raise_local(void)
 {
@@ -216,7 +209,8 @@ static int inject_init(void)
 		return -ENOMEM;
 	printk(KERN_INFO "Machine check injector initialized\n");
 	mce_chrdev_ops.write = mce_write;
-	register_die_notifier(&mce_raise_nb);
+	register_nmi_handler(NMI_LOCAL, mce_raise_notify, 0,
+				"mce_notify");
 	return 0;
 }
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index 08363b04212..fce51ad1f36 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -908,9 +908,6 @@ void do_machine_check(struct pt_regs *regs, long error_code)
 
 	percpu_inc(mce_exception_count);
 
-	if (notify_die(DIE_NMI, "machine check", regs, error_code,
-			   18, SIGKILL) == NOTIFY_STOP)
-		goto out;
 	if (!banks)
 		goto out;
 
@@ -1140,6 +1137,15 @@ static void mce_start_timer(unsigned long data)
 	add_timer_on(t, smp_processor_id());
 }
 
+/* Must not be called in IRQ context where del_timer_sync() can deadlock */
+static void mce_timer_delete_all(void)
+{
+	int cpu;
+
+	for_each_online_cpu(cpu)
+		del_timer_sync(&per_cpu(mce_timer, cpu));
+}
+
 static void mce_do_trigger(struct work_struct *work)
 {
 	call_usermodehelper(mce_helper, mce_helper_argv, NULL, UMH_NO_WAIT);
@@ -1750,7 +1756,6 @@ static struct syscore_ops mce_syscore_ops = {
 
 static void mce_cpu_restart(void *data)
 {
-	del_timer_sync(&__get_cpu_var(mce_timer));
 	if (!mce_available(__this_cpu_ptr(&cpu_info)))
 		return;
 	__mcheck_cpu_init_generic();
@@ -1760,16 +1765,15 @@ static void mce_cpu_restart(void *data)
 /* Reinit MCEs after user configuration changes */
 static void mce_restart(void)
 {
+	mce_timer_delete_all();
 	on_each_cpu(mce_cpu_restart, NULL, 1);
 }
 
 /* Toggle features for corrected errors */
-static void mce_disable_ce(void *all)
+static void mce_disable_cmci(void *data)
 {
 	if (!mce_available(__this_cpu_ptr(&cpu_info)))
 		return;
-	if (all)
-		del_timer_sync(&__get_cpu_var(mce_timer));
 	cmci_clear();
 }
 
@@ -1852,7 +1856,8 @@ static ssize_t set_ignore_ce(struct sys_device *s,
 	if (mce_ignore_ce ^ !!new) {
 		if (new) {
 			/* disable ce features */
-			on_each_cpu(mce_disable_ce, (void *)1, 1);
+			mce_timer_delete_all();
+			on_each_cpu(mce_disable_cmci, NULL, 1);
 			mce_ignore_ce = 1;
 		} else {
 			/* enable ce features */
@@ -1875,7 +1880,7 @@ static ssize_t set_cmci_disabled(struct sys_device *s,
 	if (mce_cmci_disabled ^ !!new) {
 		if (new) {
 			/* disable cmci */
-			on_each_cpu(mce_disable_ce, NULL, 1);
+			on_each_cpu(mce_disable_cmci, NULL, 1);
 			mce_cmci_disabled = 1;
 		} else {
 			/* enable cmci */
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index cfa62ec090e..640891014b2 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -32,6 +32,8 @@
 #include <asm/smp.h>
 #include <asm/alternative.h>
 
+#include "perf_event.h"
+
 #if 0
 #undef wrmsrl
 #define wrmsrl(msr, val) 					\
@@ -43,283 +45,17 @@ do {								\
 } while (0)
 #endif
 
-/*
- *          |   NHM/WSM    |      SNB     |
- * register -------------------------------
- *          |  HT  | no HT |  HT  | no HT |
- *-----------------------------------------
- * offcore  | core | core  | cpu  | core  |
- * lbr_sel  | core | core  | cpu  | core  |
- * ld_lat   | cpu  | core  | cpu  | core  |
- *-----------------------------------------
- *
- * Given that there is a small number of shared regs,
- * we can pre-allocate their slot in the per-cpu
- * per-core reg tables.
- */
-enum extra_reg_type {
-	EXTRA_REG_NONE  = -1,	/* not used */
-
-	EXTRA_REG_RSP_0 = 0,	/* offcore_response_0 */
-	EXTRA_REG_RSP_1 = 1,	/* offcore_response_1 */
-
-	EXTRA_REG_MAX		/* number of entries needed */
-};
-
-struct event_constraint {
-	union {
-		unsigned long	idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
-		u64		idxmsk64;
-	};
-	u64	code;
-	u64	cmask;
-	int	weight;
-};
-
-struct amd_nb {
-	int nb_id;  /* NorthBridge id */
-	int refcnt; /* reference count */
-	struct perf_event *owners[X86_PMC_IDX_MAX];
-	struct event_constraint event_constraints[X86_PMC_IDX_MAX];
-};
-
-struct intel_percore;
-
-#define MAX_LBR_ENTRIES		16
-
-struct cpu_hw_events {
-	/*
-	 * Generic x86 PMC bits
-	 */
-	struct perf_event	*events[X86_PMC_IDX_MAX]; /* in counter order */
-	unsigned long		active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
-	unsigned long		running[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
-	int			enabled;
-
-	int			n_events;
-	int			n_added;
-	int			n_txn;
-	int			assign[X86_PMC_IDX_MAX]; /* event to counter assignment */
-	u64			tags[X86_PMC_IDX_MAX];
-	struct perf_event	*event_list[X86_PMC_IDX_MAX]; /* in enabled order */
-
-	unsigned int		group_flag;
-
-	/*
-	 * Intel DebugStore bits
-	 */
-	struct debug_store	*ds;
-	u64			pebs_enabled;
-
-	/*
-	 * Intel LBR bits
-	 */
-	int				lbr_users;
-	void				*lbr_context;
-	struct perf_branch_stack	lbr_stack;
-	struct perf_branch_entry	lbr_entries[MAX_LBR_ENTRIES];
-
-	/*
-	 * manage shared (per-core, per-cpu) registers
-	 * used on Intel NHM/WSM/SNB
-	 */
-	struct intel_shared_regs	*shared_regs;
-
-	/*
-	 * AMD specific bits
-	 */
-	struct amd_nb		*amd_nb;
-};
-
-#define __EVENT_CONSTRAINT(c, n, m, w) {\
-	{ .idxmsk64 = (n) },		\
-	.code = (c),			\
-	.cmask = (m),			\
-	.weight = (w),			\
-}
-
-#define EVENT_CONSTRAINT(c, n, m)	\
-	__EVENT_CONSTRAINT(c, n, m, HWEIGHT(n))
-
-/*
- * Constraint on the Event code.
- */
-#define INTEL_EVENT_CONSTRAINT(c, n)	\
-	EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT)
-
-/*
- * Constraint on the Event code + UMask + fixed-mask
- *
- * filter mask to validate fixed counter events.
- * the following filters disqualify for fixed counters:
- *  - inv
- *  - edge
- *  - cnt-mask
- *  The other filters are supported by fixed counters.
- *  The any-thread option is supported starting with v3.
- */
-#define FIXED_EVENT_CONSTRAINT(c, n)	\
-	EVENT_CONSTRAINT(c, (1ULL << (32+n)), X86_RAW_EVENT_MASK)
-
-/*
- * Constraint on the Event code + UMask
- */
-#define INTEL_UEVENT_CONSTRAINT(c, n)	\
-	EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK)
-
-#define EVENT_CONSTRAINT_END		\
-	EVENT_CONSTRAINT(0, 0, 0)
-
-#define for_each_event_constraint(e, c)	\
-	for ((e) = (c); (e)->weight; (e)++)
-
-/*
- * Per register state.
- */
-struct er_account {
-	raw_spinlock_t		lock;	/* per-core: protect structure */
-	u64			config;	/* extra MSR config */
-	u64			reg;	/* extra MSR number */
-	atomic_t		ref;	/* reference count */
-};
-
-/*
- * Extra registers for specific events.
- *
- * Some events need large masks and require external MSRs.
- * Those extra MSRs end up being shared for all events on
- * a PMU and sometimes between PMU of sibling HT threads.
- * In either case, the kernel needs to handle conflicting
- * accesses to those extra, shared, regs. The data structure
- * to manage those registers is stored in cpu_hw_event.
- */
-struct extra_reg {
-	unsigned int		event;
-	unsigned int		msr;
-	u64			config_mask;
-	u64			valid_mask;
-	int			idx;  /* per_xxx->regs[] reg index */
-};
-
-#define EVENT_EXTRA_REG(e, ms, m, vm, i) {	\
-	.event = (e),		\
-	.msr = (ms),		\
-	.config_mask = (m),	\
-	.valid_mask = (vm),	\
-	.idx = EXTRA_REG_##i	\
-	}
-
-#define INTEL_EVENT_EXTRA_REG(event, msr, vm, idx)	\
-	EVENT_EXTRA_REG(event, msr, ARCH_PERFMON_EVENTSEL_EVENT, vm, idx)
-
-#define EVENT_EXTRA_END EVENT_EXTRA_REG(0, 0, 0, 0, RSP_0)
-
-union perf_capabilities {
-	struct {
-		u64	lbr_format    : 6;
-		u64	pebs_trap     : 1;
-		u64	pebs_arch_reg : 1;
-		u64	pebs_format   : 4;
-		u64	smm_freeze    : 1;
-	};
-	u64	capabilities;
-};
-
-/*
- * struct x86_pmu - generic x86 pmu
- */
-struct x86_pmu {
-	/*
-	 * Generic x86 PMC bits
-	 */
-	const char	*name;
-	int		version;
-	int		(*handle_irq)(struct pt_regs *);
-	void		(*disable_all)(void);
-	void		(*enable_all)(int added);
-	void		(*enable)(struct perf_event *);
-	void		(*disable)(struct perf_event *);
-	int		(*hw_config)(struct perf_event *event);
-	int		(*schedule_events)(struct cpu_hw_events *cpuc, int n, int *assign);
-	unsigned	eventsel;
-	unsigned	perfctr;
-	u64		(*event_map)(int);
-	int		max_events;
-	int		num_counters;
-	int		num_counters_fixed;
-	int		cntval_bits;
-	u64		cntval_mask;
-	int		apic;
-	u64		max_period;
-	struct event_constraint *
-			(*get_event_constraints)(struct cpu_hw_events *cpuc,
-						 struct perf_event *event);
-
-	void		(*put_event_constraints)(struct cpu_hw_events *cpuc,
-						 struct perf_event *event);
-	struct event_constraint *event_constraints;
-	void		(*quirks)(void);
-	int		perfctr_second_write;
-
-	int		(*cpu_prepare)(int cpu);
-	void		(*cpu_starting)(int cpu);
-	void		(*cpu_dying)(int cpu);
-	void		(*cpu_dead)(int cpu);
-
-	/*
-	 * Intel Arch Perfmon v2+
-	 */
-	u64			intel_ctrl;
-	union perf_capabilities intel_cap;
+struct x86_pmu x86_pmu __read_mostly;
 
-	/*
-	 * Intel DebugStore bits
-	 */
-	int		bts, pebs;
-	int		bts_active, pebs_active;
-	int		pebs_record_size;
-	void		(*drain_pebs)(struct pt_regs *regs);
-	struct event_constraint *pebs_constraints;
-
-	/*
-	 * Intel LBR
-	 */
-	unsigned long	lbr_tos, lbr_from, lbr_to; /* MSR base regs       */
-	int		lbr_nr;			   /* hardware stack size */
-
-	/*
-	 * Extra registers for events
-	 */
-	struct extra_reg *extra_regs;
-	unsigned int er_flags;
-};
-
-#define ERF_NO_HT_SHARING	1
-#define ERF_HAS_RSP_1		2
-
-static struct x86_pmu x86_pmu __read_mostly;
-
-static DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
+DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
 	.enabled = 1,
 };
 
-static int x86_perf_event_set_period(struct perf_event *event);
-
-/*
- * Generalized hw caching related hw_event table, filled
- * in on a per model basis. A value of 0 means
- * 'not supported', -1 means 'hw_event makes no sense on
- * this CPU', any other value means the raw hw_event
- * ID.
- */
-
-#define C(x) PERF_COUNT_HW_CACHE_##x
-
-static u64 __read_mostly hw_cache_event_ids
+u64 __read_mostly hw_cache_event_ids
 				[PERF_COUNT_HW_CACHE_MAX]
 				[PERF_COUNT_HW_CACHE_OP_MAX]
 				[PERF_COUNT_HW_CACHE_RESULT_MAX];
-static u64 __read_mostly hw_cache_extra_regs
+u64 __read_mostly hw_cache_extra_regs
 				[PERF_COUNT_HW_CACHE_MAX]
 				[PERF_COUNT_HW_CACHE_OP_MAX]
 				[PERF_COUNT_HW_CACHE_RESULT_MAX];
@@ -329,8 +65,7 @@ static u64 __read_mostly hw_cache_extra_regs
  * Can only be executed on the CPU where the event is active.
  * Returns the delta events processed.
  */
-static u64
-x86_perf_event_update(struct perf_event *event)
+u64 x86_perf_event_update(struct perf_event *event)
 {
 	struct hw_perf_event *hwc = &event->hw;
 	int shift = 64 - x86_pmu.cntval_bits;
@@ -373,30 +108,6 @@ again:
 	return new_raw_count;
 }
 
-static inline int x86_pmu_addr_offset(int index)
-{
-	int offset;
-
-	/* offset = X86_FEATURE_PERFCTR_CORE ? index << 1 : index */
-	alternative_io(ASM_NOP2,
-		       "shll $1, %%eax",
-		       X86_FEATURE_PERFCTR_CORE,
-		       "=a" (offset),
-		       "a"  (index));
-
-	return offset;
-}
-
-static inline unsigned int x86_pmu_config_addr(int index)
-{
-	return x86_pmu.eventsel + x86_pmu_addr_offset(index);
-}
-
-static inline unsigned int x86_pmu_event_addr(int index)
-{
-	return x86_pmu.perfctr + x86_pmu_addr_offset(index);
-}
-
 /*
  * Find and validate any extra registers to set up.
  */
@@ -532,9 +243,6 @@ msr_fail:
 	return false;
 }
 
-static void reserve_ds_buffers(void);
-static void release_ds_buffers(void);
-
 static void hw_perf_event_destroy(struct perf_event *event)
 {
 	if (atomic_dec_and_mutex_lock(&active_events, &pmc_reserve_mutex)) {
@@ -583,7 +291,7 @@ set_ext_hw_attr(struct hw_perf_event *hwc, struct perf_event *event)
 	return x86_pmu_extra_regs(val, event);
 }
 
-static int x86_setup_perfctr(struct perf_event *event)
+int x86_setup_perfctr(struct perf_event *event)
 {
 	struct perf_event_attr *attr = &event->attr;
 	struct hw_perf_event *hwc = &event->hw;
@@ -647,7 +355,7 @@ static int x86_setup_perfctr(struct perf_event *event)
 	return 0;
 }
 
-static int x86_pmu_hw_config(struct perf_event *event)
+int x86_pmu_hw_config(struct perf_event *event)
 {
 	if (event->attr.precise_ip) {
 		int precise = 0;
@@ -723,7 +431,7 @@ static int __x86_pmu_event_init(struct perf_event *event)
 	return x86_pmu.hw_config(event);
 }
 
-static void x86_pmu_disable_all(void)
+void x86_pmu_disable_all(void)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int idx;
@@ -758,15 +466,7 @@ static void x86_pmu_disable(struct pmu *pmu)
 	x86_pmu.disable_all();
 }
 
-static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc,
-					  u64 enable_mask)
-{
-	if (hwc->extra_reg.reg)
-		wrmsrl(hwc->extra_reg.reg, hwc->extra_reg.config);
-	wrmsrl(hwc->config_base, hwc->config | enable_mask);
-}
-
-static void x86_pmu_enable_all(int added)
+void x86_pmu_enable_all(int added)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int idx;
@@ -788,7 +488,7 @@ static inline int is_x86_event(struct perf_event *event)
 	return event->pmu == &pmu;
 }
 
-static int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
+int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign)
 {
 	struct event_constraint *c, *constraints[X86_PMC_IDX_MAX];
 	unsigned long used_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
@@ -959,7 +659,6 @@ static inline int match_prev_assignment(struct hw_perf_event *hwc,
 }
 
 static void x86_pmu_start(struct perf_event *event, int flags);
-static void x86_pmu_stop(struct perf_event *event, int flags);
 
 static void x86_pmu_enable(struct pmu *pmu)
 {
@@ -1031,21 +730,13 @@ static void x86_pmu_enable(struct pmu *pmu)
 	x86_pmu.enable_all(added);
 }
 
-static inline void x86_pmu_disable_event(struct perf_event *event)
-{
-	struct hw_perf_event *hwc = &event->hw;
-
-	wrmsrl(hwc->config_base, hwc->config);
-}
-
 static DEFINE_PER_CPU(u64 [X86_PMC_IDX_MAX], pmc_prev_left);
 
 /*
  * Set the next IRQ period, based on the hwc->period_left value.
  * To be called with the event disabled in hw:
  */
-static int
-x86_perf_event_set_period(struct perf_event *event)
+int x86_perf_event_set_period(struct perf_event *event)
 {
 	struct hw_perf_event *hwc = &event->hw;
 	s64 left = local64_read(&hwc->period_left);
@@ -1105,7 +796,7 @@ x86_perf_event_set_period(struct perf_event *event)
 	return ret;
 }
 
-static void x86_pmu_enable_event(struct perf_event *event)
+void x86_pmu_enable_event(struct perf_event *event)
 {
 	if (__this_cpu_read(cpu_hw_events.enabled))
 		__x86_pmu_enable_event(&event->hw,
@@ -1244,7 +935,7 @@ void perf_event_print_debug(void)
 	local_irq_restore(flags);
 }
 
-static void x86_pmu_stop(struct perf_event *event, int flags)
+void x86_pmu_stop(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
@@ -1297,7 +988,7 @@ static void x86_pmu_del(struct perf_event *event, int flags)
 	perf_event_update_userpage(event);
 }
 
-static int x86_pmu_handle_irq(struct pt_regs *regs)
+int x86_pmu_handle_irq(struct pt_regs *regs)
 {
 	struct perf_sample_data data;
 	struct cpu_hw_events *cpuc;
@@ -1367,109 +1058,28 @@ void perf_events_lapic_init(void)
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 }
 
-struct pmu_nmi_state {
-	unsigned int	marked;
-	int		handled;
-};
-
-static DEFINE_PER_CPU(struct pmu_nmi_state, pmu_nmi);
-
 static int __kprobes
-perf_event_nmi_handler(struct notifier_block *self,
-			 unsigned long cmd, void *__args)
+perf_event_nmi_handler(unsigned int cmd, struct pt_regs *regs)
 {
-	struct die_args *args = __args;
-	unsigned int this_nmi;
-	int handled;
-
 	if (!atomic_read(&active_events))
-		return NOTIFY_DONE;
-
-	switch (cmd) {
-	case DIE_NMI:
-		break;
-	case DIE_NMIUNKNOWN:
-		this_nmi = percpu_read(irq_stat.__nmi_count);
-		if (this_nmi != __this_cpu_read(pmu_nmi.marked))
-			/* let the kernel handle the unknown nmi */
-			return NOTIFY_DONE;
-		/*
-		 * This one is a PMU back-to-back nmi. Two events
-		 * trigger 'simultaneously' raising two back-to-back
-		 * NMIs. If the first NMI handles both, the latter
-		 * will be empty and daze the CPU. So, we drop it to
-		 * avoid false-positive 'unknown nmi' messages.
-		 */
-		return NOTIFY_STOP;
-	default:
-		return NOTIFY_DONE;
-	}
-
-	handled = x86_pmu.handle_irq(args->regs);
-	if (!handled)
-		return NOTIFY_DONE;
-
-	this_nmi = percpu_read(irq_stat.__nmi_count);
-	if ((handled > 1) ||
-		/* the next nmi could be a back-to-back nmi */
-	    ((__this_cpu_read(pmu_nmi.marked) == this_nmi) &&
-	     (__this_cpu_read(pmu_nmi.handled) > 1))) {
-		/*
-		 * We could have two subsequent back-to-back nmis: The
-		 * first handles more than one counter, the 2nd
-		 * handles only one counter and the 3rd handles no
-		 * counter.
-		 *
-		 * This is the 2nd nmi because the previous was
-		 * handling more than one counter. We will mark the
-		 * next (3rd) and then drop it if unhandled.
-		 */
-		__this_cpu_write(pmu_nmi.marked, this_nmi + 1);
-		__this_cpu_write(pmu_nmi.handled, handled);
-	}
+		return NMI_DONE;
 
-	return NOTIFY_STOP;
+	return x86_pmu.handle_irq(regs);
 }
 
-static __read_mostly struct notifier_block perf_event_nmi_notifier = {
-	.notifier_call		= perf_event_nmi_handler,
-	.next			= NULL,
-	.priority		= NMI_LOCAL_LOW_PRIOR,
-};
-
-static struct event_constraint unconstrained;
-static struct event_constraint emptyconstraint;
-
-static struct event_constraint *
-x86_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
-{
-	struct event_constraint *c;
-
-	if (x86_pmu.event_constraints) {
-		for_each_event_constraint(c, x86_pmu.event_constraints) {
-			if ((event->hw.config & c->cmask) == c->code)
-				return c;
-		}
-	}
-
-	return &unconstrained;
-}
-
-#include "perf_event_amd.c"
-#include "perf_event_p6.c"
-#include "perf_event_p4.c"
-#include "perf_event_intel_lbr.c"
-#include "perf_event_intel_ds.c"
-#include "perf_event_intel.c"
+struct event_constraint emptyconstraint;
+struct event_constraint unconstrained;
 
 static int __cpuinit
 x86_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu)
 {
 	unsigned int cpu = (long)hcpu;
+	struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu);
 	int ret = NOTIFY_OK;
 
 	switch (action & ~CPU_TASKS_FROZEN) {
 	case CPU_UP_PREPARE:
+		cpuc->kfree_on_online = NULL;
 		if (x86_pmu.cpu_prepare)
 			ret = x86_pmu.cpu_prepare(cpu);
 		break;
@@ -1479,6 +1089,10 @@ x86_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu)
 			x86_pmu.cpu_starting(cpu);
 		break;
 
+	case CPU_ONLINE:
+		kfree(cpuc->kfree_on_online);
+		break;
+
 	case CPU_DYING:
 		if (x86_pmu.cpu_dying)
 			x86_pmu.cpu_dying(cpu);
@@ -1557,7 +1171,7 @@ static int __init init_hw_perf_events(void)
 		((1LL << x86_pmu.num_counters_fixed)-1) << X86_PMC_IDX_FIXED;
 
 	perf_events_lapic_init();
-	register_die_notifier(&perf_event_nmi_notifier);
+	register_nmi_handler(NMI_LOCAL, perf_event_nmi_handler, 0, "PMI");
 
 	unconstrained = (struct event_constraint)
 		__EVENT_CONSTRAINT(0, (1ULL << x86_pmu.num_counters) - 1,
diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h
new file mode 100644
index 00000000000..b9698d40ac4
--- /dev/null
+++ b/arch/x86/kernel/cpu/perf_event.h
@@ -0,0 +1,505 @@
+/*
+ * Performance events x86 architecture header
+ *
+ *  Copyright (C) 2008 Thomas Gleixner <tglx@linutronix.de>
+ *  Copyright (C) 2008-2009 Red Hat, Inc., Ingo Molnar
+ *  Copyright (C) 2009 Jaswinder Singh Rajput
+ *  Copyright (C) 2009 Advanced Micro Devices, Inc., Robert Richter
+ *  Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
+ *  Copyright (C) 2009 Intel Corporation, <markus.t.metzger@intel.com>
+ *  Copyright (C) 2009 Google, Inc., Stephane Eranian
+ *
+ *  For licencing details see kernel-base/COPYING
+ */
+
+#include <linux/perf_event.h>
+
+/*
+ *          |   NHM/WSM    |      SNB     |
+ * register -------------------------------
+ *          |  HT  | no HT |  HT  | no HT |
+ *-----------------------------------------
+ * offcore  | core | core  | cpu  | core  |
+ * lbr_sel  | core | core  | cpu  | core  |
+ * ld_lat   | cpu  | core  | cpu  | core  |
+ *-----------------------------------------
+ *
+ * Given that there is a small number of shared regs,
+ * we can pre-allocate their slot in the per-cpu
+ * per-core reg tables.
+ */
+enum extra_reg_type {
+	EXTRA_REG_NONE  = -1,	/* not used */
+
+	EXTRA_REG_RSP_0 = 0,	/* offcore_response_0 */
+	EXTRA_REG_RSP_1 = 1,	/* offcore_response_1 */
+
+	EXTRA_REG_MAX		/* number of entries needed */
+};
+
+struct event_constraint {
+	union {
+		unsigned long	idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+		u64		idxmsk64;
+	};
+	u64	code;
+	u64	cmask;
+	int	weight;
+};
+
+struct amd_nb {
+	int nb_id;  /* NorthBridge id */
+	int refcnt; /* reference count */
+	struct perf_event *owners[X86_PMC_IDX_MAX];
+	struct event_constraint event_constraints[X86_PMC_IDX_MAX];
+};
+
+/* The maximal number of PEBS events: */
+#define MAX_PEBS_EVENTS		4
+
+/*
+ * A debug store configuration.
+ *
+ * We only support architectures that use 64bit fields.
+ */
+struct debug_store {
+	u64	bts_buffer_base;
+	u64	bts_index;
+	u64	bts_absolute_maximum;
+	u64	bts_interrupt_threshold;
+	u64	pebs_buffer_base;
+	u64	pebs_index;
+	u64	pebs_absolute_maximum;
+	u64	pebs_interrupt_threshold;
+	u64	pebs_event_reset[MAX_PEBS_EVENTS];
+};
+
+/*
+ * Per register state.
+ */
+struct er_account {
+	raw_spinlock_t		lock;	/* per-core: protect structure */
+	u64                 config;	/* extra MSR config */
+	u64                 reg;	/* extra MSR number */
+	atomic_t            ref;	/* reference count */
+};
+
+/*
+ * Per core/cpu state
+ *
+ * Used to coordinate shared registers between HT threads or
+ * among events on a single PMU.
+ */
+struct intel_shared_regs {
+	struct er_account       regs[EXTRA_REG_MAX];
+	int                     refcnt;		/* per-core: #HT threads */
+	unsigned                core_id;	/* per-core: core id */
+};
+
+#define MAX_LBR_ENTRIES		16
+
+struct cpu_hw_events {
+	/*
+	 * Generic x86 PMC bits
+	 */
+	struct perf_event	*events[X86_PMC_IDX_MAX]; /* in counter order */
+	unsigned long		active_mask[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+	unsigned long		running[BITS_TO_LONGS(X86_PMC_IDX_MAX)];
+	int			enabled;
+
+	int			n_events;
+	int			n_added;
+	int			n_txn;
+	int			assign[X86_PMC_IDX_MAX]; /* event to counter assignment */
+	u64			tags[X86_PMC_IDX_MAX];
+	struct perf_event	*event_list[X86_PMC_IDX_MAX]; /* in enabled order */
+
+	unsigned int		group_flag;
+
+	/*
+	 * Intel DebugStore bits
+	 */
+	struct debug_store	*ds;
+	u64			pebs_enabled;
+
+	/*
+	 * Intel LBR bits
+	 */
+	int				lbr_users;
+	void				*lbr_context;
+	struct perf_branch_stack	lbr_stack;
+	struct perf_branch_entry	lbr_entries[MAX_LBR_ENTRIES];
+
+	/*
+	 * Intel host/guest exclude bits
+	 */
+	u64				intel_ctrl_guest_mask;
+	u64				intel_ctrl_host_mask;
+	struct perf_guest_switch_msr	guest_switch_msrs[X86_PMC_IDX_MAX];
+
+	/*
+	 * manage shared (per-core, per-cpu) registers
+	 * used on Intel NHM/WSM/SNB
+	 */
+	struct intel_shared_regs	*shared_regs;
+
+	/*
+	 * AMD specific bits
+	 */
+	struct amd_nb		*amd_nb;
+
+	void				*kfree_on_online;
+};
+
+#define __EVENT_CONSTRAINT(c, n, m, w) {\
+	{ .idxmsk64 = (n) },		\
+	.code = (c),			\
+	.cmask = (m),			\
+	.weight = (w),			\
+}
+
+#define EVENT_CONSTRAINT(c, n, m)	\
+	__EVENT_CONSTRAINT(c, n, m, HWEIGHT(n))
+
+/*
+ * Constraint on the Event code.
+ */
+#define INTEL_EVENT_CONSTRAINT(c, n)	\
+	EVENT_CONSTRAINT(c, n, ARCH_PERFMON_EVENTSEL_EVENT)
+
+/*
+ * Constraint on the Event code + UMask + fixed-mask
+ *
+ * filter mask to validate fixed counter events.
+ * the following filters disqualify for fixed counters:
+ *  - inv
+ *  - edge
+ *  - cnt-mask
+ *  The other filters are supported by fixed counters.
+ *  The any-thread option is supported starting with v3.
+ */
+#define FIXED_EVENT_CONSTRAINT(c, n)	\
+	EVENT_CONSTRAINT(c, (1ULL << (32+n)), X86_RAW_EVENT_MASK)
+
+/*
+ * Constraint on the Event code + UMask
+ */
+#define INTEL_UEVENT_CONSTRAINT(c, n)	\
+	EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVENT_MASK)
+
+#define EVENT_CONSTRAINT_END		\
+	EVENT_CONSTRAINT(0, 0, 0)
+
+#define for_each_event_constraint(e, c)	\
+	for ((e) = (c); (e)->weight; (e)++)
+
+/*
+ * Extra registers for specific events.
+ *
+ * Some events need large masks and require external MSRs.
+ * Those extra MSRs end up being shared for all events on
+ * a PMU and sometimes between PMU of sibling HT threads.
+ * In either case, the kernel needs to handle conflicting
+ * accesses to those extra, shared, regs. The data structure
+ * to manage those registers is stored in cpu_hw_event.
+ */
+struct extra_reg {
+	unsigned int		event;
+	unsigned int		msr;
+	u64			config_mask;
+	u64			valid_mask;
+	int			idx;  /* per_xxx->regs[] reg index */
+};
+
+#define EVENT_EXTRA_REG(e, ms, m, vm, i) {	\
+	.event = (e),		\
+	.msr = (ms),		\
+	.config_mask = (m),	\
+	.valid_mask = (vm),	\
+	.idx = EXTRA_REG_##i	\
+	}
+
+#define INTEL_EVENT_EXTRA_REG(event, msr, vm, idx)	\
+	EVENT_EXTRA_REG(event, msr, ARCH_PERFMON_EVENTSEL_EVENT, vm, idx)
+
+#define EVENT_EXTRA_END EVENT_EXTRA_REG(0, 0, 0, 0, RSP_0)
+
+union perf_capabilities {
+	struct {
+		u64	lbr_format:6;
+		u64	pebs_trap:1;
+		u64	pebs_arch_reg:1;
+		u64	pebs_format:4;
+		u64	smm_freeze:1;
+	};
+	u64	capabilities;
+};
+
+/*
+ * struct x86_pmu - generic x86 pmu
+ */
+struct x86_pmu {
+	/*
+	 * Generic x86 PMC bits
+	 */
+	const char	*name;
+	int		version;
+	int		(*handle_irq)(struct pt_regs *);
+	void		(*disable_all)(void);
+	void		(*enable_all)(int added);
+	void		(*enable)(struct perf_event *);
+	void		(*disable)(struct perf_event *);
+	int		(*hw_config)(struct perf_event *event);
+	int		(*schedule_events)(struct cpu_hw_events *cpuc, int n, int *assign);
+	unsigned	eventsel;
+	unsigned	perfctr;
+	u64		(*event_map)(int);
+	int		max_events;
+	int		num_counters;
+	int		num_counters_fixed;
+	int		cntval_bits;
+	u64		cntval_mask;
+	int		apic;
+	u64		max_period;
+	struct event_constraint *
+			(*get_event_constraints)(struct cpu_hw_events *cpuc,
+						 struct perf_event *event);
+
+	void		(*put_event_constraints)(struct cpu_hw_events *cpuc,
+						 struct perf_event *event);
+	struct event_constraint *event_constraints;
+	void		(*quirks)(void);
+	int		perfctr_second_write;
+
+	int		(*cpu_prepare)(int cpu);
+	void		(*cpu_starting)(int cpu);
+	void		(*cpu_dying)(int cpu);
+	void		(*cpu_dead)(int cpu);
+
+	/*
+	 * Intel Arch Perfmon v2+
+	 */
+	u64			intel_ctrl;
+	union perf_capabilities intel_cap;
+
+	/*
+	 * Intel DebugStore bits
+	 */
+	int		bts, pebs;
+	int		bts_active, pebs_active;
+	int		pebs_record_size;
+	void		(*drain_pebs)(struct pt_regs *regs);
+	struct event_constraint *pebs_constraints;
+
+	/*
+	 * Intel LBR
+	 */
+	unsigned long	lbr_tos, lbr_from, lbr_to; /* MSR base regs       */
+	int		lbr_nr;			   /* hardware stack size */
+
+	/*
+	 * Extra registers for events
+	 */
+	struct extra_reg *extra_regs;
+	unsigned int er_flags;
+
+	/*
+	 * Intel host/guest support (KVM)
+	 */
+	struct perf_guest_switch_msr *(*guest_get_msrs)(int *nr);
+};
+
+#define ERF_NO_HT_SHARING	1
+#define ERF_HAS_RSP_1		2
+
+extern struct x86_pmu x86_pmu __read_mostly;
+
+DECLARE_PER_CPU(struct cpu_hw_events, cpu_hw_events);
+
+int x86_perf_event_set_period(struct perf_event *event);
+
+/*
+ * Generalized hw caching related hw_event table, filled
+ * in on a per model basis. A value of 0 means
+ * 'not supported', -1 means 'hw_event makes no sense on
+ * this CPU', any other value means the raw hw_event
+ * ID.
+ */
+
+#define C(x) PERF_COUNT_HW_CACHE_##x
+
+extern u64 __read_mostly hw_cache_event_ids
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX];
+extern u64 __read_mostly hw_cache_extra_regs
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX];
+
+u64 x86_perf_event_update(struct perf_event *event);
+
+static inline int x86_pmu_addr_offset(int index)
+{
+	int offset;
+
+	/* offset = X86_FEATURE_PERFCTR_CORE ? index << 1 : index */
+	alternative_io(ASM_NOP2,
+		       "shll $1, %%eax",
+		       X86_FEATURE_PERFCTR_CORE,
+		       "=a" (offset),
+		       "a"  (index));
+
+	return offset;
+}
+
+static inline unsigned int x86_pmu_config_addr(int index)
+{
+	return x86_pmu.eventsel + x86_pmu_addr_offset(index);
+}
+
+static inline unsigned int x86_pmu_event_addr(int index)
+{
+	return x86_pmu.perfctr + x86_pmu_addr_offset(index);
+}
+
+int x86_setup_perfctr(struct perf_event *event);
+
+int x86_pmu_hw_config(struct perf_event *event);
+
+void x86_pmu_disable_all(void);
+
+static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc,
+					  u64 enable_mask)
+{
+	if (hwc->extra_reg.reg)
+		wrmsrl(hwc->extra_reg.reg, hwc->extra_reg.config);
+	wrmsrl(hwc->config_base, hwc->config | enable_mask);
+}
+
+void x86_pmu_enable_all(int added);
+
+int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign);
+
+void x86_pmu_stop(struct perf_event *event, int flags);
+
+static inline void x86_pmu_disable_event(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	wrmsrl(hwc->config_base, hwc->config);
+}
+
+void x86_pmu_enable_event(struct perf_event *event);
+
+int x86_pmu_handle_irq(struct pt_regs *regs);
+
+extern struct event_constraint emptyconstraint;
+
+extern struct event_constraint unconstrained;
+
+#ifdef CONFIG_CPU_SUP_AMD
+
+int amd_pmu_init(void);
+
+#else /* CONFIG_CPU_SUP_AMD */
+
+static inline int amd_pmu_init(void)
+{
+	return 0;
+}
+
+#endif /* CONFIG_CPU_SUP_AMD */
+
+#ifdef CONFIG_CPU_SUP_INTEL
+
+int intel_pmu_save_and_restart(struct perf_event *event);
+
+struct event_constraint *
+x86_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event);
+
+struct intel_shared_regs *allocate_shared_regs(int cpu);
+
+int intel_pmu_init(void);
+
+void init_debug_store_on_cpu(int cpu);
+
+void fini_debug_store_on_cpu(int cpu);
+
+void release_ds_buffers(void);
+
+void reserve_ds_buffers(void);
+
+extern struct event_constraint bts_constraint;
+
+void intel_pmu_enable_bts(u64 config);
+
+void intel_pmu_disable_bts(void);
+
+int intel_pmu_drain_bts_buffer(void);
+
+extern struct event_constraint intel_core2_pebs_event_constraints[];
+
+extern struct event_constraint intel_atom_pebs_event_constraints[];
+
+extern struct event_constraint intel_nehalem_pebs_event_constraints[];
+
+extern struct event_constraint intel_westmere_pebs_event_constraints[];
+
+extern struct event_constraint intel_snb_pebs_event_constraints[];
+
+struct event_constraint *intel_pebs_constraints(struct perf_event *event);
+
+void intel_pmu_pebs_enable(struct perf_event *event);
+
+void intel_pmu_pebs_disable(struct perf_event *event);
+
+void intel_pmu_pebs_enable_all(void);
+
+void intel_pmu_pebs_disable_all(void);
+
+void intel_ds_init(void);
+
+void intel_pmu_lbr_reset(void);
+
+void intel_pmu_lbr_enable(struct perf_event *event);
+
+void intel_pmu_lbr_disable(struct perf_event *event);
+
+void intel_pmu_lbr_enable_all(void);
+
+void intel_pmu_lbr_disable_all(void);
+
+void intel_pmu_lbr_read(void);
+
+void intel_pmu_lbr_init_core(void);
+
+void intel_pmu_lbr_init_nhm(void);
+
+void intel_pmu_lbr_init_atom(void);
+
+int p4_pmu_init(void);
+
+int p6_pmu_init(void);
+
+#else /* CONFIG_CPU_SUP_INTEL */
+
+static inline void reserve_ds_buffers(void)
+{
+}
+
+static inline void release_ds_buffers(void)
+{
+}
+
+static inline int intel_pmu_init(void)
+{
+	return 0;
+}
+
+static inline struct intel_shared_regs *allocate_shared_regs(int cpu)
+{
+	return NULL;
+}
+
+#endif /* CONFIG_CPU_SUP_INTEL */
diff --git a/arch/x86/kernel/cpu/perf_event_amd.c b/arch/x86/kernel/cpu/perf_event_amd.c
index 941caa2e449..aeefd45697a 100644
--- a/arch/x86/kernel/cpu/perf_event_amd.c
+++ b/arch/x86/kernel/cpu/perf_event_amd.c
@@ -1,4 +1,10 @@
-#ifdef CONFIG_CPU_SUP_AMD
+#include <linux/perf_event.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <asm/apicdef.h>
+
+#include "perf_event.h"
 
 static __initconst const u64 amd_hw_cache_event_ids
 				[PERF_COUNT_HW_CACHE_MAX]
@@ -132,6 +138,19 @@ static int amd_pmu_hw_config(struct perf_event *event)
 	if (ret)
 		return ret;
 
+	if (event->attr.exclude_host && event->attr.exclude_guest)
+		/*
+		 * When HO == GO == 1 the hardware treats that as GO == HO == 0
+		 * and will count in both modes. We don't want to count in that
+		 * case so we emulate no-counting by setting US = OS = 0.
+		 */
+		event->hw.config &= ~(ARCH_PERFMON_EVENTSEL_USR |
+				      ARCH_PERFMON_EVENTSEL_OS);
+	else if (event->attr.exclude_host)
+		event->hw.config |= AMD_PERFMON_EVENTSEL_GUESTONLY;
+	else if (event->attr.exclude_guest)
+		event->hw.config |= AMD_PERFMON_EVENTSEL_HOSTONLY;
+
 	if (event->attr.type != PERF_TYPE_RAW)
 		return 0;
 
@@ -350,7 +369,7 @@ static void amd_pmu_cpu_starting(int cpu)
 			continue;
 
 		if (nb->nb_id == nb_id) {
-			kfree(cpuc->amd_nb);
+			cpuc->kfree_on_online = cpuc->amd_nb;
 			cpuc->amd_nb = nb;
 			break;
 		}
@@ -392,7 +411,7 @@ static __initconst const struct x86_pmu amd_pmu = {
 	.perfctr		= MSR_K7_PERFCTR0,
 	.event_map		= amd_pmu_event_map,
 	.max_events		= ARRAY_SIZE(amd_perfmon_event_map),
-	.num_counters		= 4,
+	.num_counters		= AMD64_NUM_COUNTERS,
 	.cntval_bits		= 48,
 	.cntval_mask		= (1ULL << 48) - 1,
 	.apic			= 1,
@@ -556,7 +575,7 @@ static __initconst const struct x86_pmu amd_pmu_f15h = {
 	.perfctr		= MSR_F15H_PERF_CTR,
 	.event_map		= amd_pmu_event_map,
 	.max_events		= ARRAY_SIZE(amd_perfmon_event_map),
-	.num_counters		= 6,
+	.num_counters		= AMD64_NUM_COUNTERS_F15H,
 	.cntval_bits		= 48,
 	.cntval_mask		= (1ULL << 48) - 1,
 	.apic			= 1,
@@ -573,7 +592,7 @@ static __initconst const struct x86_pmu amd_pmu_f15h = {
 #endif
 };
 
-static __init int amd_pmu_init(void)
+__init int amd_pmu_init(void)
 {
 	/* Performance-monitoring supported from K7 and later: */
 	if (boot_cpu_data.x86 < 6)
@@ -602,12 +621,3 @@ static __init int amd_pmu_init(void)
 
 	return 0;
 }
-
-#else /* CONFIG_CPU_SUP_AMD */
-
-static int amd_pmu_init(void)
-{
-	return 0;
-}
-
-#endif
diff --git a/arch/x86/kernel/cpu/perf_event_amd_ibs.c b/arch/x86/kernel/cpu/perf_event_amd_ibs.c
new file mode 100644
index 00000000000..ab6343d2182
--- /dev/null
+++ b/arch/x86/kernel/cpu/perf_event_amd_ibs.c
@@ -0,0 +1,294 @@
+/*
+ * Performance events - AMD IBS
+ *
+ *  Copyright (C) 2011 Advanced Micro Devices, Inc., Robert Richter
+ *
+ *  For licencing details see kernel-base/COPYING
+ */
+
+#include <linux/perf_event.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+
+#include <asm/apic.h>
+
+static u32 ibs_caps;
+
+#if defined(CONFIG_PERF_EVENTS) && defined(CONFIG_CPU_SUP_AMD)
+
+static struct pmu perf_ibs;
+
+static int perf_ibs_init(struct perf_event *event)
+{
+	if (perf_ibs.type != event->attr.type)
+		return -ENOENT;
+	return 0;
+}
+
+static int perf_ibs_add(struct perf_event *event, int flags)
+{
+	return 0;
+}
+
+static void perf_ibs_del(struct perf_event *event, int flags)
+{
+}
+
+static struct pmu perf_ibs = {
+	.event_init= perf_ibs_init,
+	.add= perf_ibs_add,
+	.del= perf_ibs_del,
+};
+
+static __init int perf_event_ibs_init(void)
+{
+	if (!ibs_caps)
+		return -ENODEV;	/* ibs not supported by the cpu */
+
+	perf_pmu_register(&perf_ibs, "ibs", -1);
+	printk(KERN_INFO "perf: AMD IBS detected (0x%08x)\n", ibs_caps);
+
+	return 0;
+}
+
+#else /* defined(CONFIG_PERF_EVENTS) && defined(CONFIG_CPU_SUP_AMD) */
+
+static __init int perf_event_ibs_init(void) { return 0; }
+
+#endif
+
+/* IBS - apic initialization, for perf and oprofile */
+
+static __init u32 __get_ibs_caps(void)
+{
+	u32 caps;
+	unsigned int max_level;
+
+	if (!boot_cpu_has(X86_FEATURE_IBS))
+		return 0;
+
+	/* check IBS cpuid feature flags */
+	max_level = cpuid_eax(0x80000000);
+	if (max_level < IBS_CPUID_FEATURES)
+		return IBS_CAPS_DEFAULT;
+
+	caps = cpuid_eax(IBS_CPUID_FEATURES);
+	if (!(caps & IBS_CAPS_AVAIL))
+		/* cpuid flags not valid */
+		return IBS_CAPS_DEFAULT;
+
+	return caps;
+}
+
+u32 get_ibs_caps(void)
+{
+	return ibs_caps;
+}
+
+EXPORT_SYMBOL(get_ibs_caps);
+
+static inline int get_eilvt(int offset)
+{
+	return !setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 1);
+}
+
+static inline int put_eilvt(int offset)
+{
+	return !setup_APIC_eilvt(offset, 0, 0, 1);
+}
+
+/*
+ * Check and reserve APIC extended interrupt LVT offset for IBS if available.
+ */
+static inline int ibs_eilvt_valid(void)
+{
+	int offset;
+	u64 val;
+	int valid = 0;
+
+	preempt_disable();
+
+	rdmsrl(MSR_AMD64_IBSCTL, val);
+	offset = val & IBSCTL_LVT_OFFSET_MASK;
+
+	if (!(val & IBSCTL_LVT_OFFSET_VALID)) {
+		pr_err(FW_BUG "cpu %d, invalid IBS interrupt offset %d (MSR%08X=0x%016llx)\n",
+		       smp_processor_id(), offset, MSR_AMD64_IBSCTL, val);
+		goto out;
+	}
+
+	if (!get_eilvt(offset)) {
+		pr_err(FW_BUG "cpu %d, IBS interrupt offset %d not available (MSR%08X=0x%016llx)\n",
+		       smp_processor_id(), offset, MSR_AMD64_IBSCTL, val);
+		goto out;
+	}
+
+	valid = 1;
+out:
+	preempt_enable();
+
+	return valid;
+}
+
+static int setup_ibs_ctl(int ibs_eilvt_off)
+{
+	struct pci_dev *cpu_cfg;
+	int nodes;
+	u32 value = 0;
+
+	nodes = 0;
+	cpu_cfg = NULL;
+	do {
+		cpu_cfg = pci_get_device(PCI_VENDOR_ID_AMD,
+					 PCI_DEVICE_ID_AMD_10H_NB_MISC,
+					 cpu_cfg);
+		if (!cpu_cfg)
+			break;
+		++nodes;
+		pci_write_config_dword(cpu_cfg, IBSCTL, ibs_eilvt_off
+				       | IBSCTL_LVT_OFFSET_VALID);
+		pci_read_config_dword(cpu_cfg, IBSCTL, &value);
+		if (value != (ibs_eilvt_off | IBSCTL_LVT_OFFSET_VALID)) {
+			pci_dev_put(cpu_cfg);
+			printk(KERN_DEBUG "Failed to setup IBS LVT offset, "
+			       "IBSCTL = 0x%08x\n", value);
+			return -EINVAL;
+		}
+	} while (1);
+
+	if (!nodes) {
+		printk(KERN_DEBUG "No CPU node configured for IBS\n");
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+/*
+ * This runs only on the current cpu. We try to find an LVT offset and
+ * setup the local APIC. For this we must disable preemption. On
+ * success we initialize all nodes with this offset. This updates then
+ * the offset in the IBS_CTL per-node msr. The per-core APIC setup of
+ * the IBS interrupt vector is handled by perf_ibs_cpu_notifier that
+ * is using the new offset.
+ */
+static int force_ibs_eilvt_setup(void)
+{
+	int offset;
+	int ret;
+
+	preempt_disable();
+	/* find the next free available EILVT entry, skip offset 0 */
+	for (offset = 1; offset < APIC_EILVT_NR_MAX; offset++) {
+		if (get_eilvt(offset))
+			break;
+	}
+	preempt_enable();
+
+	if (offset == APIC_EILVT_NR_MAX) {
+		printk(KERN_DEBUG "No EILVT entry available\n");
+		return -EBUSY;
+	}
+
+	ret = setup_ibs_ctl(offset);
+	if (ret)
+		goto out;
+
+	if (!ibs_eilvt_valid()) {
+		ret = -EFAULT;
+		goto out;
+	}
+
+	pr_err(FW_BUG "using offset %d for IBS interrupts\n", offset);
+	pr_err(FW_BUG "workaround enabled for IBS LVT offset\n");
+
+	return 0;
+out:
+	preempt_disable();
+	put_eilvt(offset);
+	preempt_enable();
+	return ret;
+}
+
+static inline int get_ibs_lvt_offset(void)
+{
+	u64 val;
+
+	rdmsrl(MSR_AMD64_IBSCTL, val);
+	if (!(val & IBSCTL_LVT_OFFSET_VALID))
+		return -EINVAL;
+
+	return val & IBSCTL_LVT_OFFSET_MASK;
+}
+
+static void setup_APIC_ibs(void *dummy)
+{
+	int offset;
+
+	offset = get_ibs_lvt_offset();
+	if (offset < 0)
+		goto failed;
+
+	if (!setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_NMI, 0))
+		return;
+failed:
+	pr_warn("perf: IBS APIC setup failed on cpu #%d\n",
+		smp_processor_id());
+}
+
+static void clear_APIC_ibs(void *dummy)
+{
+	int offset;
+
+	offset = get_ibs_lvt_offset();
+	if (offset >= 0)
+		setup_APIC_eilvt(offset, 0, APIC_EILVT_MSG_FIX, 1);
+}
+
+static int __cpuinit
+perf_ibs_cpu_notifier(struct notifier_block *self, unsigned long action, void *hcpu)
+{
+	switch (action & ~CPU_TASKS_FROZEN) {
+	case CPU_STARTING:
+		setup_APIC_ibs(NULL);
+		break;
+	case CPU_DYING:
+		clear_APIC_ibs(NULL);
+		break;
+	default:
+		break;
+	}
+
+	return NOTIFY_OK;
+}
+
+static __init int amd_ibs_init(void)
+{
+	u32 caps;
+	int ret;
+
+	caps = __get_ibs_caps();
+	if (!caps)
+		return -ENODEV;	/* ibs not supported by the cpu */
+
+	if (!ibs_eilvt_valid()) {
+		ret = force_ibs_eilvt_setup();
+		if (ret) {
+			pr_err("Failed to setup IBS, %d\n", ret);
+			return ret;
+		}
+	}
+
+	get_online_cpus();
+	ibs_caps = caps;
+	/* make ibs_caps visible to other cpus: */
+	smp_mb();
+	perf_cpu_notifier(perf_ibs_cpu_notifier);
+	smp_call_function(setup_APIC_ibs, NULL, 1);
+	put_online_cpus();
+
+	return perf_event_ibs_init();
+}
+
+/* Since we need the pci subsystem to init ibs we can't do this earlier: */
+device_initcall(amd_ibs_init);
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index f88af2c2a56..e09ca20e86e 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -1,16 +1,19 @@
-#ifdef CONFIG_CPU_SUP_INTEL
-
 /*
  * Per core/cpu state
  *
  * Used to coordinate shared registers between HT threads or
  * among events on a single PMU.
  */
-struct intel_shared_regs {
-	struct er_account       regs[EXTRA_REG_MAX];
-	int                     refcnt;		/* per-core: #HT threads */
-	unsigned                core_id;	/* per-core: core id */
-};
+
+#include <linux/stddef.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+
+#include <asm/hardirq.h>
+#include <asm/apic.h>
+
+#include "perf_event.h"
 
 /*
  * Intel PerfMon, used on Core and later.
@@ -746,7 +749,8 @@ static void intel_pmu_enable_all(int added)
 
 	intel_pmu_pebs_enable_all();
 	intel_pmu_lbr_enable_all();
-	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl);
+	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)) {
 		struct perf_event *event =
@@ -869,6 +873,7 @@ static void intel_pmu_disable_fixed(struct hw_perf_event *hwc)
 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)) {
 		intel_pmu_disable_bts();
@@ -876,6 +881,9 @@ static void intel_pmu_disable_event(struct perf_event *event)
 		return;
 	}
 
+	cpuc->intel_ctrl_guest_mask &= ~(1ull << hwc->idx);
+	cpuc->intel_ctrl_host_mask &= ~(1ull << hwc->idx);
+
 	if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) {
 		intel_pmu_disable_fixed(hwc);
 		return;
@@ -921,6 +929,7 @@ static void intel_pmu_enable_fixed(struct hw_perf_event *hwc)
 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 (!__this_cpu_read(cpu_hw_events.enabled))
@@ -930,6 +939,11 @@ static void intel_pmu_enable_event(struct perf_event *event)
 		return;
 	}
 
+	if (event->attr.exclude_host)
+		cpuc->intel_ctrl_guest_mask |= (1ull << hwc->idx);
+	if (event->attr.exclude_guest)
+		cpuc->intel_ctrl_host_mask |= (1ull << hwc->idx);
+
 	if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) {
 		intel_pmu_enable_fixed(hwc);
 		return;
@@ -945,7 +959,7 @@ static void intel_pmu_enable_event(struct perf_event *event)
  * Save and restart an expired event. Called by NMI contexts,
  * so it has to be careful about preempting normal event ops:
  */
-static int intel_pmu_save_and_restart(struct perf_event *event)
+int intel_pmu_save_and_restart(struct perf_event *event)
 {
 	x86_perf_event_update(event);
 	return x86_perf_event_set_period(event);
@@ -1197,6 +1211,21 @@ intel_shared_regs_constraints(struct cpu_hw_events *cpuc,
 	return c;
 }
 
+struct event_constraint *
+x86_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
+{
+	struct event_constraint *c;
+
+	if (x86_pmu.event_constraints) {
+		for_each_event_constraint(c, x86_pmu.event_constraints) {
+			if ((event->hw.config & c->cmask) == c->code)
+				return c;
+		}
+	}
+
+	return &unconstrained;
+}
+
 static struct event_constraint *
 intel_get_event_constraints(struct cpu_hw_events *cpuc, struct perf_event *event)
 {
@@ -1284,12 +1313,84 @@ static int intel_pmu_hw_config(struct perf_event *event)
 	return 0;
 }
 
+struct perf_guest_switch_msr *perf_guest_get_msrs(int *nr)
+{
+	if (x86_pmu.guest_get_msrs)
+		return x86_pmu.guest_get_msrs(nr);
+	*nr = 0;
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(perf_guest_get_msrs);
+
+static struct perf_guest_switch_msr *intel_guest_get_msrs(int *nr)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct perf_guest_switch_msr *arr = cpuc->guest_switch_msrs;
+
+	arr[0].msr = MSR_CORE_PERF_GLOBAL_CTRL;
+	arr[0].host = x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_guest_mask;
+	arr[0].guest = x86_pmu.intel_ctrl & ~cpuc->intel_ctrl_host_mask;
+
+	*nr = 1;
+	return arr;
+}
+
+static struct perf_guest_switch_msr *core_guest_get_msrs(int *nr)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct perf_guest_switch_msr *arr = cpuc->guest_switch_msrs;
+	int idx;
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++)  {
+		struct perf_event *event = cpuc->events[idx];
+
+		arr[idx].msr = x86_pmu_config_addr(idx);
+		arr[idx].host = arr[idx].guest = 0;
+
+		if (!test_bit(idx, cpuc->active_mask))
+			continue;
+
+		arr[idx].host = arr[idx].guest =
+			event->hw.config | ARCH_PERFMON_EVENTSEL_ENABLE;
+
+		if (event->attr.exclude_host)
+			arr[idx].host &= ~ARCH_PERFMON_EVENTSEL_ENABLE;
+		else if (event->attr.exclude_guest)
+			arr[idx].guest &= ~ARCH_PERFMON_EVENTSEL_ENABLE;
+	}
+
+	*nr = x86_pmu.num_counters;
+	return arr;
+}
+
+static void core_pmu_enable_event(struct perf_event *event)
+{
+	if (!event->attr.exclude_host)
+		x86_pmu_enable_event(event);
+}
+
+static void core_pmu_enable_all(int added)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	int idx;
+
+	for (idx = 0; idx < x86_pmu.num_counters; idx++) {
+		struct hw_perf_event *hwc = &cpuc->events[idx]->hw;
+
+		if (!test_bit(idx, cpuc->active_mask) ||
+				cpuc->events[idx]->attr.exclude_host)
+			continue;
+
+		__x86_pmu_enable_event(hwc, ARCH_PERFMON_EVENTSEL_ENABLE);
+	}
+}
+
 static __initconst const struct x86_pmu core_pmu = {
 	.name			= "core",
 	.handle_irq		= x86_pmu_handle_irq,
 	.disable_all		= x86_pmu_disable_all,
-	.enable_all		= x86_pmu_enable_all,
-	.enable			= x86_pmu_enable_event,
+	.enable_all		= core_pmu_enable_all,
+	.enable			= core_pmu_enable_event,
 	.disable		= x86_pmu_disable_event,
 	.hw_config		= x86_pmu_hw_config,
 	.schedule_events	= x86_schedule_events,
@@ -1307,9 +1408,10 @@ static __initconst const struct x86_pmu core_pmu = {
 	.get_event_constraints	= intel_get_event_constraints,
 	.put_event_constraints	= intel_put_event_constraints,
 	.event_constraints	= intel_core_event_constraints,
+	.guest_get_msrs		= core_guest_get_msrs,
 };
 
-static struct intel_shared_regs *allocate_shared_regs(int cpu)
+struct intel_shared_regs *allocate_shared_regs(int cpu)
 {
 	struct intel_shared_regs *regs;
 	int i;
@@ -1362,7 +1464,7 @@ static void intel_pmu_cpu_starting(int cpu)
 
 		pc = per_cpu(cpu_hw_events, i).shared_regs;
 		if (pc && pc->core_id == core_id) {
-			kfree(cpuc->shared_regs);
+			cpuc->kfree_on_online = cpuc->shared_regs;
 			cpuc->shared_regs = pc;
 			break;
 		}
@@ -1413,6 +1515,7 @@ static __initconst const struct x86_pmu intel_pmu = {
 	.cpu_prepare		= intel_pmu_cpu_prepare,
 	.cpu_starting		= intel_pmu_cpu_starting,
 	.cpu_dying		= intel_pmu_cpu_dying,
+	.guest_get_msrs		= intel_guest_get_msrs,
 };
 
 static void intel_clovertown_quirks(void)
@@ -1441,7 +1544,7 @@ static void intel_clovertown_quirks(void)
 	x86_pmu.pebs_constraints = NULL;
 }
 
-static __init int intel_pmu_init(void)
+__init int intel_pmu_init(void)
 {
 	union cpuid10_edx edx;
 	union cpuid10_eax eax;
@@ -1597,7 +1700,7 @@ static __init int intel_pmu_init(void)
 		intel_pmu_lbr_init_nhm();
 
 		x86_pmu.event_constraints = intel_snb_event_constraints;
-		x86_pmu.pebs_constraints = intel_snb_pebs_events;
+		x86_pmu.pebs_constraints = intel_snb_pebs_event_constraints;
 		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;
@@ -1628,16 +1731,3 @@ static __init int intel_pmu_init(void)
 	}
 	return 0;
 }
-
-#else /* CONFIG_CPU_SUP_INTEL */
-
-static int intel_pmu_init(void)
-{
-	return 0;
-}
-
-static struct intel_shared_regs *allocate_shared_regs(int cpu)
-{
-	return NULL;
-}
-#endif /* CONFIG_CPU_SUP_INTEL */
diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c
index 1b1ef3addcf..c0d238f49db 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_ds.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c
@@ -1,7 +1,10 @@
-#ifdef CONFIG_CPU_SUP_INTEL
+#include <linux/bitops.h>
+#include <linux/types.h>
+#include <linux/slab.h>
 
-/* The maximal number of PEBS events: */
-#define MAX_PEBS_EVENTS		4
+#include <asm/perf_event.h>
+
+#include "perf_event.h"
 
 /* The size of a BTS record in bytes: */
 #define BTS_RECORD_SIZE		24
@@ -37,24 +40,7 @@ struct pebs_record_nhm {
 	u64 status, dla, dse, lat;
 };
 
-/*
- * A debug store configuration.
- *
- * We only support architectures that use 64bit fields.
- */
-struct debug_store {
-	u64	bts_buffer_base;
-	u64	bts_index;
-	u64	bts_absolute_maximum;
-	u64	bts_interrupt_threshold;
-	u64	pebs_buffer_base;
-	u64	pebs_index;
-	u64	pebs_absolute_maximum;
-	u64	pebs_interrupt_threshold;
-	u64	pebs_event_reset[MAX_PEBS_EVENTS];
-};
-
-static void init_debug_store_on_cpu(int cpu)
+void init_debug_store_on_cpu(int cpu)
 {
 	struct debug_store *ds = per_cpu(cpu_hw_events, cpu).ds;
 
@@ -66,7 +52,7 @@ static void init_debug_store_on_cpu(int cpu)
 		     (u32)((u64)(unsigned long)ds >> 32));
 }
 
-static void fini_debug_store_on_cpu(int cpu)
+void fini_debug_store_on_cpu(int cpu)
 {
 	if (!per_cpu(cpu_hw_events, cpu).ds)
 		return;
@@ -175,7 +161,7 @@ static void release_ds_buffer(int cpu)
 	kfree(ds);
 }
 
-static void release_ds_buffers(void)
+void release_ds_buffers(void)
 {
 	int cpu;
 
@@ -194,7 +180,7 @@ static void release_ds_buffers(void)
 	put_online_cpus();
 }
 
-static void reserve_ds_buffers(void)
+void reserve_ds_buffers(void)
 {
 	int bts_err = 0, pebs_err = 0;
 	int cpu;
@@ -260,10 +246,10 @@ static void reserve_ds_buffers(void)
  * BTS
  */
 
-static struct event_constraint bts_constraint =
+struct event_constraint bts_constraint =
 	EVENT_CONSTRAINT(0, 1ULL << X86_PMC_IDX_FIXED_BTS, 0);
 
-static void intel_pmu_enable_bts(u64 config)
+void intel_pmu_enable_bts(u64 config)
 {
 	unsigned long debugctlmsr;
 
@@ -282,7 +268,7 @@ static void intel_pmu_enable_bts(u64 config)
 	update_debugctlmsr(debugctlmsr);
 }
 
-static void intel_pmu_disable_bts(void)
+void intel_pmu_disable_bts(void)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	unsigned long debugctlmsr;
@@ -299,7 +285,7 @@ static void intel_pmu_disable_bts(void)
 	update_debugctlmsr(debugctlmsr);
 }
 
-static int intel_pmu_drain_bts_buffer(void)
+int intel_pmu_drain_bts_buffer(void)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct debug_store *ds = cpuc->ds;
@@ -361,7 +347,7 @@ static int intel_pmu_drain_bts_buffer(void)
 /*
  * PEBS
  */
-static struct event_constraint intel_core2_pebs_event_constraints[] = {
+struct event_constraint intel_core2_pebs_event_constraints[] = {
 	INTEL_UEVENT_CONSTRAINT(0x00c0, 0x1), /* INST_RETIRED.ANY */
 	INTEL_UEVENT_CONSTRAINT(0xfec1, 0x1), /* X87_OPS_RETIRED.ANY */
 	INTEL_UEVENT_CONSTRAINT(0x00c5, 0x1), /* BR_INST_RETIRED.MISPRED */
@@ -370,14 +356,14 @@ static struct event_constraint intel_core2_pebs_event_constraints[] = {
 	EVENT_CONSTRAINT_END
 };
 
-static struct event_constraint intel_atom_pebs_event_constraints[] = {
+struct event_constraint intel_atom_pebs_event_constraints[] = {
 	INTEL_UEVENT_CONSTRAINT(0x00c0, 0x1), /* INST_RETIRED.ANY */
 	INTEL_UEVENT_CONSTRAINT(0x00c5, 0x1), /* MISPREDICTED_BRANCH_RETIRED */
 	INTEL_EVENT_CONSTRAINT(0xcb, 0x1),    /* MEM_LOAD_RETIRED.* */
 	EVENT_CONSTRAINT_END
 };
 
-static struct event_constraint intel_nehalem_pebs_event_constraints[] = {
+struct event_constraint intel_nehalem_pebs_event_constraints[] = {
 	INTEL_EVENT_CONSTRAINT(0x0b, 0xf),    /* MEM_INST_RETIRED.* */
 	INTEL_EVENT_CONSTRAINT(0x0f, 0xf),    /* MEM_UNCORE_RETIRED.* */
 	INTEL_UEVENT_CONSTRAINT(0x010c, 0xf), /* MEM_STORE_RETIRED.DTLB_MISS */
@@ -392,7 +378,7 @@ static struct event_constraint intel_nehalem_pebs_event_constraints[] = {
 	EVENT_CONSTRAINT_END
 };
 
-static struct event_constraint intel_westmere_pebs_event_constraints[] = {
+struct event_constraint intel_westmere_pebs_event_constraints[] = {
 	INTEL_EVENT_CONSTRAINT(0x0b, 0xf),    /* MEM_INST_RETIRED.* */
 	INTEL_EVENT_CONSTRAINT(0x0f, 0xf),    /* MEM_UNCORE_RETIRED.* */
 	INTEL_UEVENT_CONSTRAINT(0x010c, 0xf), /* MEM_STORE_RETIRED.DTLB_MISS */
@@ -407,7 +393,7 @@ static struct event_constraint intel_westmere_pebs_event_constraints[] = {
 	EVENT_CONSTRAINT_END
 };
 
-static struct event_constraint intel_snb_pebs_events[] = {
+struct event_constraint intel_snb_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 */
@@ -428,8 +414,7 @@ static struct event_constraint intel_snb_pebs_events[] = {
 	EVENT_CONSTRAINT_END
 };
 
-static struct event_constraint *
-intel_pebs_constraints(struct perf_event *event)
+struct event_constraint *intel_pebs_constraints(struct perf_event *event)
 {
 	struct event_constraint *c;
 
@@ -446,7 +431,7 @@ intel_pebs_constraints(struct perf_event *event)
 	return &emptyconstraint;
 }
 
-static void intel_pmu_pebs_enable(struct perf_event *event)
+void intel_pmu_pebs_enable(struct perf_event *event)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
@@ -460,7 +445,7 @@ static void intel_pmu_pebs_enable(struct perf_event *event)
 		intel_pmu_lbr_enable(event);
 }
 
-static void intel_pmu_pebs_disable(struct perf_event *event)
+void intel_pmu_pebs_disable(struct perf_event *event)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
@@ -475,7 +460,7 @@ static void intel_pmu_pebs_disable(struct perf_event *event)
 		intel_pmu_lbr_disable(event);
 }
 
-static void intel_pmu_pebs_enable_all(void)
+void intel_pmu_pebs_enable_all(void)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -483,7 +468,7 @@ static void intel_pmu_pebs_enable_all(void)
 		wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled);
 }
 
-static void intel_pmu_pebs_disable_all(void)
+void intel_pmu_pebs_disable_all(void)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -576,8 +561,6 @@ static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs)
 	return 0;
 }
 
-static int intel_pmu_save_and_restart(struct perf_event *event);
-
 static void __intel_pmu_pebs_event(struct perf_event *event,
 				   struct pt_regs *iregs, void *__pebs)
 {
@@ -716,7 +699,7 @@ static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs)
  * BTS, PEBS probe and setup
  */
 
-static void intel_ds_init(void)
+void intel_ds_init(void)
 {
 	/*
 	 * No support for 32bit formats
@@ -749,15 +732,3 @@ static void intel_ds_init(void)
 		}
 	}
 }
-
-#else /* CONFIG_CPU_SUP_INTEL */
-
-static void reserve_ds_buffers(void)
-{
-}
-
-static void release_ds_buffers(void)
-{
-}
-
-#endif /* CONFIG_CPU_SUP_INTEL */
diff --git a/arch/x86/kernel/cpu/perf_event_intel_lbr.c b/arch/x86/kernel/cpu/perf_event_intel_lbr.c
index d202c1bece1..3fab3de3ce9 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_lbr.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_lbr.c
@@ -1,4 +1,10 @@
-#ifdef CONFIG_CPU_SUP_INTEL
+#include <linux/perf_event.h>
+#include <linux/types.h>
+
+#include <asm/perf_event.h>
+#include <asm/msr.h>
+
+#include "perf_event.h"
 
 enum {
 	LBR_FORMAT_32		= 0x00,
@@ -48,7 +54,7 @@ static void intel_pmu_lbr_reset_64(void)
 	}
 }
 
-static void intel_pmu_lbr_reset(void)
+void intel_pmu_lbr_reset(void)
 {
 	if (!x86_pmu.lbr_nr)
 		return;
@@ -59,7 +65,7 @@ static void intel_pmu_lbr_reset(void)
 		intel_pmu_lbr_reset_64();
 }
 
-static void intel_pmu_lbr_enable(struct perf_event *event)
+void intel_pmu_lbr_enable(struct perf_event *event)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -81,7 +87,7 @@ static void intel_pmu_lbr_enable(struct perf_event *event)
 	cpuc->lbr_users++;
 }
 
-static void intel_pmu_lbr_disable(struct perf_event *event)
+void intel_pmu_lbr_disable(struct perf_event *event)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -95,7 +101,7 @@ static void intel_pmu_lbr_disable(struct perf_event *event)
 		__intel_pmu_lbr_disable();
 }
 
-static void intel_pmu_lbr_enable_all(void)
+void intel_pmu_lbr_enable_all(void)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -103,7 +109,7 @@ static void intel_pmu_lbr_enable_all(void)
 		__intel_pmu_lbr_enable();
 }
 
-static void intel_pmu_lbr_disable_all(void)
+void intel_pmu_lbr_disable_all(void)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -178,7 +184,7 @@ static void intel_pmu_lbr_read_64(struct cpu_hw_events *cpuc)
 	cpuc->lbr_stack.nr = i;
 }
 
-static void intel_pmu_lbr_read(void)
+void intel_pmu_lbr_read(void)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -191,7 +197,7 @@ static void intel_pmu_lbr_read(void)
 		intel_pmu_lbr_read_64(cpuc);
 }
 
-static void intel_pmu_lbr_init_core(void)
+void intel_pmu_lbr_init_core(void)
 {
 	x86_pmu.lbr_nr     = 4;
 	x86_pmu.lbr_tos    = 0x01c9;
@@ -199,7 +205,7 @@ static void intel_pmu_lbr_init_core(void)
 	x86_pmu.lbr_to     = 0x60;
 }
 
-static void intel_pmu_lbr_init_nhm(void)
+void intel_pmu_lbr_init_nhm(void)
 {
 	x86_pmu.lbr_nr     = 16;
 	x86_pmu.lbr_tos    = 0x01c9;
@@ -207,12 +213,10 @@ static void intel_pmu_lbr_init_nhm(void)
 	x86_pmu.lbr_to     = 0x6c0;
 }
 
-static void intel_pmu_lbr_init_atom(void)
+void intel_pmu_lbr_init_atom(void)
 {
 	x86_pmu.lbr_nr	   = 8;
 	x86_pmu.lbr_tos    = 0x01c9;
 	x86_pmu.lbr_from   = 0x40;
 	x86_pmu.lbr_to     = 0x60;
 }
-
-#endif /* CONFIG_CPU_SUP_INTEL */
diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c
index 7809d2bcb20..492bf1358a7 100644
--- a/arch/x86/kernel/cpu/perf_event_p4.c
+++ b/arch/x86/kernel/cpu/perf_event_p4.c
@@ -7,9 +7,13 @@
  *  For licencing details see kernel-base/COPYING
  */
 
-#ifdef CONFIG_CPU_SUP_INTEL
+#include <linux/perf_event.h>
 
 #include <asm/perf_event_p4.h>
+#include <asm/hardirq.h>
+#include <asm/apic.h>
+
+#include "perf_event.h"
 
 #define P4_CNTR_LIMIT 3
 /*
@@ -1303,7 +1307,7 @@ static __initconst const struct x86_pmu p4_pmu = {
 	.perfctr_second_write	= 1,
 };
 
-static __init int p4_pmu_init(void)
+__init int p4_pmu_init(void)
 {
 	unsigned int low, high;
 
@@ -1326,5 +1330,3 @@ static __init int p4_pmu_init(void)
 
 	return 0;
 }
-
-#endif /* CONFIG_CPU_SUP_INTEL */
diff --git a/arch/x86/kernel/cpu/perf_event_p6.c b/arch/x86/kernel/cpu/perf_event_p6.c
index 20c097e3386..c7181befecd 100644
--- a/arch/x86/kernel/cpu/perf_event_p6.c
+++ b/arch/x86/kernel/cpu/perf_event_p6.c
@@ -1,4 +1,7 @@
-#ifdef CONFIG_CPU_SUP_INTEL
+#include <linux/perf_event.h>
+#include <linux/types.h>
+
+#include "perf_event.h"
 
 /*
  * Not sure about some of these
@@ -114,7 +117,7 @@ static __initconst const struct x86_pmu p6_pmu = {
 	.event_constraints	= p6_event_constraints,
 };
 
-static __init int p6_pmu_init(void)
+__init int p6_pmu_init(void)
 {
 	switch (boot_cpu_data.x86_model) {
 	case 1:
@@ -138,5 +141,3 @@ static __init int p6_pmu_init(void)
 
 	return 0;
 }
-
-#endif /* CONFIG_CPU_SUP_INTEL */