From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932773Ab2GKPMS (ORCPT ); Wed, 11 Jul 2012 11:12:18 -0400 Received: from casper.infradead.org ([85.118.1.10]:33375 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932139Ab2GKPMR convert rfc822-to-8bit (ORCPT ); Wed, 11 Jul 2012 11:12:17 -0400 Message-ID: <1342019524.3462.167.camel@twins> Subject: Re: [PATCH] trace: add ability to set a target task for events (v2) From: Peter Zijlstra To: Frederic Weisbecker Cc: Andrew Vagin , linux-kernel@vger.kernel.org, Ingo Molnar , Steven Rostedt , Paul Mackerras , Arnaldo Carvalho de Melo , Arun Sharma Date: Wed, 11 Jul 2012 17:12:04 +0200 In-Reply-To: <1342018508.3462.163.camel@twins> References: <1342016098-213063-1-git-send-email-avagin@openvz.org> <20120711143121.GA17991@somewhere> <1342017221.3462.159.camel@twins> <20120711143656.GB17991@somewhere> <1342017499.3462.160.camel@twins> <20120711144840.GC17991@somewhere> <1342018508.3462.163.camel@twins> Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT X-Mailer: Evolution 3.2.2- Mime-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2012-07-11 at 16:55 +0200, Peter Zijlstra wrote: > Right.. back when I did that the plan was to make PERF_SAMPLE_PERIOD fix > that, of course that never seemed to have happened. > > With PERF_SAMPLE_PERIOD you can simply write the 1b into the period of 1 > event and be done with it. It did! Andrew fixed it.. --- commit 5d81e5cfb37a174e8ddc0413e2e70cdf05807ace Author: Andrew Vagin Date: Mon Nov 7 15:54:12 2011 +0300 events: Don't divide events if it has field period This patch solves the following problem: Now some samples may be lost due to throttling. The number of samples is restricted by sysctl_perf_event_sample_rate/HZ. A trace event is divided on some samples according to event's period. I don't sure, that we should generate more than one sample on each trace event. I think the better way to use SAMPLE_PERIOD. E.g.: I want to trace when a process sleeps. I created a process, which sleeps for 1ms and for 4ms. perf got 100 events in both cases. swapper 0 [000] 1141.371830: sched_stat_sleep: comm=foo pid=1801 delay=1386750 [ns] swapper 0 [000] 1141.369444: sched_stat_sleep: comm=foo pid=1801 delay=4499585 [ns] In the first case a kernel want to send 4499585 events and in the second case it wants to send 1386750 events. perf-reports shows that process sleeps in both places equal time. It's bug. With this patch kernel generates one event on each "sleep" and the time slice is saved in the field "period". Perf knows how handle it. Signed-off-by: Andrew Vagin Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1320670457-2633428-3-git-send-email-avagin@openvz.org Signed-off-by: Ingo Molnar diff --git a/kernel/events/core.c b/kernel/events/core.c index eadac69..8d9dea5 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -4528,7 +4528,6 @@ static void perf_swevent_overflow(struct perf_event *event, u64 overflow, struct hw_perf_event *hwc = &event->hw; int throttle = 0; - data->period = event->hw.last_period; if (!overflow) overflow = perf_swevent_set_period(event); @@ -4562,6 +4561,12 @@ static void perf_swevent_event(struct perf_event *event, u64 nr, if (!is_sampling_event(event)) return; + if ((event->attr.sample_type & PERF_SAMPLE_PERIOD) && !event->attr.freq) { + data->period = nr; + return perf_swevent_overflow(event, 1, data, regs); + } else + data->period = event->hw.last_period; + if (nr == 1 && hwc->sample_period == 1 && !event->attr.freq) return perf_swevent_overflow(event, 1, data, regs);