linux-kernel - Re: [PATCH v1 5/5] perf record: Make --buildid-mmap the default

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] [day] [month] [year] [list]

Message-ID: <CAP-5=fWwW9G7EpR6Bn02dm9PVsqvzyDBcAzzVF1Sc2djAcqaYw@mail.gmail.com>
Date: Fri, 25 Apr 2025 09:03:59 -0700
From: Ian Rogers <irogers@...gle.com>
To: Arnaldo Carvalho de Melo <acme@...nel.org>
Cc: Peter Zijlstra <peterz@...radead.org>, Ingo Molnar <mingo@...hat.com>, 
	Namhyung Kim <namhyung@...nel.org>, Mark Rutland <mark.rutland@....com>, 
	Alexander Shishkin <alexander.shishkin@...ux.intel.com>, Jiri Olsa <jolsa@...nel.org>, 
	Adrian Hunter <adrian.hunter@...el.com>, Kan Liang <kan.liang@...ux.intel.com>, 
	Athira Rajeev <atrajeev@...ux.ibm.com>, Kajol Jain <kjain@...ux.ibm.com>, 
	Li Huafei <lihuafei1@...wei.com>, "Steinar H. Gunderson" <sesse@...gle.com>, 
	Stephen Brennan <stephen.s.brennan@...cle.com>, James Clark <james.clark@...aro.org>, 
	Andi Kleen <ak@...ux.intel.com>, Dmitry Vyukov <dvyukov@...gle.com>, 
	Zhongqiu Han <quic_zhonhan@...cinc.com>, Yicong Yang <yangyicong@...ilicon.com>, 
	Michael Petlan <mpetlan@...hat.com>, 
	Krzysztof Łopatowski <krzysztof.m.lopatowski@...il.com>, 
	"Dr. David Alan Gilbert" <linux@...blig.org>, Leo Yan <leo.yan@....com>, 
	Steve Clevenger <scclevenger@...amperecomputing.com>, Zixian Cai <fzczx123@...il.com>, 
	Thomas Falcon <thomas.falcon@...el.com>, Martin Liska <martin.liska@....com>, 
	Martin Liška <m.liska@...link.cz>, 
	linux-perf-users@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v1 5/5] perf record: Make --buildid-mmap the default

On Fri, Apr 25, 2025 at 7:45 AM Arnaldo Carvalho de Melo
<acme@...nel.org> wrote:
>
> On Thu, Apr 24, 2025 at 12:20:44AM -0700, Ian Rogers wrote:
> > On Wed, Apr 23, 2025 at 11:19 PM Ian Rogers <irogers@...gle.com> wrote:
> > > @@ -1795,7 +1796,8 @@ record__finish_output(struct record *rec)
> > >                         data->dir.files[i].size = lseek(data->dir.files[i].fd, 0, SEEK_CUR);
> > >         }
> > >
> > > -       if (!rec->no_buildid) {
> > > +       /* Buildid scanning disabled or build ID in kernel and synthesized map events. */
> > > +       if (!rec->no_buildid && !rec->buildid_mmap) {
>
> > So I think this is wrong. It matches current behaviors, but it is
> > wrong. If we don't process the kernel's mmap events the DSOs won't be
> > loaded and the build ID cache won't contain the DSOs. There is also
> > the bug that the sample processing to find maps to find DSOs, doesn't
> > handle call chains. Given the broken nature of the build ID cache I'm
> > not sure it makes any sense for perf record to be by default
> > populating it. I think it probably makes sense to consider the default
> > population a legacy feature and make -N the default along with
> > --buildid-mmap.
>
> The first four patches are good fixes/cleanups, so I'm picking them, ok?

Thanks, happy for the clean ups to land. I'm keen to push the needle
on making build IDs the norm in the codebase so if we don't take the
switching on build ID mmap by default change, let me know what I can
do to make it so it can land.

Thanks,
Ian