2 files changed, 97 insertions, 77 deletions

diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c
index 08460a2e9f41..43b83c35cf54 100644
--- a/arch/powerpc/kernel/perf_event.c
+++ b/arch/powerpc/kernel/perf_event.c
@@ -35,6 +35,9 @@ struct cpu_hw_events {
 	u64 alternatives[MAX_HWEVENTS][MAX_EVENT_ALTERNATIVES];
 	unsigned long amasks[MAX_HWEVENTS][MAX_EVENT_ALTERNATIVES];
 	unsigned long avalues[MAX_HWEVENTS][MAX_EVENT_ALTERNATIVES];
+
+	unsigned int group_flag;
+	int n_txn_start;
 };
 DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events);
 
@@ -718,66 +721,6 @@ static int collect_events(struct perf_event *group, int max_count,
 	return n;
 }
 
-static void event_sched_in(struct perf_event *event)
-{
-	event->state = PERF_EVENT_STATE_ACTIVE;
-	event->oncpu = smp_processor_id();
-	event->tstamp_running += event->ctx->time - event->tstamp_stopped;
-	if (is_software_event(event))
-		event->pmu->enable(event);
-}
-
-/*
- * Called to enable a whole group of events.
- * Returns 1 if the group was enabled, or -EAGAIN if it could not be.
- * Assumes the caller has disabled interrupts and has
- * frozen the PMU with hw_perf_save_disable.
- */
-int hw_perf_group_sched_in(struct perf_event *group_leader,
-	       struct perf_cpu_context *cpuctx,
-	       struct perf_event_context *ctx)
-{
-	struct cpu_hw_events *cpuhw;
-	long i, n, n0;
-	struct perf_event *sub;
-
-	if (!ppmu)
-		return 0;
-	cpuhw = &__get_cpu_var(cpu_hw_events);
-	n0 = cpuhw->n_events;
-	n = collect_events(group_leader, ppmu->n_counter - n0,
-			   &cpuhw->event[n0], &cpuhw->events[n0],
-			   &cpuhw->flags[n0]);
-	if (n < 0)
-		return -EAGAIN;
-	if (check_excludes(cpuhw->event, cpuhw->flags, n0, n))
-		return -EAGAIN;
-	i = power_check_constraints(cpuhw, cpuhw->events, cpuhw->flags, n + n0);
-	if (i < 0)
-		return -EAGAIN;
-	cpuhw->n_events = n0 + n;
-	cpuhw->n_added += n;
-
-	/*
-	 * OK, this group can go on; update event states etc.,
-	 * and enable any software events
-	 */
-	for (i = n0; i < n0 + n; ++i)
-		cpuhw->event[i]->hw.config = cpuhw->events[i];
-	cpuctx->active_oncpu += n;
-	n = 1;
-	event_sched_in(group_leader);
-	list_for_each_entry(sub, &group_leader->sibling_list, group_entry) {
-		if (sub->state != PERF_EVENT_STATE_OFF) {
-			event_sched_in(sub);
-			++n;
-		}
-	}
-	ctx->nr_active += n;
-
-	return 1;
-}
-
 /*
  * Add a event to the PMU.
  * If all events are not already frozen, then we disable and
@@ -805,12 +748,22 @@ static int power_pmu_enable(struct perf_event *event)
 	cpuhw->event[n0] = event;
 	cpuhw->events[n0] = event->hw.config;
 	cpuhw->flags[n0] = event->hw.event_base;
+
+	/*
+	 * If group events scheduling transaction was started,
+	 * skip the schedulability test here, it will be peformed
+	 * at commit time(->commit_txn) as a whole
+	 */
+	if (cpuhw->group_flag & PERF_EVENT_TXN_STARTED)
+		goto nocheck;
+
 	if (check_excludes(cpuhw->event, cpuhw->flags, n0, 1))
 		goto out;
 	if (power_check_constraints(cpuhw, cpuhw->events, cpuhw->flags, n0 + 1))
 		goto out;
-
 	event->hw.config = cpuhw->events[n0];
+
+nocheck:
 	++cpuhw->n_events;
 	++cpuhw->n_added;
 
@@ -896,11 +849,65 @@ static void power_pmu_unthrottle(struct perf_event *event)
 	local_irq_restore(flags);
 }
 
+/*
+ * Start group events scheduling transaction
+ * 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)
+{
+	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
+
+	cpuhw->group_flag |= PERF_EVENT_TXN_STARTED;
+	cpuhw->n_txn_start = cpuhw->n_events;
+}
+
+/*
+ * Stop group events scheduling transaction
+ * Clear the flag and pmu::enable() will perform the
+ * schedulability test.
+ */
+void power_pmu_cancel_txn(const struct pmu *pmu)
+{
+	struct cpu_hw_events *cpuhw = &__get_cpu_var(cpu_hw_events);
+
+	cpuhw->group_flag &= ~PERF_EVENT_TXN_STARTED;
+}
+
+/*
+ * Commit group events scheduling transaction
+ * Perform the group schedulability test as a whole
+ * Return 0 if success
+ */
+int power_pmu_commit_txn(const struct pmu *pmu)
+{
+	struct cpu_hw_events *cpuhw;
+	long i, n;
+
+	if (!ppmu)
+		return -EAGAIN;
+	cpuhw = &__get_cpu_var(cpu_hw_events);
+	n = cpuhw->n_events;
+	if (check_excludes(cpuhw->event, cpuhw->flags, 0, n))
+		return -EAGAIN;
+	i = power_check_constraints(cpuhw, cpuhw->events, cpuhw->flags, n);
+	if (i < 0)
+		return -EAGAIN;
+
+	for (i = cpuhw->n_txn_start; i < n; ++i)
+		cpuhw->event[i]->hw.config = cpuhw->events[i];
+
+	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,
 };
 
 /*
diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index a9047463fd83..a4fa381db3c2 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -255,6 +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)
 {
@@ -308,8 +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--;
@@ -322,7 +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
@@ -333,6 +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
@@ -498,18 +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)
@@ -1868,6 +1872,12 @@ 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);
 	/*
 	 * There are two ways this annotation is useful:
@@ -1882,7 +1892,10 @@ int perf_event_release_kernel(struct perf_event *event)
 	 *     to trigger the AB-BA case.
 	 */
 	mutex_lock_nested(&ctx->mutex, SINGLE_DEPTH_NESTING);
-	perf_event_remove_from_context(event);
+	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);