diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c
index 85d8e4f..1cc4968 100644
--- a/arch/alpha/kernel/perf_event.c
+++ b/arch/alpha/kernel/perf_event.c
@@ -307,7 +307,7 @@
 			     new_raw_count) != prev_raw_count)
 		goto again;
 
-	delta = (new_raw_count  - (prev_raw_count & alpha_pmu->pmc_count_mask[idx])) + ovf;
+	delta = (new_raw_count - (prev_raw_count & alpha_pmu->pmc_count_mask[idx])) + ovf;
 
 	/* It is possible on very rare occasions that the PMC has overflowed
 	 * but the interrupt is yet to come.  Detect and fix this situation.
@@ -402,14 +402,13 @@
 		struct hw_perf_event *hwc = &pe->hw;
 		int idx = hwc->idx;
 
-		if (cpuc->current_idx[j] != PMC_NO_INDEX) {
-			cpuc->idx_mask |= (1<<cpuc->current_idx[j]);
-			continue;
+		if (cpuc->current_idx[j] == PMC_NO_INDEX) {
+			alpha_perf_event_set_period(pe, hwc, idx);
+			cpuc->current_idx[j] = idx;
 		}
 
-		alpha_perf_event_set_period(pe, hwc, idx);
-		cpuc->current_idx[j] = idx;
-		cpuc->idx_mask |= (1<<cpuc->current_idx[j]);
+		if (!(hwc->state & PERF_HES_STOPPED))
+			cpuc->idx_mask |= (1<<cpuc->current_idx[j]);
 	}
 	cpuc->config = cpuc->event[0]->hw.config_base;
 }
@@ -420,12 +419,13 @@
  *  - this function is called from outside this module via the pmu struct
  *    returned from perf event initialisation.
  */
-static int alpha_pmu_enable(struct perf_event *event)
+static int alpha_pmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
 	int n0;
 	int ret;
-	unsigned long flags;
+	unsigned long irq_flags;
 
 	/*
 	 * The Sparc code has the IRQ disable first followed by the perf
@@ -435,8 +435,8 @@
 	 * nevertheless we disable the PMCs first to enable a potential
 	 * final PMI to occur before we disable interrupts.
 	 */
-	perf_disable();
-	local_irq_save(flags);
+	perf_pmu_disable(event->pmu);
+	local_irq_save(irq_flags);
 
 	/* Default to error to be returned */
 	ret = -EAGAIN;
@@ -455,8 +455,12 @@
 		}
 	}
 
-	local_irq_restore(flags);
-	perf_enable();
+	hwc->state = PERF_HES_UPTODATE;
+	if (!(flags & PERF_EF_START))
+		hwc->state |= PERF_HES_STOPPED;
+
+	local_irq_restore(irq_flags);
+	perf_pmu_enable(event->pmu);
 
 	return ret;
 }
@@ -467,15 +471,15 @@
  *  - this function is called from outside this module via the pmu struct
  *    returned from perf event initialisation.
  */
-static void alpha_pmu_disable(struct perf_event *event)
+static void alpha_pmu_del(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
-	unsigned long flags;
+	unsigned long irq_flags;
 	int j;
 
-	perf_disable();
-	local_irq_save(flags);
+	perf_pmu_disable(event->pmu);
+	local_irq_save(irq_flags);
 
 	for (j = 0; j < cpuc->n_events; j++) {
 		if (event == cpuc->event[j]) {
@@ -501,8 +505,8 @@
 		}
 	}
 
-	local_irq_restore(flags);
-	perf_enable();
+	local_irq_restore(irq_flags);
+	perf_pmu_enable(event->pmu);
 }
 
 
@@ -514,13 +518,44 @@
 }
 
 
-static void alpha_pmu_unthrottle(struct perf_event *event)
+static void alpha_pmu_stop(struct perf_event *event, int flags)
 {
 	struct hw_perf_event *hwc = &event->hw;
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
+	if (!(hwc->state & PERF_HES_STOPPED)) {
+		cpuc->idx_mask &= ~(1UL<<hwc->idx);
+		hwc->state |= PERF_HES_STOPPED;
+	}
+
+	if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
+		alpha_perf_event_update(event, hwc, hwc->idx, 0);
+		hwc->state |= PERF_HES_UPTODATE;
+	}
+
+	if (cpuc->enabled)
+		wrperfmon(PERFMON_CMD_DISABLE, (1UL<<hwc->idx));
+}
+
+
+static void alpha_pmu_start(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
+	if (WARN_ON_ONCE(!(hwc->state & PERF_HES_STOPPED)))
+		return;
+
+	if (flags & PERF_EF_RELOAD) {
+		WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE));
+		alpha_perf_event_set_period(event, hwc, hwc->idx);
+	}
+
+	hwc->state = 0;
+
 	cpuc->idx_mask |= 1UL<<hwc->idx;
-	wrperfmon(PERFMON_CMD_ENABLE, (1UL<<hwc->idx));
+	if (cpuc->enabled)
+		wrperfmon(PERFMON_CMD_ENABLE, (1UL<<hwc->idx));
 }
 
 
@@ -642,39 +677,36 @@
 	return 0;
 }
 
-static const struct pmu pmu = {
-	.enable		= alpha_pmu_enable,
-	.disable	= alpha_pmu_disable,
-	.read		= alpha_pmu_read,
-	.unthrottle	= alpha_pmu_unthrottle,
-};
-
-
 /*
  * Main entry point to initialise a HW performance event.
  */
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+static int alpha_pmu_event_init(struct perf_event *event)
 {
 	int err;
 
+	switch (event->attr.type) {
+	case PERF_TYPE_RAW:
+	case PERF_TYPE_HARDWARE:
+	case PERF_TYPE_HW_CACHE:
+		break;
+
+	default:
+		return -ENOENT;
+	}
+
 	if (!alpha_pmu)
-		return ERR_PTR(-ENODEV);
+		return -ENODEV;
 
 	/* Do the real initialisation work. */
 	err = __hw_perf_event_init(event);
 
-	if (err)
-		return ERR_PTR(err);
-
-	return &pmu;
+	return err;
 }
 
-
-
 /*
  * Main entry point - enable HW performance counters.
  */
-void hw_perf_enable(void)
+static void alpha_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -700,7 +732,7 @@
  * Main entry point - disable HW performance counters.
  */
 
-void hw_perf_disable(void)
+static void alpha_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -713,6 +745,17 @@
 	wrperfmon(PERFMON_CMD_DISABLE, cpuc->idx_mask);
 }
 
+static struct pmu pmu = {
+	.pmu_enable	= alpha_pmu_enable,
+	.pmu_disable	= alpha_pmu_disable,
+	.event_init	= alpha_pmu_event_init,
+	.add		= alpha_pmu_add,
+	.del		= alpha_pmu_del,
+	.start		= alpha_pmu_start,
+	.stop		= alpha_pmu_stop,
+	.read		= alpha_pmu_read,
+};
+
 
 /*
  * Main entry point - don't know when this is called but it
@@ -766,7 +809,7 @@
 	wrperfmon(PERFMON_CMD_DISABLE, cpuc->idx_mask);
 
 	/* la_ptr is the counter that overflowed. */
-	if (unlikely(la_ptr >= perf_max_events)) {
+	if (unlikely(la_ptr >= alpha_pmu->num_pmcs)) {
 		/* This should never occur! */
 		irq_err_count++;
 		pr_warning("PMI: silly index %ld\n", la_ptr);
@@ -807,7 +850,7 @@
 			/* Interrupts coming too quickly; "throttle" the
 			 * counter, i.e., disable it for a little while.
 			 */
-			cpuc->idx_mask &= ~(1UL<<idx);
+			alpha_pmu_stop(event, 0);
 		}
 	}
 	wrperfmon(PERFMON_CMD_ENABLE, cpuc->idx_mask);
@@ -837,6 +880,7 @@
 
 	/* And set up PMU specification */
 	alpha_pmu = &ev67_pmu;
-	perf_max_events = alpha_pmu->num_pmcs;
+
+	perf_pmu_register(&pmu);
 }
 
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index ecbb028..ad19c27 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -221,27 +221,6 @@
 }
 
 static void
-armpmu_disable(struct perf_event *event)
-{
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-	struct hw_perf_event *hwc = &event->hw;
-	int idx = hwc->idx;
-
-	WARN_ON(idx < 0);
-
-	clear_bit(idx, cpuc->active_mask);
-	armpmu->disable(hwc, idx);
-
-	barrier();
-
-	armpmu_event_update(event, hwc, idx);
-	cpuc->events[idx] = NULL;
-	clear_bit(idx, cpuc->used_mask);
-
-	perf_event_update_userpage(event);
-}
-
-static void
 armpmu_read(struct perf_event *event)
 {
 	struct hw_perf_event *hwc = &event->hw;
@@ -254,13 +233,44 @@
 }
 
 static void
-armpmu_unthrottle(struct perf_event *event)
+armpmu_stop(struct perf_event *event, int flags)
 {
 	struct hw_perf_event *hwc = &event->hw;
 
+	if (!armpmu)
+		return;
+
+	/*
+	 * ARM pmu always has to update the counter, so ignore
+	 * PERF_EF_UPDATE, see comments in armpmu_start().
+	 */
+	if (!(hwc->state & PERF_HES_STOPPED)) {
+		armpmu->disable(hwc, hwc->idx);
+		barrier(); /* why? */
+		armpmu_event_update(event, hwc, hwc->idx);
+		hwc->state |= PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	}
+}
+
+static void
+armpmu_start(struct perf_event *event, int flags)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (!armpmu)
+		return;
+
+	/*
+	 * ARM pmu always has to reprogram the period, so ignore
+	 * PERF_EF_RELOAD, see the comment below.
+	 */
+	if (flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE));
+
+	hwc->state = 0;
 	/*
 	 * Set the period again. Some counters can't be stopped, so when we
-	 * were throttled we simply disabled the IRQ source and the counter
+	 * were stopped we simply disabled the IRQ source and the counter
 	 * may have been left counting. If we don't do this step then we may
 	 * get an interrupt too soon or *way* too late if the overflow has
 	 * happened since disabling.
@@ -269,14 +279,33 @@
 	armpmu->enable(hwc, hwc->idx);
 }
 
+static void
+armpmu_del(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	int idx = hwc->idx;
+
+	WARN_ON(idx < 0);
+
+	clear_bit(idx, cpuc->active_mask);
+	armpmu_stop(event, PERF_EF_UPDATE);
+	cpuc->events[idx] = NULL;
+	clear_bit(idx, cpuc->used_mask);
+
+	perf_event_update_userpage(event);
+}
+
 static int
-armpmu_enable(struct perf_event *event)
+armpmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
 	int idx;
 	int err = 0;
 
+	perf_pmu_disable(event->pmu);
+
 	/* If we don't have a space for the counter then finish early. */
 	idx = armpmu->get_event_idx(cpuc, hwc);
 	if (idx < 0) {
@@ -293,25 +322,19 @@
 	cpuc->events[idx] = event;
 	set_bit(idx, cpuc->active_mask);
 
-	/* Set the period for the event. */
-	armpmu_event_set_period(event, hwc, idx);
-
-	/* Enable the event. */
-	armpmu->enable(hwc, idx);
+	hwc->state = PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	if (flags & PERF_EF_START)
+		armpmu_start(event, PERF_EF_RELOAD);
 
 	/* Propagate our changes to the userspace mapping. */
 	perf_event_update_userpage(event);
 
 out:
+	perf_pmu_enable(event->pmu);
 	return err;
 }
 
-static struct pmu pmu = {
-	.enable	    = armpmu_enable,
-	.disable    = armpmu_disable,
-	.unthrottle = armpmu_unthrottle,
-	.read	    = armpmu_read,
-};
+static struct pmu pmu;
 
 static int
 validate_event(struct cpu_hw_events *cpuc,
@@ -491,20 +514,29 @@
 	return err;
 }
 
-const struct pmu *
-hw_perf_event_init(struct perf_event *event)
+static int armpmu_event_init(struct perf_event *event)
 {
 	int err = 0;
 
+	switch (event->attr.type) {
+	case PERF_TYPE_RAW:
+	case PERF_TYPE_HARDWARE:
+	case PERF_TYPE_HW_CACHE:
+		break;
+
+	default:
+		return -ENOENT;
+	}
+
 	if (!armpmu)
-		return ERR_PTR(-ENODEV);
+		return -ENODEV;
 
 	event->destroy = hw_perf_event_destroy;
 
 	if (!atomic_inc_not_zero(&active_events)) {
-		if (atomic_read(&active_events) > perf_max_events) {
+		if (atomic_read(&active_events) > armpmu.num_events) {
 			atomic_dec(&active_events);
-			return ERR_PTR(-ENOSPC);
+			return -ENOSPC;
 		}
 
 		mutex_lock(&pmu_reserve_mutex);
@@ -518,17 +550,16 @@
 	}
 
 	if (err)
-		return ERR_PTR(err);
+		return err;
 
 	err = __hw_perf_event_init(event);
 	if (err)
 		hw_perf_event_destroy(event);
 
-	return err ? ERR_PTR(err) : &pmu;
+	return err;
 }
 
-void
-hw_perf_enable(void)
+static void armpmu_enable(struct pmu *pmu)
 {
 	/* Enable all of the perf events on hardware. */
 	int idx;
@@ -549,13 +580,23 @@
 	armpmu->start();
 }
 
-void
-hw_perf_disable(void)
+static void armpmu_disable(struct pmu *pmu)
 {
 	if (armpmu)
 		armpmu->stop();
 }
 
+static struct pmu pmu = {
+	.pmu_enable	= armpmu_enable,
+	.pmu_disable	= armpmu_disable,
+	.event_init	= armpmu_event_init,
+	.add		= armpmu_add,
+	.del		= armpmu_del,
+	.start		= armpmu_start,
+	.stop		= armpmu_stop,
+	.read		= armpmu_read,
+};
+
 /*
  * ARMv6 Performance counter handling code.
  *
@@ -2933,14 +2974,12 @@
 			armpmu = &armv6pmu;
 			memcpy(armpmu_perf_cache_map, armv6_perf_cache_map,
 					sizeof(armv6_perf_cache_map));
-			perf_max_events	= armv6pmu.num_events;
 			break;
 		case 0xB020:	/* ARM11mpcore */
 			armpmu = &armv6mpcore_pmu;
 			memcpy(armpmu_perf_cache_map,
 			       armv6mpcore_perf_cache_map,
 			       sizeof(armv6mpcore_perf_cache_map));
-			perf_max_events = armv6mpcore_pmu.num_events;
 			break;
 		case 0xC080:	/* Cortex-A8 */
 			armv7pmu.id = ARM_PERF_PMU_ID_CA8;
@@ -2952,7 +2991,6 @@
 			/* Reset PMNC and read the nb of CNTx counters
 			    supported */
 			armv7pmu.num_events = armv7_reset_read_pmnc();
-			perf_max_events = armv7pmu.num_events;
 			break;
 		case 0xC090:	/* Cortex-A9 */
 			armv7pmu.id = ARM_PERF_PMU_ID_CA9;
@@ -2964,7 +3002,6 @@
 			/* Reset PMNC and read the nb of CNTx counters
 			    supported */
 			armv7pmu.num_events = armv7_reset_read_pmnc();
-			perf_max_events = armv7pmu.num_events;
 			break;
 		}
 	/* Intel CPUs [xscale]. */
@@ -2975,13 +3012,11 @@
 			armpmu = &xscale1pmu;
 			memcpy(armpmu_perf_cache_map, xscale_perf_cache_map,
 					sizeof(xscale_perf_cache_map));
-			perf_max_events	= xscale1pmu.num_events;
 			break;
 		case 2:
 			armpmu = &xscale2pmu;
 			memcpy(armpmu_perf_cache_map, xscale_perf_cache_map,
 					sizeof(xscale_perf_cache_map));
-			perf_max_events	= xscale2pmu.num_events;
 			break;
 		}
 	}
@@ -2991,9 +3026,10 @@
 				arm_pmu_names[armpmu->id], armpmu->num_events);
 	} else {
 		pr_info("no hardware support available\n");
-		perf_max_events = -1;
 	}
 
+	perf_pmu_register(&pmu);
+
 	return 0;
 }
 arch_initcall(init_hw_perf_events);
@@ -3001,13 +3037,6 @@
 /*
  * Callchain handling code.
  */
-static inline void
-callchain_store(struct perf_callchain_entry *entry,
-		u64 ip)
-{
-	if (entry->nr < PERF_MAX_STACK_DEPTH)
-		entry->ip[entry->nr++] = ip;
-}
 
 /*
  * The registers we're interested in are at the end of the variable
@@ -3039,7 +3068,7 @@
 	if (__copy_from_user_inatomic(&buftail, tail, sizeof(buftail)))
 		return NULL;
 
-	callchain_store(entry, buftail.lr);
+	perf_callchain_store(entry, buftail.lr);
 
 	/*
 	 * Frame pointers should strictly progress back up the stack
@@ -3051,16 +3080,11 @@
 	return buftail.fp - 1;
 }
 
-static void
-perf_callchain_user(struct pt_regs *regs,
-		    struct perf_callchain_entry *entry)
+void
+perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
 	struct frame_tail *tail;
 
-	callchain_store(entry, PERF_CONTEXT_USER);
-
-	if (!user_mode(regs))
-		regs = task_pt_regs(current);
 
 	tail = (struct frame_tail *)regs->ARM_fp - 1;
 
@@ -3078,56 +3102,18 @@
 		void *data)
 {
 	struct perf_callchain_entry *entry = data;
-	callchain_store(entry, fr->pc);
+	perf_callchain_store(entry, fr->pc);
 	return 0;
 }
 
-static void
-perf_callchain_kernel(struct pt_regs *regs,
-		      struct perf_callchain_entry *entry)
+void
+perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
 	struct stackframe fr;
 
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
 	fr.fp = regs->ARM_fp;
 	fr.sp = regs->ARM_sp;
 	fr.lr = regs->ARM_lr;
 	fr.pc = regs->ARM_pc;
 	walk_stackframe(&fr, callchain_trace, entry);
 }
-
-static void
-perf_do_callchain(struct pt_regs *regs,
-		  struct perf_callchain_entry *entry)
-{
-	int is_user;
-
-	if (!regs)
-		return;
-
-	is_user = user_mode(regs);
-
-	if (!current || !current->pid)
-		return;
-
-	if (is_user && current->state != TASK_RUNNING)
-		return;
-
-	if (!is_user)
-		perf_callchain_kernel(regs, entry);
-
-	if (current->mm)
-		perf_callchain_user(regs, entry);
-}
-
-static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry);
-
-struct perf_callchain_entry *
-perf_callchain(struct pt_regs *regs)
-{
-	struct perf_callchain_entry *entry = &__get_cpu_var(pmc_irq_entry);
-
-	entry->nr = 0;
-	perf_do_callchain(regs, entry);
-	return entry;
-}
diff --git a/arch/arm/oprofile/common.c b/arch/arm/oprofile/common.c
index 0691176..aad63e6 100644
--- a/arch/arm/oprofile/common.c
+++ b/arch/arm/oprofile/common.c
@@ -96,7 +96,7 @@
 		return ret;
 
 	pevent = perf_event_create_kernel_counter(&counter_config[event].attr,
-						  cpu, -1,
+						  cpu, NULL,
 						  op_overflow_handler);
 
 	if (IS_ERR(pevent)) {
diff --git a/arch/powerpc/kernel/perf_callchain.c b/arch/powerpc/kernel/perf_callchain.c
index 95ad9da..d05ae42 100644
--- a/arch/powerpc/kernel/perf_callchain.c
+++ b/arch/powerpc/kernel/perf_callchain.c
@@ -23,18 +23,6 @@
 #include "ppc32.h"
 #endif
 
-/*
- * Store another value in a callchain_entry.
- */
-static inline void callchain_store(struct perf_callchain_entry *entry, u64 ip)
-{
-	unsigned int nr = entry->nr;
-
-	if (nr < PERF_MAX_STACK_DEPTH) {
-		entry->ip[nr] = ip;
-		entry->nr = nr + 1;
-	}
-}
 
 /*
  * Is sp valid as the address of the next kernel stack frame after prev_sp?
@@ -58,8 +46,8 @@
 	return 0;
 }
 
-static void perf_callchain_kernel(struct pt_regs *regs,
-				  struct perf_callchain_entry *entry)
+void
+perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
 	unsigned long sp, next_sp;
 	unsigned long next_ip;
@@ -69,8 +57,7 @@
 
 	lr = regs->link;
 	sp = regs->gpr[1];
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
-	callchain_store(entry, regs->nip);
+	perf_callchain_store(entry, regs->nip);
 
 	if (!validate_sp(sp, current, STACK_FRAME_OVERHEAD))
 		return;
@@ -89,7 +76,7 @@
 			next_ip = regs->nip;
 			lr = regs->link;
 			level = 0;
-			callchain_store(entry, PERF_CONTEXT_KERNEL);
+			perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
 
 		} else {
 			if (level == 0)
@@ -111,7 +98,7 @@
 			++level;
 		}
 
-		callchain_store(entry, next_ip);
+		perf_callchain_store(entry, next_ip);
 		if (!valid_next_sp(next_sp, sp))
 			return;
 		sp = next_sp;
@@ -233,8 +220,8 @@
 		puc == (unsigned long) &sf->uc;
 }
 
-static void perf_callchain_user_64(struct pt_regs *regs,
-				   struct perf_callchain_entry *entry)
+static void perf_callchain_user_64(struct perf_callchain_entry *entry,
+				   struct pt_regs *regs)
 {
 	unsigned long sp, next_sp;
 	unsigned long next_ip;
@@ -246,8 +233,7 @@
 	next_ip = regs->nip;
 	lr = regs->link;
 	sp = regs->gpr[1];
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, next_ip);
+	perf_callchain_store(entry, next_ip);
 
 	for (;;) {
 		fp = (unsigned long __user *) sp;
@@ -276,14 +262,14 @@
 			    read_user_stack_64(&uregs[PT_R1], &sp))
 				return;
 			level = 0;
-			callchain_store(entry, PERF_CONTEXT_USER);
-			callchain_store(entry, next_ip);
+			perf_callchain_store(entry, PERF_CONTEXT_USER);
+			perf_callchain_store(entry, next_ip);
 			continue;
 		}
 
 		if (level == 0)
 			next_ip = lr;
-		callchain_store(entry, next_ip);
+		perf_callchain_store(entry, next_ip);
 		++level;
 		sp = next_sp;
 	}
@@ -315,8 +301,8 @@
 	return __get_user_inatomic(*ret, ptr);
 }
 
-static inline void perf_callchain_user_64(struct pt_regs *regs,
-					  struct perf_callchain_entry *entry)
+static inline void perf_callchain_user_64(struct perf_callchain_entry *entry,
+					  struct pt_regs *regs)
 {
 }
 
@@ -435,8 +421,8 @@
 	return mctx->mc_gregs;
 }
 
-static void perf_callchain_user_32(struct pt_regs *regs,
-				   struct perf_callchain_entry *entry)
+static void perf_callchain_user_32(struct perf_callchain_entry *entry,
+				   struct pt_regs *regs)
 {
 	unsigned int sp, next_sp;
 	unsigned int next_ip;
@@ -447,8 +433,7 @@
 	next_ip = regs->nip;
 	lr = regs->link;
 	sp = regs->gpr[1];
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, next_ip);
+	perf_callchain_store(entry, next_ip);
 
 	while (entry->nr < PERF_MAX_STACK_DEPTH) {
 		fp = (unsigned int __user *) (unsigned long) sp;
@@ -470,45 +455,24 @@
 			    read_user_stack_32(&uregs[PT_R1], &sp))
 				return;
 			level = 0;
-			callchain_store(entry, PERF_CONTEXT_USER);
-			callchain_store(entry, next_ip);
+			perf_callchain_store(entry, PERF_CONTEXT_USER);
+			perf_callchain_store(entry, next_ip);
 			continue;
 		}
 
 		if (level == 0)
 			next_ip = lr;
-		callchain_store(entry, next_ip);
+		perf_callchain_store(entry, next_ip);
 		++level;
 		sp = next_sp;
 	}
 }
 
-/*
- * Since we can't get PMU interrupts inside a PMU interrupt handler,
- * we don't need separate irq and nmi entries here.
- */
-static DEFINE_PER_CPU(struct perf_callchain_entry, cpu_perf_callchain);
-
-struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
+void
+perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
-	struct perf_callchain_entry *entry = &__get_cpu_var(cpu_perf_callchain);
-
-	entry->nr = 0;
-
-	if (!user_mode(regs)) {
-		perf_callchain_kernel(regs, entry);
-		if (current->mm)
-			regs = task_pt_regs(current);
-		else
-			regs = NULL;
-	}
-
-	if (regs) {
-		if (current_is_64bit())
-			perf_callchain_user_64(regs, entry);
-		else
-			perf_callchain_user_32(regs, entry);
-	}
-
-	return entry;
+	if (current_is_64bit())
+		perf_callchain_user_64(entry, regs);
+	else
+		perf_callchain_user_32(entry, regs);
 }
diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c
index d301a30..9cb4924 100644
--- a/arch/powerpc/kernel/perf_event.c
+++ b/arch/powerpc/kernel/perf_event.c
@@ -402,6 +402,9 @@
 {
 	s64 val, delta, prev;
 
+	if (event->hw.state & PERF_HES_STOPPED)
+		return;
+
 	if (!event->hw.idx)
 		return;
 	/*
@@ -517,7 +520,7 @@
  * Disable all events to prevent PMU interrupts and to allow
  * events to be added or removed.
  */
-void hw_perf_disable(void)
+static void power_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -565,7 +568,7 @@
  * If we were previously disabled and events were added, then
  * put the new config on the PMU.
  */
-void hw_perf_enable(void)
+static void power_pmu_enable(struct pmu *pmu)
 {
 	struct perf_event *event;
 	struct cpu_hw_events *cpuhw;
@@ -672,6 +675,8 @@
 		}
 		local64_set(&event->hw.prev_count, val);
 		event->hw.idx = idx;
+		if (event->hw.state & PERF_HES_STOPPED)
+			val = 0;
 		write_pmc(idx, val);
 		perf_event_update_userpage(event);
 	}
@@ -727,7 +732,7 @@
  * re-enable the PMU in order to get hw_perf_enable to do the
  * actual work of reconfiguring the PMU.
  */
-static int power_pmu_enable(struct perf_event *event)
+static int power_pmu_add(struct perf_event *event, int ef_flags)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -735,7 +740,7 @@
 	int ret = -EAGAIN;
 
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 
 	/*
 	 * Add the event to the list (if there is room)
@@ -749,6 +754,9 @@
 	cpuhw->events[n0] = event->hw.config;
 	cpuhw->flags[n0] = event->hw.event_base;
 
+	if (!(ef_flags & PERF_EF_START))
+		event->hw.state = PERF_HES_STOPPED | PERF_HES_UPTODATE;
+
 	/*
 	 * If group events scheduling transaction was started,
 	 * skip the schedulability test here, it will be peformed
@@ -769,7 +777,7 @@
 
 	ret = 0;
  out:
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 	return ret;
 }
@@ -777,14 +785,14 @@
 /*
  * Remove a event from the PMU.
  */
-static void power_pmu_disable(struct perf_event *event)
+static void power_pmu_del(struct perf_event *event, int ef_flags)
 {
 	struct cpu_hw_events *cpuhw;
 	long i;
 	unsigned long flags;
 
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 
 	power_pmu_read(event);
 
@@ -821,34 +829,60 @@
 		cpuhw->mmcr[0] &= ~(MMCR0_PMXE | MMCR0_FCECE);
 	}
 
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 }
 
 /*
- * Re-enable interrupts on a event after they were throttled
- * because they were coming too fast.
+ * POWER-PMU does not support disabling individual counters, hence
+ * program their cycle counter to their max value and ignore the interrupts.
  */
-static void power_pmu_unthrottle(struct perf_event *event)
+
+static void power_pmu_start(struct perf_event *event, int ef_flags)
 {
-	s64 val, left;
+	unsigned long flags;
+	s64 left;
+
+	if (!event->hw.idx || !event->hw.sample_period)
+		return;
+
+	if (!(event->hw.state & PERF_HES_STOPPED))
+		return;
+
+	if (ef_flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+
+	local_irq_save(flags);
+	perf_pmu_disable(event->pmu);
+
+	event->hw.state = 0;
+	left = local64_read(&event->hw.period_left);
+	write_pmc(event->hw.idx, left);
+
+	perf_event_update_userpage(event);
+	perf_pmu_enable(event->pmu);
+	local_irq_restore(flags);
+}
+
+static void power_pmu_stop(struct perf_event *event, int ef_flags)
+{
 	unsigned long flags;
 
 	if (!event->hw.idx || !event->hw.sample_period)
 		return;
+
+	if (event->hw.state & PERF_HES_STOPPED)
+		return;
+
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
+
 	power_pmu_read(event);
-	left = event->hw.sample_period;
-	event->hw.last_period = left;
-	val = 0;
-	if (left < 0x80000000L)
-		val = 0x80000000L - left;
-	write_pmc(event->hw.idx, val);
-	local64_set(&event->hw.prev_count, val);
-	local64_set(&event->hw.period_left, left);
+	event->hw.state |= PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	write_pmc(event->hw.idx, 0);
+
 	perf_event_update_userpage(event);
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 }
 
@@ -857,10 +891,11 @@
  * Set the flag to make pmu::enable() not perform the
  * schedulability test, it will be performed at commit time
  */
-void power_pmu_start_txn(const struct pmu *pmu)
+void power_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
+	perf_pmu_disable(pmu);
 	cpuhw->group_flag |= PERF_EVENT_TXN;
 	cpuhw->n_txn_start = cpuhw->n_events;
 }
@@ -870,11 +905,12 @@
  * Clear the flag and pmu::enable() will perform the
  * schedulability test.
  */
-void power_pmu_cancel_txn(const struct pmu *pmu)
+void power_pmu_cancel_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
 	cpuhw->group_flag &= ~PERF_EVENT_TXN;
+	perf_pmu_enable(pmu);
 }
 
 /*
@@ -882,7 +918,7 @@
  * Perform the group schedulability test as a whole
  * Return 0 if success
  */
-int power_pmu_commit_txn(const struct pmu *pmu)
+int power_pmu_commit_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	long i, n;
@@ -901,19 +937,10 @@
 		cpuhw->event[i]->hw.config = cpuhw->events[i];
 
 	cpuhw->group_flag &= ~PERF_EVENT_TXN;
+	perf_pmu_enable(pmu);
 	return 0;
 }
 
-struct pmu power_pmu = {
-	.enable		= power_pmu_enable,
-	.disable	= power_pmu_disable,
-	.read		= power_pmu_read,
-	.unthrottle	= power_pmu_unthrottle,
-	.start_txn	= power_pmu_start_txn,
-	.cancel_txn	= power_pmu_cancel_txn,
-	.commit_txn	= power_pmu_commit_txn,
-};
-
 /*
  * Return 1 if we might be able to put event on a limited PMC,
  * or 0 if not.
@@ -1014,7 +1041,7 @@
 	return 0;
 }
 
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+static int power_pmu_event_init(struct perf_event *event)
 {
 	u64 ev;
 	unsigned long flags;
@@ -1026,25 +1053,27 @@
 	struct cpu_hw_events *cpuhw;
 
 	if (!ppmu)
-		return ERR_PTR(-ENXIO);
+		return -ENOENT;
+
 	switch (event->attr.type) {
 	case PERF_TYPE_HARDWARE:
 		ev = event->attr.config;
 		if (ev >= ppmu->n_generic || ppmu->generic_events[ev] == 0)
-			return ERR_PTR(-EOPNOTSUPP);
+			return -EOPNOTSUPP;
 		ev = ppmu->generic_events[ev];
 		break;
 	case PERF_TYPE_HW_CACHE:
 		err = hw_perf_cache_event(event->attr.config, &ev);
 		if (err)
-			return ERR_PTR(err);
+			return err;
 		break;
 	case PERF_TYPE_RAW:
 		ev = event->attr.config;
 		break;
 	default:
-		return ERR_PTR(-EINVAL);
+		return -ENOENT;
 	}
+
 	event->hw.config_base = ev;
 	event->hw.idx = 0;
 
@@ -1081,7 +1110,7 @@
 			 */
 			ev = normal_pmc_alternative(ev, flags);
 			if (!ev)
-				return ERR_PTR(-EINVAL);
+				return -EINVAL;
 		}
 	}
 
@@ -1095,19 +1124,19 @@
 		n = collect_events(event->group_leader, ppmu->n_counter - 1,
 				   ctrs, events, cflags);
 		if (n < 0)
-			return ERR_PTR(-EINVAL);
+			return -EINVAL;
 	}
 	events[n] = ev;
 	ctrs[n] = event;
 	cflags[n] = flags;
 	if (check_excludes(ctrs, cflags, n, 1))
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 
 	cpuhw = &get_cpu_var(cpu_hw_events);
 	err = power_check_constraints(cpuhw, events, cflags, n + 1);
 	put_cpu_var(cpu_hw_events);
 	if (err)
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 
 	event->hw.config = events[n];
 	event->hw.event_base = cflags[n];
@@ -1132,11 +1161,23 @@
 	}
 	event->destroy = hw_perf_event_destroy;
 
-	if (err)
-		return ERR_PTR(err);
-	return &power_pmu;
+	return err;
 }
 
+struct pmu power_pmu = {
+	.pmu_enable	= power_pmu_enable,
+	.pmu_disable	= power_pmu_disable,
+	.event_init	= power_pmu_event_init,
+	.add		= power_pmu_add,
+	.del		= power_pmu_del,
+	.start		= power_pmu_start,
+	.stop		= power_pmu_stop,
+	.read		= power_pmu_read,
+	.start_txn	= power_pmu_start_txn,
+	.cancel_txn	= power_pmu_cancel_txn,
+	.commit_txn	= power_pmu_commit_txn,
+};
+
 /*
  * A counter has overflowed; update its count and record
  * things if requested.  Note that interrupts are hard-disabled
@@ -1149,6 +1190,11 @@
 	s64 prev, delta, left;
 	int record = 0;
 
+	if (event->hw.state & PERF_HES_STOPPED) {
+		write_pmc(event->hw.idx, 0);
+		return;
+	}
+
 	/* we don't have to worry about interrupts here */
 	prev = local64_read(&event->hw.prev_count);
 	delta = (val - prev) & 0xfffffffful;
@@ -1171,6 +1217,11 @@
 			val = 0x80000000LL - left;
 	}
 
+	write_pmc(event->hw.idx, val);
+	local64_set(&event->hw.prev_count, val);
+	local64_set(&event->hw.period_left, left);
+	perf_event_update_userpage(event);
+
 	/*
 	 * Finally record data if requested.
 	 */
@@ -1183,23 +1234,9 @@
 		if (event->attr.sample_type & PERF_SAMPLE_ADDR)
 			perf_get_data_addr(regs, &data.addr);
 
-		if (perf_event_overflow(event, nmi, &data, regs)) {
-			/*
-			 * Interrupts are coming too fast - throttle them
-			 * by setting the event to 0, so it will be
-			 * at least 2^30 cycles until the next interrupt
-			 * (assuming each event counts at most 2 counts
-			 * per cycle).
-			 */
-			val = 0;
-			left = ~0ULL >> 1;
-		}
+		if (perf_event_overflow(event, nmi, &data, regs))
+			power_pmu_stop(event, 0);
 	}
-
-	write_pmc(event->hw.idx, val);
-	local64_set(&event->hw.prev_count, val);
-	local64_set(&event->hw.period_left, left);
-	perf_event_update_userpage(event);
 }
 
 /*
@@ -1342,6 +1379,7 @@
 		freeze_events_kernel = MMCR0_FCHV;
 #endif /* CONFIG_PPC64 */
 
+	perf_pmu_register(&power_pmu);
 	perf_cpu_notifier(power_pmu_notifier);
 
 	return 0;
diff --git a/arch/powerpc/kernel/perf_event_fsl_emb.c b/arch/powerpc/kernel/perf_event_fsl_emb.c
index 1ba4547..7ecca59 100644
--- a/arch/powerpc/kernel/perf_event_fsl_emb.c
+++ b/arch/powerpc/kernel/perf_event_fsl_emb.c
@@ -156,6 +156,9 @@
 {
 	s64 val, delta, prev;
 
+	if (event->hw.state & PERF_HES_STOPPED)
+		return;
+
 	/*
 	 * Performance monitor interrupts come even when interrupts
 	 * are soft-disabled, as long as interrupts are hard-enabled.
@@ -177,7 +180,7 @@
  * Disable all events to prevent PMU interrupts and to allow
  * events to be added or removed.
  */
-void hw_perf_disable(void)
+static void fsl_emb_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -216,7 +219,7 @@
  * If we were previously disabled and events were added, then
  * put the new config on the PMU.
  */
-void hw_perf_enable(void)
+static void fsl_emb_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw;
 	unsigned long flags;
@@ -262,8 +265,8 @@
 	return n;
 }
 
-/* perf must be disabled, context locked on entry */
-static int fsl_emb_pmu_enable(struct perf_event *event)
+/* context locked on entry */
+static int fsl_emb_pmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuhw;
 	int ret = -EAGAIN;
@@ -271,6 +274,7 @@
 	u64 val;
 	int i;
 
+	perf_pmu_disable(event->pmu);
 	cpuhw = &get_cpu_var(cpu_hw_events);
 
 	if (event->hw.config & FSL_EMB_EVENT_RESTRICTED)
@@ -301,6 +305,12 @@
 			val = 0x80000000L - left;
 	}
 	local64_set(&event->hw.prev_count, val);
+
+	if (!(flags & PERF_EF_START)) {
+		event->hw.state = PERF_HES_STOPPED | PERF_HES_UPTODATE;
+		val = 0;
+	}
+
 	write_pmc(i, val);
 	perf_event_update_userpage(event);
 
@@ -310,15 +320,17 @@
 	ret = 0;
  out:
 	put_cpu_var(cpu_hw_events);
+	perf_pmu_enable(event->pmu);
 	return ret;
 }
 
-/* perf must be disabled, context locked on entry */
-static void fsl_emb_pmu_disable(struct perf_event *event)
+/* context locked on entry */
+static void fsl_emb_pmu_del(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuhw;
 	int i = event->hw.idx;
 
+	perf_pmu_disable(event->pmu);
 	if (i < 0)
 		goto out;
 
@@ -346,45 +358,58 @@
 	cpuhw->n_events--;
 
  out:
+	perf_pmu_enable(event->pmu);
 	put_cpu_var(cpu_hw_events);
 }
 
-/*
- * Re-enable interrupts on a event after they were throttled
- * because they were coming too fast.
- *
- * Context is locked on entry, but perf is not disabled.
- */
-static void fsl_emb_pmu_unthrottle(struct perf_event *event)
+static void fsl_emb_pmu_start(struct perf_event *event, int ef_flags)
 {
-	s64 val, left;
+	unsigned long flags;
+	s64 left;
+
+	if (event->hw.idx < 0 || !event->hw.sample_period)
+		return;
+
+	if (!(event->hw.state & PERF_HES_STOPPED))
+		return;
+
+	if (ef_flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+
+	local_irq_save(flags);
+	perf_pmu_disable(event->pmu);
+
+	event->hw.state = 0;
+	left = local64_read(&event->hw.period_left);
+	write_pmc(event->hw.idx, left);
+
+	perf_event_update_userpage(event);
+	perf_pmu_enable(event->pmu);
+	local_irq_restore(flags);
+}
+
+static void fsl_emb_pmu_stop(struct perf_event *event, int ef_flags)
+{
 	unsigned long flags;
 
 	if (event->hw.idx < 0 || !event->hw.sample_period)
 		return;
+
+	if (event->hw.state & PERF_HES_STOPPED)
+		return;
+
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
+
 	fsl_emb_pmu_read(event);
-	left = event->hw.sample_period;
-	event->hw.last_period = left;
-	val = 0;
-	if (left < 0x80000000L)
-		val = 0x80000000L - left;
-	write_pmc(event->hw.idx, val);
-	local64_set(&event->hw.prev_count, val);
-	local64_set(&event->hw.period_left, left);
+	event->hw.state |= PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	write_pmc(event->hw.idx, 0);
+
 	perf_event_update_userpage(event);
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 }
 
-static struct pmu fsl_emb_pmu = {
-	.enable		= fsl_emb_pmu_enable,
-	.disable	= fsl_emb_pmu_disable,
-	.read		= fsl_emb_pmu_read,
-	.unthrottle	= fsl_emb_pmu_unthrottle,
-};
-
 /*
  * Release the PMU if this is the last perf_event.
  */
@@ -428,7 +453,7 @@
 	return 0;
 }
 
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+static int fsl_emb_pmu_event_init(struct perf_event *event)
 {
 	u64 ev;
 	struct perf_event *events[MAX_HWEVENTS];
@@ -441,14 +466,14 @@
 	case PERF_TYPE_HARDWARE:
 		ev = event->attr.config;
 		if (ev >= ppmu->n_generic || ppmu->generic_events[ev] == 0)
-			return ERR_PTR(-EOPNOTSUPP);
+			return -EOPNOTSUPP;
 		ev = ppmu->generic_events[ev];
 		break;
 
 	case PERF_TYPE_HW_CACHE:
 		err = hw_perf_cache_event(event->attr.config, &ev);
 		if (err)
-			return ERR_PTR(err);
+			return err;
 		break;
 
 	case PERF_TYPE_RAW:
@@ -456,12 +481,12 @@
 		break;
 
 	default:
-		return ERR_PTR(-EINVAL);
+		return -ENOENT;
 	}
 
 	event->hw.config = ppmu->xlate_event(ev);
 	if (!(event->hw.config & FSL_EMB_EVENT_VALID))
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 
 	/*
 	 * If this is in a group, check if it can go on with all the
@@ -473,7 +498,7 @@
 		n = collect_events(event->group_leader,
 		                   ppmu->n_counter - 1, events);
 		if (n < 0)
-			return ERR_PTR(-EINVAL);
+			return -EINVAL;
 	}
 
 	if (event->hw.config & FSL_EMB_EVENT_RESTRICTED) {
@@ -484,7 +509,7 @@
 		}
 
 		if (num_restricted >= ppmu->n_restricted)
-			return ERR_PTR(-EINVAL);
+			return -EINVAL;
 	}
 
 	event->hw.idx = -1;
@@ -497,7 +522,7 @@
 	if (event->attr.exclude_kernel)
 		event->hw.config_base |= PMLCA_FCS;
 	if (event->attr.exclude_idle)
-		return ERR_PTR(-ENOTSUPP);
+		return -ENOTSUPP;
 
 	event->hw.last_period = event->hw.sample_period;
 	local64_set(&event->hw.period_left, event->hw.last_period);
@@ -523,11 +548,20 @@
 	}
 	event->destroy = hw_perf_event_destroy;
 
-	if (err)
-		return ERR_PTR(err);
-	return &fsl_emb_pmu;
+	return err;
 }
 
+static struct pmu fsl_emb_pmu = {
+	.pmu_enable	= fsl_emb_pmu_enable,
+	.pmu_disable	= fsl_emb_pmu_disable,
+	.event_init	= fsl_emb_pmu_event_init,
+	.add		= fsl_emb_pmu_add,
+	.del		= fsl_emb_pmu_del,
+	.start		= fsl_emb_pmu_start,
+	.stop		= fsl_emb_pmu_stop,
+	.read		= fsl_emb_pmu_read,
+};
+
 /*
  * A counter has overflowed; update its count and record
  * things if requested.  Note that interrupts are hard-disabled
@@ -540,6 +574,11 @@
 	s64 prev, delta, left;
 	int record = 0;
 
+	if (event->hw.state & PERF_HES_STOPPED) {
+		write_pmc(event->hw.idx, 0);
+		return;
+	}
+
 	/* we don't have to worry about interrupts here */
 	prev = local64_read(&event->hw.prev_count);
 	delta = (val - prev) & 0xfffffffful;
@@ -562,6 +601,11 @@
 			val = 0x80000000LL - left;
 	}
 
+	write_pmc(event->hw.idx, val);
+	local64_set(&event->hw.prev_count, val);
+	local64_set(&event->hw.period_left, left);
+	perf_event_update_userpage(event);
+
 	/*
 	 * Finally record data if requested.
 	 */
@@ -571,23 +615,9 @@
 		perf_sample_data_init(&data, 0);
 		data.period = event->hw.last_period;
 
-		if (perf_event_overflow(event, nmi, &data, regs)) {
-			/*
-			 * Interrupts are coming too fast - throttle them
-			 * by setting the event to 0, so it will be
-			 * at least 2^30 cycles until the next interrupt
-			 * (assuming each event counts at most 2 counts
-			 * per cycle).
-			 */
-			val = 0;
-			left = ~0ULL >> 1;
-		}
+		if (perf_event_overflow(event, nmi, &data, regs))
+			fsl_emb_pmu_stop(event, 0);
 	}
-
-	write_pmc(event->hw.idx, val);
-	local64_set(&event->hw.prev_count, val);
-	local64_set(&event->hw.period_left, left);
-	perf_event_update_userpage(event);
 }
 
 static void perf_event_interrupt(struct pt_regs *regs)
@@ -651,5 +681,7 @@
 	pr_info("%s performance monitor hardware support registered\n",
 		pmu->name);
 
+	perf_pmu_register(&fsl_emb_pmu);
+
 	return 0;
 }
diff --git a/arch/sh/kernel/perf_callchain.c b/arch/sh/kernel/perf_callchain.c
index a9dd3ab..d5ca1ef 100644
--- a/arch/sh/kernel/perf_callchain.c
+++ b/arch/sh/kernel/perf_callchain.c
@@ -14,11 +14,6 @@
 #include <asm/unwinder.h>
 #include <asm/ptrace.h>
 
-static inline void callchain_store(struct perf_callchain_entry *entry, u64 ip)
-{
-	if (entry->nr < PERF_MAX_STACK_DEPTH)
-		entry->ip[entry->nr++] = ip;
-}
 
 static void callchain_warning(void *data, char *msg)
 {
@@ -39,7 +34,7 @@
 	struct perf_callchain_entry *entry = data;
 
 	if (reliable)
-		callchain_store(entry, addr);
+		perf_callchain_store(entry, addr);
 }
 
 static const struct stacktrace_ops callchain_ops = {
@@ -49,47 +44,10 @@
 	.address	= callchain_address,
 };
 
-static void
-perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry)
+void
+perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
-	callchain_store(entry, regs->pc);
+	perf_callchain_store(entry, regs->pc);
 
 	unwind_stack(NULL, regs, NULL, &callchain_ops, entry);
 }
-
-static void
-perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry)
-{
-	int is_user;
-
-	if (!regs)
-		return;
-
-	is_user = user_mode(regs);
-
-	if (is_user && current->state != TASK_RUNNING)
-		return;
-
-	/*
-	 * Only the kernel side is implemented for now.
-	 */
-	if (!is_user)
-		perf_callchain_kernel(regs, entry);
-}
-
-/*
- * No need for separate IRQ and NMI entries.
- */
-static DEFINE_PER_CPU(struct perf_callchain_entry, callchain);
-
-struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
-{
-	struct perf_callchain_entry *entry = &__get_cpu_var(callchain);
-
-	entry->nr = 0;
-
-	perf_do_callchain(regs, entry);
-
-	return entry;
-}
diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c
index 7a3dc35..036f7a9 100644
--- a/arch/sh/kernel/perf_event.c
+++ b/arch/sh/kernel/perf_event.c
@@ -206,50 +206,80 @@
 	local64_add(delta, &event->count);
 }
 
-static void sh_pmu_disable(struct perf_event *event)
+static void sh_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;
 	int idx = hwc->idx;
 
-	clear_bit(idx, cpuc->active_mask);
-	sh_pmu->disable(hwc, idx);
+	if (!(event->hw.state & PERF_HES_STOPPED)) {
+		sh_pmu->disable(hwc, idx);
+		cpuc->events[idx] = NULL;
+		event->hw.state |= PERF_HES_STOPPED;
+	}
 
-	barrier();
+	if ((flags & PERF_EF_UPDATE) && !(event->hw.state & PERF_HES_UPTODATE)) {
+		sh_perf_event_update(event, &event->hw, idx);
+		event->hw.state |= PERF_HES_UPTODATE;
+	}
+}
 
-	sh_perf_event_update(event, &event->hw, idx);
+static void sh_pmu_start(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	int idx = hwc->idx;
 
-	cpuc->events[idx] = NULL;
-	clear_bit(idx, cpuc->used_mask);
+	if (WARN_ON_ONCE(idx == -1))
+		return;
+
+	if (flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+
+	cpuc->events[idx] = event;
+	event->hw.state = 0;
+	sh_pmu->enable(hwc, idx);
+}
+
+static void sh_pmu_del(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+
+	sh_pmu_stop(event, PERF_EF_UPDATE);
+	__clear_bit(event->hw.idx, cpuc->used_mask);
 
 	perf_event_update_userpage(event);
 }
 
-static int sh_pmu_enable(struct perf_event *event)
+static int sh_pmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc = &event->hw;
 	int idx = hwc->idx;
+	int ret = -EAGAIN;
 
-	if (test_and_set_bit(idx, cpuc->used_mask)) {
+	perf_pmu_disable(event->pmu);
+
+	if (__test_and_set_bit(idx, cpuc->used_mask)) {
 		idx = find_first_zero_bit(cpuc->used_mask, sh_pmu->num_events);
 		if (idx == sh_pmu->num_events)
-			return -EAGAIN;
+			goto out;
 
-		set_bit(idx, cpuc->used_mask);
+		__set_bit(idx, cpuc->used_mask);
 		hwc->idx = idx;
 	}
 
 	sh_pmu->disable(hwc, idx);
 
-	cpuc->events[idx] = event;
-	set_bit(idx, cpuc->active_mask);
-
-	sh_pmu->enable(hwc, idx);
+	event->hw.state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+	if (flags & PERF_EF_START)
+		sh_pmu_start(event, PERF_EF_RELOAD);
 
 	perf_event_update_userpage(event);
-
-	return 0;
+	ret = 0;
+out:
+	perf_pmu_enable(event->pmu);
+	return ret;
 }
 
 static void sh_pmu_read(struct perf_event *event)
@@ -257,24 +287,56 @@
 	sh_perf_event_update(event, &event->hw, event->hw.idx);
 }
 
-static const struct pmu pmu = {
-	.enable		= sh_pmu_enable,
-	.disable	= sh_pmu_disable,
-	.read		= sh_pmu_read,
-};
-
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+static int sh_pmu_event_init(struct perf_event *event)
 {
-	int err = __hw_perf_event_init(event);
+	int err;
+
+	switch (event->attr.type) {
+	case PERF_TYPE_RAW:
+	case PERF_TYPE_HW_CACHE:
+	case PERF_TYPE_HARDWARE:
+		err = __hw_perf_event_init(event);
+		break;
+
+	default:
+		return -ENOENT;
+	}
+
 	if (unlikely(err)) {
 		if (event->destroy)
 			event->destroy(event);
-		return ERR_PTR(err);
 	}
 
-	return &pmu;
+	return err;
 }
 
+static void sh_pmu_enable(struct pmu *pmu)
+{
+	if (!sh_pmu_initialized())
+		return;
+
+	sh_pmu->enable_all();
+}
+
+static void sh_pmu_disable(struct pmu *pmu)
+{
+	if (!sh_pmu_initialized())
+		return;
+
+	sh_pmu->disable_all();
+}
+
+static struct pmu pmu = {
+	.pmu_enable	= sh_pmu_enable,
+	.pmu_disable	= sh_pmu_disable,
+	.event_init	= sh_pmu_event_init,
+	.add		= sh_pmu_add,
+	.del		= sh_pmu_del,
+	.start		= sh_pmu_start,
+	.stop		= sh_pmu_stop,
+	.read		= sh_pmu_read,
+};
+
 static void sh_pmu_setup(int cpu)
 {
 	struct cpu_hw_events *cpuhw = &per_cpu(cpu_hw_events, cpu);
@@ -299,32 +361,17 @@
 	return NOTIFY_OK;
 }
 
-void hw_perf_enable(void)
-{
-	if (!sh_pmu_initialized())
-		return;
-
-	sh_pmu->enable_all();
-}
-
-void hw_perf_disable(void)
-{
-	if (!sh_pmu_initialized())
-		return;
-
-	sh_pmu->disable_all();
-}
-
-int __cpuinit register_sh_pmu(struct sh_pmu *pmu)
+int __cpuinit register_sh_pmu(struct sh_pmu *_pmu)
 {
 	if (sh_pmu)
 		return -EBUSY;
-	sh_pmu = pmu;
+	sh_pmu = _pmu;
 
-	pr_info("Performance Events: %s support registered\n", pmu->name);
+	pr_info("Performance Events: %s support registered\n", _pmu->name);
 
-	WARN_ON(pmu->num_events > MAX_HWEVENTS);
+	WARN_ON(_pmu->num_events > MAX_HWEVENTS);
 
+	perf_pmu_register(&pmu);
 	perf_cpu_notifier(sh_pmu_notifier);
 	return 0;
 }
diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c
index 6318e62..0d6deb5 100644
--- a/arch/sparc/kernel/perf_event.c
+++ b/arch/sparc/kernel/perf_event.c
@@ -658,13 +658,16 @@
 
 		enc = perf_event_get_enc(cpuc->events[i]);
 		pcr &= ~mask_for_index(idx);
-		pcr |= event_encoding(enc, idx);
+		if (hwc->state & PERF_HES_STOPPED)
+			pcr |= nop_for_index(idx);
+		else
+			pcr |= event_encoding(enc, idx);
 	}
 out:
 	return pcr;
 }
 
-void hw_perf_enable(void)
+static void sparc_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	u64 pcr;
@@ -691,7 +694,7 @@
 	pcr_ops->write(cpuc->pcr);
 }
 
-void hw_perf_disable(void)
+static void sparc_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	u64 val;
@@ -710,48 +713,6 @@
 	pcr_ops->write(cpuc->pcr);
 }
 
-static void sparc_pmu_disable(struct perf_event *event)
-{
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-	struct hw_perf_event *hwc = &event->hw;
-	unsigned long flags;
-	int i;
-
-	local_irq_save(flags);
-	perf_disable();
-
-	for (i = 0; i < cpuc->n_events; i++) {
-		if (event == cpuc->event[i]) {
-			int idx = cpuc->current_idx[i];
-
-			/* Shift remaining entries down into
-			 * the existing slot.
-			 */
-			while (++i < cpuc->n_events) {
-				cpuc->event[i - 1] = cpuc->event[i];
-				cpuc->events[i - 1] = cpuc->events[i];
-				cpuc->current_idx[i - 1] =
-					cpuc->current_idx[i];
-			}
-
-			/* Absorb the final count and turn off the
-			 * event.
-			 */
-			sparc_pmu_disable_event(cpuc, hwc, idx);
-			barrier();
-			sparc_perf_event_update(event, hwc, idx);
-
-			perf_event_update_userpage(event);
-
-			cpuc->n_events--;
-			break;
-		}
-	}
-
-	perf_enable();
-	local_irq_restore(flags);
-}
-
 static int active_event_index(struct cpu_hw_events *cpuc,
 			      struct perf_event *event)
 {
@@ -765,6 +726,74 @@
 	return cpuc->current_idx[i];
 }
 
+static void sparc_pmu_start(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	int idx = active_event_index(cpuc, event);
+
+	if (flags & PERF_EF_RELOAD) {
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+		sparc_perf_event_set_period(event, &event->hw, idx);
+	}
+
+	event->hw.state = 0;
+
+	sparc_pmu_enable_event(cpuc, &event->hw, idx);
+}
+
+static void sparc_pmu_stop(struct perf_event *event, int flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	int idx = active_event_index(cpuc, event);
+
+	if (!(event->hw.state & PERF_HES_STOPPED)) {
+		sparc_pmu_disable_event(cpuc, &event->hw, idx);
+		event->hw.state |= PERF_HES_STOPPED;
+	}
+
+	if (!(event->hw.state & PERF_HES_UPTODATE) && (flags & PERF_EF_UPDATE)) {
+		sparc_perf_event_update(event, &event->hw, idx);
+		event->hw.state |= PERF_HES_UPTODATE;
+	}
+}
+
+static void sparc_pmu_del(struct perf_event *event, int _flags)
+{
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	unsigned long flags;
+	int i;
+
+	local_irq_save(flags);
+	perf_pmu_disable(event->pmu);
+
+	for (i = 0; i < cpuc->n_events; i++) {
+		if (event == cpuc->event[i]) {
+			/* Absorb the final count and turn off the
+			 * event.
+			 */
+			sparc_pmu_stop(event, PERF_EF_UPDATE);
+
+			/* Shift remaining entries down into
+			 * the existing slot.
+			 */
+			while (++i < cpuc->n_events) {
+				cpuc->event[i - 1] = cpuc->event[i];
+				cpuc->events[i - 1] = cpuc->events[i];
+				cpuc->current_idx[i - 1] =
+					cpuc->current_idx[i];
+			}
+
+			perf_event_update_userpage(event);
+
+			cpuc->n_events--;
+			break;
+		}
+	}
+
+	perf_pmu_enable(event->pmu);
+	local_irq_restore(flags);
+}
+
 static void sparc_pmu_read(struct perf_event *event)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
@@ -774,15 +803,6 @@
 	sparc_perf_event_update(event, hwc, idx);
 }
 
-static void sparc_pmu_unthrottle(struct perf_event *event)
-{
-	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-	int idx = active_event_index(cpuc, event);
-	struct hw_perf_event *hwc = &event->hw;
-
-	sparc_pmu_enable_event(cpuc, hwc, idx);
-}
-
 static atomic_t active_events = ATOMIC_INIT(0);
 static DEFINE_MUTEX(pmc_grab_mutex);
 
@@ -877,7 +897,7 @@
 	if (!n_ev)
 		return 0;
 
-	if (n_ev > perf_max_events)
+	if (n_ev > MAX_HWEVENTS)
 		return -1;
 
 	msk0 = perf_event_get_msk(events[0]);
@@ -984,23 +1004,27 @@
 	return n;
 }
 
-static int sparc_pmu_enable(struct perf_event *event)
+static int sparc_pmu_add(struct perf_event *event, int ef_flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int n0, ret = -EAGAIN;
 	unsigned long flags;
 
 	local_irq_save(flags);
-	perf_disable();
+	perf_pmu_disable(event->pmu);
 
 	n0 = cpuc->n_events;
-	if (n0 >= perf_max_events)
+	if (n0 >= MAX_HWEVENTS)
 		goto out;
 
 	cpuc->event[n0] = event;
 	cpuc->events[n0] = event->hw.event_base;
 	cpuc->current_idx[n0] = PIC_NO_INDEX;
 
+	event->hw.state = PERF_HES_UPTODATE;
+	if (!(ef_flags & PERF_EF_START))
+		event->hw.state |= PERF_HES_STOPPED;
+
 	/*
 	 * If group events scheduling transaction was started,
 	 * skip the schedulability test here, it will be peformed
@@ -1020,12 +1044,12 @@
 
 	ret = 0;
 out:
-	perf_enable();
+	perf_pmu_enable(event->pmu);
 	local_irq_restore(flags);
 	return ret;
 }
 
-static int __hw_perf_event_init(struct perf_event *event)
+static int sparc_pmu_event_init(struct perf_event *event)
 {
 	struct perf_event_attr *attr = &event->attr;
 	struct perf_event *evts[MAX_HWEVENTS];
@@ -1038,22 +1062,33 @@
 	if (atomic_read(&nmi_active) < 0)
 		return -ENODEV;
 
-	pmap = NULL;
-	if (attr->type == PERF_TYPE_HARDWARE) {
+	switch (attr->type) {
+	case PERF_TYPE_HARDWARE:
 		if (attr->config >= sparc_pmu->max_events)
 			return -EINVAL;
 		pmap = sparc_pmu->event_map(attr->config);
-	} else if (attr->type == PERF_TYPE_HW_CACHE) {
+		break;
+
+	case PERF_TYPE_HW_CACHE:
 		pmap = sparc_map_cache_event(attr->config);
 		if (IS_ERR(pmap))
 			return PTR_ERR(pmap);
-	} else if (attr->type != PERF_TYPE_RAW)
-		return -EOPNOTSUPP;
+		break;
+
+	case PERF_TYPE_RAW:
+		pmap = NULL;
+		break;
+
+	default:
+		return -ENOENT;
+
+	}
 
 	if (pmap) {
 		hwc->event_base = perf_event_encode(pmap);
 	} else {
-		/* User gives us "(encoding << 16) | pic_mask" for
+		/*
+		 * User gives us "(encoding << 16) | pic_mask" for
 		 * PERF_TYPE_RAW events.
 		 */
 		hwc->event_base = attr->config;
@@ -1071,7 +1106,7 @@
 	n = 0;
 	if (event->group_leader != event) {
 		n = collect_events(event->group_leader,
-				   perf_max_events - 1,
+				   MAX_HWEVENTS - 1,
 				   evts, events, current_idx_dmy);
 		if (n < 0)
 			return -EINVAL;
@@ -1107,10 +1142,11 @@
  * Set the flag to make pmu::enable() not perform the
  * schedulability test, it will be performed at commit time
  */
-static void sparc_pmu_start_txn(const struct pmu *pmu)
+static void sparc_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
+	perf_pmu_disable(pmu);
 	cpuhw->group_flag |= PERF_EVENT_TXN;
 }
 
@@ -1119,11 +1155,12 @@
  * Clear the flag and pmu::enable() will perform the
  * schedulability test.
  */
-static void sparc_pmu_cancel_txn(const struct pmu *pmu)
+static void sparc_pmu_cancel_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
 
 	cpuhw->group_flag &= ~PERF_EVENT_TXN;
+	perf_pmu_enable(pmu);
 }
 
 /*
@@ -1131,7 +1168,7 @@
  * Perform the group schedulability test as a whole
  * Return 0 if success
  */
-static int sparc_pmu_commit_txn(const struct pmu *pmu)
+static int sparc_pmu_commit_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int n;
@@ -1147,28 +1184,24 @@
 		return -EAGAIN;
 
 	cpuc->group_flag &= ~PERF_EVENT_TXN;
+	perf_pmu_enable(pmu);
 	return 0;
 }
 
-static const struct pmu pmu = {
-	.enable		= sparc_pmu_enable,
-	.disable	= sparc_pmu_disable,
+static struct pmu pmu = {
+	.pmu_enable	= sparc_pmu_enable,
+	.pmu_disable	= sparc_pmu_disable,
+	.event_init	= sparc_pmu_event_init,
+	.add		= sparc_pmu_add,
+	.del		= sparc_pmu_del,
+	.start		= sparc_pmu_start,
+	.stop		= sparc_pmu_stop,
 	.read		= sparc_pmu_read,
-	.unthrottle	= sparc_pmu_unthrottle,
 	.start_txn	= sparc_pmu_start_txn,
 	.cancel_txn	= sparc_pmu_cancel_txn,
 	.commit_txn	= sparc_pmu_commit_txn,
 };
 
-const struct pmu *hw_perf_event_init(struct perf_event *event)
-{
-	int err = __hw_perf_event_init(event);
-
-	if (err)
-		return ERR_PTR(err);
-	return &pmu;
-}
-
 void perf_event_print_debug(void)
 {
 	unsigned long flags;
@@ -1244,7 +1277,7 @@
 			continue;
 
 		if (perf_event_overflow(event, 1, &data, regs))
-			sparc_pmu_disable_event(cpuc, hwc, idx);
+			sparc_pmu_stop(event, 0);
 	}
 
 	return NOTIFY_STOP;
@@ -1285,28 +1318,21 @@
 
 	pr_cont("Supported PMU type is '%s'\n", sparc_pmu_type);
 
-	/* All sparc64 PMUs currently have 2 events.  */
-	perf_max_events = 2;
-
+	perf_pmu_register(&pmu);
 	register_die_notifier(&perf_event_nmi_notifier);
 }
 
-static inline void callchain_store(struct perf_callchain_entry *entry, u64 ip)
-{
-	if (entry->nr < PERF_MAX_STACK_DEPTH)
-		entry->ip[entry->nr++] = ip;
-}
-
-static void perf_callchain_kernel(struct pt_regs *regs,
-				  struct perf_callchain_entry *entry)
+void perf_callchain_kernel(struct perf_callchain_entry *entry,
+			   struct pt_regs *regs)
 {
 	unsigned long ksp, fp;
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 	int graph = 0;
 #endif
 
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
-	callchain_store(entry, regs->tpc);
+	stack_trace_flush();
+
+	perf_callchain_store(entry, regs->tpc);
 
 	ksp = regs->u_regs[UREG_I6];
 	fp = ksp + STACK_BIAS;
@@ -1330,13 +1356,13 @@
 			pc = sf->callers_pc;
 			fp = (unsigned long)sf->fp + STACK_BIAS;
 		}
-		callchain_store(entry, pc);
+		perf_callchain_store(entry, pc);
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 		if ((pc + 8UL) == (unsigned long) &return_to_handler) {
 			int index = current->curr_ret_stack;
 			if (current->ret_stack && index >= graph) {
 				pc = current->ret_stack[index - graph].ret;
-				callchain_store(entry, pc);
+				perf_callchain_store(entry, pc);
 				graph++;
 			}
 		}
@@ -1344,13 +1370,12 @@
 	} while (entry->nr < PERF_MAX_STACK_DEPTH);
 }
 
-static void perf_callchain_user_64(struct pt_regs *regs,
-				   struct perf_callchain_entry *entry)
+static void perf_callchain_user_64(struct perf_callchain_entry *entry,
+				   struct pt_regs *regs)
 {
 	unsigned long ufp;
 
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, regs->tpc);
+	perf_callchain_store(entry, regs->tpc);
 
 	ufp = regs->u_regs[UREG_I6] + STACK_BIAS;
 	do {
@@ -1363,17 +1388,16 @@
 
 		pc = sf.callers_pc;
 		ufp = (unsigned long)sf.fp + STACK_BIAS;
-		callchain_store(entry, pc);
+		perf_callchain_store(entry, pc);
 	} while (entry->nr < PERF_MAX_STACK_DEPTH);
 }
 
-static void perf_callchain_user_32(struct pt_regs *regs,
-				   struct perf_callchain_entry *entry)
+static void perf_callchain_user_32(struct perf_callchain_entry *entry,
+				   struct pt_regs *regs)
 {
 	unsigned long ufp;
 
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, regs->tpc);
+	perf_callchain_store(entry, regs->tpc);
 
 	ufp = regs->u_regs[UREG_I6] & 0xffffffffUL;
 	do {
@@ -1386,34 +1410,16 @@
 
 		pc = sf.callers_pc;
 		ufp = (unsigned long)sf.fp;
-		callchain_store(entry, pc);
+		perf_callchain_store(entry, pc);
 	} while (entry->nr < PERF_MAX_STACK_DEPTH);
 }
 
-/* Like powerpc we can't get PMU interrupts within the PMU handler,
- * so no need for separate NMI and IRQ chains as on x86.
- */
-static DEFINE_PER_CPU(struct perf_callchain_entry, callchain);
-
-struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
+void
+perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
-	struct perf_callchain_entry *entry = &__get_cpu_var(callchain);
-
-	entry->nr = 0;
-	if (!user_mode(regs)) {
-		stack_trace_flush();
-		perf_callchain_kernel(regs, entry);
-		if (current->mm)
-			regs = task_pt_regs(current);
-		else
-			regs = NULL;
-	}
-	if (regs) {
-		flushw_user();
-		if (test_thread_flag(TIF_32BIT))
-			perf_callchain_user_32(regs, entry);
-		else
-			perf_callchain_user_64(regs, entry);
-	}
-	return entry;
+	flushw_user();
+	if (test_thread_flag(TIF_32BIT))
+		perf_callchain_user_32(entry, regs);
+	else
+		perf_callchain_user_64(entry, regs);
 }
diff --git a/arch/x86/include/asm/perf_event_p4.h b/arch/x86/include/asm/perf_event_p4.h
index def5007..a70cd21 100644
--- a/arch/x86/include/asm/perf_event_p4.h
+++ b/arch/x86/include/asm/perf_event_p4.h
@@ -36,19 +36,6 @@
 #define P4_ESCR_EMASK(v)	((v) << P4_ESCR_EVENTMASK_SHIFT)
 #define P4_ESCR_TAG(v)		((v) << P4_ESCR_TAG_SHIFT)
 
-/* Non HT mask */
-#define P4_ESCR_MASK			\
-	(P4_ESCR_EVENT_MASK	|	\
-	P4_ESCR_EVENTMASK_MASK	|	\
-	P4_ESCR_TAG_MASK	|	\
-	P4_ESCR_TAG_ENABLE	|	\
-	P4_ESCR_T0_OS		|	\
-	P4_ESCR_T0_USR)
-
-/* HT mask */
-#define P4_ESCR_MASK_HT			\
-	(P4_ESCR_MASK |	P4_ESCR_T1_OS | P4_ESCR_T1_USR)
-
 #define P4_CCCR_OVF			0x80000000U
 #define P4_CCCR_CASCADE			0x40000000U
 #define P4_CCCR_OVF_PMI_T0		0x04000000U
@@ -70,23 +57,6 @@
 #define P4_CCCR_THRESHOLD(v)		((v) << P4_CCCR_THRESHOLD_SHIFT)
 #define P4_CCCR_ESEL(v)			((v) << P4_CCCR_ESCR_SELECT_SHIFT)
 
-/* Non HT mask */
-#define P4_CCCR_MASK				\
-	(P4_CCCR_OVF			|	\
-	P4_CCCR_CASCADE			|	\
-	P4_CCCR_OVF_PMI_T0		|	\
-	P4_CCCR_FORCE_OVF		|	\
-	P4_CCCR_EDGE			|	\
-	P4_CCCR_THRESHOLD_MASK		|	\
-	P4_CCCR_COMPLEMENT		|	\
-	P4_CCCR_COMPARE			|	\
-	P4_CCCR_ESCR_SELECT_MASK	|	\
-	P4_CCCR_ENABLE)
-
-/* HT mask */
-#define P4_CCCR_MASK_HT				\
-	(P4_CCCR_MASK | P4_CCCR_OVF_PMI_T1 | P4_CCCR_THREAD_ANY)
-
 #define P4_GEN_ESCR_EMASK(class, name, bit)	\
 	class##__##name = ((1 << bit) << P4_ESCR_EVENTMASK_SHIFT)
 #define P4_ESCR_EMASK_BIT(class, name)		class##__##name
@@ -127,6 +97,28 @@
 #define P4_CONFIG_HT_SHIFT		63
 #define P4_CONFIG_HT			(1ULL << P4_CONFIG_HT_SHIFT)
 
+/*
+ * The bits we allow to pass for RAW events
+ */
+#define P4_CONFIG_MASK_ESCR		\
+	P4_ESCR_EVENT_MASK	|	\
+	P4_ESCR_EVENTMASK_MASK	|	\
+	P4_ESCR_TAG_MASK	|	\
+	P4_ESCR_TAG_ENABLE
+
+#define P4_CONFIG_MASK_CCCR		\
+	P4_CCCR_EDGE		|	\
+	P4_CCCR_THRESHOLD_MASK	|	\
+	P4_CCCR_COMPLEMENT	|	\
+	P4_CCCR_COMPARE		|	\
+	P4_CCCR_THREAD_ANY	|	\
+	P4_CCCR_RESERVED
+
+/* some dangerous bits are reserved for kernel internals */
+#define P4_CONFIG_MASK				  	  \
+	(p4_config_pack_escr(P4_CONFIG_MASK_ESCR))	| \
+	(p4_config_pack_cccr(P4_CONFIG_MASK_CCCR))
+
 static inline bool p4_is_event_cascaded(u64 config)
 {
 	u32 cccr = p4_config_unpack_cccr(config);
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 3efdf28..0fb1705 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -530,7 +530,7 @@
 /*
  * Setup the hardware configuration for a given attr_type
  */
-static int __hw_perf_event_init(struct perf_event *event)
+static int __x86_pmu_event_init(struct perf_event *event)
 {
 	int err;
 
@@ -583,7 +583,7 @@
 	}
 }
 
-void hw_perf_disable(void)
+static void x86_pmu_disable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -618,7 +618,7 @@
 	}
 }
 
-static const struct pmu pmu;
+static struct pmu pmu;
 
 static inline int is_x86_event(struct perf_event *event)
 {
@@ -800,10 +800,10 @@
 		hwc->last_tag == cpuc->tags[i];
 }
 
-static int x86_pmu_start(struct perf_event *event);
-static void x86_pmu_stop(struct perf_event *event);
+static void x86_pmu_start(struct perf_event *event, int flags);
+static void x86_pmu_stop(struct perf_event *event, int flags);
 
-void hw_perf_enable(void)
+static void x86_pmu_enable(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct perf_event *event;
@@ -839,7 +839,14 @@
 			    match_prev_assignment(hwc, cpuc, i))
 				continue;
 
-			x86_pmu_stop(event);
+			/*
+			 * Ensure we don't accidentally enable a stopped
+			 * counter simply because we rescheduled.
+			 */
+			if (hwc->state & PERF_HES_STOPPED)
+				hwc->state |= PERF_HES_ARCH;
+
+			x86_pmu_stop(event, PERF_EF_UPDATE);
 		}
 
 		for (i = 0; i < cpuc->n_events; i++) {
@@ -851,7 +858,10 @@
 			else if (i < n_running)
 				continue;
 
-			x86_pmu_start(event);
+			if (hwc->state & PERF_HES_ARCH)
+				continue;
+
+			x86_pmu_start(event, PERF_EF_RELOAD);
 		}
 		cpuc->n_added = 0;
 		perf_events_lapic_init();
@@ -952,15 +962,12 @@
 }
 
 /*
- * activate a single event
+ * Add a single event to the PMU.
  *
  * The event is added to the group of enabled events
  * but only if it can be scehduled with existing events.
- *
- * Called with PMU disabled. If successful and return value 1,
- * then guaranteed to call perf_enable() and hw_perf_enable()
  */
-static int x86_pmu_enable(struct perf_event *event)
+static int x86_pmu_add(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct hw_perf_event *hwc;
@@ -969,57 +976,66 @@
 
 	hwc = &event->hw;
 
+	perf_pmu_disable(event->pmu);
 	n0 = cpuc->n_events;
-	n = collect_events(cpuc, event, false);
-	if (n < 0)
-		return n;
+	ret = n = collect_events(cpuc, event, false);
+	if (ret < 0)
+		goto out;
+
+	hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED;
+	if (!(flags & PERF_EF_START))
+		hwc->state |= PERF_HES_ARCH;
 
 	/*
 	 * If group events scheduling transaction was started,
 	 * skip the schedulability test here, it will be peformed
-	 * at commit time(->commit_txn) as a whole
+	 * at commit time (->commit_txn) as a whole
 	 */
 	if (cpuc->group_flag & PERF_EVENT_TXN)
-		goto out;
+		goto done_collect;
 
 	ret = x86_pmu.schedule_events(cpuc, n, assign);
 	if (ret)
-		return ret;
+		goto out;
 	/*
 	 * copy new assignment, now we know it is possible
 	 * will be used by hw_perf_enable()
 	 */
 	memcpy(cpuc->assign, assign, n*sizeof(int));
 
-out:
+done_collect:
 	cpuc->n_events = n;
 	cpuc->n_added += n - n0;
 	cpuc->n_txn += n - n0;
 
-	return 0;
+	ret = 0;
+out:
+	perf_pmu_enable(event->pmu);
+	return ret;
 }
 
-static int x86_pmu_start(struct perf_event *event)
+static void x86_pmu_start(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int idx = event->hw.idx;
 
-	if (idx == -1)
-		return -EAGAIN;
+	if (WARN_ON_ONCE(!(event->hw.state & PERF_HES_STOPPED)))
+		return;
 
-	x86_perf_event_set_period(event);
+	if (WARN_ON_ONCE(idx == -1))
+		return;
+
+	if (flags & PERF_EF_RELOAD) {
+		WARN_ON_ONCE(!(event->hw.state & PERF_HES_UPTODATE));
+		x86_perf_event_set_period(event);
+	}
+
+	event->hw.state = 0;
+
 	cpuc->events[idx] = event;
 	__set_bit(idx, cpuc->active_mask);
 	x86_pmu.enable(event);
 	perf_event_update_userpage(event);
-
-	return 0;
-}
-
-static void x86_pmu_unthrottle(struct perf_event *event)
-{
-	int ret = x86_pmu_start(event);
-	WARN_ON_ONCE(ret);
 }
 
 void perf_event_print_debug(void)
@@ -1076,27 +1092,29 @@
 	local_irq_restore(flags);
 }
 
-static void x86_pmu_stop(struct perf_event *event)
+static 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;
-	int idx = hwc->idx;
 
-	if (!__test_and_clear_bit(idx, cpuc->active_mask))
-		return;
+	if (__test_and_clear_bit(hwc->idx, cpuc->active_mask)) {
+		x86_pmu.disable(event);
+		cpuc->events[hwc->idx] = NULL;
+		WARN_ON_ONCE(hwc->state & PERF_HES_STOPPED);
+		hwc->state |= PERF_HES_STOPPED;
+	}
 
-	x86_pmu.disable(event);
-
-	/*
-	 * Drain the remaining delta count out of a event
-	 * that we are disabling:
-	 */
-	x86_perf_event_update(event);
-
-	cpuc->events[idx] = NULL;
+	if ((flags & PERF_EF_UPDATE) && !(hwc->state & PERF_HES_UPTODATE)) {
+		/*
+		 * Drain the remaining delta count out of a event
+		 * that we are disabling:
+		 */
+		x86_perf_event_update(event);
+		hwc->state |= PERF_HES_UPTODATE;
+	}
 }
 
-static void x86_pmu_disable(struct perf_event *event)
+static void x86_pmu_del(struct perf_event *event, int flags)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int i;
@@ -1109,7 +1127,7 @@
 	if (cpuc->group_flag & PERF_EVENT_TXN)
 		return;
 
-	x86_pmu_stop(event);
+	x86_pmu_stop(event, PERF_EF_UPDATE);
 
 	for (i = 0; i < cpuc->n_events; i++) {
 		if (event == cpuc->event_list[i]) {
@@ -1161,7 +1179,7 @@
 			continue;
 
 		if (perf_event_overflow(event, 1, &data, regs))
-			x86_pmu_stop(event);
+			x86_pmu_stop(event, 0);
 	}
 
 	if (handled)
@@ -1378,7 +1396,6 @@
 		x86_pmu.num_counters = X86_PMC_MAX_GENERIC;
 	}
 	x86_pmu.intel_ctrl = (1 << x86_pmu.num_counters) - 1;
-	perf_max_events = x86_pmu.num_counters;
 
 	if (x86_pmu.num_counters_fixed > X86_PMC_MAX_FIXED) {
 		WARN(1, KERN_ERR "hw perf events fixed %d > max(%d), clipping!",
@@ -1414,6 +1431,7 @@
 	pr_info("... fixed-purpose events:   %d\n",     x86_pmu.num_counters_fixed);
 	pr_info("... event mask:             %016Lx\n", x86_pmu.intel_ctrl);
 
+	perf_pmu_register(&pmu);
 	perf_cpu_notifier(x86_pmu_notifier);
 }
 
@@ -1427,10 +1445,11 @@
  * Set the flag to make pmu::enable() not perform the
  * schedulability test, it will be performed at commit time
  */
-static void x86_pmu_start_txn(const struct pmu *pmu)
+static void x86_pmu_start_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
+	perf_pmu_disable(pmu);
 	cpuc->group_flag |= PERF_EVENT_TXN;
 	cpuc->n_txn = 0;
 }
@@ -1440,7 +1459,7 @@
  * Clear the flag and pmu::enable() will perform the
  * schedulability test.
  */
-static void x86_pmu_cancel_txn(const struct pmu *pmu)
+static void x86_pmu_cancel_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 
@@ -1450,6 +1469,7 @@
 	 */
 	cpuc->n_added -= cpuc->n_txn;
 	cpuc->n_events -= cpuc->n_txn;
+	perf_pmu_enable(pmu);
 }
 
 /*
@@ -1457,7 +1477,7 @@
  * Perform the group schedulability test as a whole
  * Return 0 if success
  */
-static int x86_pmu_commit_txn(const struct pmu *pmu)
+static int x86_pmu_commit_txn(struct pmu *pmu)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	int assign[X86_PMC_IDX_MAX];
@@ -1479,22 +1499,10 @@
 	memcpy(cpuc->assign, assign, n*sizeof(int));
 
 	cpuc->group_flag &= ~PERF_EVENT_TXN;
-
+	perf_pmu_enable(pmu);
 	return 0;
 }
 
-static const struct pmu pmu = {
-	.enable		= x86_pmu_enable,
-	.disable	= x86_pmu_disable,
-	.start		= x86_pmu_start,
-	.stop		= x86_pmu_stop,
-	.read		= x86_pmu_read,
-	.unthrottle	= x86_pmu_unthrottle,
-	.start_txn	= x86_pmu_start_txn,
-	.cancel_txn	= x86_pmu_cancel_txn,
-	.commit_txn	= x86_pmu_commit_txn,
-};
-
 /*
  * validate that we can schedule this event
  */
@@ -1569,12 +1577,22 @@
 	return ret;
 }
 
-const struct pmu *hw_perf_event_init(struct perf_event *event)
+int x86_pmu_event_init(struct perf_event *event)
 {
-	const struct pmu *tmp;
+	struct pmu *tmp;
 	int err;
 
-	err = __hw_perf_event_init(event);
+	switch (event->attr.type) {
+	case PERF_TYPE_RAW:
+	case PERF_TYPE_HARDWARE:
+	case PERF_TYPE_HW_CACHE:
+		break;
+
+	default:
+		return -ENOENT;
+	}
+
+	err = __x86_pmu_event_init(event);
 	if (!err) {
 		/*
 		 * we temporarily connect event to its pmu
@@ -1594,27 +1612,32 @@
 	if (err) {
 		if (event->destroy)
 			event->destroy(event);
-		return ERR_PTR(err);
 	}
 
-	return &pmu;
+	return err;
 }
 
+static struct pmu pmu = {
+	.pmu_enable	= x86_pmu_enable,
+	.pmu_disable	= x86_pmu_disable,
+
+	.event_init	= x86_pmu_event_init,
+
+	.add		= x86_pmu_add,
+	.del		= x86_pmu_del,
+	.start		= x86_pmu_start,
+	.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,
+};
+
 /*
  * callchain support
  */
 
-static inline
-void callchain_store(struct perf_callchain_entry *entry, u64 ip)
-{
-	if (entry->nr < PERF_MAX_STACK_DEPTH)
-		entry->ip[entry->nr++] = ip;
-}
-
-static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_irq_entry);
-static DEFINE_PER_CPU(struct perf_callchain_entry, pmc_nmi_entry);
-
-
 static void
 backtrace_warning_symbol(void *data, char *msg, unsigned long symbol)
 {
@@ -1635,7 +1658,7 @@
 {
 	struct perf_callchain_entry *entry = data;
 
-	callchain_store(entry, addr);
+	perf_callchain_store(entry, addr);
 }
 
 static const struct stacktrace_ops backtrace_ops = {
@@ -1646,11 +1669,15 @@
 	.walk_stack		= print_context_stack_bp,
 };
 
-static void
-perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry)
+void
+perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
-	callchain_store(entry, PERF_CONTEXT_KERNEL);
-	callchain_store(entry, regs->ip);
+	if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) {
+		/* TODO: We don't support guest os callchain now */
+		return;
+	}
+
+	perf_callchain_store(entry, regs->ip);
 
 	dump_trace(NULL, regs, NULL, regs->bp, &backtrace_ops, entry);
 }
@@ -1679,7 +1706,7 @@
 		if (fp < compat_ptr(regs->sp))
 			break;
 
-		callchain_store(entry, frame.return_address);
+		perf_callchain_store(entry, frame.return_address);
 		fp = compat_ptr(frame.next_frame);
 	}
 	return 1;
@@ -1692,19 +1719,20 @@
 }
 #endif
 
-static void
-perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry)
+void
+perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 {
 	struct stack_frame frame;
 	const void __user *fp;
 
-	if (!user_mode(regs))
-		regs = task_pt_regs(current);
+	if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) {
+		/* TODO: We don't support guest os callchain now */
+		return;
+	}
 
 	fp = (void __user *)regs->bp;
 
-	callchain_store(entry, PERF_CONTEXT_USER);
-	callchain_store(entry, regs->ip);
+	perf_callchain_store(entry, regs->ip);
 
 	if (perf_callchain_user32(regs, entry))
 		return;
@@ -1721,52 +1749,11 @@
 		if ((unsigned long)fp < regs->sp)
 			break;
 
-		callchain_store(entry, frame.return_address);
+		perf_callchain_store(entry, frame.return_address);
 		fp = frame.next_frame;
 	}
 }
 
-static void
-perf_do_callchain(struct pt_regs *regs, struct perf_callchain_entry *entry)
-{
-	int is_user;
-
-	if (!regs)
-		return;
-
-	is_user = user_mode(regs);
-
-	if (is_user && current->state != TASK_RUNNING)
-		return;
-
-	if (!is_user)
-		perf_callchain_kernel(regs, entry);
-
-	if (current->mm)
-		perf_callchain_user(regs, entry);
-}
-
-struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
-{
-	struct perf_callchain_entry *entry;
-
-	if (perf_guest_cbs && perf_guest_cbs->is_in_guest()) {
-		/* TODO: We don't support guest os callchain now */
-		return NULL;
-	}
-
-	if (in_nmi())
-		entry = &__get_cpu_var(pmc_nmi_entry);
-	else
-		entry = &__get_cpu_var(pmc_irq_entry);
-
-	entry->nr = 0;
-
-	perf_do_callchain(regs, entry);
-
-	return entry;
-}
-
 unsigned long perf_instruction_pointer(struct pt_regs *regs)
 {
 	unsigned long ip;
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index ee05c90..c8f5c08 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -713,18 +713,18 @@
 	struct cpu_hw_events *cpuc;
 	int bit, loops;
 	u64 status;
-	int handled = 0;
+	int handled;
 
 	perf_sample_data_init(&data, 0);
 
 	cpuc = &__get_cpu_var(cpu_hw_events);
 
 	intel_pmu_disable_all();
-	intel_pmu_drain_bts_buffer();
+	handled = intel_pmu_drain_bts_buffer();
 	status = intel_pmu_get_status();
 	if (!status) {
 		intel_pmu_enable_all(0);
-		return 0;
+		return handled;
 	}
 
 	loops = 0;
@@ -763,7 +763,7 @@
 		data.period = event->hw.last_period;
 
 		if (perf_event_overflow(event, 1, &data, regs))
-			x86_pmu_stop(event);
+			x86_pmu_stop(event, 0);
 	}
 
 	/*
diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c
index 18018d1..4977f9c 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_ds.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c
@@ -214,7 +214,7 @@
 	update_debugctlmsr(debugctlmsr);
 }
 
-static void intel_pmu_drain_bts_buffer(void)
+static int intel_pmu_drain_bts_buffer(void)
 {
 	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
 	struct debug_store *ds = cpuc->ds;
@@ -231,16 +231,16 @@
 	struct pt_regs regs;
 
 	if (!event)
-		return;
+		return 0;
 
 	if (!ds)
-		return;
+		return 0;
 
 	at  = (struct bts_record *)(unsigned long)ds->bts_buffer_base;
 	top = (struct bts_record *)(unsigned long)ds->bts_index;
 
 	if (top <= at)
-		return;
+		return 0;
 
 	ds->bts_index = ds->bts_buffer_base;
 
@@ -256,7 +256,7 @@
 	perf_prepare_sample(&header, &data, event, &regs);
 
 	if (perf_output_begin(&handle, event, header.size * (top - at), 1, 1))
-		return;
+		return 1;
 
 	for (; at < top; at++) {
 		data.ip		= at->from;
@@ -270,6 +270,7 @@
 	/* There's new data available. */
 	event->hw.interrupts++;
 	event->pending_kill = POLL_IN;
+	return 1;
 }
 
 /*
@@ -491,7 +492,7 @@
 		regs.flags &= ~PERF_EFLAGS_EXACT;
 
 	if (perf_event_overflow(event, 1, &data, &regs))
-		x86_pmu_stop(event);
+		x86_pmu_stop(event, 0);
 }
 
 static void intel_pmu_drain_pebs_core(struct pt_regs *iregs)
diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c
index b560db3..c70c878 100644
--- a/arch/x86/kernel/cpu/perf_event_p4.c
+++ b/arch/x86/kernel/cpu/perf_event_p4.c
@@ -18,6 +18,8 @@
 struct p4_event_bind {
 	unsigned int opcode;			/* Event code and ESCR selector */
 	unsigned int escr_msr[2];		/* ESCR MSR for this event */
+	unsigned int escr_emask;		/* valid ESCR EventMask bits */
+	unsigned int shared;			/* event is shared across threads */
 	char cntr[2][P4_CNTR_LIMIT];		/* counter index (offset), -1 on abscence */
 };
 
@@ -66,231 +68,435 @@
 	[P4_EVENT_TC_DELIVER_MODE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_TC_DELIVER_MODE),
 		.escr_msr	= { MSR_P4_TC_ESCR0, MSR_P4_TC_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DD)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DB)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, DI)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BD)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BB)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, BI)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_DELIVER_MODE, ID),
+		.shared		= 1,
 		.cntr		= { {4, 5, -1}, {6, 7, -1} },
 	},
 	[P4_EVENT_BPU_FETCH_REQUEST] = {
 		.opcode		= P4_OPCODE(P4_EVENT_BPU_FETCH_REQUEST),
 		.escr_msr	= { MSR_P4_BPU_ESCR0, MSR_P4_BPU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BPU_FETCH_REQUEST, TCMISS),
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_ITLB_REFERENCE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_ITLB_REFERENCE),
 		.escr_msr	= { MSR_P4_ITLB_ESCR0, MSR_P4_ITLB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, HIT)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, MISS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_ITLB_REFERENCE, HIT_UK),
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_MEMORY_CANCEL] = {
 		.opcode		= P4_OPCODE(P4_EVENT_MEMORY_CANCEL),
 		.escr_msr	= { MSR_P4_DAC_ESCR0, MSR_P4_DAC_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_CANCEL, ST_RB_FULL)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_CANCEL, 64K_CONF),
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_MEMORY_COMPLETE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_MEMORY_COMPLETE),
 		.escr_msr	= { MSR_P4_SAAT_ESCR0 , MSR_P4_SAAT_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_COMPLETE, LSC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MEMORY_COMPLETE, SSC),
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_LOAD_PORT_REPLAY] = {
 		.opcode		= P4_OPCODE(P4_EVENT_LOAD_PORT_REPLAY),
 		.escr_msr	= { MSR_P4_SAAT_ESCR0, MSR_P4_SAAT_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_LOAD_PORT_REPLAY, SPLIT_LD),
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_STORE_PORT_REPLAY] = {
 		.opcode		= P4_OPCODE(P4_EVENT_STORE_PORT_REPLAY),
 		.escr_msr	= { MSR_P4_SAAT_ESCR0 ,  MSR_P4_SAAT_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_STORE_PORT_REPLAY, SPLIT_ST),
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_MOB_LOAD_REPLAY] = {
 		.opcode		= P4_OPCODE(P4_EVENT_MOB_LOAD_REPLAY),
 		.escr_msr	= { MSR_P4_MOB_ESCR0, MSR_P4_MOB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, NO_STA)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, NO_STD)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, PARTIAL_DATA)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MOB_LOAD_REPLAY, UNALGN_ADDR),
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_PAGE_WALK_TYPE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_PAGE_WALK_TYPE),
 		.escr_msr	= { MSR_P4_PMH_ESCR0, MSR_P4_PMH_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_PAGE_WALK_TYPE, DTMISS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_PAGE_WALK_TYPE, ITMISS),
+		.shared		= 1,
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_BSQ_CACHE_REFERENCE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_BSQ_CACHE_REFERENCE),
 		.escr_msr	= { MSR_P4_BSU_ESCR0, MSR_P4_BSU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITS)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_HITM)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITS)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_HITM)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_2ndL_MISS)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, RD_3rdL_MISS)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_CACHE_REFERENCE, WR_2ndL_MISS),
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_IOQ_ALLOCATION] = {
 		.opcode		= P4_OPCODE(P4_EVENT_IOQ_ALLOCATION),
 		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, DEFAULT)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, ALL_READ)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, ALL_WRITE)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_UC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WT)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WP)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, MEM_WB)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, OWN)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, OTHER)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ALLOCATION, PREFETCH),
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_IOQ_ACTIVE_ENTRIES] = {	/* shared ESCR */
 		.opcode		= P4_OPCODE(P4_EVENT_IOQ_ACTIVE_ENTRIES),
 		.escr_msr	= { MSR_P4_FSB_ESCR1,  MSR_P4_FSB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, DEFAULT)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, ALL_READ)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, ALL_WRITE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_UC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WC)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WT)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WP)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, MEM_WB)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, OWN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, OTHER)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_IOQ_ACTIVE_ENTRIES, PREFETCH),
 		.cntr		= { {2, -1, -1}, {3, -1, -1} },
 	},
 	[P4_EVENT_FSB_DATA_ACTIVITY] = {
 		.opcode		= P4_OPCODE(P4_EVENT_FSB_DATA_ACTIVITY),
 		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_DRV)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OWN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DRDY_OTHER)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_DRV)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_OWN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FSB_DATA_ACTIVITY, DBSY_OTHER),
+		.shared		= 1,
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_BSQ_ALLOCATION] = {		/* shared ESCR, broken CCCR1 */
 		.opcode		= P4_OPCODE(P4_EVENT_BSQ_ALLOCATION),
 		.escr_msr	= { MSR_P4_BSU_ESCR0, MSR_P4_BSU_ESCR0 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_TYPE0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_TYPE1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LEN0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LEN1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_IO_TYPE)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_LOCK_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_CACHE_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_SPLIT_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_DEM_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, REQ_ORD_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ALLOCATION, MEM_TYPE2),
 		.cntr		= { {0, -1, -1}, {1, -1, -1} },
 	},
 	[P4_EVENT_BSQ_ACTIVE_ENTRIES] = {	/* shared ESCR */
 		.opcode		= P4_OPCODE(P4_EVENT_BSQ_ACTIVE_ENTRIES),
 		.escr_msr	= { MSR_P4_BSU_ESCR1 , MSR_P4_BSU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_TYPE0)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_TYPE1)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LEN0)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LEN1)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_IO_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_LOCK_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_CACHE_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_SPLIT_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_DEM_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, REQ_ORD_TYPE)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE0)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE1)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BSQ_ACTIVE_ENTRIES, MEM_TYPE2),
 		.cntr		= { {2, -1, -1}, {3, -1, -1} },
 	},
 	[P4_EVENT_SSE_INPUT_ASSIST] = {
 		.opcode		= P4_OPCODE(P4_EVENT_SSE_INPUT_ASSIST),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_SSE_INPUT_ASSIST, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_PACKED_SP_UOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_PACKED_SP_UOP),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_PACKED_SP_UOP, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_PACKED_DP_UOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_PACKED_DP_UOP),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_PACKED_DP_UOP, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_SCALAR_SP_UOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_SCALAR_SP_UOP),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_SCALAR_SP_UOP, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_SCALAR_DP_UOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_SCALAR_DP_UOP),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_SCALAR_DP_UOP, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_64BIT_MMX_UOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_64BIT_MMX_UOP),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_64BIT_MMX_UOP, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_128BIT_MMX_UOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_128BIT_MMX_UOP),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_128BIT_MMX_UOP, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_X87_FP_UOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_X87_FP_UOP),
 		.escr_msr	= { MSR_P4_FIRM_ESCR0, MSR_P4_FIRM_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_FP_UOP, ALL),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_TC_MISC] = {
 		.opcode		= P4_OPCODE(P4_EVENT_TC_MISC),
 		.escr_msr	= { MSR_P4_TC_ESCR0, MSR_P4_TC_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_MISC, FLUSH),
 		.cntr		= { {4, 5, -1}, {6, 7, -1} },
 	},
 	[P4_EVENT_GLOBAL_POWER_EVENTS] = {
 		.opcode		= P4_OPCODE(P4_EVENT_GLOBAL_POWER_EVENTS),
 		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_GLOBAL_POWER_EVENTS, RUNNING),
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_TC_MS_XFER] = {
 		.opcode		= P4_OPCODE(P4_EVENT_TC_MS_XFER),
 		.escr_msr	= { MSR_P4_MS_ESCR0, MSR_P4_MS_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_TC_MS_XFER, CISC),
 		.cntr		= { {4, 5, -1}, {6, 7, -1} },
 	},
 	[P4_EVENT_UOP_QUEUE_WRITES] = {
 		.opcode		= P4_OPCODE(P4_EVENT_UOP_QUEUE_WRITES),
 		.escr_msr	= { MSR_P4_MS_ESCR0, MSR_P4_MS_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_TC_BUILD)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_TC_DELIVER)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_QUEUE_WRITES, FROM_ROM),
 		.cntr		= { {4, 5, -1}, {6, 7, -1} },
 	},
 	[P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE),
 		.escr_msr	= { MSR_P4_TBPU_ESCR0 , MSR_P4_TBPU_ESCR0 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, CONDITIONAL)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, CALL)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, RETURN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_MISPRED_BRANCH_TYPE, INDIRECT),
 		.cntr		= { {4, 5, -1}, {6, 7, -1} },
 	},
 	[P4_EVENT_RETIRED_BRANCH_TYPE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_RETIRED_BRANCH_TYPE),
 		.escr_msr	= { MSR_P4_TBPU_ESCR0 , MSR_P4_TBPU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CONDITIONAL)	|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, CALL)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, RETURN)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_RETIRED_BRANCH_TYPE, INDIRECT),
 		.cntr		= { {4, 5, -1}, {6, 7, -1} },
 	},
 	[P4_EVENT_RESOURCE_STALL] = {
 		.opcode		= P4_OPCODE(P4_EVENT_RESOURCE_STALL),
 		.escr_msr	= { MSR_P4_ALF_ESCR0, MSR_P4_ALF_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_RESOURCE_STALL, SBFULL),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_WC_BUFFER] = {
 		.opcode		= P4_OPCODE(P4_EVENT_WC_BUFFER),
 		.escr_msr	= { MSR_P4_DAC_ESCR0, MSR_P4_DAC_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_WC_BUFFER, WCB_EVICTS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_WC_BUFFER, WCB_FULL_EVICTS),
+		.shared		= 1,
 		.cntr		= { {8, 9, -1}, {10, 11, -1} },
 	},
 	[P4_EVENT_B2B_CYCLES] = {
 		.opcode		= P4_OPCODE(P4_EVENT_B2B_CYCLES),
 		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	= 0,
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_BNR] = {
 		.opcode		= P4_OPCODE(P4_EVENT_BNR),
 		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	= 0,
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_SNOOP] = {
 		.opcode		= P4_OPCODE(P4_EVENT_SNOOP),
 		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	= 0,
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_RESPONSE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_RESPONSE),
 		.escr_msr	= { MSR_P4_FSB_ESCR0, MSR_P4_FSB_ESCR1 },
+		.escr_emask	= 0,
 		.cntr		= { {0, -1, -1}, {2, -1, -1} },
 	},
 	[P4_EVENT_FRONT_END_EVENT] = {
 		.opcode		= P4_OPCODE(P4_EVENT_FRONT_END_EVENT),
 		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_FRONT_END_EVENT, NBOGUS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_FRONT_END_EVENT, BOGUS),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_EXECUTION_EVENT] = {
 		.opcode		= P4_OPCODE(P4_EVENT_EXECUTION_EVENT),
 		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS2)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, NBOGUS3)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS0)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS1)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS2)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_EXECUTION_EVENT, BOGUS3),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_REPLAY_EVENT] = {
 		.opcode		= P4_OPCODE(P4_EVENT_REPLAY_EVENT),
 		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_REPLAY_EVENT, NBOGUS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_REPLAY_EVENT, BOGUS),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_INSTR_RETIRED] = {
 		.opcode		= P4_OPCODE(P4_EVENT_INSTR_RETIRED),
 		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, NBOGUSNTAG)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, NBOGUSTAG)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, BOGUSNTAG)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_RETIRED, BOGUSTAG),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_UOPS_RETIRED] = {
 		.opcode		= P4_OPCODE(P4_EVENT_UOPS_RETIRED),
 		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOPS_RETIRED, NBOGUS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOPS_RETIRED, BOGUS),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_UOP_TYPE] = {
 		.opcode		= P4_OPCODE(P4_EVENT_UOP_TYPE),
 		.escr_msr	= { MSR_P4_RAT_ESCR0, MSR_P4_RAT_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_TYPE, TAGLOADS)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_UOP_TYPE, TAGSTORES),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_BRANCH_RETIRED] = {
 		.opcode		= P4_OPCODE(P4_EVENT_BRANCH_RETIRED),
 		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMNP)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMNM)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMTP)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_BRANCH_RETIRED, MMTM),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_MISPRED_BRANCH_RETIRED] = {
 		.opcode		= P4_OPCODE(P4_EVENT_MISPRED_BRANCH_RETIRED),
 		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.escr_emask	=
+		P4_ESCR_EMASK_BIT(P4_EVENT_MISPRED_BRANCH_RETIRED, NBOGUS),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_X87_ASSIST] = {
 		.opcode		= P4_OPCODE(P4_EVENT_X87_ASSIST),
 		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, FPSU)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, FPSO)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, POAO)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, POAU)			|
+			P4_ESCR_EMASK_BIT(P4_EVENT_X87_ASSIST, PREA),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_MACHINE_CLEAR] = {
 		.opcode		= P4_OPCODE(P4_EVENT_MACHINE_CLEAR),
 		.escr_msr	= { MSR_P4_CRU_ESCR2, MSR_P4_CRU_ESCR3 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, CLEAR)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, MOCLEAR)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_MACHINE_CLEAR, SMCLEAR),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 	[P4_EVENT_INSTR_COMPLETED] = {
 		.opcode		= P4_OPCODE(P4_EVENT_INSTR_COMPLETED),
 		.escr_msr	= { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 },
+		.escr_emask	=
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_COMPLETED, NBOGUS)		|
+			P4_ESCR_EMASK_BIT(P4_EVENT_INSTR_COMPLETED, BOGUS),
 		.cntr		= { {12, 13, 16}, {14, 15, 17} },
 	},
 };
@@ -428,29 +634,73 @@
 	return config;
 }
 
-static int p4_validate_raw_event(struct perf_event *event)
+/* check cpu model specifics */
+static bool p4_event_match_cpu_model(unsigned int event_idx)
 {
-	unsigned int v;
-
-	/* user data may have out-of-bound event index */
-	v = p4_config_unpack_event(event->attr.config);
-	if (v >= ARRAY_SIZE(p4_event_bind_map)) {
-		pr_warning("P4 PMU: Unknown event code: %d\n", v);
-		return -EINVAL;
+	/* INSTR_COMPLETED event only exist for model 3, 4, 6 (Prescott) */
+	if (event_idx == P4_EVENT_INSTR_COMPLETED) {
+		if (boot_cpu_data.x86_model != 3 &&
+			boot_cpu_data.x86_model != 4 &&
+			boot_cpu_data.x86_model != 6)
+			return false;
 	}
 
 	/*
-	 * it may have some screwed PEBS bits
+	 * For info
+	 * - IQ_ESCR0, IQ_ESCR1 only for models 1 and 2
 	 */
-	if (p4_config_pebs_has(event->attr.config, P4_PEBS_CONFIG_ENABLE)) {
-		pr_warning("P4 PMU: PEBS are not supported yet\n");
+
+	return true;
+}
+
+static int p4_validate_raw_event(struct perf_event *event)
+{
+	unsigned int v, emask;
+
+	/* User data may have out-of-bound event index */
+	v = p4_config_unpack_event(event->attr.config);
+	if (v >= ARRAY_SIZE(p4_event_bind_map))
 		return -EINVAL;
+
+	/* It may be unsupported: */
+	if (!p4_event_match_cpu_model(v))
+		return -EINVAL;
+
+	/*
+	 * NOTE: P4_CCCR_THREAD_ANY has not the same meaning as
+	 * in Architectural Performance Monitoring, it means not
+	 * on _which_ logical cpu to count but rather _when_, ie it
+	 * depends on logical cpu state -- count event if one cpu active,
+	 * none, both or any, so we just allow user to pass any value
+	 * desired.
+	 *
+	 * In turn we always set Tx_OS/Tx_USR bits bound to logical
+	 * cpu without their propagation to another cpu
+	 */
+
+	/*
+	 * if an event is shared accross the logical threads
+	 * the user needs special permissions to be able to use it
+	 */
+	if (p4_event_bind_map[v].shared) {
+		if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN))
+			return -EACCES;
 	}
+
+	/* ESCR EventMask bits may be invalid */
+	emask = p4_config_unpack_escr(event->attr.config) & P4_ESCR_EVENTMASK_MASK;
+	if (emask & ~p4_event_bind_map[v].escr_emask)
+		return -EINVAL;
+
+	/*
+	 * it may have some invalid PEBS bits
+	 */
+	if (p4_config_pebs_has(event->attr.config, P4_PEBS_CONFIG_ENABLE))
+		return -EINVAL;
+
 	v = p4_config_unpack_metric(event->attr.config);
-	if (v >= ARRAY_SIZE(p4_pebs_bind_map)) {
-		pr_warning("P4 PMU: Unknown metric code: %d\n", v);
+	if (v >= ARRAY_SIZE(p4_pebs_bind_map))
 		return -EINVAL;
-	}
 
 	return 0;
 }
@@ -478,27 +728,21 @@
 
 	if (event->attr.type == PERF_TYPE_RAW) {
 
+		/*
+		 * Clear bits we reserve to be managed by kernel itself
+		 * and never allowed from a user space
+		 */
+		 event->attr.config &= P4_CONFIG_MASK;
+
 		rc = p4_validate_raw_event(event);
 		if (rc)
 			goto out;
 
 		/*
-		 * We don't control raw events so it's up to the caller
-		 * to pass sane values (and we don't count the thread number
-		 * on HT machine but allow HT-compatible specifics to be
-		 * passed on)
-		 *
 		 * Note that for RAW events we allow user to use P4_CCCR_RESERVED
 		 * bits since we keep additional info here (for cache events and etc)
-		 *
-		 * XXX: HT wide things should check perf_paranoid_cpu() &&
-		 *      CAP_SYS_ADMIN
 		 */
-		event->hw.config |= event->attr.config &
-			(p4_config_pack_escr(P4_ESCR_MASK_HT) |
-			 p4_config_pack_cccr(P4_CCCR_MASK_HT | P4_CCCR_RESERVED));
-
-		event->hw.config &= ~P4_CCCR_FORCE_OVF;
+		event->hw.config |= event->attr.config;
 	}
 
 	rc = x86_setup_perfctr(event);
diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index 770ebfb..e05952a 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -230,9 +230,6 @@
 	return 0;
 }
 
-/* Dummy buffers for kallsyms_lookup */
-static char __dummy_buf[KSYM_NAME_LEN];
-
 /* Check if paddr is at an instruction boundary */
 static int __kprobes can_probe(unsigned long paddr)
 {
@@ -241,7 +238,7 @@
 	struct insn insn;
 	kprobe_opcode_t buf[MAX_INSN_SIZE];
 
-	if (!kallsyms_lookup(paddr, NULL, &offset, NULL, __dummy_buf))
+	if (!kallsyms_lookup_size_offset(paddr, NULL, &offset))
 		return 0;
 
 	/* Decode instructions */
@@ -1129,7 +1126,7 @@
 	*(unsigned long *)addr = val;
 }
 
-void __kprobes kprobes_optinsn_template_holder(void)
+static void __used __kprobes kprobes_optinsn_template_holder(void)
 {
 	asm volatile (
 			".global optprobe_template_entry\n"
@@ -1269,11 +1266,9 @@
 	unsigned long addr, size = 0, offset = 0;
 	struct insn insn;
 	kprobe_opcode_t buf[MAX_INSN_SIZE];
-	/* Dummy buffers for lookup_symbol_attrs */
-	static char __dummy_buf[KSYM_NAME_LEN];
 
 	/* Lookup symbol including addr */
-	if (!kallsyms_lookup(paddr, &size, &offset, NULL, __dummy_buf))
+	if (!kallsyms_lookup_size_offset(paddr, &size, &offset))
 		return 0;
 
 	/* Check there is enough space for a relative jump. */
diff --git a/include/asm-generic/hardirq.h b/include/asm-generic/hardirq.h
index 62f5908..04d0a97 100644
--- a/include/asm-generic/hardirq.h
+++ b/include/asm-generic/hardirq.h
@@ -3,13 +3,13 @@
 
 #include <linux/cache.h>
 #include <linux/threads.h>
-#include <linux/irq.h>
 
 typedef struct {
 	unsigned int __softirq_pending;
 } ____cacheline_aligned irq_cpustat_t;
 
 #include <linux/irq_cpustat.h>	/* Standard mappings for irq_cpustat_t above */
+#include <linux/irq.h>
 
 #ifndef ack_bad_irq
 static inline void ack_bad_irq(unsigned int irq)
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index 02b8b24..8beabb9 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
@@ -191,8 +191,8 @@
 	unsigned int		flags;
 
 #ifdef CONFIG_PERF_EVENTS
-	int			perf_refcount;
-	struct hlist_head	*perf_events;
+	int				perf_refcount;
+	struct hlist_head __percpu	*perf_events;
 #endif
 };
 
@@ -252,8 +252,8 @@
 
 extern int  perf_trace_init(struct perf_event *event);
 extern void perf_trace_destroy(struct perf_event *event);
-extern int  perf_trace_enable(struct perf_event *event);
-extern void perf_trace_disable(struct perf_event *event);
+extern int  perf_trace_add(struct perf_event *event, int flags);
+extern void perf_trace_del(struct perf_event *event, int flags);
 extern int  ftrace_profile_set_filter(struct perf_event *event, int event_id,
 				     char *filter_str);
 extern void ftrace_profile_free_filter(struct perf_event *event);
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index a0384a4..531495d 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -18,6 +18,7 @@
 #include <asm/atomic.h>
 #include <asm/ptrace.h>
 #include <asm/system.h>
+#include <trace/events/irq.h>
 
 /*
  * These correspond to the IORESOURCE_IRQ_* defines in
@@ -407,7 +408,12 @@
 asmlinkage void __do_softirq(void);
 extern void open_softirq(int nr, void (*action)(struct softirq_action *));
 extern void softirq_init(void);
-#define __raise_softirq_irqoff(nr) do { or_softirq_pending(1UL << (nr)); } while (0)
+static inline void __raise_softirq_irqoff(unsigned int nr)
+{
+	trace_softirq_raise((struct softirq_action *)(unsigned long)nr, NULL);
+	or_softirq_pending(1UL << nr);
+}
+
 extern void raise_softirq_irqoff(unsigned int nr);
 extern void raise_softirq(unsigned int nr);
 extern void wakeup_softirqd(void);
diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index 49466b1..0eb5083 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -39,6 +39,15 @@
 	preempt_enable();				\
 } while (0)
 
+#define get_cpu_ptr(var) ({				\
+	preempt_disable();				\
+	this_cpu_ptr(var); })
+
+#define put_cpu_ptr(var) do {				\
+	(void)(var);					\
+	preempt_enable();				\
+} while (0)
+
 #ifdef CONFIG_SMP
 
 /* minimum unit size, also is the maximum supported allocation size */
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 716f99b..61b1e2d 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -529,7 +529,6 @@
 			int		last_cpu;
 		};
 		struct { /* software */
-			s64		remaining;
 			struct hrtimer	hrtimer;
 		};
 #ifdef CONFIG_HAVE_HW_BREAKPOINT
@@ -539,6 +538,7 @@
 		};
 #endif
 	};
+	int				state;
 	local64_t			prev_count;
 	u64				sample_period;
 	u64				last_period;
@@ -550,6 +550,13 @@
 #endif
 };
 
+/*
+ * hw_perf_event::state flags
+ */
+#define PERF_HES_STOPPED	0x01 /* the counter is stopped */
+#define PERF_HES_UPTODATE	0x02 /* event->count up-to-date */
+#define PERF_HES_ARCH		0x04
+
 struct perf_event;
 
 /*
@@ -561,36 +568,70 @@
  * struct pmu - generic performance monitoring unit
  */
 struct pmu {
-	int (*enable)			(struct perf_event *event);
-	void (*disable)			(struct perf_event *event);
-	int (*start)			(struct perf_event *event);
-	void (*stop)			(struct perf_event *event);
+	struct list_head		entry;
+
+	int * __percpu			pmu_disable_count;
+	struct perf_cpu_context * __percpu pmu_cpu_context;
+	int				task_ctx_nr;
+
+	/*
+	 * Fully disable/enable this PMU, can be used to protect from the PMI
+	 * as well as for lazy/batch writing of the MSRs.
+	 */
+	void (*pmu_enable)		(struct pmu *pmu); /* optional */
+	void (*pmu_disable)		(struct pmu *pmu); /* optional */
+
+	/*
+	 * Try and initialize the event for this PMU.
+	 * Should return -ENOENT when the @event doesn't match this PMU.
+	 */
+	int (*event_init)		(struct perf_event *event);
+
+#define PERF_EF_START	0x01		/* start the counter when adding    */
+#define PERF_EF_RELOAD	0x02		/* reload the counter when starting */
+#define PERF_EF_UPDATE	0x04		/* update the counter when stopping */
+
+	/*
+	 * Adds/Removes a counter to/from the PMU, can be done inside
+	 * a transaction, see the ->*_txn() methods.
+	 */
+	int  (*add)			(struct perf_event *event, int flags);
+	void (*del)			(struct perf_event *event, int flags);
+
+	/*
+	 * Starts/Stops a counter present on the PMU. The PMI handler
+	 * should stop the counter when perf_event_overflow() returns
+	 * !0. ->start() will be used to continue.
+	 */
+	void (*start)			(struct perf_event *event, int flags);
+	void (*stop)			(struct perf_event *event, int flags);
+
+	/*
+	 * Updates the counter value of the event.
+	 */
 	void (*read)			(struct perf_event *event);
-	void (*unthrottle)		(struct perf_event *event);
 
 	/*
-	 * Group events scheduling is treated as a transaction, add group
-	 * events as a whole and perform one schedulability test. If the test
-	 * fails, roll back the whole group
+	 * Group events scheduling is treated as a transaction, add
+	 * group events as a whole and perform one schedulability test.
+	 * If the test fails, roll back the whole group
+	 *
+	 * Start the transaction, after this ->add() doesn't need to
+	 * do schedulability tests.
 	 */
-
+	void (*start_txn)	(struct pmu *pmu); /* optional */
 	/*
-	 * Start the transaction, after this ->enable() doesn't need
-	 * to do schedulability tests.
-	 */
-	void (*start_txn)	(const struct pmu *pmu);
-	/*
-	 * If ->start_txn() disabled the ->enable() schedulability test
+	 * If ->start_txn() disabled the ->add() schedulability test
 	 * then ->commit_txn() is required to perform one. On success
 	 * the transaction is closed. On error the transaction is kept
 	 * open until ->cancel_txn() is called.
 	 */
-	int  (*commit_txn)	(const struct pmu *pmu);
+	int  (*commit_txn)	(struct pmu *pmu); /* optional */
 	/*
-	 * Will cancel the transaction, assumes ->disable() is called for
-	 * each successfull ->enable() during the transaction.
+	 * Will cancel the transaction, assumes ->del() is called
+	 * for each successfull ->add() during the transaction.
 	 */
-	void (*cancel_txn)	(const struct pmu *pmu);
+	void (*cancel_txn)	(struct pmu *pmu); /* optional */
 };
 
 /**
@@ -669,7 +710,7 @@
 	int				nr_siblings;
 	int				group_flags;
 	struct perf_event		*group_leader;
-	const struct pmu		*pmu;
+	struct pmu			*pmu;
 
 	enum perf_event_active_state	state;
 	unsigned int			attach_state;
@@ -763,12 +804,19 @@
 #endif /* CONFIG_PERF_EVENTS */
 };
 
+enum perf_event_context_type {
+	task_context,
+	cpu_context,
+};
+
 /**
  * struct perf_event_context - event context structure
  *
  * Used as a container for task events and CPU events as well:
  */
 struct perf_event_context {
+	enum perf_event_context_type	type;
+	struct pmu			*pmu;
 	/*
 	 * Protect the states of the events in the list,
 	 * nr_active, and the list:
@@ -808,6 +856,12 @@
 	struct rcu_head			rcu_head;
 };
 
+/*
+ * Number of contexts where an event can trigger:
+ * 	task, softirq, hardirq, nmi.
+ */
+#define PERF_NR_CONTEXTS	4
+
 /**
  * struct perf_event_cpu_context - per cpu event context structure
  */
@@ -815,18 +869,9 @@
 	struct perf_event_context	ctx;
 	struct perf_event_context	*task_ctx;
 	int				active_oncpu;
-	int				max_pertask;
 	int				exclusive;
-	struct swevent_hlist		*swevent_hlist;
-	struct mutex			hlist_mutex;
-	int				hlist_refcount;
-
-	/*
-	 * Recursion avoidance:
-	 *
-	 * task, softirq, irq, nmi context
-	 */
-	int				recursion[4];
+	struct list_head		rotation_list;
+	int				jiffies_interval;
 };
 
 struct perf_output_handle {
@@ -842,26 +887,20 @@
 
 #ifdef CONFIG_PERF_EVENTS
 
-/*
- * Set by architecture code:
- */
-extern int perf_max_events;
-
-extern const struct pmu *hw_perf_event_init(struct perf_event *event);
+extern int perf_pmu_register(struct pmu *pmu);
+extern void perf_pmu_unregister(struct pmu *pmu);
 
 extern void perf_event_task_sched_in(struct task_struct *task);
 extern void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next);
-extern void perf_event_task_tick(struct task_struct *task);
 extern int perf_event_init_task(struct task_struct *child);
 extern void perf_event_exit_task(struct task_struct *child);
 extern void perf_event_free_task(struct task_struct *task);
+extern void perf_event_delayed_put(struct task_struct *task);
 extern void set_perf_event_pending(void);
 extern void perf_event_do_pending(void);
 extern void perf_event_print_debug(void);
-extern void __perf_disable(void);
-extern bool __perf_enable(void);
-extern void perf_disable(void);
-extern void perf_enable(void);
+extern void perf_pmu_disable(struct pmu *pmu);
+extern void perf_pmu_enable(struct pmu *pmu);
 extern int perf_event_task_disable(void);
 extern int perf_event_task_enable(void);
 extern void perf_event_update_userpage(struct perf_event *event);
@@ -869,7 +908,7 @@
 extern struct perf_event *
 perf_event_create_kernel_counter(struct perf_event_attr *attr,
 				int cpu,
-				pid_t pid,
+				struct task_struct *task,
 				perf_overflow_handler_t callback);
 extern u64 perf_event_read_value(struct perf_event *event,
 				 u64 *enabled, u64 *running);
@@ -920,14 +959,7 @@
  */
 static inline int is_software_event(struct perf_event *event)
 {
-	switch (event->attr.type) {
-	case PERF_TYPE_SOFTWARE:
-	case PERF_TYPE_TRACEPOINT:
-	/* for now the breakpoint stuff also works as software event */
-	case PERF_TYPE_BREAKPOINT:
-		return 1;
-	}
-	return 0;
+	return event->pmu->task_ctx_nr == perf_sw_context;
 }
 
 extern atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX];
@@ -976,7 +1008,21 @@
 extern void perf_event_comm(struct task_struct *tsk);
 extern void perf_event_fork(struct task_struct *tsk);
 
-extern struct perf_callchain_entry *perf_callchain(struct pt_regs *regs);
+/* Callchains */
+DECLARE_PER_CPU(struct perf_callchain_entry, perf_callchain_entry);
+
+extern void perf_callchain_user(struct perf_callchain_entry *entry,
+				struct pt_regs *regs);
+extern void perf_callchain_kernel(struct perf_callchain_entry *entry,
+				  struct pt_regs *regs);
+
+
+static inline void
+perf_callchain_store(struct perf_callchain_entry *entry, u64 ip)
+{
+	if (entry->nr < PERF_MAX_STACK_DEPTH)
+		entry->ip[entry->nr++] = ip;
+}
 
 extern int sysctl_perf_event_paranoid;
 extern int sysctl_perf_event_mlock;
@@ -1019,21 +1065,19 @@
 extern void perf_swevent_put_recursion_context(int rctx);
 extern void perf_event_enable(struct perf_event *event);
 extern void perf_event_disable(struct perf_event *event);
+extern void perf_event_task_tick(void);
 #else
 static inline void
 perf_event_task_sched_in(struct task_struct *task)			{ }
 static inline void
 perf_event_task_sched_out(struct task_struct *task,
 			    struct task_struct *next)			{ }
-static inline void
-perf_event_task_tick(struct task_struct *task)				{ }
 static inline int perf_event_init_task(struct task_struct *child)	{ return 0; }
 static inline void perf_event_exit_task(struct task_struct *child)	{ }
 static inline void perf_event_free_task(struct task_struct *task)	{ }
+static inline void perf_event_delayed_put(struct task_struct *task)	{ }
 static inline void perf_event_do_pending(void)				{ }
 static inline void perf_event_print_debug(void)				{ }
-static inline void perf_disable(void)					{ }
-static inline void perf_enable(void)					{ }
 static inline int perf_event_task_disable(void)				{ return -EINVAL; }
 static inline int perf_event_task_enable(void)				{ return -EINVAL; }
 
@@ -1056,6 +1100,7 @@
 static inline void perf_swevent_put_recursion_context(int rctx)		{ }
 static inline void perf_event_enable(struct perf_event *event)		{ }
 static inline void perf_event_disable(struct perf_event *event)		{ }
+static inline void perf_event_task_tick(void)				{ }
 #endif
 
 #define perf_output_put(handle, x) \
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 1e2a6db..eb3c1ce 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1160,6 +1160,13 @@
 
 struct rcu_node;
 
+enum perf_event_task_context {
+	perf_invalid_context = -1,
+	perf_hw_context = 0,
+	perf_sw_context,
+	perf_nr_task_contexts,
+};
+
 struct task_struct {
 	volatile long state;	/* -1 unrunnable, 0 runnable, >0 stopped */
 	void *stack;
@@ -1431,7 +1438,7 @@
 	struct futex_pi_state *pi_state_cache;
 #endif
 #ifdef CONFIG_PERF_EVENTS
-	struct perf_event_context *perf_event_ctxp;
+	struct perf_event_context *perf_event_ctxp[perf_nr_task_contexts];
 	struct mutex perf_event_mutex;
 	struct list_head perf_event_list;
 #endif
diff --git a/include/trace/events/irq.h b/include/trace/events/irq.h
index 0e4cfb6..6fa7cba 100644
--- a/include/trace/events/irq.h
+++ b/include/trace/events/irq.h
@@ -5,7 +5,9 @@
 #define _TRACE_IRQ_H
 
 #include <linux/tracepoint.h>
-#include <linux/interrupt.h>
+
+struct irqaction;
+struct softirq_action;
 
 #define softirq_name(sirq) { sirq##_SOFTIRQ, #sirq }
 #define show_softirq_name(val)				\
@@ -93,7 +95,10 @@
 	),
 
 	TP_fast_assign(
-		__entry->vec = (int)(h - vec);
+		if (vec)
+			__entry->vec = (int)(h - vec);
+		else
+			__entry->vec = (int)(long)h;
 	),
 
 	TP_printk("vec=%d [action=%s]", __entry->vec,
@@ -136,6 +141,23 @@
 	TP_ARGS(h, vec)
 );
 
+/**
+ * softirq_raise - called immediately when a softirq is raised
+ * @h: pointer to struct softirq_action
+ * @vec: pointer to first struct softirq_action in softirq_vec array
+ *
+ * The @h parameter contains a pointer to the softirq vector number which is
+ * raised. @vec is NULL and it means @h includes vector number not
+ * softirq_action. When used in combination with the softirq_entry tracepoint
+ * we can determine the softirq raise latency.
+ */
+DEFINE_EVENT(softirq, softirq_raise,
+
+	TP_PROTO(struct softirq_action *h, struct softirq_action *vec),
+
+	TP_ARGS(h, vec)
+);
+
 #endif /*  _TRACE_IRQ_H */
 
 /* This part must be outside protection */
diff --git a/include/trace/events/napi.h b/include/trace/events/napi.h
index 188deca..8fe1e93 100644
--- a/include/trace/events/napi.h
+++ b/include/trace/events/napi.h
@@ -6,10 +6,31 @@
 
 #include <linux/netdevice.h>
 #include <linux/tracepoint.h>
+#include <linux/ftrace.h>
 
-DECLARE_TRACE(napi_poll,
+#define NO_DEV "(no_device)"
+
+TRACE_EVENT(napi_poll,
+
 	TP_PROTO(struct napi_struct *napi),
-	TP_ARGS(napi));
+
+	TP_ARGS(napi),
+
+	TP_STRUCT__entry(
+		__field(	struct napi_struct *,	napi)
+		__string(	dev_name, napi->dev ? napi->dev->name : NO_DEV)
+	),
+
+	TP_fast_assign(
+		__entry->napi = napi;
+		__assign_str(dev_name, napi->dev ? napi->dev->name : NO_DEV);
+	),
+
+	TP_printk("napi poll on napi struct %p for device %s",
+		__entry->napi, __get_str(dev_name))
+);
+
+#undef NO_DEV
 
 #endif /* _TRACE_NAPI_H_ */
 
diff --git a/include/trace/events/net.h b/include/trace/events/net.h
new file mode 100644
index 0000000..5f247f5
--- /dev/null
+++ b/include/trace/events/net.h
@@ -0,0 +1,82 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM net
+
+#if !defined(_TRACE_NET_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_NET_H
+
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/ip.h>
+#include <linux/tracepoint.h>
+
+TRACE_EVENT(net_dev_xmit,
+
+	TP_PROTO(struct sk_buff *skb,
+		 int rc),
+
+	TP_ARGS(skb, rc),
+
+	TP_STRUCT__entry(
+		__field(	void *,		skbaddr		)
+		__field(	unsigned int,	len		)
+		__field(	int,		rc		)
+		__string(	name,		skb->dev->name	)
+	),
+
+	TP_fast_assign(
+		__entry->skbaddr = skb;
+		__entry->len = skb->len;
+		__entry->rc = rc;
+		__assign_str(name, skb->dev->name);
+	),
+
+	TP_printk("dev=%s skbaddr=%p len=%u rc=%d",
+		__get_str(name), __entry->skbaddr, __entry->len, __entry->rc)
+);
+
+DECLARE_EVENT_CLASS(net_dev_template,
+
+	TP_PROTO(struct sk_buff *skb),
+
+	TP_ARGS(skb),
+
+	TP_STRUCT__entry(
+		__field(	void *,		skbaddr		)
+		__field(	unsigned int,	len		)
+		__string(	name,		skb->dev->name	)
+	),
+
+	TP_fast_assign(
+		__entry->skbaddr = skb;
+		__entry->len = skb->len;
+		__assign_str(name, skb->dev->name);
+	),
+
+	TP_printk("dev=%s skbaddr=%p len=%u",
+		__get_str(name), __entry->skbaddr, __entry->len)
+)
+
+DEFINE_EVENT(net_dev_template, net_dev_queue,
+
+	TP_PROTO(struct sk_buff *skb),
+
+	TP_ARGS(skb)
+);
+
+DEFINE_EVENT(net_dev_template, netif_receive_skb,
+
+	TP_PROTO(struct sk_buff *skb),
+
+	TP_ARGS(skb)
+);
+
+DEFINE_EVENT(net_dev_template, netif_rx,
+
+	TP_PROTO(struct sk_buff *skb),
+
+	TP_ARGS(skb)
+);
+#endif /* _TRACE_NET_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/include/trace/events/power.h b/include/trace/events/power.h
index 35a2a6e..286784d 100644
--- a/include/trace/events/power.h
+++ b/include/trace/events/power.h
@@ -10,12 +10,17 @@
 #ifndef _TRACE_POWER_ENUM_
 #define _TRACE_POWER_ENUM_
 enum {
-	POWER_NONE = 0,
-	POWER_CSTATE = 1,
-	POWER_PSTATE = 2,
+	POWER_NONE	= 0,
+	POWER_CSTATE	= 1,	/* C-State */
+	POWER_PSTATE	= 2,	/* Fequency change or DVFS */
+	POWER_SSTATE	= 3,	/* Suspend */
 };
 #endif
 
+/*
+ * The power events are used for cpuidle & suspend (power_start, power_end)
+ *  and for cpufreq (power_frequency)
+ */
 DECLARE_EVENT_CLASS(power,
 
 	TP_PROTO(unsigned int type, unsigned int state, unsigned int cpu_id),
@@ -70,6 +75,85 @@
 
 );
 
+/*
+ * The clock events are used for clock enable/disable and for
+ *  clock rate change
+ */
+DECLARE_EVENT_CLASS(clock,
+
+	TP_PROTO(const char *name, unsigned int state, unsigned int cpu_id),
+
+	TP_ARGS(name, state, cpu_id),
+
+	TP_STRUCT__entry(
+		__string(       name,           name            )
+		__field(        u64,            state           )
+		__field(        u64,            cpu_id          )
+	),
+
+	TP_fast_assign(
+		__assign_str(name, name);
+		__entry->state = state;
+		__entry->cpu_id = cpu_id;
+	),
+
+	TP_printk("%s state=%lu cpu_id=%lu", __get_str(name),
+		(unsigned long)__entry->state, (unsigned long)__entry->cpu_id)
+);
+
+DEFINE_EVENT(clock, clock_enable,
+
+	TP_PROTO(const char *name, unsigned int state, unsigned int cpu_id),
+
+	TP_ARGS(name, state, cpu_id)
+);
+
+DEFINE_EVENT(clock, clock_disable,
+
+	TP_PROTO(const char *name, unsigned int state, unsigned int cpu_id),
+
+	TP_ARGS(name, state, cpu_id)
+);
+
+DEFINE_EVENT(clock, clock_set_rate,
+
+	TP_PROTO(const char *name, unsigned int state, unsigned int cpu_id),
+
+	TP_ARGS(name, state, cpu_id)
+);
+
+/*
+ * The power domain events are used for power domains transitions
+ */
+DECLARE_EVENT_CLASS(power_domain,
+
+	TP_PROTO(const char *name, unsigned int state, unsigned int cpu_id),
+
+	TP_ARGS(name, state, cpu_id),
+
+	TP_STRUCT__entry(
+		__string(       name,           name            )
+		__field(        u64,            state           )
+		__field(        u64,            cpu_id          )
+	),
+
+	TP_fast_assign(
+		__assign_str(name, name);
+		__entry->state = state;
+		__entry->cpu_id = cpu_id;
+),
+
+	TP_printk("%s state=%lu cpu_id=%lu", __get_str(name),
+		(unsigned long)__entry->state, (unsigned long)__entry->cpu_id)
+);
+
+DEFINE_EVENT(power_domain, power_domain_target,
+
+	TP_PROTO(const char *name, unsigned int state, unsigned int cpu_id),
+
+	TP_ARGS(name, state, cpu_id)
+);
+
 #endif /* _TRACE_POWER_H */
 
 /* This part must be outside protection */
diff --git a/include/trace/events/skb.h b/include/trace/events/skb.h
index 4b2be6d..75ce9d5 100644
--- a/include/trace/events/skb.h
+++ b/include/trace/events/skb.h
@@ -35,6 +35,23 @@
 		__entry->skbaddr, __entry->protocol, __entry->location)
 );
 
+TRACE_EVENT(consume_skb,
+
+	TP_PROTO(struct sk_buff *skb),
+
+	TP_ARGS(skb),
+
+	TP_STRUCT__entry(
+		__field(	void *,	skbaddr	)
+	),
+
+	TP_fast_assign(
+		__entry->skbaddr = skb;
+	),
+
+	TP_printk("skbaddr=%p", __entry->skbaddr)
+);
+
 TRACE_EVENT(skb_copy_datagram_iovec,
 
 	TP_PROTO(const struct sk_buff *skb, int len),
diff --git a/kernel/exit.c b/kernel/exit.c
index 0312022..e2bdf37 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -149,9 +149,7 @@
 {
 	struct task_struct *tsk = container_of(rhp, struct task_struct, rcu);
 
-#ifdef CONFIG_PERF_EVENTS
-	WARN_ON_ONCE(tsk->perf_event_ctxp);
-#endif
+	perf_event_delayed_put(tsk);
 	trace_sched_process_free(tsk);
 	put_task_struct(tsk);
 }
diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c
index c7c2aed..3b714e8 100644
--- a/kernel/hw_breakpoint.c
+++ b/kernel/hw_breakpoint.c
@@ -433,8 +433,7 @@
 			    perf_overflow_handler_t triggered,
 			    struct task_struct *tsk)
 {
-	return perf_event_create_kernel_counter(attr, -1, task_pid_vnr(tsk),
-						triggered);
+	return perf_event_create_kernel_counter(attr, -1, tsk, triggered);
 }
 EXPORT_SYMBOL_GPL(register_user_hw_breakpoint);
 
@@ -516,7 +515,7 @@
 	get_online_cpus();
 	for_each_online_cpu(cpu) {
 		pevent = per_cpu_ptr(cpu_events, cpu);
-		bp = perf_event_create_kernel_counter(attr, cpu, -1, triggered);
+		bp = perf_event_create_kernel_counter(attr, cpu, NULL, triggered);
 
 		*pevent = bp;
 
@@ -566,6 +565,61 @@
 	.priority = 0x7fffffff
 };
 
+static void bp_perf_event_destroy(struct perf_event *event)
+{
+	release_bp_slot(event);
+}
+
+static int hw_breakpoint_event_init(struct perf_event *bp)
+{
+	int err;
+
+	if (bp->attr.type != PERF_TYPE_BREAKPOINT)
+		return -ENOENT;
+
+	err = register_perf_hw_breakpoint(bp);
+	if (err)
+		return err;
+
+	bp->destroy = bp_perf_event_destroy;
+
+	return 0;
+}
+
+static int hw_breakpoint_add(struct perf_event *bp, int flags)
+{
+	if (!(flags & PERF_EF_START))
+		bp->hw.state = PERF_HES_STOPPED;
+
+	return arch_install_hw_breakpoint(bp);
+}
+
+static void hw_breakpoint_del(struct perf_event *bp, int flags)
+{
+	arch_uninstall_hw_breakpoint(bp);
+}
+
+static void hw_breakpoint_start(struct perf_event *bp, int flags)
+{
+	bp->hw.state = 0;
+}
+
+static void hw_breakpoint_stop(struct perf_event *bp, int flags)
+{
+	bp->hw.state = PERF_HES_STOPPED;
+}
+
+static struct pmu perf_breakpoint = {
+	.task_ctx_nr	= perf_sw_context, /* could eventually get its own */
+
+	.event_init	= hw_breakpoint_event_init,
+	.add		= hw_breakpoint_add,
+	.del		= hw_breakpoint_del,
+	.start		= hw_breakpoint_start,
+	.stop		= hw_breakpoint_stop,
+	.read		= hw_breakpoint_pmu_read,
+};
+
 static int __init init_hw_breakpoint(void)
 {
 	unsigned int **task_bp_pinned;
@@ -587,6 +641,8 @@
 
 	constraints_initialized = 1;
 
+	perf_pmu_register(&perf_breakpoint);
+
 	return register_die_notifier(&hw_breakpoint_exceptions_nb);
 
  err_alloc:
@@ -602,8 +658,3 @@
 core_initcall(init_hw_breakpoint);
 
 
-struct pmu perf_ops_bp = {
-	.enable		= arch_install_hw_breakpoint,
-	.disable	= arch_uninstall_hw_breakpoint,
-	.read		= hw_breakpoint_pmu_read,
-};
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 282035f..6dd5359 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -399,7 +399,7 @@
  * Return an optimized kprobe whose optimizing code replaces
  * instructions including addr (exclude breakpoint).
  */
-struct kprobe *__kprobes get_optimized_kprobe(unsigned long addr)
+static struct kprobe *__kprobes get_optimized_kprobe(unsigned long addr)
 {
 	int i;
 	struct kprobe *p = NULL;
@@ -831,6 +831,7 @@
 
 void __kprobes kretprobe_hash_lock(struct task_struct *tsk,
 			 struct hlist_head **head, unsigned long *flags)
+__acquires(hlist_lock)
 {
 	unsigned long hash = hash_ptr(tsk, KPROBE_HASH_BITS);
 	spinlock_t *hlist_lock;
@@ -842,6 +843,7 @@
 
 static void __kprobes kretprobe_table_lock(unsigned long hash,
 	unsigned long *flags)
+__acquires(hlist_lock)
 {
 	spinlock_t *hlist_lock = kretprobe_table_lock_ptr(hash);
 	spin_lock_irqsave(hlist_lock, *flags);
@@ -849,6 +851,7 @@
 
 void __kprobes kretprobe_hash_unlock(struct task_struct *tsk,
 	unsigned long *flags)
+__releases(hlist_lock)
 {
 	unsigned long hash = hash_ptr(tsk, KPROBE_HASH_BITS);
 	spinlock_t *hlist_lock;
@@ -857,7 +860,9 @@
 	spin_unlock_irqrestore(hlist_lock, *flags);
 }
 
-void __kprobes kretprobe_table_unlock(unsigned long hash, unsigned long *flags)
+static void __kprobes kretprobe_table_unlock(unsigned long hash,
+       unsigned long *flags)
+__releases(hlist_lock)
 {
 	spinlock_t *hlist_lock = kretprobe_table_lock_ptr(hash);
 	spin_unlock_irqrestore(hlist_lock, *flags);
@@ -1339,18 +1344,19 @@
 	if (num <= 0)
 		return -EINVAL;
 	for (i = 0; i < num; i++) {
-		unsigned long addr;
+		unsigned long addr, offset;
 		jp = jps[i];
 		addr = arch_deref_entry_point(jp->entry);
 
-		if (!kernel_text_address(addr))
-			ret = -EINVAL;
-		else {
-			/* Todo: Verify probepoint is a function entry point */
+		/* Verify probepoint is a function entry point */
+		if (kallsyms_lookup_size_offset(addr, NULL, &offset) &&
+		    offset == 0) {
 			jp->kp.pre_handler = setjmp_pre_handler;
 			jp->kp.break_handler = longjmp_break_handler;
 			ret = register_kprobe(&jp->kp);
-		}
+		} else
+			ret = -EINVAL;
+
 		if (ret < 0) {
 			if (i > 0)
 				unregister_jprobes(jps, i);
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index db5b560..c16158c 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -31,24 +31,18 @@
 #include <linux/kernel_stat.h>
 #include <linux/perf_event.h>
 #include <linux/ftrace_event.h>
-#include <linux/hw_breakpoint.h>
 
 #include <asm/irq_regs.h>
 
-/*
- * Each CPU has a list of per CPU events:
- */
-static DEFINE_PER_CPU(struct perf_cpu_context, perf_cpu_context);
-
-int perf_max_events __read_mostly = 1;
-static int perf_reserved_percpu __read_mostly;
-static int perf_overcommit __read_mostly = 1;
-
 static atomic_t nr_events __read_mostly;
 static atomic_t nr_mmap_events __read_mostly;
 static atomic_t nr_comm_events __read_mostly;
 static atomic_t nr_task_events __read_mostly;
 
+static LIST_HEAD(pmus);
+static DEFINE_MUTEX(pmus_lock);
+static struct srcu_struct pmus_srcu;
+
 /*
  * perf event paranoia level:
  *  -1 - not paranoid at all
@@ -67,36 +61,38 @@
 
 static atomic64_t perf_event_id;
 
-/*
- * Lock for (sysadmin-configurable) event reservations:
- */
-static DEFINE_SPINLOCK(perf_resource_lock);
-
-/*
- * Architecture provided APIs - weak aliases:
- */
-extern __weak const struct pmu *hw_perf_event_init(struct perf_event *event)
-{
-	return NULL;
-}
-
-void __weak hw_perf_disable(void)		{ barrier(); }
-void __weak hw_perf_enable(void)		{ barrier(); }
-
 void __weak perf_event_print_debug(void)	{ }
 
-static DEFINE_PER_CPU(int, perf_disable_count);
-
-void perf_disable(void)
+void perf_pmu_disable(struct pmu *pmu)
 {
-	if (!__get_cpu_var(perf_disable_count)++)
-		hw_perf_disable();
+	int *count = this_cpu_ptr(pmu->pmu_disable_count);
+	if (!(*count)++)
+		pmu->pmu_disable(pmu);
 }
 
-void perf_enable(void)
+void perf_pmu_enable(struct pmu *pmu)
 {
-	if (!--__get_cpu_var(perf_disable_count))
-		hw_perf_enable();
+	int *count = this_cpu_ptr(pmu->pmu_disable_count);
+	if (!--(*count))
+		pmu->pmu_enable(pmu);
+}
+
+static DEFINE_PER_CPU(struct list_head, rotation_list);
+
+/*
+ * perf_pmu_rotate_start() and perf_rotate_context() are fully serialized
+ * because they're strictly cpu affine and rotate_start is called with IRQs
+ * disabled, while rotate_context is called from IRQ context.
+ */
+static void perf_pmu_rotate_start(struct pmu *pmu)
+{
+	struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
+	struct list_head *head = &__get_cpu_var(rotation_list);
+
+	WARN_ON(!irqs_disabled());
+
+	if (list_empty(&cpuctx->rotation_list))
+		list_add(&cpuctx->rotation_list, head);
 }
 
 static void get_ctx(struct perf_event_context *ctx)
@@ -151,13 +147,13 @@
  * the context could get moved to another task.
  */
 static struct perf_event_context *
-perf_lock_task_context(struct task_struct *task, unsigned long *flags)
+perf_lock_task_context(struct task_struct *task, int ctxn, unsigned long *flags)
 {
 	struct perf_event_context *ctx;
 
 	rcu_read_lock();
- retry:
-	ctx = rcu_dereference(task->perf_event_ctxp);
+retry:
+	ctx = rcu_dereference(task->perf_event_ctxp[ctxn]);
 	if (ctx) {
 		/*
 		 * If this context is a clone of another, it might
@@ -170,7 +166,7 @@
 		 * can't get swapped on us any more.
 		 */
 		raw_spin_lock_irqsave(&ctx->lock, *flags);
-		if (ctx != rcu_dereference(task->perf_event_ctxp)) {
+		if (ctx != rcu_dereference(task->perf_event_ctxp[ctxn])) {
 			raw_spin_unlock_irqrestore(&ctx->lock, *flags);
 			goto retry;
 		}
@@ -189,12 +185,13 @@
  * can't get swapped to another task.  This also increments its
  * reference count so that the context can't get freed.
  */
-static struct perf_event_context *perf_pin_task_context(struct task_struct *task)
+static struct perf_event_context *
+perf_pin_task_context(struct task_struct *task, int ctxn)
 {
 	struct perf_event_context *ctx;
 	unsigned long flags;
 
-	ctx = perf_lock_task_context(task, &flags);
+	ctx = perf_lock_task_context(task, ctxn, &flags);
 	if (ctx) {
 		++ctx->pin_count;
 		raw_spin_unlock_irqrestore(&ctx->lock, flags);
@@ -302,6 +299,8 @@
 	}
 
 	list_add_rcu(&event->event_entry, &ctx->event_list);
+	if (!ctx->nr_events)
+		perf_pmu_rotate_start(ctx->pmu);
 	ctx->nr_events++;
 	if (event->attr.inherit_stat)
 		ctx->nr_stat++;
@@ -436,7 +435,7 @@
 		event->state = PERF_EVENT_STATE_OFF;
 	}
 	event->tstamp_stopped = ctx->time;
-	event->pmu->disable(event);
+	event->pmu->del(event, 0);
 	event->oncpu = -1;
 
 	if (!is_software_event(event))
@@ -466,6 +465,12 @@
 		cpuctx->exclusive = 0;
 }
 
+static inline struct perf_cpu_context *
+__get_cpu_context(struct perf_event_context *ctx)
+{
+	return this_cpu_ptr(ctx->pmu->pmu_cpu_context);
+}
+
 /*
  * Cross CPU call to remove a performance event
  *
@@ -474,9 +479,9 @@
  */
 static void __perf_event_remove_from_context(void *info)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event *event = info;
 	struct perf_event_context *ctx = event->ctx;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	/*
 	 * If this is a task context, we need to check whether it is
@@ -487,27 +492,11 @@
 		return;
 
 	raw_spin_lock(&ctx->lock);
-	/*
-	 * Protect the list operation against NMI by disabling the
-	 * events on a global level.
-	 */
-	perf_disable();
 
 	event_sched_out(event, cpuctx, ctx);
 
 	list_del_event(event, ctx);
 
-	if (!ctx->task) {
-		/*
-		 * Allow more per task events with respect to the
-		 * reservation:
-		 */
-		cpuctx->max_pertask =
-			min(perf_max_events - ctx->nr_events,
-			    perf_max_events - perf_reserved_percpu);
-	}
-
-	perf_enable();
 	raw_spin_unlock(&ctx->lock);
 }
 
@@ -572,8 +561,8 @@
 static void __perf_event_disable(void *info)
 {
 	struct perf_event *event = info;
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event_context *ctx = event->ctx;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	/*
 	 * If this is a per-task event, need to check whether this
@@ -628,7 +617,7 @@
 		return;
 	}
 
- retry:
+retry:
 	task_oncpu_function_call(task, __perf_event_disable, event);
 
 	raw_spin_lock_irq(&ctx->lock);
@@ -667,7 +656,7 @@
 	 */
 	smp_wmb();
 
-	if (event->pmu->enable(event)) {
+	if (event->pmu->add(event, PERF_EF_START)) {
 		event->state = PERF_EVENT_STATE_INACTIVE;
 		event->oncpu = -1;
 		return -EAGAIN;
@@ -691,22 +680,15 @@
 	       struct perf_event_context *ctx)
 {
 	struct perf_event *event, *partial_group = NULL;
-	const struct pmu *pmu = group_event->pmu;
-	bool txn = false;
+	struct pmu *pmu = group_event->pmu;
 
 	if (group_event->state == PERF_EVENT_STATE_OFF)
 		return 0;
 
-	/* Check if group transaction availabe */
-	if (pmu->start_txn)
-		txn = true;
-
-	if (txn)
-		pmu->start_txn(pmu);
+	pmu->start_txn(pmu);
 
 	if (event_sched_in(group_event, cpuctx, ctx)) {
-		if (txn)
-			pmu->cancel_txn(pmu);
+		pmu->cancel_txn(pmu);
 		return -EAGAIN;
 	}
 
@@ -720,7 +702,7 @@
 		}
 	}
 
-	if (!txn || !pmu->commit_txn(pmu))
+	if (!pmu->commit_txn(pmu))
 		return 0;
 
 group_error:
@@ -735,8 +717,7 @@
 	}
 	event_sched_out(group_event, cpuctx, ctx);
 
-	if (txn)
-		pmu->cancel_txn(pmu);
+	pmu->cancel_txn(pmu);
 
 	return -EAGAIN;
 }
@@ -789,10 +770,10 @@
  */
 static void __perf_install_in_context(void *info)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event *event = info;
 	struct perf_event_context *ctx = event->ctx;
 	struct perf_event *leader = event->group_leader;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 	int err;
 
 	/*
@@ -812,12 +793,6 @@
 	ctx->is_active = 1;
 	update_context_time(ctx);
 
-	/*
-	 * Protect the list operation against NMI by disabling the
-	 * events on a global level. NOP for non NMI based events.
-	 */
-	perf_disable();
-
 	add_event_to_ctx(event, ctx);
 
 	if (event->cpu != -1 && event->cpu != smp_processor_id())
@@ -855,12 +830,7 @@
 		}
 	}
 
-	if (!err && !ctx->task && cpuctx->max_pertask)
-		cpuctx->max_pertask--;
-
- unlock:
-	perf_enable();
-
+unlock:
 	raw_spin_unlock(&ctx->lock);
 }
 
@@ -883,6 +853,8 @@
 {
 	struct task_struct *task = ctx->task;
 
+	event->ctx = ctx;
+
 	if (!task) {
 		/*
 		 * Per cpu events are installed via an smp call and
@@ -931,10 +903,12 @@
 
 	event->state = PERF_EVENT_STATE_INACTIVE;
 	event->tstamp_enabled = ctx->time - event->total_time_enabled;
-	list_for_each_entry(sub, &event->sibling_list, group_entry)
-		if (sub->state >= PERF_EVENT_STATE_INACTIVE)
+	list_for_each_entry(sub, &event->sibling_list, group_entry) {
+		if (sub->state >= PERF_EVENT_STATE_INACTIVE) {
 			sub->tstamp_enabled =
 				ctx->time - sub->total_time_enabled;
+		}
+	}
 }
 
 /*
@@ -943,9 +917,9 @@
 static void __perf_event_enable(void *info)
 {
 	struct perf_event *event = info;
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event_context *ctx = event->ctx;
 	struct perf_event *leader = event->group_leader;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 	int err;
 
 	/*
@@ -979,12 +953,10 @@
 	if (!group_can_go_on(event, cpuctx, 1)) {
 		err = -EEXIST;
 	} else {
-		perf_disable();
 		if (event == leader)
 			err = group_sched_in(event, cpuctx, ctx);
 		else
 			err = event_sched_in(event, cpuctx, ctx);
-		perf_enable();
 	}
 
 	if (err) {
@@ -1000,7 +972,7 @@
 		}
 	}
 
- unlock:
+unlock:
 	raw_spin_unlock(&ctx->lock);
 }
 
@@ -1041,7 +1013,7 @@
 	if (event->state == PERF_EVENT_STATE_ERROR)
 		event->state = PERF_EVENT_STATE_OFF;
 
- retry:
+retry:
 	raw_spin_unlock_irq(&ctx->lock);
 	task_oncpu_function_call(task, __perf_event_enable, event);
 
@@ -1061,7 +1033,7 @@
 	if (event->state == PERF_EVENT_STATE_OFF)
 		__perf_event_mark_enabled(event, ctx);
 
- out:
+out:
 	raw_spin_unlock_irq(&ctx->lock);
 }
 
@@ -1092,26 +1064,26 @@
 	struct perf_event *event;
 
 	raw_spin_lock(&ctx->lock);
+	perf_pmu_disable(ctx->pmu);
 	ctx->is_active = 0;
 	if (likely(!ctx->nr_events))
 		goto out;
 	update_context_time(ctx);
 
-	perf_disable();
 	if (!ctx->nr_active)
-		goto out_enable;
+		goto out;
 
-	if (event_type & EVENT_PINNED)
+	if (event_type & EVENT_PINNED) {
 		list_for_each_entry(event, &ctx->pinned_groups, group_entry)
 			group_sched_out(event, cpuctx, ctx);
+	}
 
-	if (event_type & EVENT_FLEXIBLE)
+	if (event_type & EVENT_FLEXIBLE) {
 		list_for_each_entry(event, &ctx->flexible_groups, group_entry)
 			group_sched_out(event, cpuctx, ctx);
-
- out_enable:
-	perf_enable();
- out:
+	}
+out:
+	perf_pmu_enable(ctx->pmu);
 	raw_spin_unlock(&ctx->lock);
 }
 
@@ -1209,34 +1181,25 @@
 	}
 }
 
-/*
- * Called from scheduler to remove the events of the current task,
- * with interrupts disabled.
- *
- * We stop each event and update the event value in event->count.
- *
- * This does not protect us against NMI, but disable()
- * sets the disabled bit in the control field of event _before_
- * accessing the event control register. If a NMI hits, then it will
- * not restart the event.
- */
-void perf_event_task_sched_out(struct task_struct *task,
-				 struct task_struct *next)
+void perf_event_context_sched_out(struct task_struct *task, int ctxn,
+				  struct task_struct *next)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	struct perf_event_context *ctx = task->perf_event_ctxp;
+	struct perf_event_context *ctx = task->perf_event_ctxp[ctxn];
 	struct perf_event_context *next_ctx;
 	struct perf_event_context *parent;
+	struct perf_cpu_context *cpuctx;
 	int do_switch = 1;
 
-	perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, NULL, 0);
+	if (likely(!ctx))
+		return;
 
-	if (likely(!ctx || !cpuctx->task_ctx))
+	cpuctx = __get_cpu_context(ctx);
+	if (!cpuctx->task_ctx)
 		return;
 
 	rcu_read_lock();
 	parent = rcu_dereference(ctx->parent_ctx);
-	next_ctx = next->perf_event_ctxp;
+	next_ctx = next->perf_event_ctxp[ctxn];
 	if (parent && next_ctx &&
 	    rcu_dereference(next_ctx->parent_ctx) == parent) {
 		/*
@@ -1255,8 +1218,8 @@
 			 * XXX do we need a memory barrier of sorts
 			 * wrt to rcu_dereference() of perf_event_ctxp
 			 */
-			task->perf_event_ctxp = next_ctx;
-			next->perf_event_ctxp = ctx;
+			task->perf_event_ctxp[ctxn] = next_ctx;
+			next->perf_event_ctxp[ctxn] = ctx;
 			ctx->task = next;
 			next_ctx->task = task;
 			do_switch = 0;
@@ -1274,10 +1237,35 @@
 	}
 }
 
+#define for_each_task_context_nr(ctxn)					\
+	for ((ctxn) = 0; (ctxn) < perf_nr_task_contexts; (ctxn)++)
+
+/*
+ * Called from scheduler to remove the events of the current task,
+ * with interrupts disabled.
+ *
+ * We stop each event and update the event value in event->count.
+ *
+ * This does not protect us against NMI, but disable()
+ * sets the disabled bit in the control field of event _before_
+ * accessing the event control register. If a NMI hits, then it will
+ * not restart the event.
+ */
+void perf_event_task_sched_out(struct task_struct *task,
+			       struct task_struct *next)
+{
+	int ctxn;
+
+	perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 1, NULL, 0);
+
+	for_each_task_context_nr(ctxn)
+		perf_event_context_sched_out(task, ctxn, next);
+}
+
 static void task_ctx_sched_out(struct perf_event_context *ctx,
 			       enum event_type_t event_type)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	if (!cpuctx->task_ctx)
 		return;
@@ -1350,9 +1338,10 @@
 		if (event->cpu != -1 && event->cpu != smp_processor_id())
 			continue;
 
-		if (group_can_go_on(event, cpuctx, can_add_hw))
+		if (group_can_go_on(event, cpuctx, can_add_hw)) {
 			if (group_sched_in(event, cpuctx, ctx))
 				can_add_hw = 0;
+		}
 	}
 }
 
@@ -1368,8 +1357,6 @@
 
 	ctx->timestamp = perf_clock();
 
-	perf_disable();
-
 	/*
 	 * First go through the list and put on any pinned groups
 	 * in order to give them the best chance of going on.
@@ -1381,8 +1368,7 @@
 	if (event_type & EVENT_FLEXIBLE)
 		ctx_flexible_sched_in(ctx, cpuctx);
 
-	perf_enable();
- out:
+out:
 	raw_spin_unlock(&ctx->lock);
 }
 
@@ -1394,43 +1380,28 @@
 	ctx_sched_in(ctx, cpuctx, event_type);
 }
 
-static void task_ctx_sched_in(struct task_struct *task,
+static void task_ctx_sched_in(struct perf_event_context *ctx,
 			      enum event_type_t event_type)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	struct perf_event_context *ctx = task->perf_event_ctxp;
+	struct perf_cpu_context *cpuctx;
 
-	if (likely(!ctx))
-		return;
+       	cpuctx = __get_cpu_context(ctx);
 	if (cpuctx->task_ctx == ctx)
 		return;
+
 	ctx_sched_in(ctx, cpuctx, event_type);
 	cpuctx->task_ctx = ctx;
 }
-/*
- * Called from scheduler to add the events of the current task
- * with interrupts disabled.
- *
- * We restore the event value and then enable it.
- *
- * This does not protect us against NMI, but enable()
- * sets the enabled bit in the control field of event _before_
- * accessing the event control register. If a NMI hits, then it will
- * keep the event running.
- */
-void perf_event_task_sched_in(struct task_struct *task)
+
+void perf_event_context_sched_in(struct perf_event_context *ctx)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	struct perf_event_context *ctx = task->perf_event_ctxp;
+	struct perf_cpu_context *cpuctx;
 
-	if (likely(!ctx))
-		return;
-
+	cpuctx = __get_cpu_context(ctx);
 	if (cpuctx->task_ctx == ctx)
 		return;
 
-	perf_disable();
-
+	perf_pmu_disable(ctx->pmu);
 	/*
 	 * We want to keep the following priority order:
 	 * cpu pinned (that don't need to move), task pinned,
@@ -1444,7 +1415,37 @@
 
 	cpuctx->task_ctx = ctx;
 
-	perf_enable();
+	/*
+	 * Since these rotations are per-cpu, we need to ensure the
+	 * cpu-context we got scheduled on is actually rotating.
+	 */
+	perf_pmu_rotate_start(ctx->pmu);
+	perf_pmu_enable(ctx->pmu);
+}
+
+/*
+ * Called from scheduler to add the events of the current task
+ * with interrupts disabled.
+ *
+ * We restore the event value and then enable it.
+ *
+ * This does not protect us against NMI, but enable()
+ * sets the enabled bit in the control field of event _before_
+ * accessing the event control register. If a NMI hits, then it will
+ * keep the event running.
+ */
+void perf_event_task_sched_in(struct task_struct *task)
+{
+	struct perf_event_context *ctx;
+	int ctxn;
+
+	for_each_task_context_nr(ctxn) {
+		ctx = task->perf_event_ctxp[ctxn];
+		if (likely(!ctx))
+			continue;
+
+		perf_event_context_sched_in(ctx);
+	}
 }
 
 #define MAX_INTERRUPTS (~0ULL)
@@ -1524,22 +1525,6 @@
 	return div64_u64(dividend, divisor);
 }
 
-static void perf_event_stop(struct perf_event *event)
-{
-	if (!event->pmu->stop)
-		return event->pmu->disable(event);
-
-	return event->pmu->stop(event);
-}
-
-static int perf_event_start(struct perf_event *event)
-{
-	if (!event->pmu->start)
-		return event->pmu->enable(event);
-
-	return event->pmu->start(event);
-}
-
 static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count)
 {
 	struct hw_perf_event *hwc = &event->hw;
@@ -1559,15 +1544,13 @@
 	hwc->sample_period = sample_period;
 
 	if (local64_read(&hwc->period_left) > 8*sample_period) {
-		perf_disable();
-		perf_event_stop(event);
+		event->pmu->stop(event, PERF_EF_UPDATE);
 		local64_set(&hwc->period_left, 0);
-		perf_event_start(event);
-		perf_enable();
+		event->pmu->start(event, PERF_EF_RELOAD);
 	}
 }
 
-static void perf_ctx_adjust_freq(struct perf_event_context *ctx)
+static void perf_ctx_adjust_freq(struct perf_event_context *ctx, u64 period)
 {
 	struct perf_event *event;
 	struct hw_perf_event *hwc;
@@ -1592,23 +1575,19 @@
 		 */
 		if (interrupts == MAX_INTERRUPTS) {
 			perf_log_throttle(event, 1);
-			perf_disable();
-			event->pmu->unthrottle(event);
-			perf_enable();
+			event->pmu->start(event, 0);
 		}
 
 		if (!event->attr.freq || !event->attr.sample_freq)
 			continue;
 
-		perf_disable();
 		event->pmu->read(event);
 		now = local64_read(&event->count);
 		delta = now - hwc->freq_count_stamp;
 		hwc->freq_count_stamp = now;
 
 		if (delta > 0)
-			perf_adjust_period(event, TICK_NSEC, delta);
-		perf_enable();
+			perf_adjust_period(event, period, delta);
 	}
 	raw_spin_unlock(&ctx->lock);
 }
@@ -1626,32 +1605,38 @@
 	raw_spin_unlock(&ctx->lock);
 }
 
-void perf_event_task_tick(struct task_struct *curr)
+/*
+ * perf_pmu_rotate_start() and perf_rotate_context() are fully serialized
+ * because they're strictly cpu affine and rotate_start is called with IRQs
+ * disabled, while rotate_context is called from IRQ context.
+ */
+static void perf_rotate_context(struct perf_cpu_context *cpuctx)
 {
-	struct perf_cpu_context *cpuctx;
-	struct perf_event_context *ctx;
-	int rotate = 0;
+	u64 interval = (u64)cpuctx->jiffies_interval * TICK_NSEC;
+	struct perf_event_context *ctx = NULL;
+	int rotate = 0, remove = 1;
 
-	if (!atomic_read(&nr_events))
-		return;
+	if (cpuctx->ctx.nr_events) {
+		remove = 0;
+		if (cpuctx->ctx.nr_events != cpuctx->ctx.nr_active)
+			rotate = 1;
+	}
 
-	cpuctx = &__get_cpu_var(perf_cpu_context);
-	if (cpuctx->ctx.nr_events &&
-	    cpuctx->ctx.nr_events != cpuctx->ctx.nr_active)
-		rotate = 1;
+	ctx = cpuctx->task_ctx;
+	if (ctx && ctx->nr_events) {
+		remove = 0;
+		if (ctx->nr_events != ctx->nr_active)
+			rotate = 1;
+	}
 
-	ctx = curr->perf_event_ctxp;
-	if (ctx && ctx->nr_events && ctx->nr_events != ctx->nr_active)
-		rotate = 1;
-
-	perf_ctx_adjust_freq(&cpuctx->ctx);
+	perf_pmu_disable(cpuctx->ctx.pmu);
+	perf_ctx_adjust_freq(&cpuctx->ctx, interval);
 	if (ctx)
-		perf_ctx_adjust_freq(ctx);
+		perf_ctx_adjust_freq(ctx, interval);
 
 	if (!rotate)
-		return;
+		goto done;
 
-	perf_disable();
 	cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE);
 	if (ctx)
 		task_ctx_sched_out(ctx, EVENT_FLEXIBLE);
@@ -1662,8 +1647,27 @@
 
 	cpu_ctx_sched_in(cpuctx, EVENT_FLEXIBLE);
 	if (ctx)
-		task_ctx_sched_in(curr, EVENT_FLEXIBLE);
-	perf_enable();
+		task_ctx_sched_in(ctx, EVENT_FLEXIBLE);
+
+done:
+	if (remove)
+		list_del_init(&cpuctx->rotation_list);
+
+	perf_pmu_enable(cpuctx->ctx.pmu);
+}
+
+void perf_event_task_tick(void)
+{
+	struct list_head *head = &__get_cpu_var(rotation_list);
+	struct perf_cpu_context *cpuctx, *tmp;
+
+	WARN_ON(!irqs_disabled());
+
+	list_for_each_entry_safe(cpuctx, tmp, head, rotation_list) {
+		if (cpuctx->jiffies_interval == 1 ||
+				!(jiffies % cpuctx->jiffies_interval))
+			perf_rotate_context(cpuctx);
+	}
 }
 
 static int event_enable_on_exec(struct perf_event *event,
@@ -1685,20 +1689,18 @@
  * Enable all of a task's events that have been marked enable-on-exec.
  * This expects task == current.
  */
-static void perf_event_enable_on_exec(struct task_struct *task)
+static void perf_event_enable_on_exec(struct perf_event_context *ctx)
 {
-	struct perf_event_context *ctx;
 	struct perf_event *event;
 	unsigned long flags;
 	int enabled = 0;
 	int ret;
 
 	local_irq_save(flags);
-	ctx = task->perf_event_ctxp;
 	if (!ctx || !ctx->nr_events)
 		goto out;
 
-	__perf_event_task_sched_out(ctx);
+	task_ctx_sched_out(ctx, EVENT_ALL);
 
 	raw_spin_lock(&ctx->lock);
 
@@ -1722,8 +1724,8 @@
 
 	raw_spin_unlock(&ctx->lock);
 
-	perf_event_task_sched_in(task);
- out:
+	perf_event_context_sched_in(ctx);
+out:
 	local_irq_restore(flags);
 }
 
@@ -1732,9 +1734,9 @@
  */
 static void __perf_event_read(void *info)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
 	struct perf_event *event = info;
 	struct perf_event_context *ctx = event->ctx;
+	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	/*
 	 * If this is a task context, we need to check whether it is
@@ -1782,11 +1784,219 @@
 }
 
 /*
+ * Callchain support
+ */
+
+struct callchain_cpus_entries {
+	struct rcu_head			rcu_head;
+	struct perf_callchain_entry	*cpu_entries[0];
+};
+
+static DEFINE_PER_CPU(int, callchain_recursion[PERF_NR_CONTEXTS]);
+static atomic_t nr_callchain_events;
+static DEFINE_MUTEX(callchain_mutex);
+struct callchain_cpus_entries *callchain_cpus_entries;
+
+
+__weak void perf_callchain_kernel(struct perf_callchain_entry *entry,
+				  struct pt_regs *regs)
+{
+}
+
+__weak void perf_callchain_user(struct perf_callchain_entry *entry,
+				struct pt_regs *regs)
+{
+}
+
+static void release_callchain_buffers_rcu(struct rcu_head *head)
+{
+	struct callchain_cpus_entries *entries;
+	int cpu;
+
+	entries = container_of(head, struct callchain_cpus_entries, rcu_head);
+
+	for_each_possible_cpu(cpu)
+		kfree(entries->cpu_entries[cpu]);
+
+	kfree(entries);
+}
+
+static void release_callchain_buffers(void)
+{
+	struct callchain_cpus_entries *entries;
+
+	entries = callchain_cpus_entries;
+	rcu_assign_pointer(callchain_cpus_entries, NULL);
+	call_rcu(&entries->rcu_head, release_callchain_buffers_rcu);
+}
+
+static int alloc_callchain_buffers(void)
+{
+	int cpu;
+	int size;
+	struct callchain_cpus_entries *entries;
+
+	/*
+	 * We can't use the percpu allocation API for data that can be
+	 * accessed from NMI. Use a temporary manual per cpu allocation
+	 * until that gets sorted out.
+	 */
+	size = sizeof(*entries) + sizeof(struct perf_callchain_entry *) *
+		num_possible_cpus();
+
+	entries = kzalloc(size, GFP_KERNEL);
+	if (!entries)
+		return -ENOMEM;
+
+	size = sizeof(struct perf_callchain_entry) * PERF_NR_CONTEXTS;
+
+	for_each_possible_cpu(cpu) {
+		entries->cpu_entries[cpu] = kmalloc_node(size, GFP_KERNEL,
+							 cpu_to_node(cpu));
+		if (!entries->cpu_entries[cpu])
+			goto fail;
+	}
+
+	rcu_assign_pointer(callchain_cpus_entries, entries);
+
+	return 0;
+
+fail:
+	for_each_possible_cpu(cpu)
+		kfree(entries->cpu_entries[cpu]);
+	kfree(entries);
+
+	return -ENOMEM;
+}
+
+static int get_callchain_buffers(void)
+{
+	int err = 0;
+	int count;
+
+	mutex_lock(&callchain_mutex);
+
+	count = atomic_inc_return(&nr_callchain_events);
+	if (WARN_ON_ONCE(count < 1)) {
+		err = -EINVAL;
+		goto exit;
+	}
+
+	if (count > 1) {
+		/* If the allocation failed, give up */
+		if (!callchain_cpus_entries)
+			err = -ENOMEM;
+		goto exit;
+	}
+
+	err = alloc_callchain_buffers();
+	if (err)
+		release_callchain_buffers();
+exit:
+	mutex_unlock(&callchain_mutex);
+
+	return err;
+}
+
+static void put_callchain_buffers(void)
+{
+	if (atomic_dec_and_mutex_lock(&nr_callchain_events, &callchain_mutex)) {
+		release_callchain_buffers();
+		mutex_unlock(&callchain_mutex);
+	}
+}
+
+static int get_recursion_context(int *recursion)
+{
+	int rctx;
+
+	if (in_nmi())
+		rctx = 3;
+	else if (in_irq())
+		rctx = 2;
+	else if (in_softirq())
+		rctx = 1;
+	else
+		rctx = 0;
+
+	if (recursion[rctx])
+		return -1;
+
+	recursion[rctx]++;
+	barrier();
+
+	return rctx;
+}
+
+static inline void put_recursion_context(int *recursion, int rctx)
+{
+	barrier();
+	recursion[rctx]--;
+}
+
+static struct perf_callchain_entry *get_callchain_entry(int *rctx)
+{
+	int cpu;
+	struct callchain_cpus_entries *entries;
+
+	*rctx = get_recursion_context(__get_cpu_var(callchain_recursion));
+	if (*rctx == -1)
+		return NULL;
+
+	entries = rcu_dereference(callchain_cpus_entries);
+	if (!entries)
+		return NULL;
+
+	cpu = smp_processor_id();
+
+	return &entries->cpu_entries[cpu][*rctx];
+}
+
+static void
+put_callchain_entry(int rctx)
+{
+	put_recursion_context(__get_cpu_var(callchain_recursion), rctx);
+}
+
+static struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
+{
+	int rctx;
+	struct perf_callchain_entry *entry;
+
+
+	entry = get_callchain_entry(&rctx);
+	if (rctx == -1)
+		return NULL;
+
+	if (!entry)
+		goto exit_put;
+
+	entry->nr = 0;
+
+	if (!user_mode(regs)) {
+		perf_callchain_store(entry, PERF_CONTEXT_KERNEL);
+		perf_callchain_kernel(entry, regs);
+		if (current->mm)
+			regs = task_pt_regs(current);
+		else
+			regs = NULL;
+	}
+
+	if (regs) {
+		perf_callchain_store(entry, PERF_CONTEXT_USER);
+		perf_callchain_user(entry, regs);
+	}
+
+exit_put:
+	put_callchain_entry(rctx);
+
+	return entry;
+}
+
+/*
  * Initialize the perf_event context in a task_struct:
  */
-static void
-__perf_event_init_context(struct perf_event_context *ctx,
-			    struct task_struct *task)
+static void __perf_event_init_context(struct perf_event_context *ctx)
 {
 	raw_spin_lock_init(&ctx->lock);
 	mutex_init(&ctx->mutex);
@@ -1794,45 +2004,38 @@
 	INIT_LIST_HEAD(&ctx->flexible_groups);
 	INIT_LIST_HEAD(&ctx->event_list);
 	atomic_set(&ctx->refcount, 1);
-	ctx->task = task;
 }
 
-static struct perf_event_context *find_get_context(pid_t pid, int cpu)
+static struct perf_event_context *
+alloc_perf_context(struct pmu *pmu, struct task_struct *task)
 {
 	struct perf_event_context *ctx;
-	struct perf_cpu_context *cpuctx;
+
+	ctx = kzalloc(sizeof(struct perf_event_context), GFP_KERNEL);
+	if (!ctx)
+		return NULL;
+
+	__perf_event_init_context(ctx);
+	if (task) {
+		ctx->task = task;
+		get_task_struct(task);
+	}
+	ctx->pmu = pmu;
+
+	return ctx;
+}
+
+static struct task_struct *
+find_lively_task_by_vpid(pid_t vpid)
+{
 	struct task_struct *task;
-	unsigned long flags;
 	int err;
 
-	if (pid == -1 && cpu != -1) {
-		/* Must be root to operate on a CPU event: */
-		if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN))
-			return ERR_PTR(-EACCES);
-
-		if (cpu < 0 || cpu >= nr_cpumask_bits)
-			return ERR_PTR(-EINVAL);
-
-		/*
-		 * We could be clever and allow to attach a event to an
-		 * offline CPU and activate it when the CPU comes up, but
-		 * that's for later.
-		 */
-		if (!cpu_online(cpu))
-			return ERR_PTR(-ENODEV);
-
-		cpuctx = &per_cpu(perf_cpu_context, cpu);
-		ctx = &cpuctx->ctx;
-		get_ctx(ctx);
-
-		return ctx;
-	}
-
 	rcu_read_lock();
-	if (!pid)
+	if (!vpid)
 		task = current;
 	else
-		task = find_task_by_vpid(pid);
+		task = find_task_by_vpid(vpid);
 	if (task)
 		get_task_struct(task);
 	rcu_read_unlock();
@@ -1852,35 +2055,79 @@
 	if (!ptrace_may_access(task, PTRACE_MODE_READ))
 		goto errout;
 
- retry:
-	ctx = perf_lock_task_context(task, &flags);
+	return task;
+errout:
+	put_task_struct(task);
+	return ERR_PTR(err);
+
+}
+
+static struct perf_event_context *
+find_get_context(struct pmu *pmu, struct task_struct *task, int cpu)
+{
+	struct perf_event_context *ctx;
+	struct perf_cpu_context *cpuctx;
+	unsigned long flags;
+	int ctxn, err;
+
+	if (!task && cpu != -1) {
+		/* Must be root to operate on a CPU event: */
+		if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN))
+			return ERR_PTR(-EACCES);
+
+		if (cpu < 0 || cpu >= nr_cpumask_bits)
+			return ERR_PTR(-EINVAL);
+
+		/*
+		 * We could be clever and allow to attach a event to an
+		 * offline CPU and activate it when the CPU comes up, but
+		 * that's for later.
+		 */
+		if (!cpu_online(cpu))
+			return ERR_PTR(-ENODEV);
+
+		cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu);
+		ctx = &cpuctx->ctx;
+		get_ctx(ctx);
+
+		return ctx;
+	}
+
+	err = -EINVAL;
+	ctxn = pmu->task_ctx_nr;
+	if (ctxn < 0)
+		goto errout;
+
+retry:
+	ctx = perf_lock_task_context(task, ctxn, &flags);
 	if (ctx) {
 		unclone_ctx(ctx);
 		raw_spin_unlock_irqrestore(&ctx->lock, flags);
 	}
 
 	if (!ctx) {
-		ctx = kzalloc(sizeof(struct perf_event_context), GFP_KERNEL);
+		ctx = alloc_perf_context(pmu, task);
 		err = -ENOMEM;
 		if (!ctx)
 			goto errout;
-		__perf_event_init_context(ctx, task);
+
 		get_ctx(ctx);
-		if (cmpxchg(&task->perf_event_ctxp, NULL, ctx)) {
+
+		if (cmpxchg(&task->perf_event_ctxp[ctxn], NULL, ctx)) {
 			/*
 			 * We raced with some other task; use
 			 * the context they set.
 			 */
+			put_task_struct(task);
 			kfree(ctx);
 			goto retry;
 		}
-		get_task_struct(task);
 	}
 
 	put_task_struct(task);
 	return ctx;
 
- errout:
+errout:
 	put_task_struct(task);
 	return ERR_PTR(err);
 }
@@ -1913,6 +2160,8 @@
 			atomic_dec(&nr_comm_events);
 		if (event->attr.task)
 			atomic_dec(&nr_task_events);
+		if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN)
+			put_callchain_buffers();
 	}
 
 	if (event->buffer) {
@@ -1923,7 +2172,9 @@
 	if (event->destroy)
 		event->destroy(event);
 
-	put_ctx(event->ctx);
+	if (event->ctx)
+		put_ctx(event->ctx);
+
 	call_rcu(&event->rcu_head, free_event_rcu);
 }
 
@@ -2344,6 +2595,9 @@
 
 static int perf_event_index(struct perf_event *event)
 {
+	if (event->hw.state & PERF_HES_STOPPED)
+		return 0;
+
 	if (event->state != PERF_EVENT_STATE_ACTIVE)
 		return 0;
 
@@ -2956,16 +3210,6 @@
 }
 
 /*
- * Callchain support -- arch specific
- */
-
-__weak struct perf_callchain_entry *perf_callchain(struct pt_regs *regs)
-{
-	return NULL;
-}
-
-
-/*
  * We assume there is only KVM supporting the callbacks.
  * Later on, we might change it to a list if there is
  * another virtualization implementation supporting the callbacks.
@@ -3071,7 +3315,7 @@
 	if (handle->wakeup != local_read(&buffer->wakeup))
 		perf_output_wakeup(handle);
 
- out:
+out:
 	preempt_enable();
 }
 
@@ -3459,14 +3703,20 @@
 	struct perf_output_handle handle;
 	struct perf_event_header header;
 
+	/* protect the callchain buffers */
+	rcu_read_lock();
+
 	perf_prepare_sample(&header, data, event, regs);
 
 	if (perf_output_begin(&handle, event, header.size, nmi, 1))
-		return;
+		goto exit;
 
 	perf_output_sample(&handle, &header, data, event);
 
 	perf_output_end(&handle);
+
+exit:
+	rcu_read_unlock();
 }
 
 /*
@@ -3580,16 +3830,27 @@
 static void perf_event_task_event(struct perf_task_event *task_event)
 {
 	struct perf_cpu_context *cpuctx;
-	struct perf_event_context *ctx = task_event->task_ctx;
+	struct perf_event_context *ctx;
+	struct pmu *pmu;
+	int ctxn;
 
 	rcu_read_lock();
-	cpuctx = &get_cpu_var(perf_cpu_context);
-	perf_event_task_ctx(&cpuctx->ctx, task_event);
-	if (!ctx)
-		ctx = rcu_dereference(current->perf_event_ctxp);
-	if (ctx)
-		perf_event_task_ctx(ctx, task_event);
-	put_cpu_var(perf_cpu_context);
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		cpuctx = get_cpu_ptr(pmu->pmu_cpu_context);
+		perf_event_task_ctx(&cpuctx->ctx, task_event);
+
+		ctx = task_event->task_ctx;
+		if (!ctx) {
+			ctxn = pmu->task_ctx_nr;
+			if (ctxn < 0)
+				goto next;
+			ctx = rcu_dereference(current->perf_event_ctxp[ctxn]);
+		}
+		if (ctx)
+			perf_event_task_ctx(ctx, task_event);
+next:
+		put_cpu_ptr(pmu->pmu_cpu_context);
+	}
 	rcu_read_unlock();
 }
 
@@ -3694,8 +3955,10 @@
 {
 	struct perf_cpu_context *cpuctx;
 	struct perf_event_context *ctx;
-	unsigned int size;
 	char comm[TASK_COMM_LEN];
+	unsigned int size;
+	struct pmu *pmu;
+	int ctxn;
 
 	memset(comm, 0, sizeof(comm));
 	strlcpy(comm, comm_event->task->comm, sizeof(comm));
@@ -3707,21 +3970,36 @@
 	comm_event->event_id.header.size = sizeof(comm_event->event_id) + size;
 
 	rcu_read_lock();
-	cpuctx = &get_cpu_var(perf_cpu_context);
-	perf_event_comm_ctx(&cpuctx->ctx, comm_event);
-	ctx = rcu_dereference(current->perf_event_ctxp);
-	if (ctx)
-		perf_event_comm_ctx(ctx, comm_event);
-	put_cpu_var(perf_cpu_context);
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		cpuctx = get_cpu_ptr(pmu->pmu_cpu_context);
+		perf_event_comm_ctx(&cpuctx->ctx, comm_event);
+
+		ctxn = pmu->task_ctx_nr;
+		if (ctxn < 0)
+			goto next;
+
+		ctx = rcu_dereference(current->perf_event_ctxp[ctxn]);
+		if (ctx)
+			perf_event_comm_ctx(ctx, comm_event);
+next:
+		put_cpu_ptr(pmu->pmu_cpu_context);
+	}
 	rcu_read_unlock();
 }
 
 void perf_event_comm(struct task_struct *task)
 {
 	struct perf_comm_event comm_event;
+	struct perf_event_context *ctx;
+	int ctxn;
 
-	if (task->perf_event_ctxp)
-		perf_event_enable_on_exec(task);
+	for_each_task_context_nr(ctxn) {
+		ctx = task->perf_event_ctxp[ctxn];
+		if (!ctx)
+			continue;
+
+		perf_event_enable_on_exec(ctx);
+	}
 
 	if (!atomic_read(&nr_comm_events))
 		return;
@@ -3823,6 +4101,8 @@
 	char tmp[16];
 	char *buf = NULL;
 	const char *name;
+	struct pmu *pmu;
+	int ctxn;
 
 	memset(tmp, 0, sizeof(tmp));
 
@@ -3875,12 +4155,23 @@
 	mmap_event->event_id.header.size = sizeof(mmap_event->event_id) + size;
 
 	rcu_read_lock();
-	cpuctx = &get_cpu_var(perf_cpu_context);
-	perf_event_mmap_ctx(&cpuctx->ctx, mmap_event, vma->vm_flags & VM_EXEC);
-	ctx = rcu_dereference(current->perf_event_ctxp);
-	if (ctx)
-		perf_event_mmap_ctx(ctx, mmap_event, vma->vm_flags & VM_EXEC);
-	put_cpu_var(perf_cpu_context);
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		cpuctx = get_cpu_ptr(pmu->pmu_cpu_context);
+		perf_event_mmap_ctx(&cpuctx->ctx, mmap_event,
+					vma->vm_flags & VM_EXEC);
+
+		ctxn = pmu->task_ctx_nr;
+		if (ctxn < 0)
+			goto next;
+
+		ctx = rcu_dereference(current->perf_event_ctxp[ctxn]);
+		if (ctx) {
+			perf_event_mmap_ctx(ctx, mmap_event,
+					vma->vm_flags & VM_EXEC);
+		}
+next:
+		put_cpu_ptr(pmu->pmu_cpu_context);
+	}
 	rcu_read_unlock();
 
 	kfree(buf);
@@ -3962,8 +4253,6 @@
 	struct hw_perf_event *hwc = &event->hw;
 	int ret = 0;
 
-	throttle = (throttle && event->pmu->unthrottle != NULL);
-
 	if (!throttle) {
 		hwc->interrupts++;
 	} else {
@@ -4031,6 +4320,17 @@
  * Generic software event infrastructure
  */
 
+struct swevent_htable {
+	struct swevent_hlist		*swevent_hlist;
+	struct mutex			hlist_mutex;
+	int				hlist_refcount;
+
+	/* Recursion avoidance in each contexts */
+	int				recursion[PERF_NR_CONTEXTS];
+};
+
+static DEFINE_PER_CPU(struct swevent_htable, swevent_htable);
+
 /*
  * We directly increment event->count and keep a second value in
  * event->hw.period_left to count intervals. This period event
@@ -4088,7 +4388,7 @@
 	}
 }
 
-static void perf_swevent_add(struct perf_event *event, u64 nr,
+static void perf_swevent_event(struct perf_event *event, u64 nr,
 			       int nmi, struct perf_sample_data *data,
 			       struct pt_regs *regs)
 {
@@ -4114,6 +4414,9 @@
 static int perf_exclude_event(struct perf_event *event,
 			      struct pt_regs *regs)
 {
+	if (event->hw.state & PERF_HES_STOPPED)
+		return 0;
+
 	if (regs) {
 		if (event->attr.exclude_user && user_mode(regs))
 			return 1;
@@ -4160,11 +4463,11 @@
 
 /* For the read side: events when they trigger */
 static inline struct hlist_head *
-find_swevent_head_rcu(struct perf_cpu_context *ctx, u64 type, u32 event_id)
+find_swevent_head_rcu(struct swevent_htable *swhash, u64 type, u32 event_id)
 {
 	struct swevent_hlist *hlist;
 
-	hlist = rcu_dereference(ctx->swevent_hlist);
+	hlist = rcu_dereference(swhash->swevent_hlist);
 	if (!hlist)
 		return NULL;
 
@@ -4173,7 +4476,7 @@
 
 /* For the event head insertion and removal in the hlist */
 static inline struct hlist_head *
-find_swevent_head(struct perf_cpu_context *ctx, struct perf_event *event)
+find_swevent_head(struct swevent_htable *swhash, struct perf_event *event)
 {
 	struct swevent_hlist *hlist;
 	u32 event_id = event->attr.config;
@@ -4184,7 +4487,7 @@
 	 * and release. Which makes the protected version suitable here.
 	 * The context lock guarantees that.
 	 */
-	hlist = rcu_dereference_protected(ctx->swevent_hlist,
+	hlist = rcu_dereference_protected(swhash->swevent_hlist,
 					  lockdep_is_held(&event->ctx->lock));
 	if (!hlist)
 		return NULL;
@@ -4197,23 +4500,19 @@
 				    struct perf_sample_data *data,
 				    struct pt_regs *regs)
 {
-	struct perf_cpu_context *cpuctx;
+	struct swevent_htable *swhash = &__get_cpu_var(swevent_htable);
 	struct perf_event *event;
 	struct hlist_node *node;
 	struct hlist_head *head;
 
-	cpuctx = &__get_cpu_var(perf_cpu_context);
-
 	rcu_read_lock();
-
-	head = find_swevent_head_rcu(cpuctx, type, event_id);
-
+	head = find_swevent_head_rcu(swhash, type, event_id);
 	if (!head)
 		goto end;
 
 	hlist_for_each_entry_rcu(event, node, head, hlist_entry) {
 		if (perf_swevent_match(event, type, event_id, data, regs))
-			perf_swevent_add(event, nr, nmi, data, regs);
+			perf_swevent_event(event, nr, nmi, data, regs);
 	}
 end:
 	rcu_read_unlock();
@@ -4221,33 +4520,17 @@
 
 int perf_swevent_get_recursion_context(void)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	int rctx;
+	struct swevent_htable *swhash = &__get_cpu_var(swevent_htable);
 
-	if (in_nmi())
-		rctx = 3;
-	else if (in_irq())
-		rctx = 2;
-	else if (in_softirq())
-		rctx = 1;
-	else
-		rctx = 0;
-
-	if (cpuctx->recursion[rctx])
-		return -1;
-
-	cpuctx->recursion[rctx]++;
-	barrier();
-
-	return rctx;
+	return get_recursion_context(swhash->recursion);
 }
 EXPORT_SYMBOL_GPL(perf_swevent_get_recursion_context);
 
 void inline perf_swevent_put_recursion_context(int rctx)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	barrier();
-	cpuctx->recursion[rctx]--;
+	struct swevent_htable *swhash = &__get_cpu_var(swevent_htable);
+
+	put_recursion_context(swhash->recursion, rctx);
 }
 
 void __perf_sw_event(u32 event_id, u64 nr, int nmi,
@@ -4273,20 +4556,20 @@
 {
 }
 
-static int perf_swevent_enable(struct perf_event *event)
+static int perf_swevent_add(struct perf_event *event, int flags)
 {
+	struct swevent_htable *swhash = &__get_cpu_var(swevent_htable);
 	struct hw_perf_event *hwc = &event->hw;
-	struct perf_cpu_context *cpuctx;
 	struct hlist_head *head;
 
-	cpuctx = &__get_cpu_var(perf_cpu_context);
-
 	if (hwc->sample_period) {
 		hwc->last_period = hwc->sample_period;
 		perf_swevent_set_period(event);
 	}
 
-	head = find_swevent_head(cpuctx, event);
+	hwc->state = !(flags & PERF_EF_START);
+
+	head = find_swevent_head(swhash, event);
 	if (WARN_ON_ONCE(!head))
 		return -EINVAL;
 
@@ -4295,202 +4578,27 @@
 	return 0;
 }
 
-static void perf_swevent_disable(struct perf_event *event)
+static void perf_swevent_del(struct perf_event *event, int flags)
 {
 	hlist_del_rcu(&event->hlist_entry);
 }
 
-static void perf_swevent_void(struct perf_event *event)
+static void perf_swevent_start(struct perf_event *event, int flags)
 {
+	event->hw.state = 0;
 }
 
-static int perf_swevent_int(struct perf_event *event)
+static void perf_swevent_stop(struct perf_event *event, int flags)
 {
-	return 0;
+	event->hw.state = PERF_HES_STOPPED;
 }
 
-static const struct pmu perf_ops_generic = {
-	.enable		= perf_swevent_enable,
-	.disable	= perf_swevent_disable,
-	.start		= perf_swevent_int,
-	.stop		= perf_swevent_void,
-	.read		= perf_swevent_read,
-	.unthrottle	= perf_swevent_void, /* hwc->interrupts already reset */
-};
-
-/*
- * hrtimer based swevent callback
- */
-
-static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer)
-{
-	enum hrtimer_restart ret = HRTIMER_RESTART;
-	struct perf_sample_data data;
-	struct pt_regs *regs;
-	struct perf_event *event;
-	u64 period;
-
-	event = container_of(hrtimer, struct perf_event, hw.hrtimer);
-	event->pmu->read(event);
-
-	perf_sample_data_init(&data, 0);
-	data.period = event->hw.last_period;
-	regs = get_irq_regs();
-
-	if (regs && !perf_exclude_event(event, regs)) {
-		if (!(event->attr.exclude_idle && current->pid == 0))
-			if (perf_event_overflow(event, 0, &data, regs))
-				ret = HRTIMER_NORESTART;
-	}
-
-	period = max_t(u64, 10000, event->hw.sample_period);
-	hrtimer_forward_now(hrtimer, ns_to_ktime(period));
-
-	return ret;
-}
-
-static void perf_swevent_start_hrtimer(struct perf_event *event)
-{
-	struct hw_perf_event *hwc = &event->hw;
-
-	hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
-	hwc->hrtimer.function = perf_swevent_hrtimer;
-	if (hwc->sample_period) {
-		u64 period;
-
-		if (hwc->remaining) {
-			if (hwc->remaining < 0)
-				period = 10000;
-			else
-				period = hwc->remaining;
-			hwc->remaining = 0;
-		} else {
-			period = max_t(u64, 10000, hwc->sample_period);
-		}
-		__hrtimer_start_range_ns(&hwc->hrtimer,
-				ns_to_ktime(period), 0,
-				HRTIMER_MODE_REL, 0);
-	}
-}
-
-static void perf_swevent_cancel_hrtimer(struct perf_event *event)
-{
-	struct hw_perf_event *hwc = &event->hw;
-
-	if (hwc->sample_period) {
-		ktime_t remaining = hrtimer_get_remaining(&hwc->hrtimer);
-		hwc->remaining = ktime_to_ns(remaining);
-
-		hrtimer_cancel(&hwc->hrtimer);
-	}
-}
-
-/*
- * Software event: cpu wall time clock
- */
-
-static void cpu_clock_perf_event_update(struct perf_event *event)
-{
-	int cpu = raw_smp_processor_id();
-	s64 prev;
-	u64 now;
-
-	now = cpu_clock(cpu);
-	prev = local64_xchg(&event->hw.prev_count, now);
-	local64_add(now - prev, &event->count);
-}
-
-static int cpu_clock_perf_event_enable(struct perf_event *event)
-{
-	struct hw_perf_event *hwc = &event->hw;
-	int cpu = raw_smp_processor_id();
-
-	local64_set(&hwc->prev_count, cpu_clock(cpu));
-	perf_swevent_start_hrtimer(event);
-
-	return 0;
-}
-
-static void cpu_clock_perf_event_disable(struct perf_event *event)
-{
-	perf_swevent_cancel_hrtimer(event);
-	cpu_clock_perf_event_update(event);
-}
-
-static void cpu_clock_perf_event_read(struct perf_event *event)
-{
-	cpu_clock_perf_event_update(event);
-}
-
-static const struct pmu perf_ops_cpu_clock = {
-	.enable		= cpu_clock_perf_event_enable,
-	.disable	= cpu_clock_perf_event_disable,
-	.read		= cpu_clock_perf_event_read,
-};
-
-/*
- * Software event: task time clock
- */
-
-static void task_clock_perf_event_update(struct perf_event *event, u64 now)
-{
-	u64 prev;
-	s64 delta;
-
-	prev = local64_xchg(&event->hw.prev_count, now);
-	delta = now - prev;
-	local64_add(delta, &event->count);
-}
-
-static int task_clock_perf_event_enable(struct perf_event *event)
-{
-	struct hw_perf_event *hwc = &event->hw;
-	u64 now;
-
-	now = event->ctx->time;
-
-	local64_set(&hwc->prev_count, now);
-
-	perf_swevent_start_hrtimer(event);
-
-	return 0;
-}
-
-static void task_clock_perf_event_disable(struct perf_event *event)
-{
-	perf_swevent_cancel_hrtimer(event);
-	task_clock_perf_event_update(event, event->ctx->time);
-
-}
-
-static void task_clock_perf_event_read(struct perf_event *event)
-{
-	u64 time;
-
-	if (!in_nmi()) {
-		update_context_time(event->ctx);
-		time = event->ctx->time;
-	} else {
-		u64 now = perf_clock();
-		u64 delta = now - event->ctx->timestamp;
-		time = event->ctx->time + delta;
-	}
-
-	task_clock_perf_event_update(event, time);
-}
-
-static const struct pmu perf_ops_task_clock = {
-	.enable		= task_clock_perf_event_enable,
-	.disable	= task_clock_perf_event_disable,
-	.read		= task_clock_perf_event_read,
-};
-
 /* Deref the hlist from the update side */
 static inline struct swevent_hlist *
-swevent_hlist_deref(struct perf_cpu_context *cpuctx)
+swevent_hlist_deref(struct swevent_htable *swhash)
 {
-	return rcu_dereference_protected(cpuctx->swevent_hlist,
-					 lockdep_is_held(&cpuctx->hlist_mutex));
+	return rcu_dereference_protected(swhash->swevent_hlist,
+					 lockdep_is_held(&swhash->hlist_mutex));
 }
 
 static void swevent_hlist_release_rcu(struct rcu_head *rcu_head)
@@ -4501,27 +4609,27 @@
 	kfree(hlist);
 }
 
-static void swevent_hlist_release(struct perf_cpu_context *cpuctx)
+static void swevent_hlist_release(struct swevent_htable *swhash)
 {
-	struct swevent_hlist *hlist = swevent_hlist_deref(cpuctx);
+	struct swevent_hlist *hlist = swevent_hlist_deref(swhash);
 
 	if (!hlist)
 		return;
 
-	rcu_assign_pointer(cpuctx->swevent_hlist, NULL);
+	rcu_assign_pointer(swhash->swevent_hlist, NULL);
 	call_rcu(&hlist->rcu_head, swevent_hlist_release_rcu);
 }
 
 static void swevent_hlist_put_cpu(struct perf_event *event, int cpu)
 {
-	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
+	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 
-	mutex_lock(&cpuctx->hlist_mutex);
+	mutex_lock(&swhash->hlist_mutex);
 
-	if (!--cpuctx->hlist_refcount)
-		swevent_hlist_release(cpuctx);
+	if (!--swhash->hlist_refcount)
+		swevent_hlist_release(swhash);
 
-	mutex_unlock(&cpuctx->hlist_mutex);
+	mutex_unlock(&swhash->hlist_mutex);
 }
 
 static void swevent_hlist_put(struct perf_event *event)
@@ -4539,12 +4647,12 @@
 
 static int swevent_hlist_get_cpu(struct perf_event *event, int cpu)
 {
-	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
+	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 	int err = 0;
 
-	mutex_lock(&cpuctx->hlist_mutex);
+	mutex_lock(&swhash->hlist_mutex);
 
-	if (!swevent_hlist_deref(cpuctx) && cpu_online(cpu)) {
+	if (!swevent_hlist_deref(swhash) && cpu_online(cpu)) {
 		struct swevent_hlist *hlist;
 
 		hlist = kzalloc(sizeof(*hlist), GFP_KERNEL);
@@ -4552,11 +4660,11 @@
 			err = -ENOMEM;
 			goto exit;
 		}
-		rcu_assign_pointer(cpuctx->swevent_hlist, hlist);
+		rcu_assign_pointer(swhash->swevent_hlist, hlist);
 	}
-	cpuctx->hlist_refcount++;
- exit:
-	mutex_unlock(&cpuctx->hlist_mutex);
+	swhash->hlist_refcount++;
+exit:
+	mutex_unlock(&swhash->hlist_mutex);
 
 	return err;
 }
@@ -4580,7 +4688,7 @@
 	put_online_cpus();
 
 	return 0;
- fail:
+fail:
 	for_each_possible_cpu(cpu) {
 		if (cpu == failed_cpu)
 			break;
@@ -4591,17 +4699,64 @@
 	return err;
 }
 
-#ifdef CONFIG_EVENT_TRACING
+atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX];
 
-static const struct pmu perf_ops_tracepoint = {
-	.enable		= perf_trace_enable,
-	.disable	= perf_trace_disable,
-	.start		= perf_swevent_int,
-	.stop		= perf_swevent_void,
+static void sw_perf_event_destroy(struct perf_event *event)
+{
+	u64 event_id = event->attr.config;
+
+	WARN_ON(event->parent);
+
+	atomic_dec(&perf_swevent_enabled[event_id]);
+	swevent_hlist_put(event);
+}
+
+static int perf_swevent_init(struct perf_event *event)
+{
+	int event_id = event->attr.config;
+
+	if (event->attr.type != PERF_TYPE_SOFTWARE)
+		return -ENOENT;
+
+	switch (event_id) {
+	case PERF_COUNT_SW_CPU_CLOCK:
+	case PERF_COUNT_SW_TASK_CLOCK:
+		return -ENOENT;
+
+	default:
+		break;
+	}
+
+	if (event_id > PERF_COUNT_SW_MAX)
+		return -ENOENT;
+
+	if (!event->parent) {
+		int err;
+
+		err = swevent_hlist_get(event);
+		if (err)
+			return err;
+
+		atomic_inc(&perf_swevent_enabled[event_id]);
+		event->destroy = sw_perf_event_destroy;
+	}
+
+	return 0;
+}
+
+static struct pmu perf_swevent = {
+	.task_ctx_nr	= perf_sw_context,
+
+	.event_init	= perf_swevent_init,
+	.add		= perf_swevent_add,
+	.del		= perf_swevent_del,
+	.start		= perf_swevent_start,
+	.stop		= perf_swevent_stop,
 	.read		= perf_swevent_read,
-	.unthrottle	= perf_swevent_void,
 };
 
+#ifdef CONFIG_EVENT_TRACING
+
 static int perf_tp_filter_match(struct perf_event *event,
 				struct perf_sample_data *data)
 {
@@ -4645,7 +4800,7 @@
 
 	hlist_for_each_entry_rcu(event, node, head, hlist_entry) {
 		if (perf_tp_event_match(event, &data, regs))
-			perf_swevent_add(event, count, 1, &data, regs);
+			perf_swevent_event(event, count, 1, &data, regs);
 	}
 
 	perf_swevent_put_recursion_context(rctx);
@@ -4657,10 +4812,13 @@
 	perf_trace_destroy(event);
 }
 
-static const struct pmu *tp_perf_event_init(struct perf_event *event)
+static int perf_tp_event_init(struct perf_event *event)
 {
 	int err;
 
+	if (event->attr.type != PERF_TYPE_TRACEPOINT)
+		return -ENOENT;
+
 	/*
 	 * Raw tracepoint data is a severe data leak, only allow root to
 	 * have these.
@@ -4668,15 +4826,31 @@
 	if ((event->attr.sample_type & PERF_SAMPLE_RAW) &&
 			perf_paranoid_tracepoint_raw() &&
 			!capable(CAP_SYS_ADMIN))
-		return ERR_PTR(-EPERM);
+		return -EPERM;
 
 	err = perf_trace_init(event);
 	if (err)
-		return NULL;
+		return err;
 
 	event->destroy = tp_perf_event_destroy;
 
-	return &perf_ops_tracepoint;
+	return 0;
+}
+
+static struct pmu perf_tracepoint = {
+	.task_ctx_nr	= perf_sw_context,
+
+	.event_init	= perf_tp_event_init,
+	.add		= perf_trace_add,
+	.del		= perf_trace_del,
+	.start		= perf_swevent_start,
+	.stop		= perf_swevent_stop,
+	.read		= perf_swevent_read,
+};
+
+static inline void perf_tp_register(void)
+{
+	perf_pmu_register(&perf_tracepoint);
 }
 
 static int perf_event_set_filter(struct perf_event *event, void __user *arg)
@@ -4704,9 +4878,8 @@
 
 #else
 
-static const struct pmu *tp_perf_event_init(struct perf_event *event)
+static inline void perf_tp_register(void)
 {
-	return NULL;
 }
 
 static int perf_event_set_filter(struct perf_event *event, void __user *arg)
@@ -4721,24 +4894,6 @@
 #endif /* CONFIG_EVENT_TRACING */
 
 #ifdef CONFIG_HAVE_HW_BREAKPOINT
-static void bp_perf_event_destroy(struct perf_event *event)
-{
-	release_bp_slot(event);
-}
-
-static const struct pmu *bp_perf_event_init(struct perf_event *bp)
-{
-	int err;
-
-	err = register_perf_hw_breakpoint(bp);
-	if (err)
-		return ERR_PTR(err);
-
-	bp->destroy = bp_perf_event_destroy;
-
-	return &perf_ops_bp;
-}
-
 void perf_bp_event(struct perf_event *bp, void *data)
 {
 	struct perf_sample_data sample;
@@ -4746,80 +4901,382 @@
 
 	perf_sample_data_init(&sample, bp->attr.bp_addr);
 
-	if (!perf_exclude_event(bp, regs))
-		perf_swevent_add(bp, 1, 1, &sample, regs);
-}
-#else
-static const struct pmu *bp_perf_event_init(struct perf_event *bp)
-{
-	return NULL;
-}
-
-void perf_bp_event(struct perf_event *bp, void *regs)
-{
+	if (!bp->hw.state && !perf_exclude_event(bp, regs))
+		perf_swevent_event(bp, 1, 1, &sample, regs);
 }
 #endif
 
-atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX];
+/*
+ * hrtimer based swevent callback
+ */
 
-static void sw_perf_event_destroy(struct perf_event *event)
+static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer)
 {
-	u64 event_id = event->attr.config;
+	enum hrtimer_restart ret = HRTIMER_RESTART;
+	struct perf_sample_data data;
+	struct pt_regs *regs;
+	struct perf_event *event;
+	u64 period;
 
-	WARN_ON(event->parent);
+	event = container_of(hrtimer, struct perf_event, hw.hrtimer);
+	event->pmu->read(event);
 
-	atomic_dec(&perf_swevent_enabled[event_id]);
-	swevent_hlist_put(event);
+	perf_sample_data_init(&data, 0);
+	data.period = event->hw.last_period;
+	regs = get_irq_regs();
+
+	if (regs && !perf_exclude_event(event, regs)) {
+		if (!(event->attr.exclude_idle && current->pid == 0))
+			if (perf_event_overflow(event, 0, &data, regs))
+				ret = HRTIMER_NORESTART;
+	}
+
+	period = max_t(u64, 10000, event->hw.sample_period);
+	hrtimer_forward_now(hrtimer, ns_to_ktime(period));
+
+	return ret;
 }
 
-static const struct pmu *sw_perf_event_init(struct perf_event *event)
+static void perf_swevent_start_hrtimer(struct perf_event *event)
 {
-	const struct pmu *pmu = NULL;
-	u64 event_id = event->attr.config;
+	struct hw_perf_event *hwc = &event->hw;
+
+	hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	hwc->hrtimer.function = perf_swevent_hrtimer;
+	if (hwc->sample_period) {
+		s64 period = local64_read(&hwc->period_left);
+
+		if (period) {
+			if (period < 0)
+				period = 10000;
+
+			local64_set(&hwc->period_left, 0);
+		} else {
+			period = max_t(u64, 10000, hwc->sample_period);
+		}
+		__hrtimer_start_range_ns(&hwc->hrtimer,
+				ns_to_ktime(period), 0,
+				HRTIMER_MODE_REL_PINNED, 0);
+	}
+}
+
+static void perf_swevent_cancel_hrtimer(struct perf_event *event)
+{
+	struct hw_perf_event *hwc = &event->hw;
+
+	if (hwc->sample_period) {
+		ktime_t remaining = hrtimer_get_remaining(&hwc->hrtimer);
+		local64_set(&hwc->period_left, ktime_to_ns(remaining));
+
+		hrtimer_cancel(&hwc->hrtimer);
+	}
+}
+
+/*
+ * Software event: cpu wall time clock
+ */
+
+static void cpu_clock_event_update(struct perf_event *event)
+{
+	s64 prev;
+	u64 now;
+
+	now = local_clock();
+	prev = local64_xchg(&event->hw.prev_count, now);
+	local64_add(now - prev, &event->count);
+}
+
+static void cpu_clock_event_start(struct perf_event *event, int flags)
+{
+	local64_set(&event->hw.prev_count, local_clock());
+	perf_swevent_start_hrtimer(event);
+}
+
+static void cpu_clock_event_stop(struct perf_event *event, int flags)
+{
+	perf_swevent_cancel_hrtimer(event);
+	cpu_clock_event_update(event);
+}
+
+static int cpu_clock_event_add(struct perf_event *event, int flags)
+{
+	if (flags & PERF_EF_START)
+		cpu_clock_event_start(event, flags);
+
+	return 0;
+}
+
+static void cpu_clock_event_del(struct perf_event *event, int flags)
+{
+	cpu_clock_event_stop(event, flags);
+}
+
+static void cpu_clock_event_read(struct perf_event *event)
+{
+	cpu_clock_event_update(event);
+}
+
+static int cpu_clock_event_init(struct perf_event *event)
+{
+	if (event->attr.type != PERF_TYPE_SOFTWARE)
+		return -ENOENT;
+
+	if (event->attr.config != PERF_COUNT_SW_CPU_CLOCK)
+		return -ENOENT;
+
+	return 0;
+}
+
+static struct pmu perf_cpu_clock = {
+	.task_ctx_nr	= perf_sw_context,
+
+	.event_init	= cpu_clock_event_init,
+	.add		= cpu_clock_event_add,
+	.del		= cpu_clock_event_del,
+	.start		= cpu_clock_event_start,
+	.stop		= cpu_clock_event_stop,
+	.read		= cpu_clock_event_read,
+};
+
+/*
+ * Software event: task time clock
+ */
+
+static void task_clock_event_update(struct perf_event *event, u64 now)
+{
+	u64 prev;
+	s64 delta;
+
+	prev = local64_xchg(&event->hw.prev_count, now);
+	delta = now - prev;
+	local64_add(delta, &event->count);
+}
+
+static void task_clock_event_start(struct perf_event *event, int flags)
+{
+	local64_set(&event->hw.prev_count, event->ctx->time);
+	perf_swevent_start_hrtimer(event);
+}
+
+static void task_clock_event_stop(struct perf_event *event, int flags)
+{
+	perf_swevent_cancel_hrtimer(event);
+	task_clock_event_update(event, event->ctx->time);
+}
+
+static int task_clock_event_add(struct perf_event *event, int flags)
+{
+	if (flags & PERF_EF_START)
+		task_clock_event_start(event, flags);
+
+	return 0;
+}
+
+static void task_clock_event_del(struct perf_event *event, int flags)
+{
+	task_clock_event_stop(event, PERF_EF_UPDATE);
+}
+
+static void task_clock_event_read(struct perf_event *event)
+{
+	u64 time;
+
+	if (!in_nmi()) {
+		update_context_time(event->ctx);
+		time = event->ctx->time;
+	} else {
+		u64 now = perf_clock();
+		u64 delta = now - event->ctx->timestamp;
+		time = event->ctx->time + delta;
+	}
+
+	task_clock_event_update(event, time);
+}
+
+static int task_clock_event_init(struct perf_event *event)
+{
+	if (event->attr.type != PERF_TYPE_SOFTWARE)
+		return -ENOENT;
+
+	if (event->attr.config != PERF_COUNT_SW_TASK_CLOCK)
+		return -ENOENT;
+
+	return 0;
+}
+
+static struct pmu perf_task_clock = {
+	.task_ctx_nr	= perf_sw_context,
+
+	.event_init	= task_clock_event_init,
+	.add		= task_clock_event_add,
+	.del		= task_clock_event_del,
+	.start		= task_clock_event_start,
+	.stop		= task_clock_event_stop,
+	.read		= task_clock_event_read,
+};
+
+static void perf_pmu_nop_void(struct pmu *pmu)
+{
+}
+
+static int perf_pmu_nop_int(struct pmu *pmu)
+{
+	return 0;
+}
+
+static void perf_pmu_start_txn(struct pmu *pmu)
+{
+	perf_pmu_disable(pmu);
+}
+
+static int perf_pmu_commit_txn(struct pmu *pmu)
+{
+	perf_pmu_enable(pmu);
+	return 0;
+}
+
+static void perf_pmu_cancel_txn(struct pmu *pmu)
+{
+	perf_pmu_enable(pmu);
+}
+
+/*
+ * Ensures all contexts with the same task_ctx_nr have the same
+ * pmu_cpu_context too.
+ */
+static void *find_pmu_context(int ctxn)
+{
+	struct pmu *pmu;
+
+	if (ctxn < 0)
+		return NULL;
+
+	list_for_each_entry(pmu, &pmus, entry) {
+		if (pmu->task_ctx_nr == ctxn)
+			return pmu->pmu_cpu_context;
+	}
+
+	return NULL;
+}
+
+static void free_pmu_context(void * __percpu cpu_context)
+{
+	struct pmu *pmu;
+
+	mutex_lock(&pmus_lock);
+	/*
+	 * Like a real lame refcount.
+	 */
+	list_for_each_entry(pmu, &pmus, entry) {
+		if (pmu->pmu_cpu_context == cpu_context)
+			goto out;
+	}
+
+	free_percpu(cpu_context);
+out:
+	mutex_unlock(&pmus_lock);
+}
+
+int perf_pmu_register(struct pmu *pmu)
+{
+	int cpu, ret;
+
+	mutex_lock(&pmus_lock);
+	ret = -ENOMEM;
+	pmu->pmu_disable_count = alloc_percpu(int);
+	if (!pmu->pmu_disable_count)
+		goto unlock;
+
+	pmu->pmu_cpu_context = find_pmu_context(pmu->task_ctx_nr);
+	if (pmu->pmu_cpu_context)
+		goto got_cpu_context;
+
+	pmu->pmu_cpu_context = alloc_percpu(struct perf_cpu_context);
+	if (!pmu->pmu_cpu_context)
+		goto free_pdc;
+
+	for_each_possible_cpu(cpu) {
+		struct perf_cpu_context *cpuctx;
+
+		cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu);
+		__perf_event_init_context(&cpuctx->ctx);
+		cpuctx->ctx.type = cpu_context;
+		cpuctx->ctx.pmu = pmu;
+		cpuctx->jiffies_interval = 1;
+		INIT_LIST_HEAD(&cpuctx->rotation_list);
+	}
+
+got_cpu_context:
+	if (!pmu->start_txn) {
+		if (pmu->pmu_enable) {
+			/*
+			 * If we have pmu_enable/pmu_disable calls, install
+			 * transaction stubs that use that to try and batch
+			 * hardware accesses.
+			 */
+			pmu->start_txn  = perf_pmu_start_txn;
+			pmu->commit_txn = perf_pmu_commit_txn;
+			pmu->cancel_txn = perf_pmu_cancel_txn;
+		} else {
+			pmu->start_txn  = perf_pmu_nop_void;
+			pmu->commit_txn = perf_pmu_nop_int;
+			pmu->cancel_txn = perf_pmu_nop_void;
+		}
+	}
+
+	if (!pmu->pmu_enable) {
+		pmu->pmu_enable  = perf_pmu_nop_void;
+		pmu->pmu_disable = perf_pmu_nop_void;
+	}
+
+	list_add_rcu(&pmu->entry, &pmus);
+	ret = 0;
+unlock:
+	mutex_unlock(&pmus_lock);
+
+	return ret;
+
+free_pdc:
+	free_percpu(pmu->pmu_disable_count);
+	goto unlock;
+}
+
+void perf_pmu_unregister(struct pmu *pmu)
+{
+	mutex_lock(&pmus_lock);
+	list_del_rcu(&pmu->entry);
+	mutex_unlock(&pmus_lock);
 
 	/*
-	 * Software events (currently) can't in general distinguish
-	 * between user, kernel and hypervisor events.
-	 * However, context switches and cpu migrations are considered
-	 * to be kernel events, and page faults are never hypervisor
-	 * events.
+	 * We dereference the pmu list under both SRCU and regular RCU, so
+	 * synchronize against both of those.
 	 */
-	switch (event_id) {
-	case PERF_COUNT_SW_CPU_CLOCK:
-		pmu = &perf_ops_cpu_clock;
+	synchronize_srcu(&pmus_srcu);
+	synchronize_rcu();
 
-		break;
-	case PERF_COUNT_SW_TASK_CLOCK:
-		/*
-		 * If the user instantiates this as a per-cpu event,
-		 * use the cpu_clock event instead.
-		 */
-		if (event->ctx->task)
-			pmu = &perf_ops_task_clock;
-		else
-			pmu = &perf_ops_cpu_clock;
+	free_percpu(pmu->pmu_disable_count);
+	free_pmu_context(pmu->pmu_cpu_context);
+}
 
-		break;
-	case PERF_COUNT_SW_PAGE_FAULTS:
-	case PERF_COUNT_SW_PAGE_FAULTS_MIN:
-	case PERF_COUNT_SW_PAGE_FAULTS_MAJ:
-	case PERF_COUNT_SW_CONTEXT_SWITCHES:
-	case PERF_COUNT_SW_CPU_MIGRATIONS:
-	case PERF_COUNT_SW_ALIGNMENT_FAULTS:
-	case PERF_COUNT_SW_EMULATION_FAULTS:
-		if (!event->parent) {
-			int err;
+struct pmu *perf_init_event(struct perf_event *event)
+{
+	struct pmu *pmu = NULL;
+	int idx;
 
-			err = swevent_hlist_get(event);
-			if (err)
-				return ERR_PTR(err);
+	idx = srcu_read_lock(&pmus_srcu);
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		int ret = pmu->event_init(event);
+		if (!ret)
+			goto unlock;
 
-			atomic_inc(&perf_swevent_enabled[event_id]);
-			event->destroy = sw_perf_event_destroy;
+		if (ret != -ENOENT) {
+			pmu = ERR_PTR(ret);
+			goto unlock;
 		}
-		pmu = &perf_ops_generic;
-		break;
 	}
+	pmu = ERR_PTR(-ENOENT);
+unlock:
+	srcu_read_unlock(&pmus_srcu, idx);
 
 	return pmu;
 }
@@ -4828,20 +5285,17 @@
  * Allocate and initialize a event structure
  */
 static struct perf_event *
-perf_event_alloc(struct perf_event_attr *attr,
-		   int cpu,
-		   struct perf_event_context *ctx,
+perf_event_alloc(struct perf_event_attr *attr, int cpu,
 		   struct perf_event *group_leader,
 		   struct perf_event *parent_event,
-		   perf_overflow_handler_t overflow_handler,
-		   gfp_t gfpflags)
+		   perf_overflow_handler_t overflow_handler)
 {
-	const struct pmu *pmu;
+	struct pmu *pmu;
 	struct perf_event *event;
 	struct hw_perf_event *hwc;
 	long err;
 
-	event = kzalloc(sizeof(*event), gfpflags);
+	event = kzalloc(sizeof(*event), GFP_KERNEL);
 	if (!event)
 		return ERR_PTR(-ENOMEM);
 
@@ -4866,7 +5320,6 @@
 	event->attr		= *attr;
 	event->group_leader	= group_leader;
 	event->pmu		= NULL;
-	event->ctx		= ctx;
 	event->oncpu		= -1;
 
 	event->parent		= parent_event;
@@ -4900,29 +5353,8 @@
 	if (attr->inherit && (attr->read_format & PERF_FORMAT_GROUP))
 		goto done;
 
-	switch (attr->type) {
-	case PERF_TYPE_RAW:
-	case PERF_TYPE_HARDWARE:
-	case PERF_TYPE_HW_CACHE:
-		pmu = hw_perf_event_init(event);
-		break;
+	pmu = perf_init_event(event);
 
-	case PERF_TYPE_SOFTWARE:
-		pmu = sw_perf_event_init(event);
-		break;
-
-	case PERF_TYPE_TRACEPOINT:
-		pmu = tp_perf_event_init(event);
-		break;
-
-	case PERF_TYPE_BREAKPOINT:
-		pmu = bp_perf_event_init(event);
-		break;
-
-
-	default:
-		break;
-	}
 done:
 	err = 0;
 	if (!pmu)
@@ -4947,6 +5379,13 @@
 			atomic_inc(&nr_comm_events);
 		if (event->attr.task)
 			atomic_inc(&nr_task_events);
+		if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN) {
+			err = get_callchain_buffers();
+			if (err) {
+				free_event(event);
+				return ERR_PTR(err);
+			}
+		}
 	}
 
 	return event;
@@ -5094,12 +5533,16 @@
 		struct perf_event_attr __user *, attr_uptr,
 		pid_t, pid, int, cpu, int, group_fd, unsigned long, flags)
 {
-	struct perf_event *event, *group_leader = NULL, *output_event = NULL;
+	struct perf_event *group_leader = NULL, *output_event = NULL;
+	struct perf_event *event, *sibling;
 	struct perf_event_attr attr;
 	struct perf_event_context *ctx;
 	struct file *event_file = NULL;
 	struct file *group_file = NULL;
+	struct task_struct *task = NULL;
+	struct pmu *pmu;
 	int event_fd;
+	int move_group = 0;
 	int fput_needed = 0;
 	int err;
 
@@ -5125,20 +5568,11 @@
 	if (event_fd < 0)
 		return event_fd;
 
-	/*
-	 * Get the target context (task or percpu):
-	 */
-	ctx = find_get_context(pid, cpu);
-	if (IS_ERR(ctx)) {
-		err = PTR_ERR(ctx);
-		goto err_fd;
-	}
-
 	if (group_fd != -1) {
 		group_leader = perf_fget_light(group_fd, &fput_needed);
 		if (IS_ERR(group_leader)) {
 			err = PTR_ERR(group_leader);
-			goto err_put_context;
+			goto err_fd;
 		}
 		group_file = group_leader->filp;
 		if (flags & PERF_FLAG_FD_OUTPUT)
@@ -5147,6 +5581,53 @@
 			group_leader = NULL;
 	}
 
+	event = perf_event_alloc(&attr, cpu, group_leader, NULL, NULL);
+	if (IS_ERR(event)) {
+		err = PTR_ERR(event);
+		goto err_fd;
+	}
+
+	/*
+	 * Special case software events and allow them to be part of
+	 * any hardware group.
+	 */
+	pmu = event->pmu;
+
+	if (group_leader &&
+	    (is_software_event(event) != is_software_event(group_leader))) {
+		if (is_software_event(event)) {
+			/*
+			 * If event and group_leader are not both a software
+			 * event, and event is, then group leader is not.
+			 *
+			 * Allow the addition of software events to !software
+			 * groups, this is safe because software events never
+			 * fail to schedule.
+			 */
+			pmu = group_leader->pmu;
+		} else if (is_software_event(group_leader) &&
+			   (group_leader->group_flags & PERF_GROUP_SOFTWARE)) {
+			/*
+			 * In case the group is a pure software group, and we
+			 * try to add a hardware event, move the whole group to
+			 * the hardware context.
+			 */
+			move_group = 1;
+		}
+	}
+
+	if (pid != -1)
+		task = find_lively_task_by_vpid(pid);
+
+	/*
+	 * Get the target context (task or percpu):
+	 */
+	ctx = find_get_context(pmu, task, cpu);
+	if (IS_ERR(ctx)) {
+		err = PTR_ERR(ctx);
+		goto err_group_fd;
+	}
+
 	/*
 	 * Look up the group leader (we will attach this event to it):
 	 */
@@ -5158,42 +5639,66 @@
 		 * becoming part of another group-sibling):
 		 */
 		if (group_leader->group_leader != group_leader)
-			goto err_put_context;
+			goto err_context;
 		/*
 		 * Do not allow to attach to a group in a different
 		 * task or CPU context:
 		 */
-		if (group_leader->ctx != ctx)
-			goto err_put_context;
+		if (move_group) {
+			if (group_leader->ctx->type != ctx->type)
+				goto err_context;
+		} else {
+			if (group_leader->ctx != ctx)
+				goto err_context;
+		}
+
 		/*
 		 * Only a group leader can be exclusive or pinned
 		 */
 		if (attr.exclusive || attr.pinned)
-			goto err_put_context;
-	}
-
-	event = perf_event_alloc(&attr, cpu, ctx, group_leader,
-				     NULL, NULL, GFP_KERNEL);
-	if (IS_ERR(event)) {
-		err = PTR_ERR(event);
-		goto err_put_context;
+			goto err_context;
 	}
 
 	if (output_event) {
 		err = perf_event_set_output(event, output_event);
 		if (err)
-			goto err_free_put_context;
+			goto err_context;
 	}
 
 	event_file = anon_inode_getfile("[perf_event]", &perf_fops, event, O_RDWR);
 	if (IS_ERR(event_file)) {
 		err = PTR_ERR(event_file);
-		goto err_free_put_context;
+		goto err_context;
+	}
+
+	if (move_group) {
+		struct perf_event_context *gctx = group_leader->ctx;
+
+		mutex_lock(&gctx->mutex);
+		perf_event_remove_from_context(group_leader);
+		list_for_each_entry(sibling, &group_leader->sibling_list,
+				    group_entry) {
+			perf_event_remove_from_context(sibling);
+			put_ctx(gctx);
+		}
+		mutex_unlock(&gctx->mutex);
+		put_ctx(gctx);
 	}
 
 	event->filp = event_file;
 	WARN_ON_ONCE(ctx->parent_ctx);
 	mutex_lock(&ctx->mutex);
+
+	if (move_group) {
+		perf_install_in_context(ctx, group_leader, cpu);
+		get_ctx(ctx);
+		list_for_each_entry(sibling, &group_leader->sibling_list,
+				    group_entry) {
+			perf_install_in_context(ctx, sibling, cpu);
+			get_ctx(ctx);
+		}
+	}
+
 	perf_install_in_context(ctx, event, cpu);
 	++ctx->generation;
 	mutex_unlock(&ctx->mutex);
@@ -5214,11 +5719,11 @@
 	fd_install(event_fd, event_file);
 	return event_fd;
 
-err_free_put_context:
-	free_event(event);
-err_put_context:
-	fput_light(group_file, fput_needed);
+err_context:
 	put_ctx(ctx);
+err_group_fd:
+	fput_light(group_file, fput_needed);
+	free_event(event);
 err_fd:
 	put_unused_fd(event_fd);
 	return err;
@@ -5229,32 +5734,31 @@
  *
  * @attr: attributes of the counter to create
  * @cpu: cpu in which the counter is bound
- * @pid: task to profile
+ * @task: task to profile (NULL for percpu)
  */
 struct perf_event *
 perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu,
-				 pid_t pid,
+				 struct task_struct *task,
 				 perf_overflow_handler_t overflow_handler)
 {
-	struct perf_event *event;
 	struct perf_event_context *ctx;
+	struct perf_event *event;
 	int err;
 
 	/*
 	 * Get the target context (task or percpu):
 	 */
 
-	ctx = find_get_context(pid, cpu);
-	if (IS_ERR(ctx)) {
-		err = PTR_ERR(ctx);
-		goto err_exit;
-	}
-
-	event = perf_event_alloc(attr, cpu, ctx, NULL,
-				 NULL, overflow_handler, GFP_KERNEL);
+	event = perf_event_alloc(attr, cpu, NULL, NULL, overflow_handler);
 	if (IS_ERR(event)) {
 		err = PTR_ERR(event);
-		goto err_put_context;
+		goto err;
+	}
+
+	ctx = find_get_context(event->pmu, task, cpu);
+	if (IS_ERR(ctx)) {
+		err = PTR_ERR(ctx);
+		goto err_free;
 	}
 
 	event->filp = NULL;
@@ -5272,112 +5776,13 @@
 
 	return event;
 
- err_put_context:
-	put_ctx(ctx);
- err_exit:
+err_free:
+	free_event(event);
+err:
 	return ERR_PTR(err);
 }
 EXPORT_SYMBOL_GPL(perf_event_create_kernel_counter);
 
-/*
- * inherit a event from parent task to child task:
- */
-static struct perf_event *
-inherit_event(struct perf_event *parent_event,
-	      struct task_struct *parent,
-	      struct perf_event_context *parent_ctx,
-	      struct task_struct *child,
-	      struct perf_event *group_leader,
-	      struct perf_event_context *child_ctx)
-{
-	struct perf_event *child_event;
-
-	/*
-	 * Instead of creating recursive hierarchies of events,
-	 * we link inherited events back to the original parent,
-	 * which has a filp for sure, which we use as the reference
-	 * count:
-	 */
-	if (parent_event->parent)
-		parent_event = parent_event->parent;
-
-	child_event = perf_event_alloc(&parent_event->attr,
-					   parent_event->cpu, child_ctx,
-					   group_leader, parent_event,
-					   NULL, GFP_KERNEL);
-	if (IS_ERR(child_event))
-		return child_event;
-	get_ctx(child_ctx);
-
-	/*
-	 * Make the child state follow the state of the parent event,
-	 * not its attr.disabled bit.  We hold the parent's mutex,
-	 * so we won't race with perf_event_{en, dis}able_family.
-	 */
-	if (parent_event->state >= PERF_EVENT_STATE_INACTIVE)
-		child_event->state = PERF_EVENT_STATE_INACTIVE;
-	else
-		child_event->state = PERF_EVENT_STATE_OFF;
-
-	if (parent_event->attr.freq) {
-		u64 sample_period = parent_event->hw.sample_period;
-		struct hw_perf_event *hwc = &child_event->hw;
-
-		hwc->sample_period = sample_period;
-		hwc->last_period   = sample_period;
-
-		local64_set(&hwc->period_left, sample_period);
-	}
-
-	child_event->overflow_handler = parent_event->overflow_handler;
-
-	/*
-	 * Link it up in the child's context:
-	 */
-	add_event_to_ctx(child_event, child_ctx);
-
-	/*
-	 * Get a reference to the parent filp - we will fput it
-	 * when the child event exits. This is safe to do because
-	 * we are in the parent and we know that the filp still
-	 * exists and has a nonzero count:
-	 */
-	atomic_long_inc(&parent_event->filp->f_count);
-
-	/*
-	 * Link this into the parent event's child list
-	 */
-	WARN_ON_ONCE(parent_event->ctx->parent_ctx);
-	mutex_lock(&parent_event->child_mutex);
-	list_add_tail(&child_event->child_list, &parent_event->child_list);
-	mutex_unlock(&parent_event->child_mutex);
-
-	return child_event;
-}
-
-static int inherit_group(struct perf_event *parent_event,
-	      struct task_struct *parent,
-	      struct perf_event_context *parent_ctx,
-	      struct task_struct *child,
-	      struct perf_event_context *child_ctx)
-{
-	struct perf_event *leader;
-	struct perf_event *sub;
-	struct perf_event *child_ctr;
-
-	leader = inherit_event(parent_event, parent, parent_ctx,
-				 child, NULL, child_ctx);
-	if (IS_ERR(leader))
-		return PTR_ERR(leader);
-	list_for_each_entry(sub, &parent_event->sibling_list, group_entry) {
-		child_ctr = inherit_event(sub, parent, parent_ctx,
-					    child, leader, child_ctx);
-		if (IS_ERR(child_ctr))
-			return PTR_ERR(child_ctr);
-	}
-	return 0;
-}
-
 static void sync_child_event(struct perf_event *child_event,
 			       struct task_struct *child)
 {
@@ -5434,16 +5839,13 @@
 	}
 }
 
-/*
- * When a child task exits, feed back event values to parent events.
- */
-void perf_event_exit_task(struct task_struct *child)
+static void perf_event_exit_task_context(struct task_struct *child, int ctxn)
 {
 	struct perf_event *child_event, *tmp;
 	struct perf_event_context *child_ctx;
 	unsigned long flags;
 
-	if (likely(!child->perf_event_ctxp)) {
+	if (likely(!child->perf_event_ctxp[ctxn])) {
 		perf_event_task(child, NULL, 0);
 		return;
 	}
@@ -5455,7 +5857,7 @@
 	 * scheduled, so we are now safe from rescheduling changing
 	 * our context.
 	 */
-	child_ctx = child->perf_event_ctxp;
+	child_ctx = child->perf_event_ctxp[ctxn];
 	__perf_event_task_sched_out(child_ctx);
 
 	/*
@@ -5464,7 +5866,7 @@
 	 * incremented the context's refcount before we do put_ctx below.
 	 */
 	raw_spin_lock(&child_ctx->lock);
-	child->perf_event_ctxp = NULL;
+	child->perf_event_ctxp[ctxn] = NULL;
 	/*
 	 * If this context is a clone; unclone it so it can't get
 	 * swapped to another process while we're removing all
@@ -5517,6 +5919,17 @@
 	put_ctx(child_ctx);
 }
 
+/*
+ * When a child task exits, feed back event values to parent events.
+ */
+void perf_event_exit_task(struct task_struct *child)
+{
+	int ctxn;
+
+	for_each_task_context_nr(ctxn)
+		perf_event_exit_task_context(child, ctxn);
+}
+
 static void perf_free_event(struct perf_event *event,
 			    struct perf_event_context *ctx)
 {
@@ -5538,48 +5951,165 @@
 
 /*
  * free an unexposed, unused context as created by inheritance by
- * init_task below, used by fork() in case of fail.
+ * perf_event_init_task below, used by fork() in case of fail.
  */
 void perf_event_free_task(struct task_struct *task)
 {
-	struct perf_event_context *ctx = task->perf_event_ctxp;
+	struct perf_event_context *ctx;
 	struct perf_event *event, *tmp;
+	int ctxn;
 
-	if (!ctx)
-		return;
+	for_each_task_context_nr(ctxn) {
+		ctx = task->perf_event_ctxp[ctxn];
+		if (!ctx)
+			continue;
 
-	mutex_lock(&ctx->mutex);
+		mutex_lock(&ctx->mutex);
 again:
-	list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, group_entry)
-		perf_free_event(event, ctx);
+		list_for_each_entry_safe(event, tmp, &ctx->pinned_groups,
+				group_entry)
+			perf_free_event(event, ctx);
 
-	list_for_each_entry_safe(event, tmp, &ctx->flexible_groups,
-				 group_entry)
-		perf_free_event(event, ctx);
+		list_for_each_entry_safe(event, tmp, &ctx->flexible_groups,
+				group_entry)
+			perf_free_event(event, ctx);
 
-	if (!list_empty(&ctx->pinned_groups) ||
-	    !list_empty(&ctx->flexible_groups))
-		goto again;
+		if (!list_empty(&ctx->pinned_groups) ||
+				!list_empty(&ctx->flexible_groups))
+			goto again;
 
-	mutex_unlock(&ctx->mutex);
+		mutex_unlock(&ctx->mutex);
 
-	put_ctx(ctx);
+		put_ctx(ctx);
+	}
+}
+
+void perf_event_delayed_put(struct task_struct *task)
+{
+	int ctxn;
+
+	for_each_task_context_nr(ctxn)
+		WARN_ON_ONCE(task->perf_event_ctxp[ctxn]);
+}
+
+/*
+ * inherit a event from parent task to child task:
+ */
+static struct perf_event *
+inherit_event(struct perf_event *parent_event,
+	      struct task_struct *parent,
+	      struct perf_event_context *parent_ctx,
+	      struct task_struct *child,
+	      struct perf_event *group_leader,
+	      struct perf_event_context *child_ctx)
+{
+	struct perf_event *child_event;
+	unsigned long flags;
+
+	/*
+	 * Instead of creating recursive hierarchies of events,
+	 * we link inherited events back to the original parent,
+	 * which has a filp for sure, which we use as the reference
+	 * count:
+	 */
+	if (parent_event->parent)
+		parent_event = parent_event->parent;
+
+	child_event = perf_event_alloc(&parent_event->attr,
+					   parent_event->cpu,
+					   group_leader, parent_event,
+					   NULL);
+	if (IS_ERR(child_event))
+		return child_event;
+	get_ctx(child_ctx);
+
+	/*
+	 * Make the child state follow the state of the parent event,
+	 * not its attr.disabled bit.  We hold the parent's mutex,
+	 * so we won't race with perf_event_{en, dis}able_family.
+	 */
+	if (parent_event->state >= PERF_EVENT_STATE_INACTIVE)
+		child_event->state = PERF_EVENT_STATE_INACTIVE;
+	else
+		child_event->state = PERF_EVENT_STATE_OFF;
+
+	if (parent_event->attr.freq) {
+		u64 sample_period = parent_event->hw.sample_period;
+		struct hw_perf_event *hwc = &child_event->hw;
+
+		hwc->sample_period = sample_period;
+		hwc->last_period   = sample_period;
+
+		local64_set(&hwc->period_left, sample_period);
+	}
+
+	child_event->ctx = child_ctx;
+	child_event->overflow_handler = parent_event->overflow_handler;
+
+	/*
+	 * Link it up in the child's context:
+	 */
+	raw_spin_lock_irqsave(&child_ctx->lock, flags);
+	add_event_to_ctx(child_event, child_ctx);
+	raw_spin_unlock_irqrestore(&child_ctx->lock, flags);
+
+	/*
+	 * Get a reference to the parent filp - we will fput it
+	 * when the child event exits. This is safe to do because
+	 * we are in the parent and we know that the filp still
+	 * exists and has a nonzero count:
+	 */
+	atomic_long_inc(&parent_event->filp->f_count);
+
+	/*
+	 * Link this into the parent event's child list
+	 */
+	WARN_ON_ONCE(parent_event->ctx->parent_ctx);
+	mutex_lock(&parent_event->child_mutex);
+	list_add_tail(&child_event->child_list, &parent_event->child_list);
+	mutex_unlock(&parent_event->child_mutex);
+
+	return child_event;
+}
+
+static int inherit_group(struct perf_event *parent_event,
+	      struct task_struct *parent,
+	      struct perf_event_context *parent_ctx,
+	      struct task_struct *child,
+	      struct perf_event_context *child_ctx)
+{
+	struct perf_event *leader;
+	struct perf_event *sub;
+	struct perf_event *child_ctr;
+
+	leader = inherit_event(parent_event, parent, parent_ctx,
+				 child, NULL, child_ctx);
+	if (IS_ERR(leader))
+		return PTR_ERR(leader);
+	list_for_each_entry(sub, &parent_event->sibling_list, group_entry) {
+		child_ctr = inherit_event(sub, parent, parent_ctx,
+					    child, leader, child_ctx);
+		if (IS_ERR(child_ctr))
+			return PTR_ERR(child_ctr);
+	}
+	return 0;
 }
 
 static int
 inherit_task_group(struct perf_event *event, struct task_struct *parent,
 		   struct perf_event_context *parent_ctx,
-		   struct task_struct *child,
+		   struct task_struct *child, int ctxn,
 		   int *inherited_all)
 {
 	int ret;
-	struct perf_event_context *child_ctx = child->perf_event_ctxp;
+	struct perf_event_context *child_ctx;
 
 	if (!event->attr.inherit) {
 		*inherited_all = 0;
 		return 0;
 	}
 
+       	child_ctx = child->perf_event_ctxp[ctxn];
 	if (!child_ctx) {
 		/*
 		 * This is executed from the parent task context, so
@@ -5588,14 +6118,11 @@
 		 * child.
 		 */
 
-		child_ctx = kzalloc(sizeof(struct perf_event_context),
-				    GFP_KERNEL);
+		child_ctx = alloc_perf_context(event->pmu, child);
 		if (!child_ctx)
 			return -ENOMEM;
 
-		__perf_event_init_context(child_ctx, child);
-		child->perf_event_ctxp = child_ctx;
-		get_task_struct(child);
+		child->perf_event_ctxp[ctxn] = child_ctx;
 	}
 
 	ret = inherit_group(event, parent, parent_ctx,
@@ -5607,11 +6134,10 @@
 	return ret;
 }
 
-
 /*
  * Initialize the perf_event context in task_struct
  */
-int perf_event_init_task(struct task_struct *child)
+int perf_event_init_context(struct task_struct *child, int ctxn)
 {
 	struct perf_event_context *child_ctx, *parent_ctx;
 	struct perf_event_context *cloned_ctx;
@@ -5620,19 +6146,19 @@
 	int inherited_all = 1;
 	int ret = 0;
 
-	child->perf_event_ctxp = NULL;
+	child->perf_event_ctxp[ctxn] = NULL;
 
 	mutex_init(&child->perf_event_mutex);
 	INIT_LIST_HEAD(&child->perf_event_list);
 
-	if (likely(!parent->perf_event_ctxp))
+	if (likely(!parent->perf_event_ctxp[ctxn]))
 		return 0;
 
 	/*
 	 * If the parent's context is a clone, pin it so it won't get
 	 * swapped under us.
 	 */
-	parent_ctx = perf_pin_task_context(parent);
+	parent_ctx = perf_pin_task_context(parent, ctxn);
 
 	/*
 	 * No need to check if parent_ctx != NULL here; since we saw
@@ -5652,20 +6178,20 @@
 	 * the list, not manipulating it:
 	 */
 	list_for_each_entry(event, &parent_ctx->pinned_groups, group_entry) {
-		ret = inherit_task_group(event, parent, parent_ctx, child,
-					 &inherited_all);
+		ret = inherit_task_group(event, parent, parent_ctx,
+					 child, ctxn, &inherited_all);
 		if (ret)
 			break;
 	}
 
 	list_for_each_entry(event, &parent_ctx->flexible_groups, group_entry) {
-		ret = inherit_task_group(event, parent, parent_ctx, child,
-					 &inherited_all);
+		ret = inherit_task_group(event, parent, parent_ctx,
+					 child, ctxn, &inherited_all);
 		if (ret)
 			break;
 	}
 
-	child_ctx = child->perf_event_ctxp;
+	child_ctx = child->perf_event_ctxp[ctxn];
 
 	if (child_ctx && inherited_all) {
 		/*
@@ -5694,63 +6220,98 @@
 	return ret;
 }
 
+/*
+ * Initialize the perf_event context in task_struct
+ */
+int perf_event_init_task(struct task_struct *child)
+{
+	int ctxn, ret;
+
+	for_each_task_context_nr(ctxn) {
+		ret = perf_event_init_context(child, ctxn);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
 static void __init perf_event_init_all_cpus(void)
 {
+	struct swevent_htable *swhash;
 	int cpu;
-	struct perf_cpu_context *cpuctx;
 
 	for_each_possible_cpu(cpu) {
-		cpuctx = &per_cpu(perf_cpu_context, cpu);
-		mutex_init(&cpuctx->hlist_mutex);
-		__perf_event_init_context(&cpuctx->ctx, NULL);
+		swhash = &per_cpu(swevent_htable, cpu);
+		mutex_init(&swhash->hlist_mutex);
+		INIT_LIST_HEAD(&per_cpu(rotation_list, cpu));
 	}
 }
 
 static void __cpuinit perf_event_init_cpu(int cpu)
 {
-	struct perf_cpu_context *cpuctx;
+	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 
-	cpuctx = &per_cpu(perf_cpu_context, cpu);
-
-	spin_lock(&perf_resource_lock);
-	cpuctx->max_pertask = perf_max_events - perf_reserved_percpu;
-	spin_unlock(&perf_resource_lock);
-
-	mutex_lock(&cpuctx->hlist_mutex);
-	if (cpuctx->hlist_refcount > 0) {
+	mutex_lock(&swhash->hlist_mutex);
+	if (swhash->hlist_refcount > 0) {
 		struct swevent_hlist *hlist;
 
-		hlist = kzalloc(sizeof(*hlist), GFP_KERNEL);
-		WARN_ON_ONCE(!hlist);
-		rcu_assign_pointer(cpuctx->swevent_hlist, hlist);
+		hlist = kzalloc_node(sizeof(*hlist), GFP_KERNEL, cpu_to_node(cpu));
+		WARN_ON(!hlist);
+		rcu_assign_pointer(swhash->swevent_hlist, hlist);
 	}
-	mutex_unlock(&cpuctx->hlist_mutex);
+	mutex_unlock(&swhash->hlist_mutex);
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-static void __perf_event_exit_cpu(void *info)
+static void perf_pmu_rotate_stop(struct pmu *pmu)
 {
-	struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
-	struct perf_event_context *ctx = &cpuctx->ctx;
+	struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
+
+	WARN_ON(!irqs_disabled());
+
+	list_del_init(&cpuctx->rotation_list);
+}
+
+static void __perf_event_exit_context(void *__info)
+{
+	struct perf_event_context *ctx = __info;
 	struct perf_event *event, *tmp;
 
+	perf_pmu_rotate_stop(ctx->pmu);
+
 	list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, group_entry)
 		__perf_event_remove_from_context(event);
 	list_for_each_entry_safe(event, tmp, &ctx->flexible_groups, group_entry)
 		__perf_event_remove_from_context(event);
 }
+
+static void perf_event_exit_cpu_context(int cpu)
+{
+	struct perf_event_context *ctx;
+	struct pmu *pmu;
+	int idx;
+
+	idx = srcu_read_lock(&pmus_srcu);
+	list_for_each_entry_rcu(pmu, &pmus, entry) {
+		ctx = &per_cpu_ptr(pmu->pmu_cpu_context, cpu)->ctx;
+
+		mutex_lock(&ctx->mutex);
+		smp_call_function_single(cpu, __perf_event_exit_context, ctx, 1);
+		mutex_unlock(&ctx->mutex);
+	}
+	srcu_read_unlock(&pmus_srcu, idx);
+}
+
 static void perf_event_exit_cpu(int cpu)
 {
-	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
-	struct perf_event_context *ctx = &cpuctx->ctx;
+	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 
-	mutex_lock(&cpuctx->hlist_mutex);
-	swevent_hlist_release(cpuctx);
-	mutex_unlock(&cpuctx->hlist_mutex);
+	mutex_lock(&swhash->hlist_mutex);
+	swevent_hlist_release(swhash);
+	mutex_unlock(&swhash->hlist_mutex);
 
-	mutex_lock(&ctx->mutex);
-	smp_call_function_single(cpu, __perf_event_exit_cpu, NULL, 1);
-	mutex_unlock(&ctx->mutex);
+	perf_event_exit_cpu_context(cpu);
 }
 #else
 static inline void perf_event_exit_cpu(int cpu) { }
@@ -5780,118 +6341,13 @@
 	return NOTIFY_OK;
 }
 
-/*
- * This has to have a higher priority than migration_notifier in sched.c.
- */
-static struct notifier_block __cpuinitdata perf_cpu_nb = {
-	.notifier_call		= perf_cpu_notify,
-	.priority		= 20,
-};
-
 void __init perf_event_init(void)
 {
 	perf_event_init_all_cpus();
-	perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_UP_PREPARE,
-			(void *)(long)smp_processor_id());
-	perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_ONLINE,
-			(void *)(long)smp_processor_id());
-	register_cpu_notifier(&perf_cpu_nb);
+	init_srcu_struct(&pmus_srcu);
+	perf_pmu_register(&perf_swevent);
+	perf_pmu_register(&perf_cpu_clock);
+	perf_pmu_register(&perf_task_clock);
+	perf_tp_register();
+	perf_cpu_notifier(perf_cpu_notify);
 }
-
-static ssize_t perf_show_reserve_percpu(struct sysdev_class *class,
-					struct sysdev_class_attribute *attr,
-					char *buf)
-{
-	return sprintf(buf, "%d\n", perf_reserved_percpu);
-}
-
-static ssize_t
-perf_set_reserve_percpu(struct sysdev_class *class,
-			struct sysdev_class_attribute *attr,
-			const char *buf,
-			size_t count)
-{
-	struct perf_cpu_context *cpuctx;
-	unsigned long val;
-	int err, cpu, mpt;
-
-	err = strict_strtoul(buf, 10, &val);
-	if (err)
-		return err;
-	if (val > perf_max_events)
-		return -EINVAL;
-
-	spin_lock(&perf_resource_lock);
-	perf_reserved_percpu = val;
-	for_each_online_cpu(cpu) {
-		cpuctx = &per_cpu(perf_cpu_context, cpu);
-		raw_spin_lock_irq(&cpuctx->ctx.lock);
-		mpt = min(perf_max_events - cpuctx->ctx.nr_events,
-			  perf_max_events - perf_reserved_percpu);
-		cpuctx->max_pertask = mpt;
-		raw_spin_unlock_irq(&cpuctx->ctx.lock);
-	}
-	spin_unlock(&perf_resource_lock);
-
-	return count;
-}
-
-static ssize_t perf_show_overcommit(struct sysdev_class *class,
-				    struct sysdev_class_attribute *attr,
-				    char *buf)
-{
-	return sprintf(buf, "%d\n", perf_overcommit);
-}
-
-static ssize_t
-perf_set_overcommit(struct sysdev_class *class,
-		    struct sysdev_class_attribute *attr,
-		    const char *buf, size_t count)
-{
-	unsigned long val;
-	int err;
-
-	err = strict_strtoul(buf, 10, &val);
-	if (err)
-		return err;
-	if (val > 1)
-		return -EINVAL;
-
-	spin_lock(&perf_resource_lock);
-	perf_overcommit = val;
-	spin_unlock(&perf_resource_lock);
-
-	return count;
-}
-
-static SYSDEV_CLASS_ATTR(
-				reserve_percpu,
-				0644,
-				perf_show_reserve_percpu,
-				perf_set_reserve_percpu
-			);
-
-static SYSDEV_CLASS_ATTR(
-				overcommit,
-				0644,
-				perf_show_overcommit,
-				perf_set_overcommit
-			);
-
-static struct attribute *perfclass_attrs[] = {
-	&attr_reserve_percpu.attr,
-	&attr_overcommit.attr,
-	NULL
-};
-
-static struct attribute_group perfclass_attr_group = {
-	.attrs			= perfclass_attrs,
-	.name			= "perf_events",
-};
-
-static int __init perf_event_sysfs_init(void)
-{
-	return sysfs_create_group(&cpu_sysdev_class.kset.kobj,
-				  &perfclass_attr_group);
-}
-device_initcall(perf_event_sysfs_init);
diff --git a/kernel/sched.c b/kernel/sched.c
index dc85ceb..c0d2067 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -3584,7 +3584,7 @@
 	curr->sched_class->task_tick(rq, curr, 0);
 	raw_spin_unlock(&rq->lock);
 
-	perf_event_task_tick(curr);
+	perf_event_task_tick();
 
 #ifdef CONFIG_SMP
 	rq->idle_at_tick = idle_cpu(cpu);
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index fa7ece6..65fb077 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -884,10 +884,8 @@
 	FTRACE_ENABLE_CALLS		= (1 << 0),
 	FTRACE_DISABLE_CALLS		= (1 << 1),
 	FTRACE_UPDATE_TRACE_FUNC	= (1 << 2),
-	FTRACE_ENABLE_MCOUNT		= (1 << 3),
-	FTRACE_DISABLE_MCOUNT		= (1 << 4),
-	FTRACE_START_FUNC_RET		= (1 << 5),
-	FTRACE_STOP_FUNC_RET		= (1 << 6),
+	FTRACE_START_FUNC_RET		= (1 << 3),
+	FTRACE_STOP_FUNC_RET		= (1 << 4),
 };
 
 static int ftrace_filtered;
@@ -1226,8 +1224,6 @@
 
 static void ftrace_startup_sysctl(void)
 {
-	int command = FTRACE_ENABLE_MCOUNT;
-
 	if (unlikely(ftrace_disabled))
 		return;
 
@@ -1235,23 +1231,17 @@
 	saved_ftrace_func = NULL;
 	/* ftrace_start_up is true if we want ftrace running */
 	if (ftrace_start_up)
-		command |= FTRACE_ENABLE_CALLS;
-
-	ftrace_run_update_code(command);
+		ftrace_run_update_code(FTRACE_ENABLE_CALLS);
 }
 
 static void ftrace_shutdown_sysctl(void)
 {
-	int command = FTRACE_DISABLE_MCOUNT;
-
 	if (unlikely(ftrace_disabled))
 		return;
 
 	/* ftrace_start_up is true if ftrace is running */
 	if (ftrace_start_up)
-		command |= FTRACE_DISABLE_CALLS;
-
-	ftrace_run_update_code(command);
+		ftrace_run_update_code(FTRACE_DISABLE_CALLS);
 }
 
 static cycle_t		ftrace_update_time;
@@ -1368,24 +1358,29 @@
 #define FTRACE_BUFF_MAX (KSYM_SYMBOL_LEN+4) /* room for wildcards */
 
 struct ftrace_iterator {
-	struct ftrace_page	*pg;
-	int			hidx;
-	int			idx;
-	unsigned		flags;
-	struct trace_parser	parser;
+	loff_t				pos;
+	loff_t				func_pos;
+	struct ftrace_page		*pg;
+	struct dyn_ftrace		*func;
+	struct ftrace_func_probe	*probe;
+	struct trace_parser		parser;
+	int				hidx;
+	int				idx;
+	unsigned			flags;
 };
 
 static void *
-t_hash_next(struct seq_file *m, void *v, loff_t *pos)
+t_hash_next(struct seq_file *m, loff_t *pos)
 {
 	struct ftrace_iterator *iter = m->private;
-	struct hlist_node *hnd = v;
+	struct hlist_node *hnd = NULL;
 	struct hlist_head *hhd;
 
-	WARN_ON(!(iter->flags & FTRACE_ITER_HASH));
-
 	(*pos)++;
+	iter->pos = *pos;
 
+	if (iter->probe)
+		hnd = &iter->probe->node;
  retry:
 	if (iter->hidx >= FTRACE_FUNC_HASHSIZE)
 		return NULL;
@@ -1408,7 +1403,12 @@
 		}
 	}
 
-	return hnd;
+	if (WARN_ON_ONCE(!hnd))
+		return NULL;
+
+	iter->probe = hlist_entry(hnd, struct ftrace_func_probe, node);
+
+	return iter;
 }
 
 static void *t_hash_start(struct seq_file *m, loff_t *pos)
@@ -1417,26 +1417,32 @@
 	void *p = NULL;
 	loff_t l;
 
-	if (!(iter->flags & FTRACE_ITER_HASH))
-		*pos = 0;
-
-	iter->flags |= FTRACE_ITER_HASH;
+	if (iter->func_pos > *pos)
+		return NULL;
 
 	iter->hidx = 0;
-	for (l = 0; l <= *pos; ) {
-		p = t_hash_next(m, p, &l);
+	for (l = 0; l <= (*pos - iter->func_pos); ) {
+		p = t_hash_next(m, &l);
 		if (!p)
 			break;
 	}
-	return p;
+	if (!p)
+		return NULL;
+
+	/* Only set this if we have an item */
+	iter->flags |= FTRACE_ITER_HASH;
+
+	return iter;
 }
 
-static int t_hash_show(struct seq_file *m, void *v)
+static int
+t_hash_show(struct seq_file *m, struct ftrace_iterator *iter)
 {
 	struct ftrace_func_probe *rec;
-	struct hlist_node *hnd = v;
 
-	rec = hlist_entry(hnd, struct ftrace_func_probe, node);
+	rec = iter->probe;
+	if (WARN_ON_ONCE(!rec))
+		return -EIO;
 
 	if (rec->ops->print)
 		return rec->ops->print(m, rec->ip, rec->ops, rec->data);
@@ -1457,12 +1463,13 @@
 	struct dyn_ftrace *rec = NULL;
 
 	if (iter->flags & FTRACE_ITER_HASH)
-		return t_hash_next(m, v, pos);
+		return t_hash_next(m, pos);
 
 	(*pos)++;
+	iter->pos = *pos;
 
 	if (iter->flags & FTRACE_ITER_PRINTALL)
-		return NULL;
+		return t_hash_start(m, pos);
 
  retry:
 	if (iter->idx >= iter->pg->index) {
@@ -1491,7 +1498,20 @@
 		}
 	}
 
-	return rec;
+	if (!rec)
+		return t_hash_start(m, pos);
+
+	iter->func_pos = *pos;
+	iter->func = rec;
+
+	return iter;
+}
+
+static void reset_iter_read(struct ftrace_iterator *iter)
+{
+	iter->pos = 0;
+	iter->func_pos = 0;
+	iter->flags &= ~(FTRACE_ITER_PRINTALL & FTRACE_ITER_HASH);
 }
 
 static void *t_start(struct seq_file *m, loff_t *pos)
@@ -1502,6 +1522,12 @@
 
 	mutex_lock(&ftrace_lock);
 	/*
+	 * If an lseek was done, then reset and start from beginning.
+	 */
+	if (*pos < iter->pos)
+		reset_iter_read(iter);
+
+	/*
 	 * For set_ftrace_filter reading, if we have the filter
 	 * off, we can short cut and just print out that all
 	 * functions are enabled.
@@ -1518,6 +1544,11 @@
 	if (iter->flags & FTRACE_ITER_HASH)
 		return t_hash_start(m, pos);
 
+	/*
+	 * Unfortunately, we need to restart at ftrace_pages_start
+	 * every time we let go of the ftrace_mutex. This is because
+	 * those pointers can change without the lock.
+	 */
 	iter->pg = ftrace_pages_start;
 	iter->idx = 0;
 	for (l = 0; l <= *pos; ) {
@@ -1526,10 +1557,14 @@
 			break;
 	}
 
-	if (!p && iter->flags & FTRACE_ITER_FILTER)
-		return t_hash_start(m, pos);
+	if (!p) {
+		if (iter->flags & FTRACE_ITER_FILTER)
+			return t_hash_start(m, pos);
 
-	return p;
+		return NULL;
+	}
+
+	return iter;
 }
 
 static void t_stop(struct seq_file *m, void *p)
@@ -1540,16 +1575,18 @@
 static int t_show(struct seq_file *m, void *v)
 {
 	struct ftrace_iterator *iter = m->private;
-	struct dyn_ftrace *rec = v;
+	struct dyn_ftrace *rec;
 
 	if (iter->flags & FTRACE_ITER_HASH)
-		return t_hash_show(m, v);
+		return t_hash_show(m, iter);
 
 	if (iter->flags & FTRACE_ITER_PRINTALL) {
 		seq_printf(m, "#### all functions enabled ####\n");
 		return 0;
 	}
 
+	rec = iter->func;
+
 	if (!rec)
 		return 0;
 
@@ -2418,7 +2455,7 @@
 	.open = ftrace_filter_open,
 	.read = seq_read,
 	.write = ftrace_filter_write,
-	.llseek = no_llseek,
+	.llseek = ftrace_regex_lseek,
 	.release = ftrace_filter_release,
 };
 
diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 492197e..4e2f034 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -2606,6 +2606,19 @@
 }
 EXPORT_SYMBOL_GPL(ring_buffer_record_enable_cpu);
 
+/*
+ * The total entries in the ring buffer is the running counter
+ * of entries entered into the ring buffer, minus the sum of
+ * the entries read from the ring buffer and the number of
+ * entries that were overwritten.
+ */
+static inline unsigned long
+rb_num_of_entries(struct ring_buffer_per_cpu *cpu_buffer)
+{
+	return local_read(&cpu_buffer->entries) -
+		(local_read(&cpu_buffer->overrun) + cpu_buffer->read);
+}
+
 /**
  * ring_buffer_entries_cpu - get the number of entries in a cpu buffer
  * @buffer: The ring buffer
@@ -2614,16 +2627,13 @@
 unsigned long ring_buffer_entries_cpu(struct ring_buffer *buffer, int cpu)
 {
 	struct ring_buffer_per_cpu *cpu_buffer;
-	unsigned long ret;
 
 	if (!cpumask_test_cpu(cpu, buffer->cpumask))
 		return 0;
 
 	cpu_buffer = buffer->buffers[cpu];
-	ret = (local_read(&cpu_buffer->entries) - local_read(&cpu_buffer->overrun))
-		- cpu_buffer->read;
 
-	return ret;
+	return rb_num_of_entries(cpu_buffer);
 }
 EXPORT_SYMBOL_GPL(ring_buffer_entries_cpu);
 
@@ -2684,8 +2694,7 @@
 	/* if you care about this being correct, lock the buffer */
 	for_each_buffer_cpu(buffer, cpu) {
 		cpu_buffer = buffer->buffers[cpu];
-		entries += (local_read(&cpu_buffer->entries) -
-			    local_read(&cpu_buffer->overrun)) - cpu_buffer->read;
+		entries += rb_num_of_entries(cpu_buffer);
 	}
 
 	return entries;
diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c
index 31cc4cb..39c059c 100644
--- a/kernel/trace/trace_event_perf.c
+++ b/kernel/trace/trace_event_perf.c
@@ -9,7 +9,7 @@
 #include <linux/kprobes.h>
 #include "trace.h"
 
-static char *perf_trace_buf[4];
+static char __percpu *perf_trace_buf[PERF_NR_CONTEXTS];
 
 /*
  * Force it to be aligned to unsigned long to avoid misaligned accesses
@@ -24,7 +24,7 @@
 static int perf_trace_event_init(struct ftrace_event_call *tp_event,
 				 struct perf_event *p_event)
 {
-	struct hlist_head *list;
+	struct hlist_head __percpu *list;
 	int ret = -ENOMEM;
 	int cpu;
 
@@ -42,11 +42,11 @@
 	tp_event->perf_events = list;
 
 	if (!total_ref_count) {
-		char *buf;
+		char __percpu *buf;
 		int i;
 
-		for (i = 0; i < 4; i++) {
-			buf = (char *)alloc_percpu(perf_trace_t);
+		for (i = 0; i < PERF_NR_CONTEXTS; i++) {
+			buf = (char __percpu *)alloc_percpu(perf_trace_t);
 			if (!buf)
 				goto fail;
 
@@ -65,7 +65,7 @@
 	if (!total_ref_count) {
 		int i;
 
-		for (i = 0; i < 4; i++) {
+		for (i = 0; i < PERF_NR_CONTEXTS; i++) {
 			free_percpu(perf_trace_buf[i]);
 			perf_trace_buf[i] = NULL;
 		}
@@ -101,22 +101,26 @@
 	return ret;
 }
 
-int perf_trace_enable(struct perf_event *p_event)
+int perf_trace_add(struct perf_event *p_event, int flags)
 {
 	struct ftrace_event_call *tp_event = p_event->tp_event;
+	struct hlist_head __percpu *pcpu_list;
 	struct hlist_head *list;
 
-	list = tp_event->perf_events;
-	if (WARN_ON_ONCE(!list))
+	pcpu_list = tp_event->perf_events;
+	if (WARN_ON_ONCE(!pcpu_list))
 		return -EINVAL;
 
-	list = this_cpu_ptr(list);
+	if (!(flags & PERF_EF_START))
+		p_event->hw.state = PERF_HES_STOPPED;
+
+	list = this_cpu_ptr(pcpu_list);
 	hlist_add_head_rcu(&p_event->hlist_entry, list);
 
 	return 0;
 }
 
-void perf_trace_disable(struct perf_event *p_event)
+void perf_trace_del(struct perf_event *p_event, int flags)
 {
 	hlist_del_rcu(&p_event->hlist_entry);
 }
@@ -142,7 +146,7 @@
 	tp_event->perf_events = NULL;
 
 	if (!--total_ref_count) {
-		for (i = 0; i < 4; i++) {
+		for (i = 0; i < PERF_NR_CONTEXTS; i++) {
 			free_percpu(perf_trace_buf[i]);
 			perf_trace_buf[i] = NULL;
 		}
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index 4c758f1..398c0e8 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -600,21 +600,29 @@
 
 enum {
 	FORMAT_HEADER		= 1,
-	FORMAT_PRINTFMT		= 2,
+	FORMAT_FIELD_SEPERATOR	= 2,
+	FORMAT_PRINTFMT		= 3,
 };
 
 static void *f_next(struct seq_file *m, void *v, loff_t *pos)
 {
 	struct ftrace_event_call *call = m->private;
 	struct ftrace_event_field *field;
-	struct list_head *head;
+	struct list_head *common_head = &ftrace_common_fields;
+	struct list_head *head = trace_get_fields(call);
 
 	(*pos)++;
 
 	switch ((unsigned long)v) {
 	case FORMAT_HEADER:
-		head = &ftrace_common_fields;
+		if (unlikely(list_empty(common_head)))
+			return NULL;
 
+		field = list_entry(common_head->prev,
+				   struct ftrace_event_field, link);
+		return field;
+
+	case FORMAT_FIELD_SEPERATOR:
 		if (unlikely(list_empty(head)))
 			return NULL;
 
@@ -626,31 +634,10 @@
 		return NULL;
 	}
 
-	head = trace_get_fields(call);
-
-	/*
-	 * To separate common fields from event fields, the
-	 * LSB is set on the first event field. Clear it in case.
-	 */
-	v = (void *)((unsigned long)v & ~1L);
-
 	field = v;
-	/*
-	 * If this is a common field, and at the end of the list, then
-	 * continue with main list.
-	 */
-	if (field->link.prev == &ftrace_common_fields) {
-		if (unlikely(list_empty(head)))
-			return NULL;
-		field = list_entry(head->prev, struct ftrace_event_field, link);
-		/* Set the LSB to notify f_show to print an extra newline */
-		field = (struct ftrace_event_field *)
-			((unsigned long)field | 1);
-		return field;
-	}
-
-	/* If we are done tell f_show to print the format */
-	if (field->link.prev == head)
+	if (field->link.prev == common_head)
+		return (void *)FORMAT_FIELD_SEPERATOR;
+	else if (field->link.prev == head)
 		return (void *)FORMAT_PRINTFMT;
 
 	field = list_entry(field->link.prev, struct ftrace_event_field, link);
@@ -688,22 +675,16 @@
 		seq_printf(m, "format:\n");
 		return 0;
 
+	case FORMAT_FIELD_SEPERATOR:
+		seq_putc(m, '\n');
+		return 0;
+
 	case FORMAT_PRINTFMT:
 		seq_printf(m, "\nprint fmt: %s\n",
 			   call->print_fmt);
 		return 0;
 	}
 
-	/*
-	 * To separate common fields from event fields, the
-	 * LSB is set on the first event field. Clear it and
-	 * print a newline if it is set.
-	 */
-	if ((unsigned long)v & 1) {
-		seq_putc(m, '\n');
-		v = (void *)((unsigned long)v & ~1L);
-	}
-
 	field = v;
 
 	/*
diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c
index 6f23369..02c708a 100644
--- a/kernel/trace/trace_functions_graph.c
+++ b/kernel/trace/trace_functions_graph.c
@@ -15,15 +15,19 @@
 #include "trace.h"
 #include "trace_output.h"
 
+/* When set, irq functions will be ignored */
+static int ftrace_graph_skip_irqs;
+
 struct fgraph_cpu_data {
 	pid_t		last_pid;
 	int		depth;
+	int		depth_irq;
 	int		ignore;
 	unsigned long	enter_funcs[FTRACE_RETFUNC_DEPTH];
 };
 
 struct fgraph_data {
-	struct fgraph_cpu_data		*cpu_data;
+	struct fgraph_cpu_data __percpu *cpu_data;
 
 	/* Place to preserve last processed entry. */
 	struct ftrace_graph_ent_entry	ent;
@@ -41,6 +45,7 @@
 #define TRACE_GRAPH_PRINT_PROC		0x8
 #define TRACE_GRAPH_PRINT_DURATION	0x10
 #define TRACE_GRAPH_PRINT_ABS_TIME	0x20
+#define TRACE_GRAPH_PRINT_IRQS		0x40
 
 static struct tracer_opt trace_opts[] = {
 	/* Display overruns? (for self-debug purpose) */
@@ -55,13 +60,15 @@
 	{ TRACER_OPT(funcgraph-duration, TRACE_GRAPH_PRINT_DURATION) },
 	/* Display absolute time of an entry */
 	{ TRACER_OPT(funcgraph-abstime, TRACE_GRAPH_PRINT_ABS_TIME) },
+	/* Display interrupts */
+	{ TRACER_OPT(funcgraph-irqs, TRACE_GRAPH_PRINT_IRQS) },
 	{ } /* Empty entry */
 };
 
 static struct tracer_flags tracer_flags = {
 	/* Don't display overruns and proc by default */
 	.val = TRACE_GRAPH_PRINT_CPU | TRACE_GRAPH_PRINT_OVERHEAD |
-	       TRACE_GRAPH_PRINT_DURATION,
+	       TRACE_GRAPH_PRINT_DURATION | TRACE_GRAPH_PRINT_IRQS,
 	.opts = trace_opts
 };
 
@@ -204,6 +211,14 @@
 	return 1;
 }
 
+static inline int ftrace_graph_ignore_irqs(void)
+{
+	if (!ftrace_graph_skip_irqs)
+		return 0;
+
+	return in_irq();
+}
+
 int trace_graph_entry(struct ftrace_graph_ent *trace)
 {
 	struct trace_array *tr = graph_array;
@@ -218,7 +233,8 @@
 		return 0;
 
 	/* trace it when it is-nested-in or is a function enabled. */
-	if (!(trace->depth || ftrace_graph_addr(trace->func)))
+	if (!(trace->depth || ftrace_graph_addr(trace->func)) ||
+	      ftrace_graph_ignore_irqs())
 		return 0;
 
 	local_irq_save(flags);
@@ -855,6 +871,92 @@
 	return 0;
 }
 
+/*
+ * Entry check for irq code
+ *
+ * returns 1 if
+ *  - we are inside irq code
+ *  - we just extered irq code
+ *
+ * retunns 0 if
+ *  - funcgraph-interrupts option is set
+ *  - we are not inside irq code
+ */
+static int
+check_irq_entry(struct trace_iterator *iter, u32 flags,
+		unsigned long addr, int depth)
+{
+	int cpu = iter->cpu;
+	struct fgraph_data *data = iter->private;
+	int *depth_irq = &(per_cpu_ptr(data->cpu_data, cpu)->depth_irq);
+
+	if (flags & TRACE_GRAPH_PRINT_IRQS)
+		return 0;
+
+	/*
+	 * We are inside the irq code
+	 */
+	if (*depth_irq >= 0)
+		return 1;
+
+	if ((addr < (unsigned long)__irqentry_text_start) ||
+	    (addr >= (unsigned long)__irqentry_text_end))
+		return 0;
+
+	/*
+	 * We are entering irq code.
+	 */
+	*depth_irq = depth;
+	return 1;
+}
+
+/*
+ * Return check for irq code
+ *
+ * returns 1 if
+ *  - we are inside irq code
+ *  - we just left irq code
+ *
+ * returns 0 if
+ *  - funcgraph-interrupts option is set
+ *  - we are not inside irq code
+ */
+static int
+check_irq_return(struct trace_iterator *iter, u32 flags, int depth)
+{
+	int cpu = iter->cpu;
+	struct fgraph_data *data = iter->private;
+	int *depth_irq = &(per_cpu_ptr(data->cpu_data, cpu)->depth_irq);
+
+	if (flags & TRACE_GRAPH_PRINT_IRQS)
+		return 0;
+
+	/*
+	 * We are not inside the irq code.
+	 */
+	if (*depth_irq == -1)
+		return 0;
+
+	/*
+	 * We are inside the irq code, and this is returning entry.
+	 * Let's not trace it and clear the entry depth, since
+	 * we are out of irq code.
+	 *
+	 * This condition ensures that we 'leave the irq code' once
+	 * we are out of the entry depth. Thus protecting us from
+	 * the RETURN entry loss.
+	 */
+	if (*depth_irq >= depth) {
+		*depth_irq = -1;
+		return 1;
+	}
+
+	/*
+	 * We are inside the irq code, and this is not the entry.
+	 */
+	return 1;
+}
+
 static enum print_line_t
 print_graph_entry(struct ftrace_graph_ent_entry *field, struct trace_seq *s,
 			struct trace_iterator *iter, u32 flags)
@@ -865,6 +967,9 @@
 	static enum print_line_t ret;
 	int cpu = iter->cpu;
 
+	if (check_irq_entry(iter, flags, call->func, call->depth))
+		return TRACE_TYPE_HANDLED;
+
 	if (print_graph_prologue(iter, s, TRACE_GRAPH_ENT, call->func, flags))
 		return TRACE_TYPE_PARTIAL_LINE;
 
@@ -902,6 +1007,9 @@
 	int ret;
 	int i;
 
+	if (check_irq_return(iter, flags, trace->depth))
+		return TRACE_TYPE_HANDLED;
+
 	if (data) {
 		struct fgraph_cpu_data *cpu_data;
 		int cpu = iter->cpu;
@@ -1210,9 +1318,12 @@
 		pid_t *pid = &(per_cpu_ptr(data->cpu_data, cpu)->last_pid);
 		int *depth = &(per_cpu_ptr(data->cpu_data, cpu)->depth);
 		int *ignore = &(per_cpu_ptr(data->cpu_data, cpu)->ignore);
+		int *depth_irq = &(per_cpu_ptr(data->cpu_data, cpu)->depth_irq);
+
 		*pid = -1;
 		*depth = 0;
 		*ignore = 0;
+		*depth_irq = -1;
 	}
 
 	iter->private = data;
@@ -1235,6 +1346,14 @@
 	}
 }
 
+static int func_graph_set_flag(u32 old_flags, u32 bit, int set)
+{
+	if (bit == TRACE_GRAPH_PRINT_IRQS)
+		ftrace_graph_skip_irqs = !set;
+
+	return 0;
+}
+
 static struct trace_event_functions graph_functions = {
 	.trace		= print_graph_function_event,
 };
@@ -1261,6 +1380,7 @@
 	.print_line	= print_graph_function,
 	.print_header	= print_graph_headers,
 	.flags		= &tracer_flags,
+	.set_flag	= func_graph_set_flag,
 #ifdef CONFIG_FTRACE_SELFTEST
 	.selftest	= trace_selftest_startup_function_graph,
 #endif
diff --git a/kernel/watchdog.c b/kernel/watchdog.c
index 7f9c3c5..dc8e168 100644
--- a/kernel/watchdog.c
+++ b/kernel/watchdog.c
@@ -43,7 +43,6 @@
 static DEFINE_PER_CPU(struct perf_event *, watchdog_ev);
 #endif
 
-static int __read_mostly did_panic;
 static int __initdata no_watchdog;
 
 
@@ -187,18 +186,6 @@
 	return 0;
 }
 
-static int
-watchdog_panic(struct notifier_block *this, unsigned long event, void *ptr)
-{
-	did_panic = 1;
-
-	return NOTIFY_DONE;
-}
-
-static struct notifier_block panic_block = {
-	.notifier_call = watchdog_panic,
-};
-
 #ifdef CONFIG_HARDLOCKUP_DETECTOR
 static struct perf_event_attr wd_hw_attr = {
 	.type		= PERF_TYPE_HARDWARE,
@@ -371,14 +358,14 @@
 	/* Try to register using hardware perf events */
 	wd_attr = &wd_hw_attr;
 	wd_attr->sample_period = hw_nmi_get_sample_period();
-	event = perf_event_create_kernel_counter(wd_attr, cpu, -1, watchdog_overflow_callback);
+	event = perf_event_create_kernel_counter(wd_attr, cpu, NULL, watchdog_overflow_callback);
 	if (!IS_ERR(event)) {
 		printk(KERN_INFO "NMI watchdog enabled, takes one hw-pmu counter.\n");
 		goto out_save;
 	}
 
 	printk(KERN_ERR "NMI watchdog failed to create perf event on cpu%i: %p\n", cpu, event);
-	return -1;
+	return PTR_ERR(event);
 
 	/* success path */
 out_save:
@@ -422,17 +409,19 @@
 static int watchdog_enable(int cpu)
 {
 	struct task_struct *p = per_cpu(softlockup_watchdog, cpu);
+	int err;
 
 	/* enable the perf event */
-	if (watchdog_nmi_enable(cpu) != 0)
-		return -1;
+	err = watchdog_nmi_enable(cpu);
+	if (err)
+		return err;
 
 	/* create the watchdog thread */
 	if (!p) {
 		p = kthread_create(watchdog, (void *)(unsigned long)cpu, "watchdog/%d", cpu);
 		if (IS_ERR(p)) {
 			printk(KERN_ERR "softlockup watchdog for %i failed\n", cpu);
-			return -1;
+			return PTR_ERR(p);
 		}
 		kthread_bind(p, cpu);
 		per_cpu(watchdog_touch_ts, cpu) = 0;
@@ -484,6 +473,9 @@
 {
 	int cpu;
 
+	if (no_watchdog)
+		return;
+
 	for_each_online_cpu(cpu)
 		watchdog_disable(cpu);
 
@@ -526,17 +518,16 @@
 cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
 {
 	int hotcpu = (unsigned long)hcpu;
+	int err = 0;
 
 	switch (action) {
 	case CPU_UP_PREPARE:
 	case CPU_UP_PREPARE_FROZEN:
-		if (watchdog_prepare_cpu(hotcpu))
-			return NOTIFY_BAD;
+		err = watchdog_prepare_cpu(hotcpu);
 		break;
 	case CPU_ONLINE:
 	case CPU_ONLINE_FROZEN:
-		if (watchdog_enable(hotcpu))
-			return NOTIFY_BAD;
+		err = watchdog_enable(hotcpu);
 		break;
 #ifdef CONFIG_HOTPLUG_CPU
 	case CPU_UP_CANCELED:
@@ -549,7 +540,7 @@
 		break;
 #endif /* CONFIG_HOTPLUG_CPU */
 	}
-	return NOTIFY_OK;
+	return notifier_from_errno(err);
 }
 
 static struct notifier_block __cpuinitdata cpu_nfb = {
@@ -565,13 +556,11 @@
 		return 0;
 
 	err = cpu_callback(&cpu_nfb, CPU_UP_PREPARE, cpu);
-	WARN_ON(err == NOTIFY_BAD);
+	WARN_ON(notifier_to_errno(err));
 
 	cpu_callback(&cpu_nfb, CPU_ONLINE, cpu);
 	register_cpu_notifier(&cpu_nfb);
 
-	atomic_notifier_chain_register(&panic_notifier_list, &panic_block);
-
 	return 0;
 }
 early_initcall(spawn_watchdog_task);
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 1b4afd2..e85d549 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -482,6 +482,7 @@
 	select DEBUG_SPINLOCK
 	select DEBUG_MUTEXES
 	select DEBUG_LOCK_ALLOC
+	select TRACE_IRQFLAGS
 	default n
 	help
 	 This feature enables the kernel to prove that all locking
@@ -579,11 +580,10 @@
 	  of more runtime overhead.
 
 config TRACE_IRQFLAGS
-	depends on DEBUG_KERNEL
 	bool
-	default y
-	depends on TRACE_IRQFLAGS_SUPPORT
-	depends on PROVE_LOCKING
+	help
+	  Enables hooks to interrupt enabling and disabling for
+	  either tracing or lock debugging.
 
 config DEBUG_SPINLOCK_SLEEP
 	bool "Spinlock debugging: sleep-inside-spinlock checking"
diff --git a/net/core/datagram.c b/net/core/datagram.c
index 251997a..282806b 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -243,6 +243,7 @@
 	unlock_sock_fast(sk, slow);
 
 	/* skb is now orphaned, can be freed outside of locked section */
+	trace_kfree_skb(skb, skb_free_datagram_locked);
 	__kfree_skb(skb);
 }
 EXPORT_SYMBOL(skb_free_datagram_locked);
diff --git a/net/core/dev.c b/net/core/dev.c
index 660dd41..7ec85e2 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -128,6 +128,8 @@
 #include <linux/jhash.h>
 #include <linux/random.h>
 #include <trace/events/napi.h>
+#include <trace/events/net.h>
+#include <trace/events/skb.h>
 #include <linux/pci.h>
 
 #include "net-sysfs.h"
@@ -1978,6 +1980,7 @@
 		}
 
 		rc = ops->ndo_start_xmit(skb, dev);
+		trace_net_dev_xmit(skb, rc);
 		if (rc == NETDEV_TX_OK)
 			txq_trans_update(txq);
 		return rc;
@@ -1998,6 +2001,7 @@
 			skb_dst_drop(nskb);
 
 		rc = ops->ndo_start_xmit(nskb, dev);
+		trace_net_dev_xmit(nskb, rc);
 		if (unlikely(rc != NETDEV_TX_OK)) {
 			if (rc & ~NETDEV_TX_MASK)
 				goto out_kfree_gso_skb;
@@ -2186,6 +2190,7 @@
 #ifdef CONFIG_NET_CLS_ACT
 	skb->tc_verd = SET_TC_AT(skb->tc_verd, AT_EGRESS);
 #endif
+	trace_net_dev_queue(skb);
 	if (q->enqueue) {
 		rc = __dev_xmit_skb(skb, q, dev, txq);
 		goto out;
@@ -2512,6 +2517,7 @@
 	if (netdev_tstamp_prequeue)
 		net_timestamp_check(skb);
 
+	trace_netif_rx(skb);
 #ifdef CONFIG_RPS
 	{
 		struct rps_dev_flow voidflow, *rflow = &voidflow;
@@ -2571,6 +2577,7 @@
 			clist = clist->next;
 
 			WARN_ON(atomic_read(&skb->users));
+			trace_kfree_skb(skb, net_tx_action);
 			__kfree_skb(skb);
 		}
 	}
@@ -2828,6 +2835,7 @@
 	if (!netdev_tstamp_prequeue)
 		net_timestamp_check(skb);
 
+	trace_netif_receive_skb(skb);
 	if (vlan_tx_tag_present(skb) && vlan_hwaccel_do_receive(skb))
 		return NET_RX_SUCCESS;
 
diff --git a/net/core/net-traces.c b/net/core/net-traces.c
index afa6380..7f1bb2a 100644
--- a/net/core/net-traces.c
+++ b/net/core/net-traces.c
@@ -26,6 +26,7 @@
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/skb.h>
+#include <trace/events/net.h>
 #include <trace/events/napi.h>
 
 EXPORT_TRACEPOINT_SYMBOL_GPL(kfree_skb);
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index c83b421..56ba3c4 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -466,6 +466,7 @@
 		smp_rmb();
 	else if (likely(!atomic_dec_and_test(&skb->users)))
 		return;
+	trace_consume_skb(skb);
 	__kfree_skb(skb);
 }
 EXPORT_SYMBOL(consume_skb);
diff --git a/tools/perf/Documentation/perf-annotate.txt b/tools/perf/Documentation/perf-annotate.txt
index 5164a65..b2c6330 100644
--- a/tools/perf/Documentation/perf-annotate.txt
+++ b/tools/perf/Documentation/perf-annotate.txt
@@ -8,7 +8,7 @@
 SYNOPSIS
 --------
 [verse]
-'perf annotate' [-i <file> | --input=file] symbol_name
+'perf annotate' [-i <file> | --input=file] [symbol_name]
 
 DESCRIPTION
 -----------
@@ -24,6 +24,13 @@
 --input=::
         Input file name. (default: perf.data)
 
+--stdio:: Use the stdio interface.
+
+--tui:: Use the TUI interface Use of --tui requires a tty, if one is not
+	present, as when piping to other commands, the stdio interface is
+	used. This interfaces starts by centering on the line with more
+	samples, TAB/UNTAB cycles thru the lines with more samples.
+
 SEE ALSO
 --------
-linkperf:perf-record[1]
+linkperf:perf-record[1], linkperf:perf-report[1]
diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt
index abfabe9..12052c9 100644
--- a/tools/perf/Documentation/perf-report.txt
+++ b/tools/perf/Documentation/perf-report.txt
@@ -65,6 +65,13 @@
 		 the tree is considered as a new profiled object. +
 	Default: fractal,0.5.
 
+--stdio:: Use the stdio interface.
+
+--tui:: Use the TUI interface, that is integrated with annotate and allows
+        zooming into DSOs or threads, among other features. Use of --tui
+	requires a tty, if one is not present, as when piping to other
+	commands, the stdio interface is used.
+
 SEE ALSO
 --------
 linkperf:perf-stat[1]
diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index 4f1fa77..fe1e307 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -313,6 +313,9 @@
 
 SCRIPT_SH += perf-archive.sh
 
+grep-libs = $(filter -l%,$(1))
+strip-libs = $(filter-out -l%,$(1))
+
 #
 # No Perl scripts right now:
 #
@@ -588,14 +591,17 @@
 ifdef NO_LIBPERL
 	BASIC_CFLAGS += -DNO_LIBPERL
 else
-	PERL_EMBED_LDOPTS = `perl -MExtUtils::Embed -e ldopts 2>/dev/null`
+       PERL_EMBED_LDOPTS = $(shell perl -MExtUtils::Embed -e ldopts 2>/dev/null)
+       PERL_EMBED_LDFLAGS = $(call strip-libs,$(PERL_EMBED_LDOPTS))
+       PERL_EMBED_LIBADD = $(call grep-libs,$(PERL_EMBED_LDOPTS))
 	PERL_EMBED_CCOPTS = `perl -MExtUtils::Embed -e ccopts 2>/dev/null`
 	FLAGS_PERL_EMBED=$(PERL_EMBED_CCOPTS) $(PERL_EMBED_LDOPTS)
 
 	ifneq ($(call try-cc,$(SOURCE_PERL_EMBED),$(FLAGS_PERL_EMBED)),y)
 		BASIC_CFLAGS += -DNO_LIBPERL
 	else
-		ALL_LDFLAGS += $(PERL_EMBED_LDOPTS)
+               ALL_LDFLAGS += $(PERL_EMBED_LDFLAGS)
+               EXTLIBS += $(PERL_EMBED_LIBADD)
 		LIB_OBJS += $(OUTPUT)util/scripting-engines/trace-event-perl.o
 		LIB_OBJS += $(OUTPUT)scripts/perl/Perf-Trace-Util/Context.o
 	endif
@@ -604,13 +610,16 @@
 ifdef NO_LIBPYTHON
 	BASIC_CFLAGS += -DNO_LIBPYTHON
 else
-	PYTHON_EMBED_LDOPTS = `python-config --ldflags 2>/dev/null`
+       PYTHON_EMBED_LDOPTS = $(shell python-config --ldflags 2>/dev/null)
+       PYTHON_EMBED_LDFLAGS = $(call strip-libs,$(PYTHON_EMBED_LDOPTS))
+       PYTHON_EMBED_LIBADD = $(call grep-libs,$(PYTHON_EMBED_LDOPTS))
 	PYTHON_EMBED_CCOPTS = `python-config --cflags 2>/dev/null`
 	FLAGS_PYTHON_EMBED=$(PYTHON_EMBED_CCOPTS) $(PYTHON_EMBED_LDOPTS)
 	ifneq ($(call try-cc,$(SOURCE_PYTHON_EMBED),$(FLAGS_PYTHON_EMBED)),y)
 		BASIC_CFLAGS += -DNO_LIBPYTHON
 	else
-		ALL_LDFLAGS += $(PYTHON_EMBED_LDOPTS)
+               ALL_LDFLAGS += $(PYTHON_EMBED_LDFLAGS)
+               EXTLIBS += $(PYTHON_EMBED_LIBADD)
 		LIB_OBJS += $(OUTPUT)util/scripting-engines/trace-event-python.o
 		LIB_OBJS += $(OUTPUT)scripts/python/Perf-Trace-Util/Context.o
 	endif
@@ -653,6 +662,15 @@
 	endif
 endif
 
+
+ifdef NO_STRLCPY
+	BASIC_CFLAGS += -DNO_STRLCPY
+else
+	ifneq ($(call try-cc,$(SOURCE_STRLCPY),),y)
+		BASIC_CFLAGS += -DNO_STRLCPY
+	endif
+endif
+
 ifndef CC_LD_DYNPATH
 	ifdef NO_R_TO_GCC_LINKER
 		# Some gcc does not accept and pass -R to the linker to specify
@@ -910,8 +928,8 @@
 		$(ALL_CFLAGS) -c $(filter %.c,$^) -o $@
 
 $(OUTPUT)perf$X: $(OUTPUT)perf.o $(BUILTIN_OBJS) $(PERFLIBS)
-	$(QUIET_LINK)$(CC) $(ALL_CFLAGS) -o $@ $(OUTPUT)perf.o \
-		$(BUILTIN_OBJS) $(ALL_LDFLAGS) $(LIBS)
+	$(QUIET_LINK)$(CC) $(ALL_CFLAGS) $(ALL_LDFLAGS) $(OUTPUT)perf.o \
+               $(BUILTIN_OBJS) $(LIBS) -o $@
 
 $(OUTPUT)builtin-help.o: builtin-help.c $(OUTPUT)common-cmds.h $(OUTPUT)PERF-CFLAGS
 	$(QUIET_CC)$(CC) -o $@ -c $(ALL_CFLAGS) \
diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
index 1478dc6..6d5604d8d 100644
--- a/tools/perf/builtin-annotate.c
+++ b/tools/perf/builtin-annotate.c
@@ -28,7 +28,7 @@
 
 static char		const *input_name = "perf.data";
 
-static bool		force;
+static bool		force, use_tui, use_stdio;
 
 static bool		full_paths;
 
@@ -321,7 +321,7 @@
 
 static void hists__find_annotations(struct hists *self)
 {
-	struct rb_node *first = rb_first(&self->entries), *nd = first;
+	struct rb_node *nd = rb_first(&self->entries), *next;
 	int key = KEY_RIGHT;
 
 	while (nd) {
@@ -343,20 +343,19 @@
 
 		if (use_browser > 0) {
 			key = hist_entry__tui_annotate(he);
-			if (is_exit_key(key))
-				break;
 			switch (key) {
 			case KEY_RIGHT:
-			case '\t':
-				nd = rb_next(nd);
+				next = rb_next(nd);
 				break;
 			case KEY_LEFT:
-				if (nd == first)
-					continue;
-				nd = rb_prev(nd);
-			default:
+				next = rb_prev(nd);
 				break;
+			default:
+				return;
 			}
+
+			if (next != NULL)
+				nd = next;
 		} else {
 			hist_entry__tty_annotate(he);
 			nd = rb_next(nd);
@@ -428,6 +427,8 @@
 		    "be more verbose (show symbol address, etc)"),
 	OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace,
 		    "dump raw trace in ASCII"),
+	OPT_BOOLEAN(0, "tui", &use_tui, "Use the TUI interface"),
+	OPT_BOOLEAN(0, "stdio", &use_stdio, "Use the stdio interface"),
 	OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name,
 		   "file", "vmlinux pathname"),
 	OPT_BOOLEAN('m', "modules", &symbol_conf.use_modules,
@@ -443,6 +444,11 @@
 {
 	argc = parse_options(argc, argv, options, annotate_usage, 0);
 
+	if (use_stdio)
+		use_browser = 0;
+	else if (use_tui)
+		use_browser = 1;
+
 	setup_browser();
 
 	symbol_conf.priv_size = sizeof(struct sym_priv);
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 55fc1f4..5de405d 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -32,7 +32,7 @@
 
 static char		const *input_name = "perf.data";
 
-static bool		force;
+static bool		force, use_tui, use_stdio;
 static bool		hide_unresolved;
 static bool		dont_use_callchains;
 
@@ -107,7 +107,8 @@
 		goto out_free_syms;
 	err = 0;
 	if (symbol_conf.use_callchain) {
-		err = append_chain(he->callchain, data->callchain, syms, data->period);
+		err = callchain_append(he->callchain, data->callchain, syms,
+				       data->period);
 		if (err)
 			goto out_free_syms;
 	}
@@ -450,6 +451,8 @@
 		    "Show per-thread event counters"),
 	OPT_STRING(0, "pretty", &pretty_printing_style, "key",
 		   "pretty printing style key: normal raw"),
+	OPT_BOOLEAN(0, "tui", &use_tui, "Use the TUI interface"),
+	OPT_BOOLEAN(0, "stdio", &use_stdio, "Use the stdio interface"),
 	OPT_STRING('s', "sort", &sort_order, "key[,key2...]",
 		   "sort by key(s): pid, comm, dso, symbol, parent"),
 	OPT_BOOLEAN(0, "showcpuutilization", &symbol_conf.show_cpu_utilization,
@@ -482,8 +485,15 @@
 {
 	argc = parse_options(argc, argv, options, report_usage, 0);
 
+	if (use_stdio)
+		use_browser = 0;
+	else if (use_tui)
+		use_browser = 1;
+
 	if (strcmp(input_name, "-") != 0)
 		setup_browser();
+	else
+		use_browser = 0;
 	/*
 	 * Only in the newt browser we are doing integrated annotation,
 	 * so don't allocate extra space that won't be used in the stdio
diff --git a/tools/perf/feature-tests.mak b/tools/perf/feature-tests.mak
index 7a7b608..b253db6 100644
--- a/tools/perf/feature-tests.mak
+++ b/tools/perf/feature-tests.mak
@@ -110,6 +110,17 @@
 }
 endef
 
+define SOURCE_STRLCPY
+#include <stdlib.h>
+extern size_t strlcpy(char *dest, const char *src, size_t size);
+
+int main(void)
+{
+	strlcpy(NULL, NULL, 0);
+	return 0;
+}
+endef
+
 # try-cc
 # Usage: option = $(call try-cc, source-to-build, cc-options)
 try-cc = $(shell sh -c						  \
diff --git a/tools/perf/scripts/python/bin/netdev-times-record b/tools/perf/scripts/python/bin/netdev-times-record
new file mode 100644
index 0000000..d931a82
--- /dev/null
+++ b/tools/perf/scripts/python/bin/netdev-times-record
@@ -0,0 +1,8 @@
+#!/bin/bash
+perf record -a -e net:net_dev_xmit -e net:net_dev_queue		\
+		-e net:netif_receive_skb -e net:netif_rx		\
+		-e skb:consume_skb -e skb:kfree_skb			\
+		-e skb:skb_copy_datagram_iovec -e napi:napi_poll	\
+		-e irq:irq_handler_entry -e irq:irq_handler_exit	\
+		-e irq:softirq_entry -e irq:softirq_exit		\
+		-e irq:softirq_raise $@
diff --git a/tools/perf/scripts/python/bin/netdev-times-report b/tools/perf/scripts/python/bin/netdev-times-report
new file mode 100644
index 0000000..c3d0a63
--- /dev/null
+++ b/tools/perf/scripts/python/bin/netdev-times-report
@@ -0,0 +1,5 @@
+#!/bin/bash
+# description: display a process of packet and processing time
+# args: [tx] [rx] [dev=] [debug]
+
+perf trace -s ~/libexec/perf-core/scripts/python/netdev-times.py $@
diff --git a/tools/perf/scripts/python/netdev-times.py b/tools/perf/scripts/python/netdev-times.py
new file mode 100644
index 0000000..9aa0a32
--- /dev/null
+++ b/tools/perf/scripts/python/netdev-times.py
@@ -0,0 +1,464 @@
+# Display a process of packets and processed time.
+# It helps us to investigate networking or network device.
+#
+# options
+# tx: show only tx chart
+# rx: show only rx chart
+# dev=: show only thing related to specified device
+# debug: work with debug mode. It shows buffer status.
+
+import os
+import sys
+
+sys.path.append(os.environ['PERF_EXEC_PATH'] + \
+	'/scripts/python/Perf-Trace-Util/lib/Perf/Trace')
+
+from perf_trace_context import *
+from Core import *
+from Util import *
+
+all_event_list = []; # insert all tracepoint event related with this script
+irq_dic = {}; # key is cpu and value is a list which stacks irqs
+              # which raise NET_RX softirq
+net_rx_dic = {}; # key is cpu and value include time of NET_RX softirq-entry
+		 # and a list which stacks receive
+receive_hunk_list = []; # a list which include a sequence of receive events
+rx_skb_list = []; # received packet list for matching
+		       # skb_copy_datagram_iovec
+
+buffer_budget = 65536; # the budget of rx_skb_list, tx_queue_list and
+		       # tx_xmit_list
+of_count_rx_skb_list = 0; # overflow count
+
+tx_queue_list = []; # list of packets which pass through dev_queue_xmit
+of_count_tx_queue_list = 0; # overflow count
+
+tx_xmit_list = [];  # list of packets which pass through dev_hard_start_xmit
+of_count_tx_xmit_list = 0; # overflow count
+
+tx_free_list = [];  # list of packets which is freed
+
+# options
+show_tx = 0;
+show_rx = 0;
+dev = 0; # store a name of device specified by option "dev="
+debug = 0;
+
+# indices of event_info tuple
+EINFO_IDX_NAME=   0
+EINFO_IDX_CONTEXT=1
+EINFO_IDX_CPU=    2
+EINFO_IDX_TIME=   3
+EINFO_IDX_PID=    4
+EINFO_IDX_COMM=   5
+
+# Calculate a time interval(msec) from src(nsec) to dst(nsec)
+def diff_msec(src, dst):
+	return (dst - src) / 1000000.0
+
+# Display a process of transmitting a packet
+def print_transmit(hunk):
+	if dev != 0 and hunk['dev'].find(dev) < 0:
+		return
+	print "%7s %5d %6d.%06dsec %12.3fmsec      %12.3fmsec" % \
+		(hunk['dev'], hunk['len'],
+		nsecs_secs(hunk['queue_t']),
+		nsecs_nsecs(hunk['queue_t'])/1000,
+		diff_msec(hunk['queue_t'], hunk['xmit_t']),
+		diff_msec(hunk['xmit_t'], hunk['free_t']))
+
+# Format for displaying rx packet processing
+PF_IRQ_ENTRY= "  irq_entry(+%.3fmsec irq=%d:%s)"
+PF_SOFT_ENTRY="  softirq_entry(+%.3fmsec)"
+PF_NAPI_POLL= "  napi_poll_exit(+%.3fmsec %s)"
+PF_JOINT=     "         |"
+PF_WJOINT=    "         |            |"
+PF_NET_RECV=  "         |---netif_receive_skb(+%.3fmsec skb=%x len=%d)"
+PF_NET_RX=    "         |---netif_rx(+%.3fmsec skb=%x)"
+PF_CPY_DGRAM= "         |      skb_copy_datagram_iovec(+%.3fmsec %d:%s)"
+PF_KFREE_SKB= "         |      kfree_skb(+%.3fmsec location=%x)"
+PF_CONS_SKB=  "         |      consume_skb(+%.3fmsec)"
+
+# Display a process of received packets and interrputs associated with
+# a NET_RX softirq
+def print_receive(hunk):
+	show_hunk = 0
+	irq_list = hunk['irq_list']
+	cpu = irq_list[0]['cpu']
+	base_t = irq_list[0]['irq_ent_t']
+	# check if this hunk should be showed
+	if dev != 0:
+		for i in range(len(irq_list)):
+			if irq_list[i]['name'].find(dev) >= 0:
+				show_hunk = 1
+				break
+	else:
+		show_hunk = 1
+	if show_hunk == 0:
+		return
+
+	print "%d.%06dsec cpu=%d" % \
+		(nsecs_secs(base_t), nsecs_nsecs(base_t)/1000, cpu)
+	for i in range(len(irq_list)):
+		print PF_IRQ_ENTRY % \
+			(diff_msec(base_t, irq_list[i]['irq_ent_t']),
+			irq_list[i]['irq'], irq_list[i]['name'])
+		print PF_JOINT
+		irq_event_list = irq_list[i]['event_list']
+		for j in range(len(irq_event_list)):
+			irq_event = irq_event_list[j]
+			if irq_event['event'] == 'netif_rx':
+				print PF_NET_RX % \
+					(diff_msec(base_t, irq_event['time']),
+					irq_event['skbaddr'])
+				print PF_JOINT
+	print PF_SOFT_ENTRY % \
+		diff_msec(base_t, hunk['sirq_ent_t'])
+	print PF_JOINT
+	event_list = hunk['event_list']
+	for i in range(len(event_list)):
+		event = event_list[i]
+		if event['event_name'] == 'napi_poll':
+			print PF_NAPI_POLL % \
+			    (diff_msec(base_t, event['event_t']), event['dev'])
+			if i == len(event_list) - 1:
+				print ""
+			else:
+				print PF_JOINT
+		else:
+			print PF_NET_RECV % \
+			    (diff_msec(base_t, event['event_t']), event['skbaddr'],
+				event['len'])
+			if 'comm' in event.keys():
+				print PF_WJOINT
+				print PF_CPY_DGRAM % \
+					(diff_msec(base_t, event['comm_t']),
+					event['pid'], event['comm'])
+			elif 'handle' in event.keys():
+				print PF_WJOINT
+				if event['handle'] == "kfree_skb":
+					print PF_KFREE_SKB % \
+						(diff_msec(base_t,
+						event['comm_t']),
+						event['location'])
+				elif event['handle'] == "consume_skb":
+					print PF_CONS_SKB % \
+						diff_msec(base_t,
+							event['comm_t'])
+			print PF_JOINT
+
+def trace_begin():
+	global show_tx
+	global show_rx
+	global dev
+	global debug
+
+	for i in range(len(sys.argv)):
+		if i == 0:
+			continue
+		arg = sys.argv[i]
+		if arg == 'tx':
+			show_tx = 1
+		elif arg =='rx':
+			show_rx = 1
+		elif arg.find('dev=',0, 4) >= 0:
+			dev = arg[4:]
+		elif arg == 'debug':
+			debug = 1
+	if show_tx == 0  and show_rx == 0:
+		show_tx = 1
+		show_rx = 1
+
+def trace_end():
+	# order all events in time
+	all_event_list.sort(lambda a,b :cmp(a[EINFO_IDX_TIME],
+					    b[EINFO_IDX_TIME]))
+	# process all events
+	for i in range(len(all_event_list)):
+		event_info = all_event_list[i]
+		name = event_info[EINFO_IDX_NAME]
+		if name == 'irq__softirq_exit':
+			handle_irq_softirq_exit(event_info)
+		elif name == 'irq__softirq_entry':
+			handle_irq_softirq_entry(event_info)
+		elif name == 'irq__softirq_raise':
+			handle_irq_softirq_raise(event_info)
+		elif name == 'irq__irq_handler_entry':
+			handle_irq_handler_entry(event_info)
+		elif name == 'irq__irq_handler_exit':
+			handle_irq_handler_exit(event_info)
+		elif name == 'napi__napi_poll':
+			handle_napi_poll(event_info)
+		elif name == 'net__netif_receive_skb':
+			handle_netif_receive_skb(event_info)
+		elif name == 'net__netif_rx':
+			handle_netif_rx(event_info)
+		elif name == 'skb__skb_copy_datagram_iovec':
+			handle_skb_copy_datagram_iovec(event_info)
+		elif name == 'net__net_dev_queue':
+			handle_net_dev_queue(event_info)
+		elif name == 'net__net_dev_xmit':
+			handle_net_dev_xmit(event_info)
+		elif name == 'skb__kfree_skb':
+			handle_kfree_skb(event_info)
+		elif name == 'skb__consume_skb':
+			handle_consume_skb(event_info)
+	# display receive hunks
+	if show_rx:
+		for i in range(len(receive_hunk_list)):
+			print_receive(receive_hunk_list[i])
+	# display transmit hunks
+	if show_tx:
+		print "   dev    len      Qdisc        " \
+			"       netdevice             free"
+		for i in range(len(tx_free_list)):
+			print_transmit(tx_free_list[i])
+	if debug:
+		print "debug buffer status"
+		print "----------------------------"
+		print "xmit Qdisc:remain:%d overflow:%d" % \
+			(len(tx_queue_list), of_count_tx_queue_list)
+		print "xmit netdevice:remain:%d overflow:%d" % \
+			(len(tx_xmit_list), of_count_tx_xmit_list)
+		print "receive:remain:%d overflow:%d" % \
+			(len(rx_skb_list), of_count_rx_skb_list)
+
+# called from perf, when it finds a correspoinding event
+def irq__softirq_entry(name, context, cpu, sec, nsec, pid, comm, vec):
+	if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
+		return
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
+	all_event_list.append(event_info)
+
+def irq__softirq_exit(name, context, cpu, sec, nsec, pid, comm, vec):
+	if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
+		return
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
+	all_event_list.append(event_info)
+
+def irq__softirq_raise(name, context, cpu, sec, nsec, pid, comm, vec):
+	if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
+		return
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
+	all_event_list.append(event_info)
+
+def irq__irq_handler_entry(name, context, cpu, sec, nsec, pid, comm,
+			irq, irq_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			irq, irq_name)
+	all_event_list.append(event_info)
+
+def irq__irq_handler_exit(name, context, cpu, sec, nsec, pid, comm, irq, ret):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, irq, ret)
+	all_event_list.append(event_info)
+
+def napi__napi_poll(name, context, cpu, sec, nsec, pid, comm, napi, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			napi, dev_name)
+	all_event_list.append(event_info)
+
+def net__netif_receive_skb(name, context, cpu, sec, nsec, pid, comm, skbaddr,
+			skblen, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen, dev_name)
+	all_event_list.append(event_info)
+
+def net__netif_rx(name, context, cpu, sec, nsec, pid, comm, skbaddr,
+			skblen, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen, dev_name)
+	all_event_list.append(event_info)
+
+def net__net_dev_queue(name, context, cpu, sec, nsec, pid, comm,
+			skbaddr, skblen, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen, dev_name)
+	all_event_list.append(event_info)
+
+def net__net_dev_xmit(name, context, cpu, sec, nsec, pid, comm,
+			skbaddr, skblen, rc, dev_name):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen, rc ,dev_name)
+	all_event_list.append(event_info)
+
+def skb__kfree_skb(name, context, cpu, sec, nsec, pid, comm,
+			skbaddr, protocol, location):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, protocol, location)
+	all_event_list.append(event_info)
+
+def skb__consume_skb(name, context, cpu, sec, nsec, pid, comm, skbaddr):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr)
+	all_event_list.append(event_info)
+
+def skb__skb_copy_datagram_iovec(name, context, cpu, sec, nsec, pid, comm,
+	skbaddr, skblen):
+	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
+			skbaddr, skblen)
+	all_event_list.append(event_info)
+
+def handle_irq_handler_entry(event_info):
+	(name, context, cpu, time, pid, comm, irq, irq_name) = event_info
+	if cpu not in irq_dic.keys():
+		irq_dic[cpu] = []
+	irq_record = {'irq':irq, 'name':irq_name, 'cpu':cpu, 'irq_ent_t':time}
+	irq_dic[cpu].append(irq_record)
+
+def handle_irq_handler_exit(event_info):
+	(name, context, cpu, time, pid, comm, irq, ret) = event_info
+	if cpu not in irq_dic.keys():
+		return
+	irq_record = irq_dic[cpu].pop()
+	if irq != irq_record['irq']:
+		return
+	irq_record.update({'irq_ext_t':time})
+	# if an irq doesn't include NET_RX softirq, drop.
+	if 'event_list' in irq_record.keys():
+		irq_dic[cpu].append(irq_record)
+
+def handle_irq_softirq_raise(event_info):
+	(name, context, cpu, time, pid, comm, vec) = event_info
+	if cpu not in irq_dic.keys() \
+	or len(irq_dic[cpu]) == 0:
+		return
+	irq_record = irq_dic[cpu].pop()
+	if 'event_list' in irq_record.keys():
+		irq_event_list = irq_record['event_list']
+	else:
+		irq_event_list = []
+	irq_event_list.append({'time':time, 'event':'sirq_raise'})
+	irq_record.update({'event_list':irq_event_list})
+	irq_dic[cpu].append(irq_record)
+
+def handle_irq_softirq_entry(event_info):
+	(name, context, cpu, time, pid, comm, vec) = event_info
+	net_rx_dic[cpu] = {'sirq_ent_t':time, 'event_list':[]}
+
+def handle_irq_softirq_exit(event_info):
+	(name, context, cpu, time, pid, comm, vec) = event_info
+	irq_list = []
+	event_list = 0
+	if cpu in irq_dic.keys():
+		irq_list = irq_dic[cpu]
+		del irq_dic[cpu]
+	if cpu in net_rx_dic.keys():
+		sirq_ent_t = net_rx_dic[cpu]['sirq_ent_t']
+		event_list = net_rx_dic[cpu]['event_list']
+		del net_rx_dic[cpu]
+	if irq_list == [] or event_list == 0:
+		return
+	rec_data = {'sirq_ent_t':sirq_ent_t, 'sirq_ext_t':time,
+		    'irq_list':irq_list, 'event_list':event_list}
+	# merge information realted to a NET_RX softirq
+	receive_hunk_list.append(rec_data)
+
+def handle_napi_poll(event_info):
+	(name, context, cpu, time, pid, comm, napi, dev_name) = event_info
+	if cpu in net_rx_dic.keys():
+		event_list = net_rx_dic[cpu]['event_list']
+		rec_data = {'event_name':'napi_poll',
+				'dev':dev_name, 'event_t':time}
+		event_list.append(rec_data)
+
+def handle_netif_rx(event_info):
+	(name, context, cpu, time, pid, comm,
+		skbaddr, skblen, dev_name) = event_info
+	if cpu not in irq_dic.keys() \
+	or len(irq_dic[cpu]) == 0:
+		return
+	irq_record = irq_dic[cpu].pop()
+	if 'event_list' in irq_record.keys():
+		irq_event_list = irq_record['event_list']
+	else:
+		irq_event_list = []
+	irq_event_list.append({'time':time, 'event':'netif_rx',
+		'skbaddr':skbaddr, 'skblen':skblen, 'dev_name':dev_name})
+	irq_record.update({'event_list':irq_event_list})
+	irq_dic[cpu].append(irq_record)
+
+def handle_netif_receive_skb(event_info):
+	global of_count_rx_skb_list
+
+	(name, context, cpu, time, pid, comm,
+		skbaddr, skblen, dev_name) = event_info
+	if cpu in net_rx_dic.keys():
+		rec_data = {'event_name':'netif_receive_skb',
+			    'event_t':time, 'skbaddr':skbaddr, 'len':skblen}
+		event_list = net_rx_dic[cpu]['event_list']
+		event_list.append(rec_data)
+		rx_skb_list.insert(0, rec_data)
+		if len(rx_skb_list) > buffer_budget:
+			rx_skb_list.pop()
+			of_count_rx_skb_list += 1
+
+def handle_net_dev_queue(event_info):
+	global of_count_tx_queue_list
+
+	(name, context, cpu, time, pid, comm,
+		skbaddr, skblen, dev_name) = event_info
+	skb = {'dev':dev_name, 'skbaddr':skbaddr, 'len':skblen, 'queue_t':time}
+	tx_queue_list.insert(0, skb)
+	if len(tx_queue_list) > buffer_budget:
+		tx_queue_list.pop()
+		of_count_tx_queue_list += 1
+
+def handle_net_dev_xmit(event_info):
+	global of_count_tx_xmit_list
+
+	(name, context, cpu, time, pid, comm,
+		skbaddr, skblen, rc, dev_name) = event_info
+	if rc == 0: # NETDEV_TX_OK
+		for i in range(len(tx_queue_list)):
+			skb = tx_queue_list[i]
+			if skb['skbaddr'] == skbaddr:
+				skb['xmit_t'] = time
+				tx_xmit_list.insert(0, skb)
+				del tx_queue_list[i]
+				if len(tx_xmit_list) > buffer_budget:
+					tx_xmit_list.pop()
+					of_count_tx_xmit_list += 1
+				return
+
+def handle_kfree_skb(event_info):
+	(name, context, cpu, time, pid, comm,
+		skbaddr, protocol, location) = event_info
+	for i in range(len(tx_queue_list)):
+		skb = tx_queue_list[i]
+		if skb['skbaddr'] == skbaddr:
+			del tx_queue_list[i]
+			return
+	for i in range(len(tx_xmit_list)):
+		skb = tx_xmit_list[i]
+		if skb['skbaddr'] == skbaddr:
+			skb['free_t'] = time
+			tx_free_list.append(skb)
+			del tx_xmit_list[i]
+			return
+	for i in range(len(rx_skb_list)):
+		rec_data = rx_skb_list[i]
+		if rec_data['skbaddr'] == skbaddr:
+			rec_data.update({'handle':"kfree_skb",
+					'comm':comm, 'pid':pid, 'comm_t':time})
+			del rx_skb_list[i]
+			return
+
+def handle_consume_skb(event_info):
+	(name, context, cpu, time, pid, comm, skbaddr) = event_info
+	for i in range(len(tx_xmit_list)):
+		skb = tx_xmit_list[i]
+		if skb['skbaddr'] == skbaddr:
+			skb['free_t'] = time
+			tx_free_list.append(skb)
+			del tx_xmit_list[i]
+			return
+
+def handle_skb_copy_datagram_iovec(event_info):
+	(name, context, cpu, time, pid, comm, skbaddr, skblen) = event_info
+	for i in range(len(rx_skb_list)):
+		rec_data = rx_skb_list[i]
+		if skbaddr == rec_data['skbaddr']:
+			rec_data.update({'handle':"skb_copy_datagram_iovec",
+					'comm':comm, 'pid':pid, 'comm_t':time})
+			del rx_skb_list[i]
+			return
diff --git a/tools/perf/util/cache.h b/tools/perf/util/cache.h
index 27e9ebe..a772979 100644
--- a/tools/perf/util/cache.h
+++ b/tools/perf/util/cache.h
@@ -82,6 +82,8 @@
 extern char *perf_pathdup(const char *fmt, ...)
 	__attribute__((format (printf, 1, 2)));
 
+#ifdef NO_STRLCPY
 extern size_t strlcpy(char *dest, const char *src, size_t size);
+#endif
 
 #endif /* __PERF_CACHE_H */
diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c
index f231f43..e12d539 100644
--- a/tools/perf/util/callchain.c
+++ b/tools/perf/util/callchain.c
@@ -28,6 +28,9 @@
 #define chain_for_each_child(child, parent)	\
 	list_for_each_entry(child, &parent->children, brothers)
 
+#define chain_for_each_child_safe(child, next, parent)	\
+	list_for_each_entry_safe(child, next, &parent->children, brothers)
+
 static void
 rb_insert_callchain(struct rb_root *root, struct callchain_node *chain,
 		    enum chain_mode mode)
@@ -86,10 +89,10 @@
  * sort them by hit
  */
 static void
-sort_chain_flat(struct rb_root *rb_root, struct callchain_node *node,
+sort_chain_flat(struct rb_root *rb_root, struct callchain_root *root,
 		u64 min_hit, struct callchain_param *param __used)
 {
-	__sort_chain_flat(rb_root, node, min_hit);
+	__sort_chain_flat(rb_root, &root->node, min_hit);
 }
 
 static void __sort_chain_graph_abs(struct callchain_node *node,
@@ -108,11 +111,11 @@
 }
 
 static void
-sort_chain_graph_abs(struct rb_root *rb_root, struct callchain_node *chain_root,
+sort_chain_graph_abs(struct rb_root *rb_root, struct callchain_root *chain_root,
 		     u64 min_hit, struct callchain_param *param __used)
 {
-	__sort_chain_graph_abs(chain_root, min_hit);
-	rb_root->rb_node = chain_root->rb_root.rb_node;
+	__sort_chain_graph_abs(&chain_root->node, min_hit);
+	rb_root->rb_node = chain_root->node.rb_root.rb_node;
 }
 
 static void __sort_chain_graph_rel(struct callchain_node *node,
@@ -133,11 +136,11 @@
 }
 
 static void
-sort_chain_graph_rel(struct rb_root *rb_root, struct callchain_node *chain_root,
+sort_chain_graph_rel(struct rb_root *rb_root, struct callchain_root *chain_root,
 		     u64 min_hit __used, struct callchain_param *param)
 {
-	__sort_chain_graph_rel(chain_root, param->min_percent / 100.0);
-	rb_root->rb_node = chain_root->rb_root.rb_node;
+	__sort_chain_graph_rel(&chain_root->node, param->min_percent / 100.0);
+	rb_root->rb_node = chain_root->node.rb_root.rb_node;
 }
 
 int register_callchain_param(struct callchain_param *param)
@@ -284,19 +287,18 @@
 }
 
 static int
-__append_chain(struct callchain_node *root, struct resolved_chain *chain,
-	       unsigned int start, u64 period);
+append_chain(struct callchain_node *root, struct resolved_chain *chain,
+	     unsigned int start, u64 period);
 
 static void
-__append_chain_children(struct callchain_node *root,
-			struct resolved_chain *chain,
-			unsigned int start, u64 period)
+append_chain_children(struct callchain_node *root, struct resolved_chain *chain,
+		      unsigned int start, u64 period)
 {
 	struct callchain_node *rnode;
 
 	/* lookup in childrens */
 	chain_for_each_child(rnode, root) {
-		unsigned int ret = __append_chain(rnode, chain, start, period);
+		unsigned int ret = append_chain(rnode, chain, start, period);
 
 		if (!ret)
 			goto inc_children_hit;
@@ -309,8 +311,8 @@
 }
 
 static int
-__append_chain(struct callchain_node *root, struct resolved_chain *chain,
-	       unsigned int start, u64 period)
+append_chain(struct callchain_node *root, struct resolved_chain *chain,
+	     unsigned int start, u64 period)
 {
 	struct callchain_list *cnode;
 	unsigned int i = start;
@@ -357,7 +359,7 @@
 	}
 
 	/* We match the node and still have a part remaining */
-	__append_chain_children(root, chain, i, period);
+	append_chain_children(root, chain, i, period);
 
 	return 0;
 }
@@ -380,8 +382,8 @@
 }
 
 
-int append_chain(struct callchain_node *root, struct ip_callchain *chain,
-		 struct map_symbol *syms, u64 period)
+int callchain_append(struct callchain_root *root, struct ip_callchain *chain,
+		     struct map_symbol *syms, u64 period)
 {
 	struct resolved_chain *filtered;
 
@@ -398,9 +400,65 @@
 	if (!filtered->nr)
 		goto end;
 
-	__append_chain_children(root, filtered, 0, period);
+	append_chain_children(&root->node, filtered, 0, period);
+
+	if (filtered->nr > root->max_depth)
+		root->max_depth = filtered->nr;
 end:
 	free(filtered);
 
 	return 0;
 }
+
+static int
+merge_chain_branch(struct callchain_node *dst, struct callchain_node *src,
+		   struct resolved_chain *chain)
+{
+	struct callchain_node *child, *next_child;
+	struct callchain_list *list, *next_list;
+	int old_pos = chain->nr;
+	int err = 0;
+
+	list_for_each_entry_safe(list, next_list, &src->val, list) {
+		chain->ips[chain->nr].ip = list->ip;
+		chain->ips[chain->nr].ms = list->ms;
+		chain->nr++;
+		list_del(&list->list);
+		free(list);
+	}
+
+	if (src->hit)
+		append_chain_children(dst, chain, 0, src->hit);
+
+	chain_for_each_child_safe(child, next_child, src) {
+		err = merge_chain_branch(dst, child, chain);
+		if (err)
+			break;
+
+		list_del(&child->brothers);
+		free(child);
+	}
+
+	chain->nr = old_pos;
+
+	return err;
+}
+
+int callchain_merge(struct callchain_root *dst, struct callchain_root *src)
+{
+	struct resolved_chain *chain;
+	int err;
+
+	chain = malloc(sizeof(*chain) +
+		       src->max_depth * sizeof(struct resolved_ip));
+	if (!chain)
+		return -ENOMEM;
+
+	chain->nr = 0;
+
+	err = merge_chain_branch(&dst->node, &src->node, chain);
+
+	free(chain);
+
+	return err;
+}
diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h
index 6de4313..c15fb8c 100644
--- a/tools/perf/util/callchain.h
+++ b/tools/perf/util/callchain.h
@@ -26,9 +26,14 @@
 	u64			children_hit;
 };
 
+struct callchain_root {
+	u64			max_depth;
+	struct callchain_node	node;
+};
+
 struct callchain_param;
 
-typedef void (*sort_chain_func_t)(struct rb_root *, struct callchain_node *,
+typedef void (*sort_chain_func_t)(struct rb_root *, struct callchain_root *,
 				 u64, struct callchain_param *);
 
 struct callchain_param {
@@ -44,15 +49,16 @@
 	struct list_head	list;
 };
 
-static inline void callchain_init(struct callchain_node *node)
+static inline void callchain_init(struct callchain_root *root)
 {
-	INIT_LIST_HEAD(&node->brothers);
-	INIT_LIST_HEAD(&node->children);
-	INIT_LIST_HEAD(&node->val);
+	INIT_LIST_HEAD(&root->node.brothers);
+	INIT_LIST_HEAD(&root->node.children);
+	INIT_LIST_HEAD(&root->node.val);
 
-	node->children_hit = 0;
-	node->parent = NULL;
-	node->hit = 0;
+	root->node.parent = NULL;
+	root->node.hit = 0;
+	root->node.children_hit = 0;
+	root->max_depth = 0;
 }
 
 static inline u64 cumul_hits(struct callchain_node *node)
@@ -61,8 +67,9 @@
 }
 
 int register_callchain_param(struct callchain_param *param);
-int append_chain(struct callchain_node *root, struct ip_callchain *chain,
-		 struct map_symbol *syms, u64 period);
+int callchain_append(struct callchain_root *root, struct ip_callchain *chain,
+		     struct map_symbol *syms, u64 period);
+int callchain_merge(struct callchain_root *dst, struct callchain_root *src);
 
 bool ip_callchain__valid(struct ip_callchain *chain, const event_t *event);
 #endif	/* __PERF_CALLCHAIN_H */
diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c
index be22ae6..2022e87 100644
--- a/tools/perf/util/hist.c
+++ b/tools/perf/util/hist.c
@@ -87,7 +87,7 @@
 
 static struct hist_entry *hist_entry__new(struct hist_entry *template)
 {
-	size_t callchain_size = symbol_conf.use_callchain ? sizeof(struct callchain_node) : 0;
+	size_t callchain_size = symbol_conf.use_callchain ? sizeof(struct callchain_root) : 0;
 	struct hist_entry *self = malloc(sizeof(*self) + callchain_size);
 
 	if (self != NULL) {
@@ -226,6 +226,8 @@
 
 		if (!cmp) {
 			iter->period += he->period;
+			if (symbol_conf.use_callchain)
+				callchain_merge(iter->callchain, he->callchain);
 			hist_entry__free(he);
 			return false;
 		}
diff --git a/tools/perf/util/path.c b/tools/perf/util/path.c
index 58a470d..bd74977 100644
--- a/tools/perf/util/path.c
+++ b/tools/perf/util/path.c
@@ -22,6 +22,7 @@
 	return ".";
 }
 
+#ifdef NO_STRLCPY
 size_t strlcpy(char *dest, const char *src, size_t size)
 {
 	size_t ret = strlen(src);
@@ -33,7 +34,7 @@
 	}
 	return ret;
 }
-
+#endif
 
 static char *get_pathname(void)
 {
diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h
index 46e531d..0b91053 100644
--- a/tools/perf/util/sort.h
+++ b/tools/perf/util/sort.h
@@ -70,7 +70,7 @@
 		struct hist_entry *pair;
 		struct rb_root	  sorted_chain;
 	};
-	struct callchain_node	callchain[0];
+	struct callchain_root	callchain[0];
 };
 
 enum sort_type {
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index b2f5ae9..b39f499 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -388,6 +388,20 @@
 	return fprintf(fp, "%s", sbuild_id);
 }
 
+size_t dso__fprintf_symbols_by_name(struct dso *self, enum map_type type, FILE *fp)
+{
+	size_t ret = 0;
+	struct rb_node *nd;
+	struct symbol_name_rb_node *pos;
+
+	for (nd = rb_first(&self->symbol_names[type]); nd; nd = rb_next(nd)) {
+		pos = rb_entry(nd, struct symbol_name_rb_node, rb_node);
+		fprintf(fp, "%s\n", pos->sym.name);
+	}
+
+	return ret;
+}
+
 size_t dso__fprintf(struct dso *self, enum map_type type, FILE *fp)
 {
 	struct rb_node *nd;
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index ea95c27..038f220 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -182,6 +182,7 @@
 size_t machines__fprintf_dsos_buildid(struct rb_root *self, FILE *fp, bool with_hits);
 
 size_t dso__fprintf_buildid(struct dso *self, FILE *fp);
+size_t dso__fprintf_symbols_by_name(struct dso *self, enum map_type type, FILE *fp);
 size_t dso__fprintf(struct dso *self, enum map_type type, FILE *fp);
 
 enum dso_origin {
diff --git a/tools/perf/util/ui/browser.c b/tools/perf/util/ui/browser.c
index 66f2d58..6d0df80 100644
--- a/tools/perf/util/ui/browser.c
+++ b/tools/perf/util/ui/browser.c
@@ -1,16 +1,6 @@
-#define _GNU_SOURCE
-#include <stdio.h>
-#undef _GNU_SOURCE
-/*
- * slang versions <= 2.0.6 have a "#if HAVE_LONG_LONG" that breaks
- * the build if it isn't defined. Use the equivalent one that glibc
- * has on features.h.
- */
-#include <features.h>
-#ifndef HAVE_LONG_LONG
-#define HAVE_LONG_LONG __GLIBC_HAVE_LONG_LONG
-#endif
 #include <slang.h>
+#include "libslang.h"
+#include <linux/compiler.h>
 #include <linux/list.h>
 #include <linux/rbtree.h>
 #include <stdlib.h>
@@ -19,17 +9,9 @@
 #include "helpline.h"
 #include "../color.h"
 #include "../util.h"
+#include <stdio.h>
 
-#if SLANG_VERSION < 20104
-#define sltt_set_color(obj, name, fg, bg) \
-	SLtt_set_color(obj,(char *)name, (char *)fg, (char *)bg)
-#else
-#define sltt_set_color SLtt_set_color
-#endif
-
-newtComponent newt_form__new(void);
-
-int ui_browser__percent_color(double percent, bool current)
+static int ui_browser__percent_color(double percent, bool current)
 {
 	if (current)
 		return HE_COLORSET_SELECTED;
@@ -40,6 +22,23 @@
 	return HE_COLORSET_NORMAL;
 }
 
+void ui_browser__set_color(struct ui_browser *self __used, int color)
+{
+	SLsmg_set_color(color);
+}
+
+void ui_browser__set_percent_color(struct ui_browser *self,
+				   double percent, bool current)
+{
+	 int color = ui_browser__percent_color(percent, current);
+	 ui_browser__set_color(self, color);
+}
+
+void ui_browser__gotorc(struct ui_browser *self, int y, int x)
+{
+	SLsmg_gotorc(self->y + y, self->x + x);
+}
+
 void ui_browser__list_head_seek(struct ui_browser *self, off_t offset, int whence)
 {
 	struct list_head *head = self->entries;
@@ -111,7 +110,7 @@
 	nd = self->top;
 
 	while (nd != NULL) {
-		SLsmg_gotorc(self->y + row, self->x);
+		ui_browser__gotorc(self, row, 0);
 		self->write(self, nd, row);
 		if (++row == self->height)
 			break;
@@ -131,13 +130,10 @@
 	int cols, rows;
 	newtGetScreenSize(&cols, &rows);
 
-	if (self->width > cols - 4)
-		self->width = cols - 4;
-	self->height = rows - 5;
-	if (self->height > self->nr_entries)
-		self->height = self->nr_entries;
-	self->y  = (rows - self->height) / 2;
-	self->x = (cols - self->width) / 2;
+	self->width = cols - 1;
+	self->height = rows - 2;
+	self->y = 1;
+	self->x = 0;
 }
 
 void ui_browser__reset_index(struct ui_browser *self)
@@ -146,34 +142,48 @@
 	self->seek(self, 0, SEEK_SET);
 }
 
+void ui_browser__add_exit_key(struct ui_browser *self, int key)
+{
+	newtFormAddHotKey(self->form, key);
+}
+
+void ui_browser__add_exit_keys(struct ui_browser *self, int keys[])
+{
+	int i = 0;
+
+	while (keys[i] && i < 64) {
+		ui_browser__add_exit_key(self, keys[i]);
+		++i;
+	}
+}
+
 int ui_browser__show(struct ui_browser *self, const char *title,
 		     const char *helpline, ...)
 {
 	va_list ap;
+	int keys[] = { NEWT_KEY_UP, NEWT_KEY_DOWN, NEWT_KEY_PGUP,
+		       NEWT_KEY_PGDN, NEWT_KEY_HOME, NEWT_KEY_END, ' ',
+		       NEWT_KEY_LEFT, NEWT_KEY_ESCAPE, 'q', CTRL('c'), 0 };
 
-	if (self->form != NULL) {
+	if (self->form != NULL)
 		newtFormDestroy(self->form);
-		newtPopWindow();
-	}
+
 	ui_browser__refresh_dimensions(self);
-	newtCenteredWindow(self->width, self->height, title);
-	self->form = newt_form__new();
+	self->form = newtForm(NULL, NULL, 0);
 	if (self->form == NULL)
 		return -1;
 
-	self->sb = newtVerticalScrollbar(self->width, 0, self->height,
+	self->sb = newtVerticalScrollbar(self->width, 1, self->height,
 					 HE_COLORSET_NORMAL,
 					 HE_COLORSET_SELECTED);
 	if (self->sb == NULL)
 		return -1;
 
-	newtFormAddHotKey(self->form, NEWT_KEY_UP);
-	newtFormAddHotKey(self->form, NEWT_KEY_DOWN);
-	newtFormAddHotKey(self->form, NEWT_KEY_PGUP);
-	newtFormAddHotKey(self->form, NEWT_KEY_PGDN);
-	newtFormAddHotKey(self->form, NEWT_KEY_HOME);
-	newtFormAddHotKey(self->form, NEWT_KEY_END);
-	newtFormAddHotKey(self->form, ' ');
+	SLsmg_gotorc(0, 0);
+	ui_browser__set_color(self, NEWT_COLORSET_ROOT);
+	slsmg_write_nstring(title, self->width);
+
+	ui_browser__add_exit_keys(self, keys);
 	newtFormAddComponent(self->form, self->sb);
 
 	va_start(ap, helpline);
@@ -185,7 +195,6 @@
 void ui_browser__hide(struct ui_browser *self)
 {
 	newtFormDestroy(self->form);
-	newtPopWindow();
 	self->form = NULL;
 	ui_helpline__pop();
 }
@@ -196,28 +205,28 @@
 
 	newtScrollbarSet(self->sb, self->index, self->nr_entries - 1);
 	row = self->refresh(self);
-	SLsmg_set_color(HE_COLORSET_NORMAL);
+	ui_browser__set_color(self, HE_COLORSET_NORMAL);
 	SLsmg_fill_region(self->y + row, self->x,
 			  self->height - row, self->width, ' ');
 
 	return 0;
 }
 
-int ui_browser__run(struct ui_browser *self, struct newtExitStruct *es)
+int ui_browser__run(struct ui_browser *self)
 {
+	struct newtExitStruct es;
+
 	if (ui_browser__refresh(self) < 0)
 		return -1;
 
 	while (1) {
 		off_t offset;
 
-		newtFormRun(self->form, es);
+		newtFormRun(self->form, &es);
 
-		if (es->reason != NEWT_EXIT_HOTKEY)
+		if (es.reason != NEWT_EXIT_HOTKEY)
 			break;
-		if (is_exit_key(es->u.key))
-			return es->u.key;
-		switch (es->u.key) {
+		switch (es.u.key) {
 		case NEWT_KEY_DOWN:
 			if (self->index == self->nr_entries - 1)
 				break;
@@ -274,12 +283,12 @@
 			self->seek(self, -offset, SEEK_END);
 			break;
 		default:
-			return es->u.key;
+			return es.u.key;
 		}
 		if (ui_browser__refresh(self) < 0)
 			return -1;
 	}
-	return 0;
+	return -1;
 }
 
 unsigned int ui_browser__list_head_refresh(struct ui_browser *self)
@@ -294,7 +303,7 @@
 	pos = self->top;
 
 	list_for_each_from(pos, head) {
-		SLsmg_gotorc(self->y + row, self->x);
+		ui_browser__gotorc(self, row, 0);
 		self->write(self, pos, row);
 		if (++row == self->height)
 			break;
diff --git a/tools/perf/util/ui/browser.h b/tools/perf/util/ui/browser.h
index 0b9f829..0dc7e4d 100644
--- a/tools/perf/util/ui/browser.h
+++ b/tools/perf/util/ui/browser.h
@@ -25,16 +25,21 @@
 };
 
 
-int ui_browser__percent_color(double percent, bool current);
+void ui_browser__set_color(struct ui_browser *self, int color);
+void ui_browser__set_percent_color(struct ui_browser *self,
+				   double percent, bool current);
 bool ui_browser__is_current_entry(struct ui_browser *self, unsigned row);
 void ui_browser__refresh_dimensions(struct ui_browser *self);
 void ui_browser__reset_index(struct ui_browser *self);
 
+void ui_browser__gotorc(struct ui_browser *self, int y, int x);
+void ui_browser__add_exit_key(struct ui_browser *self, int key);
+void ui_browser__add_exit_keys(struct ui_browser *self, int keys[]);
 int ui_browser__show(struct ui_browser *self, const char *title,
 		     const char *helpline, ...);
 void ui_browser__hide(struct ui_browser *self);
 int ui_browser__refresh(struct ui_browser *self);
-int ui_browser__run(struct ui_browser *self, struct newtExitStruct *es);
+int ui_browser__run(struct ui_browser *self);
 
 void ui_browser__rb_tree_seek(struct ui_browser *self, off_t offset, int whence);
 unsigned int ui_browser__rb_tree_refresh(struct ui_browser *self);
diff --git a/tools/perf/util/ui/browsers/annotate.c b/tools/perf/util/ui/browsers/annotate.c
index a90273e..82b78f9 100644
--- a/tools/perf/util/ui/browsers/annotate.c
+++ b/tools/perf/util/ui/browsers/annotate.c
@@ -40,14 +40,12 @@
 
 	if (ol->offset != -1) {
 		struct objdump_line_rb_node *olrb = objdump_line__rb(ol);
-		int color = ui_browser__percent_color(olrb->percent, current_entry);
-		SLsmg_set_color(color);
+		ui_browser__set_percent_color(self, olrb->percent, current_entry);
 		slsmg_printf(" %7.2f ", olrb->percent);
 		if (!current_entry)
-			SLsmg_set_color(HE_COLORSET_CODE);
+			ui_browser__set_color(self, HE_COLORSET_CODE);
 	} else {
-		int color = ui_browser__percent_color(0, current_entry);
-		SLsmg_set_color(color);
+		ui_browser__set_percent_color(self, 0, current_entry);
 		slsmg_write_nstring(" ", 9);
 	}
 
@@ -135,32 +133,31 @@
 	self->curr_hot = nd;
 }
 
-static int annotate_browser__run(struct annotate_browser *self,
-				 struct newtExitStruct *es)
+static int annotate_browser__run(struct annotate_browser *self)
 {
 	struct rb_node *nd;
 	struct hist_entry *he = self->b.priv;
+	int key;
 
 	if (ui_browser__show(&self->b, he->ms.sym->name,
-			     "<- or ESC: exit, TAB/shift+TAB: cycle thru samples") < 0)
+			     "<-, -> or ESC: exit, TAB/shift+TAB: cycle thru samples") < 0)
 		return -1;
-
-	newtFormAddHotKey(self->b.form, NEWT_KEY_LEFT);
-	newtFormAddHotKey(self->b.form, NEWT_KEY_RIGHT);
+	/*
+	 * To allow builtin-annotate to cycle thru multiple symbols by
+	 * examining the exit key for this function.
+	 */
+	ui_browser__add_exit_key(&self->b, NEWT_KEY_RIGHT);
 
 	nd = self->curr_hot;
 	if (nd) {
-		newtFormAddHotKey(self->b.form, NEWT_KEY_TAB);
-		newtFormAddHotKey(self->b.form, NEWT_KEY_UNTAB);
+		int tabs[] = { NEWT_KEY_TAB, NEWT_KEY_UNTAB, 0 };
+		ui_browser__add_exit_keys(&self->b, tabs);
 	}
 
 	while (1) {
-		ui_browser__run(&self->b, es);
+		key = ui_browser__run(&self->b);
 
-		if (es->reason != NEWT_EXIT_HOTKEY)
-			break;
-
-		switch (es->u.key) {
+		switch (key) {
 		case NEWT_KEY_TAB:
 			nd = rb_prev(nd);
 			if (nd == NULL)
@@ -179,12 +176,11 @@
 	}
 out:
 	ui_browser__hide(&self->b);
-	return es->u.key;
+	return key;
 }
 
 int hist_entry__tui_annotate(struct hist_entry *self)
 {
-	struct newtExitStruct es;
 	struct objdump_line *pos, *n;
 	struct objdump_line_rb_node *rbpos;
 	LIST_HEAD(head);
@@ -232,7 +228,7 @@
 		annotate_browser__set_top(&browser, browser.curr_hot);
 
 	browser.b.width += 18; /* Percentage */
-	ret = annotate_browser__run(&browser, &es);
+	ret = annotate_browser__run(&browser);
 	list_for_each_entry_safe(pos, n, &head, node) {
 		list_del(&pos->node);
 		objdump_line__free(pos);
diff --git a/tools/perf/util/ui/browsers/hists.c b/tools/perf/util/ui/browsers/hists.c
index dafdf67..ebda8c3 100644
--- a/tools/perf/util/ui/browsers/hists.c
+++ b/tools/perf/util/ui/browsers/hists.c
@@ -58,6 +58,11 @@
 	return map_symbol__folded(&self->ms);
 }
 
+static void map_symbol__set_folding(struct map_symbol *self, bool unfold)
+{
+	self->unfolded = unfold ? self->has_children : false;
+}
+
 static int callchain_node__count_rows_rb_tree(struct callchain_node *self)
 {
 	int n = 0;
@@ -129,16 +134,16 @@
 	for (nd = rb_first(&self->rb_root); nd; nd = rb_next(nd)) {
 		struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node);
 		struct callchain_list *chain;
-		int first = true;
+		bool first = true;
 
 		list_for_each_entry(chain, &child->val, list) {
 			if (first) {
 				first = false;
 				chain->ms.has_children = chain->list.next != &child->val ||
-							 rb_first(&child->rb_root) != NULL;
+							 !RB_EMPTY_ROOT(&child->rb_root);
 			} else
 				chain->ms.has_children = chain->list.next == &child->val &&
-							 rb_first(&child->rb_root) != NULL;
+							 !RB_EMPTY_ROOT(&child->rb_root);
 		}
 
 		callchain_node__init_have_children_rb_tree(child);
@@ -150,7 +155,7 @@
 	struct callchain_list *chain;
 
 	list_for_each_entry(chain, &self->val, list)
-		chain->ms.has_children = rb_first(&self->rb_root) != NULL;
+		chain->ms.has_children = !RB_EMPTY_ROOT(&self->rb_root);
 
 	callchain_node__init_have_children_rb_tree(self);
 }
@@ -168,6 +173,7 @@
 static void hist_entry__init_have_children(struct hist_entry *self)
 {
 	if (!self->init_have_children) {
+		self->ms.has_children = !RB_EMPTY_ROOT(&self->sorted_chain);
 		callchain__init_have_children(&self->sorted_chain);
 		self->init_have_children = true;
 	}
@@ -195,43 +201,114 @@
 	return false;
 }
 
-static int hist_browser__run(struct hist_browser *self, const char *title,
-			     struct newtExitStruct *es)
+static int callchain_node__set_folding_rb_tree(struct callchain_node *self, bool unfold)
 {
-	char str[256], unit;
-	unsigned long nr_events = self->hists->stats.nr_events[PERF_RECORD_SAMPLE];
+	int n = 0;
+	struct rb_node *nd;
+
+	for (nd = rb_first(&self->rb_root); nd; nd = rb_next(nd)) {
+		struct callchain_node *child = rb_entry(nd, struct callchain_node, rb_node);
+		struct callchain_list *chain;
+		bool has_children = false;
+
+		list_for_each_entry(chain, &child->val, list) {
+			++n;
+			map_symbol__set_folding(&chain->ms, unfold);
+			has_children = chain->ms.has_children;
+		}
+
+		if (has_children)
+			n += callchain_node__set_folding_rb_tree(child, unfold);
+	}
+
+	return n;
+}
+
+static int callchain_node__set_folding(struct callchain_node *node, bool unfold)
+{
+	struct callchain_list *chain;
+	bool has_children = false;
+	int n = 0;
+
+	list_for_each_entry(chain, &node->val, list) {
+		++n;
+		map_symbol__set_folding(&chain->ms, unfold);
+		has_children = chain->ms.has_children;
+	}
+
+	if (has_children)
+		n += callchain_node__set_folding_rb_tree(node, unfold);
+
+	return n;
+}
+
+static int callchain__set_folding(struct rb_root *chain, bool unfold)
+{
+	struct rb_node *nd;
+	int n = 0;
+
+	for (nd = rb_first(chain); nd; nd = rb_next(nd)) {
+		struct callchain_node *node = rb_entry(nd, struct callchain_node, rb_node);
+		n += callchain_node__set_folding(node, unfold);
+	}
+
+	return n;
+}
+
+static void hist_entry__set_folding(struct hist_entry *self, bool unfold)
+{
+	hist_entry__init_have_children(self);
+	map_symbol__set_folding(&self->ms, unfold);
+
+	if (self->ms.has_children) {
+		int n = callchain__set_folding(&self->sorted_chain, unfold);
+		self->nr_rows = unfold ? n : 0;
+	} else
+		self->nr_rows = 0;
+}
+
+static void hists__set_folding(struct hists *self, bool unfold)
+{
+	struct rb_node *nd;
+
+	self->nr_entries = 0;
+
+	for (nd = rb_first(&self->entries); nd; nd = rb_next(nd)) {
+		struct hist_entry *he = rb_entry(nd, struct hist_entry, rb_node);
+		hist_entry__set_folding(he, unfold);
+		self->nr_entries += 1 + he->nr_rows;
+	}
+}
+
+static void hist_browser__set_folding(struct hist_browser *self, bool unfold)
+{
+	hists__set_folding(self->hists, unfold);
+	self->b.nr_entries = self->hists->nr_entries;
+	/* Go to the start, we may be way after valid entries after a collapse */
+	ui_browser__reset_index(&self->b);
+}
+
+static int hist_browser__run(struct hist_browser *self, const char *title)
+{
+	int key;
+	int exit_keys[] = { 'a', '?', 'h', 'C', 'd', 'D', 'E', 't',
+			    NEWT_KEY_ENTER, NEWT_KEY_RIGHT, NEWT_KEY_LEFT, 0, };
 
 	self->b.entries = &self->hists->entries;
 	self->b.nr_entries = self->hists->nr_entries;
 
 	hist_browser__refresh_dimensions(self);
 
-	nr_events = convert_unit(nr_events, &unit);
-	snprintf(str, sizeof(str), "Events: %lu%c                            ",
-		 nr_events, unit);
-	newtDrawRootText(0, 0, str);
-
 	if (ui_browser__show(&self->b, title,
 			     "Press '?' for help on key bindings") < 0)
 		return -1;
 
-	newtFormAddHotKey(self->b.form, 'a');
-	newtFormAddHotKey(self->b.form, '?');
-	newtFormAddHotKey(self->b.form, 'h');
-	newtFormAddHotKey(self->b.form, 'd');
-	newtFormAddHotKey(self->b.form, 'D');
-	newtFormAddHotKey(self->b.form, 't');
-
-	newtFormAddHotKey(self->b.form, NEWT_KEY_LEFT);
-	newtFormAddHotKey(self->b.form, NEWT_KEY_RIGHT);
-	newtFormAddHotKey(self->b.form, NEWT_KEY_ENTER);
+	ui_browser__add_exit_keys(&self->b, exit_keys);
 
 	while (1) {
-		ui_browser__run(&self->b, es);
+		key = ui_browser__run(&self->b);
 
-		if (es->reason != NEWT_EXIT_HOTKEY)
-			break;
-		switch (es->u.key) {
+		switch (key) {
 		case 'D': { /* Debug */
 			static int seq;
 			struct hist_entry *h = rb_entry(self->b.top,
@@ -245,18 +322,26 @@
 					   self->b.top_idx,
 					   h->row_offset, h->nr_rows);
 		}
-			continue;
+			break;
+		case 'C':
+			/* Collapse the whole world. */
+			hist_browser__set_folding(self, false);
+			break;
+		case 'E':
+			/* Expand the whole world. */
+			hist_browser__set_folding(self, true);
+			break;
 		case NEWT_KEY_ENTER:
 			if (hist_browser__toggle_fold(self))
 				break;
 			/* fall thru */
 		default:
-			return 0;
+			goto out;
 		}
 	}
-
+out:
 	ui_browser__hide(&self->b);
-	return 0;
+	return key;
 }
 
 static char *callchain_list__sym_name(struct callchain_list *self,
@@ -306,15 +391,10 @@
 			int color;
 			bool was_first = first;
 
-			if (first) {
+			if (first)
 				first = false;
-				chain->ms.has_children = chain->list.next != &child->val ||
-							 rb_first(&child->rb_root) != NULL;
-			} else {
+			else
 				extra_offset = LEVEL_OFFSET_STEP;
-				chain->ms.has_children = chain->list.next == &child->val &&
-							 rb_first(&child->rb_root) != NULL;
-			}
 
 			folded_sign = callchain_list__folded(chain);
 			if (*row_offset != 0) {
@@ -341,8 +421,8 @@
 				*is_current_entry = true;
 			}
 
-			SLsmg_set_color(color);
-			SLsmg_gotorc(self->b.y + row, self->b.x);
+			ui_browser__set_color(&self->b, color);
+			ui_browser__gotorc(&self->b, row, 0);
 			slsmg_write_nstring(" ", offset + extra_offset);
 			slsmg_printf("%c ", folded_sign);
 			slsmg_write_nstring(str, width);
@@ -384,12 +464,7 @@
 	list_for_each_entry(chain, &node->val, list) {
 		char ipstr[BITS_PER_LONG / 4 + 1], *s;
 		int color;
-		/*
-		 * FIXME: This should be moved to somewhere else,
-		 * probably when the callchain is created, so as not to
-		 * traverse it all over again
-		 */
-		chain->ms.has_children = rb_first(&node->rb_root) != NULL;
+
 		folded_sign = callchain_list__folded(chain);
 
 		if (*row_offset != 0) {
@@ -405,8 +480,8 @@
 		}
 
 		s = callchain_list__sym_name(chain, ipstr, sizeof(ipstr));
-		SLsmg_gotorc(self->b.y + row, self->b.x);
-		SLsmg_set_color(color);
+		ui_browser__gotorc(&self->b, row, 0);
+		ui_browser__set_color(&self->b, color);
 		slsmg_write_nstring(" ", offset);
 		slsmg_printf("%c ", folded_sign);
 		slsmg_write_nstring(s, width - 2);
@@ -465,7 +540,7 @@
 	}
 
 	if (symbol_conf.use_callchain) {
-		entry->ms.has_children = !RB_EMPTY_ROOT(&entry->sorted_chain);
+		hist_entry__init_have_children(entry);
 		folded_sign = hist_entry__folded(entry);
 	}
 
@@ -484,8 +559,8 @@
 				color = HE_COLORSET_NORMAL;
 		}
 
-		SLsmg_set_color(color);
-		SLsmg_gotorc(self->b.y + row, self->b.x);
+		ui_browser__set_color(&self->b, color);
+		ui_browser__gotorc(&self->b, row, 0);
 		if (symbol_conf.use_callchain) {
 			slsmg_printf("%c ", folded_sign);
 			width -= 2;
@@ -687,8 +762,6 @@
 
 static void hist_browser__delete(struct hist_browser *self)
 {
-	newtFormDestroy(self->b.form);
-	newtPopWindow();
 	free(self);
 }
 
@@ -702,21 +775,26 @@
 	return self->he_selection->thread;
 }
 
-static int hist_browser__title(char *bf, size_t size, const char *ev_name,
-			       const struct dso *dso, const struct thread *thread)
+static int hists__browser_title(struct hists *self, char *bf, size_t size,
+				const char *ev_name, const struct dso *dso,
+				const struct thread *thread)
 {
-	int printed = 0;
+	char unit;
+	int printed;
+	unsigned long nr_events = self->stats.nr_events[PERF_RECORD_SAMPLE];
+
+	nr_events = convert_unit(nr_events, &unit);
+	printed = snprintf(bf, size, "Events: %lu%c %s", nr_events, unit, ev_name);
 
 	if (thread)
 		printed += snprintf(bf + printed, size - printed,
-				    "Thread: %s(%d)",
-				    (thread->comm_set ?  thread->comm : ""),
+				    ", Thread: %s(%d)",
+				    (thread->comm_set ? thread->comm : ""),
 				    thread->pid);
 	if (dso)
 		printed += snprintf(bf + printed, size - printed,
-				    "%sDSO: %s", thread ? " " : "",
-				    dso->short_name);
-	return printed ?: snprintf(bf, size, "Event: %s", ev_name);
+				    ", DSO: %s", dso->short_name);
+	return printed;
 }
 
 int hists__browse(struct hists *self, const char *helpline, const char *ev_name)
@@ -725,7 +803,6 @@
 	struct pstack *fstack;
 	const struct thread *thread_filter = NULL;
 	const struct dso *dso_filter = NULL;
-	struct newtExitStruct es;
 	char msg[160];
 	int key = -1;
 
@@ -738,9 +815,8 @@
 
 	ui_helpline__push(helpline);
 
-	hist_browser__title(msg, sizeof(msg), ev_name,
-			    dso_filter, thread_filter);
-
+	hists__browser_title(self, msg, sizeof(msg), ev_name,
+			     dso_filter, thread_filter);
 	while (1) {
 		const struct thread *thread;
 		const struct dso *dso;
@@ -749,70 +825,63 @@
 		    annotate = -2, zoom_dso = -2, zoom_thread = -2,
 		    browse_map = -2;
 
-		if (hist_browser__run(browser, msg, &es))
-			break;
+		key = hist_browser__run(browser, msg);
 
 		thread = hist_browser__selected_thread(browser);
 		dso = browser->selection->map ? browser->selection->map->dso : NULL;
 
-		if (es.reason == NEWT_EXIT_HOTKEY) {
-			key = es.u.key;
-
-			switch (key) {
-			case NEWT_KEY_F1:
-				goto do_help;
-			case NEWT_KEY_TAB:
-			case NEWT_KEY_UNTAB:
-				/*
-				 * Exit the browser, let hists__browser_tree
-				 * go to the next or previous
-				 */
-				goto out_free_stack;
-			default:;
-			}
-
-			switch (key) {
-			case 'a':
-				if (browser->selection->map == NULL &&
-				    browser->selection->map->dso->annotate_warned)
-					continue;
-				goto do_annotate;
-			case 'd':
-				goto zoom_dso;
-			case 't':
-				goto zoom_thread;
-			case 'h':
-			case '?':
-do_help:
-				ui__help_window("->        Zoom into DSO/Threads & Annotate current symbol\n"
-						"<-        Zoom out\n"
-						"a         Annotate current symbol\n"
-						"h/?/F1    Show this window\n"
-						"d         Zoom into current DSO\n"
-						"t         Zoom into current Thread\n"
-						"q/CTRL+C  Exit browser");
+		switch (key) {
+		case NEWT_KEY_TAB:
+		case NEWT_KEY_UNTAB:
+			/*
+			 * Exit the browser, let hists__browser_tree
+			 * go to the next or previous
+			 */
+			goto out_free_stack;
+		case 'a':
+			if (browser->selection->map == NULL &&
+			    browser->selection->map->dso->annotate_warned)
 				continue;
-			default:;
-			}
-			if (is_exit_key(key)) {
-				if (key == NEWT_KEY_ESCAPE &&
-				    !ui__dialog_yesno("Do you really want to exit?"))
-					continue;
-				break;
-			}
+			goto do_annotate;
+		case 'd':
+			goto zoom_dso;
+		case 't':
+			goto zoom_thread;
+		case NEWT_KEY_F1:
+		case 'h':
+		case '?':
+			ui__help_window("->        Zoom into DSO/Threads & Annotate current symbol\n"
+					"<-        Zoom out\n"
+					"a         Annotate current symbol\n"
+					"h/?/F1    Show this window\n"
+					"C         Collapse all callchains\n"
+					"E         Expand all callchains\n"
+					"d         Zoom into current DSO\n"
+					"t         Zoom into current Thread\n"
+					"q/CTRL+C  Exit browser");
+			continue;
+		case NEWT_KEY_ENTER:
+		case NEWT_KEY_RIGHT:
+			/* menu */
+			break;
+		case NEWT_KEY_LEFT: {
+			const void *top;
 
-			if (es.u.key == NEWT_KEY_LEFT) {
-				const void *top;
-
-				if (pstack__empty(fstack))
-					continue;
-				top = pstack__pop(fstack);
-				if (top == &dso_filter)
-					goto zoom_out_dso;
-				if (top == &thread_filter)
-					goto zoom_out_thread;
+			if (pstack__empty(fstack))
 				continue;
-			}
+			top = pstack__pop(fstack);
+			if (top == &dso_filter)
+				goto zoom_out_dso;
+			if (top == &thread_filter)
+				goto zoom_out_thread;
+			continue;
+		}
+		case NEWT_KEY_ESCAPE:
+			if (!ui__dialog_yesno("Do you really want to exit?"))
+				continue;
+			/* Fall thru */
+		default:
+			goto out_free_stack;
 		}
 
 		if (browser->selection->sym != NULL &&
@@ -885,8 +954,8 @@
 				pstack__push(fstack, &dso_filter);
 			}
 			hists__filter_by_dso(self, dso_filter);
-			hist_browser__title(msg, sizeof(msg), ev_name,
-					    dso_filter, thread_filter);
+			hists__browser_title(self, msg, sizeof(msg), ev_name,
+					     dso_filter, thread_filter);
 			hist_browser__reset(browser);
 		} else if (choice == zoom_thread) {
 zoom_thread:
@@ -903,8 +972,8 @@
 				pstack__push(fstack, &thread_filter);
 			}
 			hists__filter_by_thread(self, thread_filter);
-			hist_browser__title(msg, sizeof(msg), ev_name,
-					    dso_filter, thread_filter);
+			hists__browser_title(self, msg, sizeof(msg), ev_name,
+					     dso_filter, thread_filter);
 			hist_browser__reset(browser);
 		}
 	}
@@ -925,10 +994,6 @@
 		const char *ev_name = __event_name(hists->type, hists->config);
 
 		key = hists__browse(hists, help, ev_name);
-
-		if (is_exit_key(key))
-			break;
-
 		switch (key) {
 		case NEWT_KEY_TAB:
 			next = rb_next(nd);
@@ -940,7 +1005,7 @@
 				continue;
 			nd = rb_prev(nd);
 		default:
-			break;
+			return key;
 		}
 	}
 
diff --git a/tools/perf/util/ui/browsers/map.c b/tools/perf/util/ui/browsers/map.c
index 142b825..e35437d 100644
--- a/tools/perf/util/ui/browsers/map.c
+++ b/tools/perf/util/ui/browsers/map.c
@@ -1,6 +1,5 @@
 #include "../libslang.h"
 #include <elf.h>
-#include <newt.h>
 #include <sys/ttydefaults.h>
 #include <ctype.h>
 #include <string.h>
@@ -47,7 +46,6 @@
 struct map_browser {
 	struct ui_browser b;
 	struct map	  *map;
-	u16		  namelen;
 	u8		  addrlen;
 };
 
@@ -56,14 +54,16 @@
 	struct symbol *sym = rb_entry(nd, struct symbol, rb_node);
 	struct map_browser *mb = container_of(self, struct map_browser, b);
 	bool current_entry = ui_browser__is_current_entry(self, row);
-	int color = ui_browser__percent_color(0, current_entry);
+	int width;
 
-	SLsmg_set_color(color);
+	ui_browser__set_percent_color(self, 0, current_entry);
 	slsmg_printf("%*llx %*llx %c ",
 		     mb->addrlen, sym->start, mb->addrlen, sym->end,
 		     sym->binding == STB_GLOBAL ? 'g' :
 		     sym->binding == STB_LOCAL  ? 'l' : 'w');
-	slsmg_write_nstring(sym->name, mb->namelen);
+	width = self->width - ((mb->addrlen * 2) + 4);
+	if (width > 0)
+		slsmg_write_nstring(sym->name, width);
 }
 
 /* FIXME uber-kludgy, see comment on cmd_report... */
@@ -98,31 +98,29 @@
 	return 0;
 }
 
-static int map_browser__run(struct map_browser *self, struct newtExitStruct *es)
+static int map_browser__run(struct map_browser *self)
 {
+	int key;
+
 	if (ui_browser__show(&self->b, self->map->dso->long_name,
 			     "Press <- or ESC to exit, %s / to search",
 			     verbose ? "" : "restart with -v to use") < 0)
 		return -1;
 
-	newtFormAddHotKey(self->b.form, NEWT_KEY_LEFT);
-	newtFormAddHotKey(self->b.form, NEWT_KEY_ENTER);
 	if (verbose)
-		newtFormAddHotKey(self->b.form, '/');
+		ui_browser__add_exit_key(&self->b, '/');
 
 	while (1) {
-		ui_browser__run(&self->b, es);
+		key = ui_browser__run(&self->b);
 
-		if (es->reason != NEWT_EXIT_HOTKEY)
-			break;
-		if (verbose && es->u.key == '/')
+		if (verbose && key == '/')
 			map_browser__search(self);
 		else
 			break;
 	}
 
 	ui_browser__hide(&self->b);
-	return 0;
+	return key;
 }
 
 int map__browse(struct map *self)
@@ -136,7 +134,6 @@
 		},
 		.map = self,
 	};
-	struct newtExitStruct es;
 	struct rb_node *nd;
 	char tmp[BITS_PER_LONG / 4];
 	u64 maxaddr = 0;
@@ -144,8 +141,6 @@
 	for (nd = rb_first(mb.b.entries); nd; nd = rb_next(nd)) {
 		struct symbol *pos = rb_entry(nd, struct symbol, rb_node);
 
-		if (mb.namelen < pos->namelen)
-			mb.namelen = pos->namelen;
 		if (maxaddr < pos->end)
 			maxaddr = pos->end;
 		if (verbose) {
@@ -156,6 +151,5 @@
 	}
 
 	mb.addrlen = snprintf(tmp, sizeof(tmp), "%llx", maxaddr);
-	mb.b.width += mb.addrlen * 2 + 4 + mb.namelen;
-	return map_browser__run(&mb, &es);
+	return map_browser__run(&mb);
 }
diff --git a/tools/perf/util/ui/util.c b/tools/perf/util/ui/util.c
index 04600e2..9706d9d 100644
--- a/tools/perf/util/ui/util.c
+++ b/tools/perf/util/ui/util.c
@@ -11,8 +11,6 @@
 #include "helpline.h"
 #include "util.h"
 
-newtComponent newt_form__new(void);
-
 static void newt_form__set_exit_keys(newtComponent self)
 {
 	newtFormAddHotKey(self, NEWT_KEY_LEFT);
@@ -22,7 +20,7 @@
 	newtFormAddHotKey(self, CTRL('c'));
 }
 
-newtComponent newt_form__new(void)
+static newtComponent newt_form__new(void)
 {
 	newtComponent self = newtForm(NULL, NULL, 0);
 	if (self)
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index f380fed..7562707 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -266,19 +266,6 @@
 bool strlazymatch(const char *str, const char *pat);
 unsigned long convert_unit(unsigned long value, char *unit);
 
-#ifndef ESC
-#define ESC 27
-#endif
-
-static inline bool is_exit_key(int key)
-{
-	char up;
-	if (key == CTRL('c') || key == ESC)
-		return true;
-	up = toupper(key);
-	return up == 'Q';
-}
-
 #define _STR(x) #x
 #define STR(x) _STR(x)