perf_events: Fix sample_period transfer on inherit

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>
Cc: eranian@google.com
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <1264752266.4283.2121.camel@laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

1 files changed, 9 insertions, 2 deletions

diff --git a/kernel/perf_event.c b/kernel/perf_event.c
index 251fb9552492..53dc2a362111 100644
--- a/kernel/perf_event.c
+++ b/kernel/perf_event.c
@@ -5002,8 +5002,15 @@ inherit_event(struct perf_event *parent_event,
         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;