public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Mike Galbraith <bitbucket@online.de>
To: David Ahern <dsahern@gmail.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>,
	Stephane Eranian <eranian@google.com>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: perf: perf sched lat segfault - bisected
Date: Tue, 08 Oct 2013 15:44:36 +0200	[thread overview]
Message-ID: <1381239876.15686.56.camel@marge.simpson.net> (raw)
In-Reply-To: <52540685.2000306@gmail.com>

On Tue, 2013-10-08 at 07:20 -0600, David Ahern wrote: 
> On 10/8/13 6:35 AM, Mike Galbraith wrote:
> >
> > (except replay segfaults, but lord knows when that last worked)
> 
> Fairly recently actually. I ran it early summer adding my timehist 
> command to builtin-sched.c. I'll take a look later today.

pr_err() blows up in self_open_counters() when you hit 1024 fds.  Too
many threads to play back after recording make in perf directory.
Playing a static load back works fine. 

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fffedbc8700 (LWP 5270)]
buffered_vfprintf (s=0x7ffff301f700, format=0x65b9e0 "Error: sys_perf_event_open() syscall returned with %d (%s)\n", args=0x7fffedbc7a90) at vfprintf.c:2265
2265      result = INTUSE(_IO_vfprintf) (hp, format, args);
(gdb) where
#0  buffered_vfprintf (s=0x7ffff301f700, format=0x65b9e0 "Error: sys_perf_event_open() syscall returned with %d (%s)\n", args=0x7fffedbc7a90) at vfprintf.c:2265
#1  0x00007ffff2cd9dbe in _IO_vfprintf_internal (s=0x7ffff301f700, format=0x65b9e0 "Error: sys_perf_event_open() syscall returned with %d (%s)\n", ap=0x7fffedbc7a90) at vfprintf.c:1291
#2  0x00007ffff2d84148 in ___vfprintf_chk (fp=0x7ffff301f700, flag=1, format=<optimized out>, ap=<optimized out>) at vfprintf_chk.c:35
#3  0x000000000048d4f8 in vfprintf (__ap=0x7fffedbc7a90, __fmt=<optimized out>, __stream=<optimized out>) at /usr/include/bits/stdio2.h:128
#4  eprintf (level=<optimized out>, fmt=<optimized out>) at util/debug.c:29
#5  0x000000000043aa10 in self_open_counters () at builtin-sched.c:441
#6  thread_func (ctx=<optimized out>) at builtin-sched.c:476
#7  0x00007ffff79b1f05 in start_thread (arg=0x7fffedbc8700) at pthread_create.c:301
#8  0x00007ffff2d7010d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:115
(gdb)


      reply	other threads:[~2013-10-08 13:44 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-10-08  8:41 perf: perf sched lat segfault - bisected Mike Galbraith
2013-10-08  9:23 ` Stephane Eranian
2013-10-08 10:14   ` Mike Galbraith
2013-10-08 10:20     ` Mike Galbraith
2013-10-08 10:25       ` Stephane Eranian
2013-10-08 10:36         ` Stephane Eranian
2013-10-08 10:50           ` Mike Galbraith
2013-10-08 10:51             ` Stephane Eranian
2013-10-08 10:56               ` Mike Galbraith
2013-10-08 10:57                 ` Stephane Eranian
2013-10-08 10:41         ` Mike Galbraith
2013-10-08 12:17 ` Arnaldo Carvalho de Melo
2013-10-08 12:27   ` Mike Galbraith
2013-10-08 12:35     ` Mike Galbraith
2013-10-08 13:20       ` David Ahern
2013-10-08 13:44         ` Mike Galbraith [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1381239876.15686.56.camel@marge.simpson.net \
    --to=bitbucket@online.de \
    --cc=acme@redhat.com \
    --cc=dsahern@gmail.com \
    --cc=eranian@google.com \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox