Merge branch 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (311 commits)
  perf tools: Add mode to build without newt support
  perf symbols: symbol inconsistency message should be done only at verbose=1
  perf tui: Add explicit -lslang option
  perf options: Type check all the remaining OPT_ variants
  perf options: Type check OPT_BOOLEAN and fix the offenders
  perf options: Check v type in OPT_U?INTEGER
  perf options: Introduce OPT_UINTEGER
  perf tui: Add workaround for slang < 2.1.4
  perf record: Fix bug mismatch with -c option definition
  perf options: Introduce OPT_U64
  perf tui: Add help window to show key associations
  perf tui: Make <- exit menus too
  perf newt: Add single key shortcuts for zoom into DSO and threads
  perf newt: Exit browser unconditionally when CTRL+C, q or Q is pressed
  perf newt: Fix the 'A'/'a' shortcut for annotate
  perf newt: Make <- exit the ui_browser
  x86, perf: P4 PMU - fix counters management logic
  perf newt: Make <- zoom out filters
  perf report: Report number of events, not samples
  perf hist: Clarify events_stats fields usage
  ...

Fix up trivial conflicts in kernel/fork.c and tools/perf/builtin-record.c

16 files changed, 829 insertions, 902 deletions

diff --git a/kernel/fork.c b/kernel/fork.c
index 4c14942a0ee..4d57d9e3a6e 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1112,8 +1112,6 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	p->memcg_batch.memcg = NULL;
 #endif
 
-	p->bts = NULL;
-
 	/* Perform scheduler related setup. Assign this task to a CPU. */
 	sched_fork(p, clone_flags);
 
diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c
index 03808ed342a..7a56b22e060 100644
--- a/kernel/hw_breakpoint.c
+++ b/kernel/hw_breakpoint.c
@@ -40,23 +40,29 @@
 #include <linux/percpu.h>
 #include <linux/sched.h>
 #include <linux/init.h>
+#include <linux/slab.h>
 #include <linux/cpu.h>
 #include <linux/smp.h>
 
 #include <linux/hw_breakpoint.h>
 
+
 /*
  * Constraints data
  */
 
 /* Number of pinned cpu breakpoints in a cpu */
-static DEFINE_PER_CPU(unsigned int, nr_cpu_bp_pinned);
+static DEFINE_PER_CPU(unsigned int, nr_cpu_bp_pinned[TYPE_MAX]);
 
 /* Number of pinned task breakpoints in a cpu */
-static DEFINE_PER_CPU(unsigned int, nr_task_bp_pinned[HBP_NUM]);
+static DEFINE_PER_CPU(unsigned int *, nr_task_bp_pinned[TYPE_MAX]);
 
 /* Number of non-pinned cpu/task breakpoints in a cpu */
-static DEFINE_PER_CPU(unsigned int, nr_bp_flexible);
+static DEFINE_PER_CPU(unsigned int, nr_bp_flexible[TYPE_MAX]);
+
+static int nr_slots[TYPE_MAX];
+
+static int constraints_initialized;
 
 /* Gather the number of total pinned and un-pinned bp in a cpuset */
 struct bp_busy_slots {
@@ -67,16 +73,29 @@ struct bp_busy_slots {
 /* Serialize accesses to the above constraints */
 static DEFINE_MUTEX(nr_bp_mutex);
 
+__weak int hw_breakpoint_weight(struct perf_event *bp)
+{
+	return 1;
+}
+
+static inline enum bp_type_idx find_slot_idx(struct perf_event *bp)
+{
+	if (bp->attr.bp_type & HW_BREAKPOINT_RW)
+		return TYPE_DATA;
+
+	return TYPE_INST;
+}
+
 /*
  * Report the maximum number of pinned breakpoints a task
  * have in this cpu
  */
-static unsigned int max_task_bp_pinned(int cpu)
+static unsigned int max_task_bp_pinned(int cpu, enum bp_type_idx type)
 {
 	int i;
-	unsigned int *tsk_pinned = per_cpu(nr_task_bp_pinned, cpu);
+	unsigned int *tsk_pinned = per_cpu(nr_task_bp_pinned[type], cpu);
 
-	for (i = HBP_NUM -1; i >= 0; i--) {
+	for (i = nr_slots[type] - 1; i >= 0; i--) {
 		if (tsk_pinned[i] > 0)
 			return i + 1;
 	}
@@ -84,7 +103,7 @@ static unsigned int max_task_bp_pinned(int cpu)
 	return 0;
 }
 
-static int task_bp_pinned(struct task_struct *tsk)
+static int task_bp_pinned(struct task_struct *tsk, enum bp_type_idx type)
 {
 	struct perf_event_context *ctx = tsk->perf_event_ctxp;
 	struct list_head *list;
@@ -105,7 +124,8 @@ static int task_bp_pinned(struct task_struct *tsk)
 	 */
 	list_for_each_entry(bp, list, event_entry) {
 		if (bp->attr.type == PERF_TYPE_BREAKPOINT)
-			count++;
+			if (find_slot_idx(bp) == type)
+				count += hw_breakpoint_weight(bp);
 	}
 
 	raw_spin_unlock_irqrestore(&ctx->lock, flags);
@@ -118,18 +138,19 @@ static int task_bp_pinned(struct task_struct *tsk)
  * a given cpu (cpu > -1) or in all of them (cpu = -1).
  */
 static void
-fetch_bp_busy_slots(struct bp_busy_slots *slots, struct perf_event *bp)
+fetch_bp_busy_slots(struct bp_busy_slots *slots, struct perf_event *bp,
+		    enum bp_type_idx type)
 {
 	int cpu = bp->cpu;
 	struct task_struct *tsk = bp->ctx->task;
 
 	if (cpu >= 0) {
-		slots->pinned = per_cpu(nr_cpu_bp_pinned, cpu);
+		slots->pinned = per_cpu(nr_cpu_bp_pinned[type], cpu);
 		if (!tsk)
-			slots->pinned += max_task_bp_pinned(cpu);
+			slots->pinned += max_task_bp_pinned(cpu, type);
 		else
-			slots->pinned += task_bp_pinned(tsk);
-		slots->flexible = per_cpu(nr_bp_flexible, cpu);
+			slots->pinned += task_bp_pinned(tsk, type);
+		slots->flexible = per_cpu(nr_bp_flexible[type], cpu);
 
 		return;
 	}
@@ -137,16 +158,16 @@ fetch_bp_busy_slots(struct bp_busy_slots *slots, struct perf_event *bp)
 	for_each_online_cpu(cpu) {
 		unsigned int nr;
 
-		nr = per_cpu(nr_cpu_bp_pinned, cpu);
+		nr = per_cpu(nr_cpu_bp_pinned[type], cpu);
 		if (!tsk)
-			nr += max_task_bp_pinned(cpu);
+			nr += max_task_bp_pinned(cpu, type);
 		else
-			nr += task_bp_pinned(tsk);
+			nr += task_bp_pinned(tsk, type);
 
 		if (nr > slots->pinned)
 			slots->pinned = nr;
 
-		nr = per_cpu(nr_bp_flexible, cpu);
+		nr = per_cpu(nr_bp_flexible[type], cpu);
 
 		if (nr > slots->flexible)
 			slots->flexible = nr;
@@ -154,31 +175,49 @@ fetch_bp_busy_slots(struct bp_busy_slots *slots, struct perf_event *bp)
 }
 
 /*
+ * For now, continue to consider flexible as pinned, until we can
+ * ensure no flexible event can ever be scheduled before a pinned event
+ * in a same cpu.
+ */
+static void
+fetch_this_slot(struct bp_busy_slots *slots, int weight)
+{
+	slots->pinned += weight;
+}
+
+/*
  * Add a pinned breakpoint for the given task in our constraint table
  */
-static void toggle_bp_task_slot(struct task_struct *tsk, int cpu, bool enable)
+static void toggle_bp_task_slot(struct task_struct *tsk, int cpu, bool enable,
+				enum bp_type_idx type, int weight)
 {
 	unsigned int *tsk_pinned;
-	int count = 0;
+	int old_count = 0;
+	int old_idx = 0;
+	int idx = 0;
 
-	count = task_bp_pinned(tsk);
+	old_count = task_bp_pinned(tsk, type);
+	old_idx = old_count - 1;
+	idx = old_idx + weight;
 
-	tsk_pinned = per_cpu(nr_task_bp_pinned, cpu);
+	tsk_pinned = per_cpu(nr_task_bp_pinned[type], cpu);
 	if (enable) {
-		tsk_pinned[count]++;
-		if (count > 0)
-			tsk_pinned[count-1]--;
+		tsk_pinned[idx]++;
+		if (old_count > 0)
+			tsk_pinned[old_idx]--;
 	} else {
-		tsk_pinned[count]--;
-		if (count > 0)
-			tsk_pinned[count-1]++;
+		tsk_pinned[idx]--;
+		if (old_count > 0)
+			tsk_pinned[old_idx]++;
 	}
 }
 
 /*
  * Add/remove the given breakpoint in our constraint table
  */
-static void toggle_bp_slot(struct perf_event *bp, bool enable)
+static void
+toggle_bp_slot(struct perf_event *bp, bool enable, enum bp_type_idx type,
+	       int weight)
 {
 	int cpu = bp->cpu;
 	struct task_struct *tsk = bp->ctx->task;
@@ -186,20 +225,20 @@ static void toggle_bp_slot(struct perf_event *bp, bool enable)
 	/* Pinned counter task profiling */
 	if (tsk) {
 		if (cpu >= 0) {
-			toggle_bp_task_slot(tsk, cpu, enable);
+			toggle_bp_task_slot(tsk, cpu, enable, type, weight);
 			return;
 		}
 
 		for_each_online_cpu(cpu)
-			toggle_bp_task_slot(tsk, cpu, enable);
+			toggle_bp_task_slot(tsk, cpu, enable, type, weight);
 		return;
 	}
 
 	/* Pinned counter cpu profiling */
 	if (enable)
-		per_cpu(nr_cpu_bp_pinned, bp->cpu)++;
+		per_cpu(nr_cpu_bp_pinned[type], bp->cpu) += weight;
 	else
-		per_cpu(nr_cpu_bp_pinned, bp->cpu)--;
+		per_cpu(nr_cpu_bp_pinned[type], bp->cpu) -= weight;
 }
 
 /*
@@ -246,14 +285,29 @@ static void toggle_bp_slot(struct perf_event *bp, bool enable)
 static int __reserve_bp_slot(struct perf_event *bp)
 {
 	struct bp_busy_slots slots = {0};
+	enum bp_type_idx type;
+	int weight;
 
-	fetch_bp_busy_slots(&slots, bp);
+	/* We couldn't initialize breakpoint constraints on boot */
+	if (!constraints_initialized)
+		return -ENOMEM;
+
+	/* Basic checks */
+	if (bp->attr.bp_type == HW_BREAKPOINT_EMPTY ||
+	    bp->attr.bp_type == HW_BREAKPOINT_INVALID)
+		return -EINVAL;
+
+	type = find_slot_idx(bp);
+	weight = hw_breakpoint_weight(bp);
+
+	fetch_bp_busy_slots(&slots, bp, type);
+	fetch_this_slot(&slots, weight);
 
 	/* Flexible counters need to keep at least one slot */
-	if (slots.pinned + (!!slots.flexible) == HBP_NUM)
+	if (slots.pinned + (!!slots.flexible) > nr_slots[type])
 		return -ENOSPC;
 
-	toggle_bp_slot(bp, true);
+	toggle_bp_slot(bp, true, type, weight);
 
 	return 0;
 }
@@ -273,7 +327,12 @@ int reserve_bp_slot(struct perf_event *bp)
 
 static void __release_bp_slot(struct perf_event *bp)
 {
-	toggle_bp_slot(bp, false);
+	enum bp_type_idx type;
+	int weight;
+
+	type = find_slot_idx(bp);
+	weight = hw_breakpoint_weight(bp);
+	toggle_bp_slot(bp, false, type, weight);
 }
 
 void release_bp_slot(struct perf_event *bp)
@@ -308,6 +367,28 @@ int dbg_release_bp_slot(struct perf_event *bp)
 	return 0;
 }
 
+static int validate_hw_breakpoint(struct perf_event *bp)
+{
+	int ret;
+
+	ret = arch_validate_hwbkpt_settings(bp);
+	if (ret)
+		return ret;
+
+	if (arch_check_bp_in_kernelspace(bp)) {
+		if (bp->attr.exclude_kernel)
+			return -EINVAL;
+		/*
+		 * Don't let unprivileged users set a breakpoint in the trap
+		 * path to avoid trap recursion attacks.
+		 */
+		if (!capable(CAP_SYS_ADMIN))
+			return -EPERM;
+	}
+
+	return 0;
+}
+
 int register_perf_hw_breakpoint(struct perf_event *bp)
 {
 	int ret;
@@ -316,17 +397,7 @@ int register_perf_hw_breakpoint(struct perf_event *bp)
 	if (ret)
 		return ret;
 
-	/*
-	 * Ptrace breakpoints can be temporary perf events only
-	 * meant to reserve a slot. In this case, it is created disabled and
-	 * we don't want to check the params right now (as we put a null addr)
-	 * But perf tools create events as disabled and we want to check
-	 * the params for them.
-	 * This is a quick hack that will be removed soon, once we remove
-	 * the tmp breakpoints from ptrace
-	 */
-	if (!bp->attr.disabled || !bp->overflow_handler)
-		ret = arch_validate_hwbkpt_settings(bp, bp->ctx->task);
+	ret = validate_hw_breakpoint(bp);
 
 	/* if arch_validate_hwbkpt_settings() fails then release bp slot */
 	if (ret)
@@ -373,7 +444,7 @@ int modify_user_hw_breakpoint(struct perf_event *bp, struct perf_event_attr *att
 	if (attr->disabled)
 		goto end;
 
-	err = arch_validate_hwbkpt_settings(bp, bp->ctx->task);
+	err = validate_hw_breakpoint(bp);
 	if (!err)
 		perf_event_enable(bp);
 
@@ -480,7 +551,36 @@ static struct notifier_block hw_breakpoint_exceptions_nb = {
 
 static int __init init_hw_breakpoint(void)
 {
+	unsigned int **task_bp_pinned;
+	int cpu, err_cpu;
+	int i;
+
+	for (i = 0; i < TYPE_MAX; i++)
+		nr_slots[i] = hw_breakpoint_slots(i);
+
+	for_each_possible_cpu(cpu) {
+		for (i = 0; i < TYPE_MAX; i++) {
+			task_bp_pinned = &per_cpu(nr_task_bp_pinned[i], cpu);
+			*task_bp_pinned = kzalloc(sizeof(int) * nr_slots[i],
+						  GFP_KERNEL);
+			if (!*task_bp_pinned)
+				goto err_alloc;
+		}
+	}
+
+	constraints_initialized = 1;
+
 	return register_die_notifier(&hw_breakpoint_exceptions_nb);
+
+ err_alloc:
+	for_each_possible_cpu(err_cpu) {
+		if (err_cpu == cpu)
+			break;
+		for (i = 0; i < TYPE_MAX; i++)
+			kfree(per_cpu(nr_task_bp_pinned[i], cpu));
+	}
+
+	return -ENOMEM;
 }
 core_initcall(init_hw_breakpoint);
 
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 0ed46f3e51e..282035f3ae9 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -1588,6 +1588,72 @@ static void __kprobes kill_kprobe(struct kprobe *p)
 	arch_remove_kprobe(p);
 }
 
+/* Disable one kprobe */
+int __kprobes disable_kprobe(struct kprobe *kp)
+{
+	int ret = 0;
+	struct kprobe *p;
+
+	mutex_lock(&kprobe_mutex);
+
+	/* Check whether specified probe is valid. */
+	p = __get_valid_kprobe(kp);
+	if (unlikely(p == NULL)) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	/* If the probe is already disabled (or gone), just return */
+	if (kprobe_disabled(kp))
+		goto out;
+
+	kp->flags |= KPROBE_FLAG_DISABLED;
+	if (p != kp)
+		/* When kp != p, p is always enabled. */
+		try_to_disable_aggr_kprobe(p);
+
+	if (!kprobes_all_disarmed && kprobe_disabled(p))
+		disarm_kprobe(p);
+out:
+	mutex_unlock(&kprobe_mutex);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(disable_kprobe);
+
+/* Enable one kprobe */
+int __kprobes enable_kprobe(struct kprobe *kp)
+{
+	int ret = 0;
+	struct kprobe *p;
+
+	mutex_lock(&kprobe_mutex);
+
+	/* Check whether specified probe is valid. */
+	p = __get_valid_kprobe(kp);
+	if (unlikely(p == NULL)) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (kprobe_gone(kp)) {
+		/* This kprobe has gone, we couldn't enable it. */
+		ret = -EINVAL;
+		goto out;
+	}
+
+	if (p != kp)
+		kp->flags &= ~KPROBE_FLAG_DISABLED;
+
+	if (!kprobes_all_disarmed && kprobe_disabled(p)) {
+		p->flags &= ~KPROBE_FLAG_DISABLED;
+		arm_kprobe(p);
+	}
+out:
+	mutex_unlock(&kprobe_mutex);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(enable_kprobe);
+
 void __kprobes dump_kprobe(struct kprobe *kp)
 {
 	printk(KERN_WARNING "Dumping kprobe:\n");
@@ -1805,72 +1871,6 @@ static const struct file_operations debugfs_kprobes_operations = {
 	.release        = seq_release,
 };
 
-/* Disable one kprobe */
-int __kprobes disable_kprobe(struct kprobe *kp)
-{
-	int ret = 0;
-	struct kprobe *p;
-
-	mutex_lock(&kprobe_mutex);
-
-	/* Check whether specified probe is valid. */
-	p = __get_valid_kprobe(kp);
-	if (unlikely(p == NULL)) {
-		ret = -EINVAL;
-		goto out;
-	}
-
-	/* If the probe is already disabled (or gone), just return */
-	if (kprobe_disabled(kp))
-		goto out;
-
-	kp->flags |= KPROBE_FLAG_DISABLED;
-	if (p != kp)
-		/* When kp != p, p is always enabled. */
-		try_to_disable_aggr_kprobe(p);
-
-	if (!kprobes_all_disarmed && kprobe_disabled(p))
-		disarm_kprobe(p);
-out:
-	mutex_unlock(&kprobe_mutex);
-	return ret;
-}
-EXPORT_SYMBOL_GPL(disable_kprobe);
-
-/* Enable one kprobe */
-int __kprobes enable_kprobe(struct kprobe *kp)
-{
-	int ret = 0;
-	struct kprobe *p;
-
-	mutex_lock(&kprobe_mutex);
-
-	/* Check whether specified probe is valid. */
-	p = __get_valid_kprobe(kp);
-	if (unlikely(p == NULL)) {
-		ret = -EINVAL;
-		goto out;
-	}
-
-	if (kprobe_gone(kp)) {
-		/* This kprobe has gone, we couldn't enable it. */
-		ret = -EINVAL;
-		goto out;
-	}
-
-	if (p != kp)
-		kp->flags &= ~KPROBE_FLAG_DISABLED;
-
-	if (!kprobes_all_disarmed && kprobe_disabled(p)) {
-		p->flags &= ~KPROBE_FLAG_DISABLED;
-		arm_kprobe(p);
-	}
-out:
-	mutex_unlock(&kprobe_mutex);
-	return ret;
-}
-EXPORT_SYMBOL_GPL(enable_kprobe);
-
 static void __kprobes arm_all_kprobes(void)
 {
 	struct hlist_head *head;
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index 4349e979341..ec21304856d 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -3232,7 +3232,7 @@ void lock_release(struct lockdep_map *lock, int nested,
 	raw_local_irq_save(flags);
 	check_flags(flags);
 	current->lockdep_recursion = 1;
-	trace_lock_release(lock, nested, ip);
+	trace_lock_release(lock, ip);
 	__lock_release(lock, nested, ip);
 	current->lockdep_recursion = 0;
 	raw_local_irq_restore(flags);
@@ -3385,7 +3385,7 @@ found_it:
 		hlock->holdtime_stamp = now;
 	}
 
-	trace_lock_acquired(lock, ip, waittime);
+	trace_lock_acquired(lock, ip);
 
 	stats = get_lock_stats(hlock_class(hlock));
 	if (waittime) {
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 3d1552d3c12..a4fa381db3c 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -16,6 +16,7 @@
 #include <linux/file.h>
 #include <linux/poll.h>
 #include <linux/slab.h>
+#include <linux/hash.h>
 #include <linux/sysfs.h>
 #include <linux/dcache.h>
 #include <linux/percpu.h>
@@ -82,14 +83,6 @@ extern __weak const struct pmu *hw_perf_event_init(struct perf_event *event)
 void __weak hw_perf_disable(void)		{ barrier(); }
 void __weak hw_perf_enable(void)		{ barrier(); }
 
-int __weak
-hw_perf_group_sched_in(struct perf_event *group_leader,
-	       struct perf_cpu_context *cpuctx,
-	       struct perf_event_context *ctx)
-{
-	return 0;
-}
-
 void __weak perf_event_print_debug(void)	{ }
 
 static DEFINE_PER_CPU(int, perf_disable_count);
@@ -262,6 +255,18 @@ static void update_event_times(struct perf_event *event)
 	event->total_time_running = run_end - event->tstamp_running;
 }
 
+/*
+ * Update total_time_enabled and total_time_running for all events in a group.
+ */
+static void update_group_times(struct perf_event *leader)
+{
+	struct perf_event *event;
+
+	update_event_times(leader);
+	list_for_each_entry(event, &leader->sibling_list, group_entry)
+		update_event_times(event);
+}
+
 static struct list_head *
 ctx_group_list(struct perf_event *event, struct perf_event_context *ctx)
 {
@@ -315,8 +320,6 @@ list_add_event(struct perf_event *event, struct perf_event_context *ctx)
 static void
 list_del_event(struct perf_event *event, struct perf_event_context *ctx)
 {
-	struct perf_event *sibling, *tmp;
-
 	if (list_empty(&event->group_entry))
 		return;
 	ctx->nr_events--;
@@ -329,7 +332,7 @@ list_del_event(struct perf_event *event, struct perf_event_context *ctx)
 	if (event->group_leader != event)
 		event->group_leader->nr_siblings--;
 
-	update_event_times(event);
+	update_group_times(event);
 
 	/*
 	 * If event was in error state, then keep it
@@ -340,6 +343,12 @@ list_del_event(struct perf_event *event, struct perf_event_context *ctx)
 	 */
 	if (event->state > PERF_EVENT_STATE_OFF)
 		event->state = PERF_EVENT_STATE_OFF;
+}
+
+static void
+perf_destroy_group(struct perf_event *event, struct perf_event_context *ctx)
+{
+	struct perf_event *sibling, *tmp;
 
 	/*
 	 * If this was a group event with sibling events then
@@ -505,18 +514,6 @@ retry:
 }
 
 /*
- * Update total_time_enabled and total_time_running for all events in a group.
- */
-static void update_group_times(struct perf_event *leader)
-{
-	struct perf_event *event;
-
-	update_event_times(leader);
-	list_for_each_entry(event, &leader->sibling_list, group_entry)
-		update_event_times(event);
-}
-
-/*
  * Cross CPU call to disable a performance event
  */
 static void __perf_event_disable(void *info)
@@ -640,15 +637,20 @@ group_sched_in(struct perf_event *group_event,
 	       struct perf_cpu_context *cpuctx,
 	       struct perf_event_context *ctx)
 {
-	struct perf_event *event, *partial_group;
+	struct perf_event *event, *partial_group = NULL;
+	const struct pmu *pmu = group_event->pmu;
+	bool txn = false;
 	int ret;
 
 	if (group_event->state == PERF_EVENT_STATE_OFF)
 		return 0;
 
-	ret = hw_perf_group_sched_in(group_event, cpuctx, ctx);
-	if (ret)
-		return ret < 0 ? ret : 0;
+	/* Check if group transaction availabe */
+	if (pmu->start_txn)
+		txn = true;
+
+	if (txn)
+		pmu->start_txn(pmu);
 
 	if (event_sched_in(group_event, cpuctx, ctx))
 		return -EAGAIN;
@@ -663,9 +665,19 @@ group_sched_in(struct perf_event *group_event,
 		}
 	}
 
-	return 0;
+	if (!txn)
+		return 0;
+
+	ret = pmu->commit_txn(pmu);
+	if (!ret) {
+		pmu->cancel_txn(pmu);
+		return 0;
+	}
 
 group_error:
+	if (txn)
+		pmu->cancel_txn(pmu);
+
 	/*
 	 * Groups can be scheduled in as one unit only, so undo any
 	 * partial group before returning:
@@ -1367,6 +1379,8 @@ void perf_event_task_sched_in(struct task_struct *task)
 	if (cpuctx->task_ctx == ctx)
 		return;
 
+	perf_disable();
+
 	/*
 	 * We want to keep the following priority order:
 	 * cpu pinned (that don't need to move), task pinned,
@@ -1379,6 +1393,8 @@ void perf_event_task_sched_in(struct task_struct *task)
 	ctx_sched_in(ctx, cpuctx, EVENT_FLEXIBLE);
 
 	cpuctx->task_ctx = ctx;
+
+	perf_enable();
 }
 
 #define MAX_INTERRUPTS (~0ULL)
@@ -1856,9 +1872,30 @@ int perf_event_release_kernel(struct perf_event *event)
 {
 	struct perf_event_context *ctx = event->ctx;
 
+	/*
+	 * Remove from the PMU, can't get re-enabled since we got
+	 * here because the last ref went.
+	 */
+	perf_event_disable(event);
+
 	WARN_ON_ONCE(ctx->parent_ctx);
-	mutex_lock(&ctx->mutex);
-	perf_event_remove_from_context(event);
+	/*
+	 * There are two ways this annotation is useful:
+	 *
+	 *  1) there is a lock recursion from perf_event_exit_task
+	 *     see the comment there.
+	 *
+	 *  2) there is a lock-inversion with mmap_sem through
+	 *     perf_event_read_group(), which takes faults while
+	 *     holding ctx->mutex, however this is called after
+	 *     the last filedesc died, so there is no possibility
+	 *     to trigger the AB-BA case.
+	 */
+	mutex_lock_nested(&ctx->mutex, SINGLE_DEPTH_NESTING);
+	raw_spin_lock_irq(&ctx->lock);
+	list_del_event(event, ctx);
+	perf_destroy_group(event, ctx);
+	raw_spin_unlock_irq(&ctx->lock);
 	mutex_unlock(&ctx->mutex);
 
 	mutex_lock(&event->owner->perf_event_mutex);
@@ -2642,6 +2679,7 @@ static int perf_fasync(int fd, struct file *filp, int on)
 }
 
 static const struct file_operations perf_fops = {
+	.llseek			= no_llseek,
 	.release		= perf_release,
 	.read			= perf_read,
 	.poll			= perf_poll,
@@ -2792,6 +2830,27 @@ void perf_arch_fetch_caller_regs(struct pt_regs *regs, unsigned long ip, int ski
 
 
 /*
+ * 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.
+ */
+struct perf_guest_info_callbacks *perf_guest_cbs;
+
+int perf_register_guest_info_callbacks(struct perf_guest_info_callbacks *cbs)
+{
+	perf_guest_cbs = cbs;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(perf_register_guest_info_callbacks);
+
+int perf_unregister_guest_info_callbacks(struct perf_guest_info_callbacks *cbs)
+{
+	perf_guest_cbs = NULL;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(perf_unregister_guest_info_callbacks);
+
+/*
  * Output
  */
 static bool perf_output_space(struct perf_mmap_data *data, unsigned long tail,
@@ -3743,7 +3802,7 @@ void __perf_event_mmap(struct vm_area_struct *vma)
 		.event_id  = {
 			.header = {
 				.type = PERF_RECORD_MMAP,
-				.misc = 0,
+				.misc = PERF_RECORD_MISC_USER,
 				/* .size */
 			},
 			/* .pid */
@@ -3961,36 +4020,6 @@ static void perf_swevent_add(struct perf_event *event, u64 nr,
 	perf_swevent_overflow(event, 0, nmi, data, regs);
 }
 
-static int perf_swevent_is_counting(struct perf_event *event)
-{
-	/*
-	 * The event is active, we're good!
-	 */
-	if (event->state == PERF_EVENT_STATE_ACTIVE)
-		return 1;
-
-	/*
-	 * The event is off/error, not counting.
-	 */
-	if (event->state != PERF_EVENT_STATE_INACTIVE)
-		return 0;
-
-	/*
-	 * The event is inactive, if the context is active
-	 * we're part of a group that didn't make it on the 'pmu',
-	 * not counting.
-	 */
-	if (event->ctx->is_active)
-		return 0;
-
-	/*
-	 * We're inactive and the context is too, this means the
-	 * task is scheduled out, we're counting events that happen
-	 * to us, like migration events.
-	 */
-	return 1;
-}
-
 static int perf_tp_event_match(struct perf_event *event,
 				struct perf_sample_data *data);
 
@@ -4014,12 +4043,6 @@ static int perf_swevent_match(struct perf_event *event,
 				struct perf_sample_data *data,
 				struct pt_regs *regs)
 {
-	if (event->cpu != -1 && event->cpu != smp_processor_id())
-		return 0;
-
-	if (!perf_swevent_is_counting(event))
-		return 0;
-
 	if (event->attr.type != type)
 		return 0;
 
@@ -4036,18 +4059,53 @@ static int perf_swevent_match(struct perf_event *event,
 	return 1;
 }
 
-static void perf_swevent_ctx_event(struct perf_event_context *ctx,
-				     enum perf_type_id type,
-				     u32 event_id, u64 nr, int nmi,
-				     struct perf_sample_data *data,
-				     struct pt_regs *regs)
+static inline u64 swevent_hash(u64 type, u32 event_id)
 {
+	u64 val = event_id | (type << 32);
+
+	return hash_64(val, SWEVENT_HLIST_BITS);
+}
+
+static struct hlist_head *
+find_swevent_head(struct perf_cpu_context *ctx, u64 type, u32 event_id)
+{
+	u64 hash;
+	struct swevent_hlist *hlist;
+
+	hash = swevent_hash(type, event_id);
+
+	hlist = rcu_dereference(ctx->swevent_hlist);
+	if (!hlist)
+		return NULL;
+
+	return &hlist->heads[hash];
+}
+
+static void do_perf_sw_event(enum perf_type_id type, u32 event_id,
+				    u64 nr, int nmi,
+				    struct perf_sample_data *data,
+				    struct pt_regs *regs)
+{
+	struct perf_cpu_context *cpuctx;
 	struct perf_event *event;
+	struct hlist_node *node;
+	struct hlist_head *head;
 
-	list_for_each_entry_rcu(event, &ctx->event_list, event_entry) {
+	cpuctx = &__get_cpu_var(perf_cpu_context);
+
+	rcu_read_lock();
+
+	head = find_swevent_head(cpuctx, 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);
 	}
+end:
+	rcu_read_unlock();
 }
 
 int perf_swevent_get_recursion_context(void)
@@ -4085,27 +4143,6 @@ void perf_swevent_put_recursion_context(int rctx)
 }
 EXPORT_SYMBOL_GPL(perf_swevent_put_recursion_context);
 
-static void do_perf_sw_event(enum perf_type_id type, u32 event_id,
-				    u64 nr, int nmi,
-				    struct perf_sample_data *data,
-				    struct pt_regs *regs)
-{
-	struct perf_cpu_context *cpuctx;
-	struct perf_event_context *ctx;
-
-	cpuctx = &__get_cpu_var(perf_cpu_context);
-	rcu_read_lock();
-	perf_swevent_ctx_event(&cpuctx->ctx, type, event_id,
-				 nr, nmi, data, regs);
-	/*
-	 * doesn't really matter which of the child contexts the
-	 * events ends up in.
-	 */
-	ctx = rcu_dereference(current->perf_event_ctxp);
-	if (ctx)
-		perf_swevent_ctx_event(ctx, type, event_id, nr, nmi, data, regs);
-	rcu_read_unlock();
-}
 
 void __perf_sw_event(u32 event_id, u64 nr, int nmi,
 			    struct pt_regs *regs, u64 addr)
@@ -4131,16 +4168,28 @@ static void perf_swevent_read(struct perf_event *event)
 static int perf_swevent_enable(struct perf_event *event)
 {
 	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->attr.type, event->attr.config);
+	if (WARN_ON_ONCE(!head))
+		return -EINVAL;
+
+	hlist_add_head_rcu(&event->hlist_entry, head);
+
 	return 0;
 }
 
 static void perf_swevent_disable(struct perf_event *event)
 {
+	hlist_del_rcu(&event->hlist_entry);
 }
 
 static const struct pmu perf_ops_generic = {
@@ -4168,15 +4217,8 @@ static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer)
 	perf_sample_data_init(&data, 0);
 	data.period = event->hw.last_period;
 	regs = get_irq_regs();
-	/*
-	 * In case we exclude kernel IPs or are somehow not in interrupt
-	 * context, provide the next best thing, the user IP.
-	 */
-	if ((event->attr.exclude_kernel || !regs) &&
-			!event->attr.exclude_user)
-		regs = task_pt_regs(current);
 
-	if (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;
@@ -4324,6 +4366,105 @@ static const struct pmu perf_ops_task_clock = {
 	.read		= task_clock_perf_event_read,
 };
 
+static void swevent_hlist_release_rcu(struct rcu_head *rcu_head)
+{
+	struct swevent_hlist *hlist;
+
+	hlist = container_of(rcu_head, struct swevent_hlist, rcu_head);
+	kfree(hlist);
+}
+
+static void swevent_hlist_release(struct perf_cpu_context *cpuctx)
+{
+	struct swevent_hlist *hlist;
+
+	if (!cpuctx->swevent_hlist)
+		return;
+
+	hlist = cpuctx->swevent_hlist;
+	rcu_assign_pointer(cpuctx->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);
+
+	mutex_lock(&cpuctx->hlist_mutex);
+
+	if (!--cpuctx->hlist_refcount)
+		swevent_hlist_release(cpuctx);
+
+	mutex_unlock(&cpuctx->hlist_mutex);
+}
+
+static void swevent_hlist_put(struct perf_event *event)
+{
+	int cpu;
+
+	if (event->cpu != -1) {
+		swevent_hlist_put_cpu(event, event->cpu);
+		return;
+	}
+
+	for_each_possible_cpu(cpu)
+		swevent_hlist_put_cpu(event, cpu);
+}
+
+static int swevent_hlist_get_cpu(struct perf_event *event, int cpu)
+{
+	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
+	int err = 0;
+
+	mutex_lock(&cpuctx->hlist_mutex);
+
+	if (!cpuctx->swevent_hlist && cpu_online(cpu)) {