[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <CABPqkBQa8ndydqkMDvc3kzu84_96FbK77TUkrfO3eSPXmv3dhQ@mail.gmail.com>
Date: Wed, 13 Jun 2012 14:51:03 +0200
From: Stephane Eranian <eranian@...gle.com>
To: Peter Zijlstra <peterz@...radead.org>
Cc: LKML <linux-kernel@...r.kernel.org>, mingo@...e.hu
Subject: [BUG] perf: perf_output_sample() merge issue?
Peter,
I was looking at perf_output_sample() today, when I ran
into the following weird statement coming from:
commit a7ac67ea021b4603095d2aa458bc41641238f22c
Author: Peter Zijlstra <a.p.zijlstra@...llo.nl>
Date: Mon Jun 27 16:47:16 2011 +0200
perf: Remove the perf_output_begin(.sample) argument
Since only samples call perf_output_sample() its much saner (and more
I think the intent of this commit was to add the watermark check
at the very end of the perf_output_sample() and NOT between processing
of RAW and BRANCH_STACK. Something may have gone wrong
in the merge.
Can you confirm this?
perf_output_sample()
{
if (sample_type & PERF_SAMPLE_RAW) {
}
if (!event->attr.watermark) {
int wakeup_events = event->attr.wakeup_events;
if (wakeup_events) {
struct ring_buffer *rb = handle->rb;
int events = local_inc_return(&rb->events);
if (events >= wakeup_events) {
local_sub(wakeup_events, &rb->events);
local_inc(&rb->wakeup);
}
}
}
if (sample_type & PERF_SAMPLE_BRANCH_STACK) {
}
}
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists