[PATCH 10/15] perf: simplify __perf_event_sync_stat

[Peter Zijlstra <a.p.zijlstra@chello.nl>]

[-- Attachment #1: perf-time-3.patch --]
[-- Type: text/plain, Size: 1079 bytes --]

Removes constraints from __perf_event_read() by leaving it with a
single callsite; this callsite had ctx->lock held, the other one
does not.

Removes some superfluous code from __perf_event_sync_stat().

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/perf_event.c |    6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

Index: linux-2.6/kernel/perf_event.c
===================================================================
--- linux-2.6.orig/kernel/perf_event.c
+++ linux-2.6/kernel/perf_event.c
@@ -1061,8 +1061,6 @@ static int context_equiv(struct perf_eve
 		&& !ctx1->pin_count && !ctx2->pin_count;
 }
 
-static void __perf_event_read(void *event);
-
 static void __perf_event_sync_stat(struct perf_event *event,
 				     struct perf_event *next_event)
 {
@@ -1080,8 +1078,8 @@ static void __perf_event_sync_stat(struc
 	 */
 	switch (event->state) {
 	case PERF_EVENT_STATE_ACTIVE:
-		__perf_event_read(event);
-		break;
+		event->pmu->read(event);
+		/* fall-through */
 
 	case PERF_EVENT_STATE_INACTIVE:
 		update_event_times(event);

--