| Date | Fri, 20 Nov 2009 22:19:50 +0100 | From | Peter Zijlstra <> | Subject | [PATCH 08/15] perf: optimize perf_event_task_sched_out |
| |
Remove an update_context_time() call from the perf_event_task_sched_out() path and into the branch its needed.
The call was both superfluous, because __perf_event_sched_out() already does it, and wrong, because it was done without holding ctx->lock.
Place it in perf_event_sync_stat(), which is the only place it is needed and which does already hold ctx->lock.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> --- kernel/perf_event.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
Index: linux-2.6/kernel/perf_event.c =================================================================== --- linux-2.6.orig/kernel/perf_event.c +++ linux-2.6/kernel/perf_event.c @@ -1120,6 +1120,8 @@ static void perf_event_sync_stat(struct if (!ctx->nr_stat) return; + update_context_time(ctx); + event = list_first_entry(&ctx->event_list, struct perf_event, event_entry); @@ -1163,8 +1165,6 @@ void perf_event_task_sched_out(struct ta if (likely(!ctx || !cpuctx->task_ctx)) return; - update_context_time(ctx); - rcu_read_lock(); parent = rcu_dereference(ctx->parent_ctx); next_ctx = next->perf_event_ctxp; --
|