linux-kernel - Re: [PATCH] perf tools: Add a dso_size option to perf report --sort

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:   Tue, 27 Mar 2018 12:45:07 -0300
From:   Arnaldo Carvalho de Melo <acme@...nel.org>
To:     Kim Phillips <kim.phillips@....com>
Cc:     Andi Kleen <ak@...ux.intel.com>,
        Maxim Kuvyrkov <maxim.kuvyrkov@...aro.org>,
        linux-kernel@...r.kernel.org, linux-perf-users@...r.kernel.org,
        Peter Zijlstra <peterz@...radead.org>,
        Ingo Molnar <mingo@...hat.com>,
        Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        Jiri Olsa <jolsa@...hat.com>,
        Namhyung Kim <namhyung@...nel.org>,
        Jin Yao <yao.jin@...ux.intel.com>,
        Milian Wolff <milian.wolff@...b.com>
Subject: Re: [PATCH] perf tools: Add a dso_size option to perf report --sort

Em Tue, Mar 27, 2018 at 09:05:07AM -0500, Kim Phillips escreveu:
> On Tue, 27 Mar 2018 04:27:02 -0700
> Andi Kleen <ak@...ux.intel.com> wrote:
> 
> > On Tue, Mar 27, 2018 at 06:09:56AM -0500, Kim Phillips wrote:
> > > Add DSO size to perf report/top sort output list.
> > > 
> > > This includes adding a map__size fn to map.h, which is
> > > approximately equal to the DSO data file_size:
> > 
> > What's the point of knowing the size?  
 
> I think it's just an extra convenience over symbol_size (added in
> commit 7768f8dada66d6  "perf tools: Allow sorting by symbol size") for
> tools developers' aggregation statistics.

Yeah, for completeness, may be useful for someone, probably for Maxim?
:-)
 
> > Most of the DSO may not be faulted in. If anything the working
> > set would be more interesting. This would need more changes
> > though.
> 
> Indeed.  Maxim?

Right, that would be another interesting metric to have,

- Arnaldo