linux-kernel - Re: "perf hists browser: Support flat callchains" appears to have broken parent reporting

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, 30 Mar 2016 10:46:34 -0300
From:	Arnaldo Carvalho de Melo <acme@...nel.org>
To:	Andres Freund <andres@...razel.de>
Cc:	Namhyung Kim <namhyung@...nel.org>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	Ingo Molnar <mingo@...hat.com>, linux-kernel@...r.kernel.org
Subject: Re: "perf hists browser: Support flat callchains" appears to have
 broken parent reporting

Em Wed, Mar 30, 2016 at 02:34:18PM +0200, Andres Freund escreveu:
> Hi,
> 
> 4b3a3212233a - "perf hists browser: Support flat callchains" seems to
> have broken callchain display in tui mode when using !flat mode, or at
> least changed it in an unintended manner.

humm, at first I thought this would be related to --percent-limit...
What tree/branch are you using? Can you try pressing 'L' to play with
the percent limit?

- Arnaldo
 
> Llooking at the same perf.data file:
> perf report --tui --no-children -g
> before:
> -   16.69%  swapper          [kernel.kallsyms]             [k] poll_idle
>    - poll_idle
>       + 16.65% cpuidle_enter_state
>       + 0.04% cpuidle_enter
> -    3.51%  swapper          [kernel.kallsyms]             [k] intel_idle
>    - intel_idle
>       + 3.49% cpuidle_enter_state
>       + 0.03% cpuidle_enter
> -    1.35%  postgres         postgres                      [.] hash_search_with_hash_value
>    - hash_search_with_hash_value
>       - 0.32% BufTableLookup
>          + ReadBuffer_common
>       - 0.18% LockAcquireExtended
>          + 0.16% LockRelationOid
>          + 0.02% XactLockTableInsert
>       + 0.14% FetchPreparedStatement
>       + 0.13% CreatePortal
>       + 0.11% RelationIdGetRelation
>       + 0.10% GetPortalByName
> after:
> -   16.69%  swapper          [kernel.kallsyms]             [k] poll_idle
>      poll_idle
> -    3.51%  swapper          [kernel.kallsyms]             [k] intel_idle
>      intel_idle
> -    1.35%  postgres         postgres                      [.] hash_search_with_hash_value
>      hash_search_with_hash_value
> -    1.18%  postgres         postgres                      [.] AllocSetAlloc
>      AllocSetAlloc
> -    0.85%  postgres         libc-2.22.so                  [.] __memcpy_sse2_unaligned
>      __memcpy_sse2_unaligned
> 
> as you can see after the aforementioned commit there's only one level of
> callers reported.
> 
> --stdio output isn't affected.
> 
> I can provide an example perf.data file, but it apears to reproduce with
> just about any profile here.
> 
> Greetings,
> 
> Andres Freund