From 2e2af50b1fab3c40636839a7f439c167ae559533 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 23 Nov 2009 11:37:25 +0100
Subject: perf_events: Disable events when we detach them

If we leave the event in STATE_INACTIVE, any read of the event
after the detach will increase the running count but not the
enabled count and cause funny scaling artefacts.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20091123103819.689055515@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 1f14481..fb851ec 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -294,6 +294,8 @@ list_del_event(struct perf_event *event, struct perf_event_context *ctx)
 	if (event->group_leader != event)
 		event->group_leader->nr_siblings--;
 
+	event->state = PERF_EVENT_STATE_OFF;
+
 	/*
 	 * If this was a group event with sibling events then
 	 * upgrade the siblings to singleton events by adding them
-- 
cgit v0.10.2