[PATCH 5/4] perf_events: Fix sample_period transfer on inherit

[Peter Zijlstra <peterz@infradead.org>]

Subject: perf_events: Fix sample_period transfer on inherit
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Thu Jan 28 12:34:07 CET 2010

One problem with frequency driven counters is that we cannot predict
the rate at which they trigger, therefore we have to start them at
period=1, this causes a ramp up effect. However, if we fail to
propagate the stable state on fork each new child will have to ramp up
again. This can lead to significant artifacts in sample data.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/perf_event.c |   11 +++++++++--
 1 file changed, 9 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
@@ -5011,8 +5011,15 @@ inherit_event(struct perf_event *parent_
 	else
 		child_event->state = PERF_EVENT_STATE_OFF;
 
-	if (parent_event->attr.freq)
-		child_event->hw.sample_period = parent_event->hw.sample_period;
+	if (parent_event->attr.freq) {
+		u64 sample_period = parent_event->hw.sample_period;
+		struct hw_perf_event *hwc = &child_event->hw;
+
+		hwc->sample_period = sample_period;
+		hwc->last_period   = sample_period;
+
+		atomic64_set(&hwc->period_left, sample_period);
+	}
 
 	child_event->overflow_handler = parent_event->overflow_handler;