Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756153Ab3JHKgP (ORCPT ); Tue, 8 Oct 2013 06:36:15 -0400 Received: from mail-qa0-f49.google.com ([209.85.216.49]:42795 "EHLO mail-qa0-f49.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754016Ab3JHKgM (ORCPT ); Tue, 8 Oct 2013 06:36:12 -0400 MIME-Version: 1.0 In-Reply-To: References: <1381221715.15686.3.camel@marge.simpson.net> <1381227252.15686.9.camel@marge.simpson.net> <1381227659.15686.11.camel@marge.simpson.net> Date: Tue, 8 Oct 2013 12:36:11 +0200 Message-ID: Subject: Re: perf: perf sched lat segfault - bisected From: Stephane Eranian To: Mike Galbraith Cc: LKML , Arnaldo Carvalho de Melo Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 5162 Lines: 95 Mike, I found the problem in perf. The tool->mmap2 pointer is NULL for sched tool. Let me send you a patch to fix this. On Tue, Oct 8, 2013 at 12:25 PM, Stephane Eranian wrote: > Mike, > > Ok, I missed the sched record, let me track this down. > Note that there was a second commit to fix this one > by Arnaldo. Do you also have it in your kernel? > > 51131c4 perf: Fix up MMAP2 buffer space reservation > > I will try to reproduce your problem ASAP. > > > On Tue, Oct 8, 2013 at 12:20 PM, Mike Galbraith wrote: >> On Tue, 2013-10-08 at 12:14 +0200, Mike Galbraith wrote: >>> On Tue, 2013-10-08 at 11:23 +0200, Stephane Eranian wrote: >>> > On Tue, Oct 8, 2013 at 10:41 AM, Mike Galbraith wrote: >>> > > Box: Q6600 >>> > > >>> > > Reproducer: >>> > > perf sched record -- sleep 1 >>> > > perf sched lat (boom) >>> > > >>> > I don't understand the test case here. >>> > If I do this, perf sched lat complains of no trace records >>> > and no kernel crash. Could you provide the full cmdline? >>> >>> Hm. That was it, just like so. >>> >>> marge:/usr/local/src/kernel/linux-3.x.git/tools/perf # ./perf sched record -- sleep 1 >>> [ perf record: Woken up 1 times to write data ] >>> [ perf record: Captured and wrote 0.763 MB perf.data (~33348 samples) ] >>> marge:/usr/local/src/kernel/linux-3.x.git/tools/perf # ./perf sched lat >>> Segmentation fault >>> marge:/usr/local/src/kernel/linux-3.x.git/tools/perf # >> >> commit 72f3bdf7fd1003373b454bf923d3a196bbc36600 >> Author: Mike Galbraith >> Date: Tue Oct 8 12:18:37 2013 +0200 >> >> Revert "perf tools: Add attr->mmap2 support" >> >> This reverts commit 5c5e854bc760a2e2c878df3cfcf2afa4febcd511. >> >> commit a8af8531d08d3edb37a7d67d1a4dc2e1f2f4da58 >> Author: Mike Galbraith >> Date: Tue Oct 8 12:18:23 2013 +0200 >> >> Revert "perf trace: Add mmap2 handler" >> >> This reverts commit 384c671e33a7ddf905f2c60b433b9883f0e5a605. >> >> marge:/usr/local/src/kernel/linux-3.x.git/tools/perf # ./perf sched >> record -- sleep 1 >> [ perf record: Woken up 1 times to write data ] >> [ perf record: Captured and wrote 0.647 MB perf.data (~28289 samples) ] >> marge:/usr/local/src/kernel/linux-3.x.git/tools/perf # perf sched lat >> marge:/usr/local/src/kernel/linux-3.x.git/tools/perf # >> >> (emitted this) >> >> --------------------------------------------------------------------------------------------------------------- >> Task | Runtime ms | Switches | Average delay ms | Maximum delay ms | Maximum delay at | >> --------------------------------------------------------------------------------------------------------------- >> kworker/3:2:383 | 0.073 ms | 1 | avg: 0.020 ms | max: 0.020 ms | max at: 7197.856052 s >> sleep:24717 | 1.163 ms | 3 | avg: 0.011 ms | max: 0.022 ms | max at: 7196.884445 s >> kworker/2:1:36 | 0.097 ms | 1 | avg: 0.007 ms | max: 0.007 ms | max at: 7197.796023 s >> haveged:3547 | 0.023 ms | 2 | avg: 0.006 ms | max: 0.009 ms | max at: 7197.796345 s >> rcu_sched:9 | 0.126 ms | 14 | avg: 0.006 ms | max: 0.021 ms | max at: 7196.897051 s >> ntpd:4185 | 0.034 ms | 1 | avg: 0.004 ms | max: 0.004 ms | max at: 7197.746409 s >> kded4:4594 | 0.058 ms | 1 | avg: 0.004 ms | max: 0.004 ms | max at: 7197.422130 s >> evolution:15686 | 0.065 ms | 1 | avg: 0.004 ms | max: 0.004 ms | max at: 7196.915715 s >> pickup:21579 | 0.062 ms | 1 | avg: 0.004 ms | max: 0.004 ms | max at: 7197.329185 s >> JS Watchdog:20053 | 0.020 ms | 1 | avg: 0.004 ms | max: 0.004 ms | max at: 7197.623752 s >> perf:24716 | 3.330 ms | 1 | avg: 0.004 ms | max: 0.004 ms | max at: 7197.885690 s >> plasma-desktop:4685 | 1.632 ms | 2 | avg: 0.004 ms | max: 0.005 ms | max at: 7196.933762 s >> master:4307 | 0.104 ms | 2 | avg: 0.004 ms | max: 0.005 ms | max at: 7197.329136 s >> Xorg:4331 | 0.485 ms | 4 | avg: 0.003 ms | max: 0.005 ms | max at: 7197.010801 s >> ksoftirqd/2:18 | 0.014 ms | 1 | avg: 0.003 ms | max: 0.003 ms | max at: 7196.897039 s >> alsa-sink:5282 | 0.000 ms | 200 | avg: 0.003 ms | max: 0.003 ms | max at: 7196.905149 s >> migration/3:22 | 0.000 ms | 1 | avg: 0.002 ms | max: 0.002 ms | max at: 7196.884423 s >> ----------------------------------------------------------------------------------------- >> TOTAL: | 7.285 ms | 237 | >> --------------------------------------------------- >> -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/