The Linux Kernel Mailing List
 help / color / mirror / Atom feed
From: Pekka Enberg <penberg@cs.helsinki.fi>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@elte.hu>,
	Arnaldo Carvalho de Melo <acme@redhat.com>,
	paulus@samba.org, linux-kernel@vger.kernel.org
Subject: Re: [PROBLEM] perf report is broken in latest linus git
Date: Thu, 13 Aug 2009 15:24:04 +0300	[thread overview]
Message-ID: <1250166244.4796.0.camel@penberg-laptop> (raw)
In-Reply-To: <1250164938.4941.2.camel@penberg-laptop>

On Thu, 2009-08-13 at 15:02 +0300, Pekka Enberg wrote:
> On Thu, 2009-08-13 at 13:33 +0200, Peter Zijlstra wrote:
> > > > Perhaps it's this commit:
> > > > 
> > > > 4d1e00a: perf symbol: Fix symbol parsing in certain cases: use the build-id as a symlink
> > > > 
> > > > it does not revert cleanly, so you might want to check out these two 
> > > > commits:
> > > > 
> > > >   4d1e00a
> > > >   4d1e00a~1
> > > > 
> > > > and build perf on both, and check with your anomalous perf.data 
> > > > (without changing it) whether the output makes most sense.
> > > 
> > > The commit seems fine. Like I said, the perf.data file is busted even
> > > with 2.6.31-rc4 userspace so I think it's a kernel bug.
> > 
> > Hrmm, does a -tip kernel work as expected? I'll go audit the kernel
> > changes between -rc4 and -rc5.
> 
> tip/master seems to be broken too:
> 
> penberg@penberg-laptop:~/testing/jato$ perf report
> # Samples: 65806
> #
> # Overhead  Command                            Shared Object  Symbol
> # ........  .......  .......................................  ......
> #
>     64.31%     perf  /home/penberg/bin/perf                   [.] strlist__load
>      3.10%     perf  /home/penberg/bin/jato                   [.] 0x0000000000c6e8
>      1.61%     perf  [kernel]                                 [k] get_page_from_freelist
>      1.13%     perf  /lib/tls/i686/cmov/libc-2.8.90.so        [.] _int_malloc
>      1.11%     perf  /lib/tls/i686/cmov/libpthread-2.8.90.so  [.] 0x00000000012532
>      1.08%     perf  /lib/tls/i686/cmov/libc-2.8.90.so        [.] _fitoa
>      0.96%     perf                                 a7d23660  [.] 0x000000a7d23660
>      0.86%     perf  /lib/tls/i686/cmov/libc-2.8.90.so        [.] ____strtold_l_internal
>      0.85%     perf  /usr/lib/libz.so.1.2.3.3                 [.] 0x0000000000eef2
>      0.83%     perf  /lib/tls/i686/cmov/libc-2.8.90.so        [.] __GI_strcmp
>      0.74%     perf  /home/penberg/bin/jato                   [.] zalloc
>      0.69%     perf  [kernel]                                 [k] scsi_dispatch_cmd
>      0.66%     perf  /lib/tls/i686/cmov/libc-2.8.90.so        [.] getsubopt
>      0.56%     perf  /lib/tls/i686/cmov/libc-2.8.90.so        [.] __realpath
>      0.50%     perf  /usr/lib/libzip.so.1.0.0                 [.] _zip_get_name
> 
> penberg@penberg-laptop:~/src/linux/tools/perf$ git log|head -10
> commit d1fa7782b090a3ab0f55c9003d7315b5f0f5e9d6
> Merge: e93673b... daac07b...
> Author: Ingo Molnar <mingo@elte.hu>
> Date:   Thu Aug 13 10:37:33 2009 +0200
> 
>     Merge branch 'perfcounters/urgent'
> 
> commit daac07b2e6b77f1bd44104aa2f0593e5505f27d4
> Author: Frederic Weisbecker <fweisbec@gmail.com>
> Date:   Thu Aug 13 10:27:19 2009 +0200

v2.6.31-rc5 is fine:

# Samples: 62768
#
# Overhead  Command                                  Shared Object  Symbol
# ........  .......  .............................................  ......
#
    68.45%     jato  /home/penberg/bin/jato                         [.] gc_map_init
     2.08%     jato  /lib/tls/i686/cmov/libc-2.8.90.so              [.] _int_malloc
     1.79%     jato  /lib/tls/i686/cmov/libc-2.8.90.so              [.] __GI_strcmp
     1.49%     jato  [kernel]                                       [k] get_page_from_freelist
     1.33%     jato  /lib/tls/i686/cmov/libc-2.8.90.so              [.] __GI___libc_malloc
     1.11%     jato  /home/penberg/bin/jato                         [.] zalloc
     1.04%     jato                                       a7c02842  [.] 0x000000a7c02842
     0.92%     jato  /usr/lib/libzip.so.1.0.0                       [.] _zip_get_name
     0.76%     jato  /home/penberg/bin/jato                         [.] cafebabe_constant_info_utf8_compare
     0.72%     jato  /lib/tls/i686/cmov/libpthread-2.8.90.so        [.] 0x00000000007a2e
     0.64%     jato  /usr/lib/libz.so.1.2.3.3                       [.] 0x00000000001a0a
     0.53%     jato  /home/penberg/bin/jato                         [.] allocate_registers
     0.51%     jato  /home/penberg/bin/jato                         [.] analyze_liveness
     0.51%     jato  /home/penberg/bin/jato                         [.] split_interval_at



  reply	other threads:[~2009-08-13 12:24 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-08-13  9:28 [PROBLEM] perf report is broken in latest linus git Pekka Enberg
2009-08-13  9:36 ` Ingo Molnar
2009-08-13  9:47   ` Pekka Enberg
2009-08-13 11:33     ` Peter Zijlstra
2009-08-13 12:02       ` Pekka Enberg
2009-08-13 12:24         ` Pekka Enberg [this message]
2009-08-13 14:14 ` [PATCH] perf_counter: Report the cloning task as parent on perf_counter_fork() Peter Zijlstra
2009-08-13 14:18   ` [tip:perfcounters/urgent] " tip-bot for Peter Zijlstra
2009-08-15  8:58   ` [PATCH] " Pekka Enberg
2009-08-15  9:13     ` Ingo Molnar

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=1250166244.4796.0.camel@penberg-laptop \
    --to=penberg@cs.helsinki.fi \
    --cc=acme@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=paulus@samba.org \
    --cc=peterz@infradead.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