linux-kernel - Re: [PATCH 0/2] perf: measure page fault duration in perf trace

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 26 Oct 2016 16:47:06 -0200
From:   Arnaldo Carvalho de Melo <acme@...nel.org>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Alexis Berlemont <alexis.berlemont@...il.com>,
        linux-kernel@...r.kernel.org, mingo@...hat.com,
        alexander.shishkin@...ux.intel.com
Subject: Re: [PATCH 0/2] perf: measure page fault duration in perf trace

Em Wed, Oct 26, 2016 at 10:46:18AM +0200, Peter Zijlstra escreveu:
> On Wed, Oct 26, 2016 at 01:51:58AM +0200, Alexis Berlemont wrote:
> > Hi,
> > 
> > Here are 2 small patches which try to fulfill a point in the perf todo
> > list:
> 
> There's a todo list?

https://perf.wiki.kernel.org/index.php/Todo
 
> > * Forward port the page fault tracepoints and use it in 'trace'.
> > http://git.kernel.org/?p=linux/kernel/git/acme/linux.git;a=commitdiff;h=d53b11976093b6d8afeb8181db53aaffc754920d;hp=32ba4abf60ae1b710d22a75725491815de649bc5
> 
> dead link

I guess this is the one:

http://git.kernel.org/cgit/linux/kernel/git/acme/linux.git/commit/?id=eea86c6e06c241667c96c8e87e43c0870e1d6285

author	Frederic Weisbecker <fweisbec@...il.com>	2010-11-12 04:35:06 (GMT)
committer	Ingo Molnar <mingo@...e.hu>	2011-05-07 16:10:43 (GMT)
commit	eea86c6e06c241667c96c8e87e43c0870e1d6285 (patch)
tree	b00ce7ef9a6219f4df6b0e1ffe453d2d5baf3593
parent	57d524154ffe99d27fb55e0e30ddbad9f4c35806 (diff)
perf, mm: Add fault tracing
Part of that are two modified patches from Jiri Olsa who added the fault
tracepoints. I had to split them in two tracepoints so that we get the
faults
handling duration.

Originally-from: Frederic Weisbecker <fweisbec@...il.com>
Signed-off-by: Thomas Gleixner <tglx@...utronix.de>
Signed-off-by: Ingo Molnar <mingo@...e.hu>

--------------------

After that entry was added to the todo we got --pf, but no duration:

[root@...et ~]# perf trace --pf maj --no-syscalls
     0.000 ( 0.000 ms): perf/20412 majfault [__memcmp_sse4_1+0xbc6] => 0x7f1e0010c000 (?.)
     1.156 ( 0.000 ms): perf/20412 majfault [__memcmp_sse4_1+0xbc6] => /usr/bin/mv@0x0 (d.)
    56.907 ( 0.000 ms): perf/20412 majfault [__memcpy_avx_unaligned+0x1e7] => /usr/bin/mv@...07a0 (d.)
    57.683 ( 0.000 ms): perf/20412 majfault [__memcmp_sse4_1+0xbc6] => /usr/lib/debug/usr/bin/mv.debug@0x0 (d.)
    60.916 ( 0.000 ms): perf/20412 majfault [__memcpy_avx_unaligned+0x2b6] => /usr/lib/debug/usr/bin/mv.debug@...ace0 (d.)
18446744073708.766 ( 0.000 ms): systemd-journa/578 majfault [0x14972] => 0x7f25bdd9ebd0 (?.)
    70.193 ( 0.000 ms): perf/20412 majfault [__memcmp_sse4_1+0xbc6] => /usr/lib/debug/usr/lib/systemd/systemd-journald.debug@0x0 (d.)
    71.176 ( 0.000 ms): perf/20412 majfault [__memcpy_avx_unaligned+0x2b6] => /usr/lib/debug/usr/lib/systemd/systemd-journald.debug@...221e0 (d.)
^C[root@...et ~]# trace -h --pf

 Usage: perf trace [<options>] [<command>]
    or: perf trace [<options>] -- <command> [<options>]
    or: perf trace record [<options>] [<command>]
    or: perf trace record [<options>] -- <command> [<options>]

    -F, --pf <all|maj|min>
                          Trace pagefaults

[root@...et ~]#