linux-kernel - Re: [PATCH] perf/test: reduce run time of perf test Test java symbol

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]

Message-ID: <ZFqMsjJg00zmQyJG@kernel.org>
Date:   Tue, 9 May 2023 15:10:58 -0300
From:   Arnaldo Carvalho de Melo <acme@...nel.org>
To:     Ian Rogers <irogers@...gle.com>
Cc:     Thomas Richter <tmricht@...ux.ibm.com>, jolsa@...nel.org,
        linux-kernel@...r.kernel.org, linux-perf-users@...r.kernel.org,
        leo.yan@...aro.org, svens@...ux.ibm.com, gor@...ux.ibm.com,
        sumanthk@...ux.ibm.com, hca@...ux.ibm.com
Subject: Re: [PATCH] perf/test: reduce run time of perf test Test java symbol

Em Tue, May 09, 2023 at 10:34:44AM -0700, Ian Rogers escreveu:
> On Tue, May 9, 2023 at 6:19 AM Thomas Richter <tmricht@...ux.ibm.com> wrote:
> >
> > Test case Test java symbol might run a long time. On Fedora 38 the
> > run time is very, very long:
> >
> >  Output before:
> >  # time ./perf test 108
> >  108: Test java symbol                  : Ok
> >  real   22m15.775s
> >  user   3m42.584s
> >  sys    4m30.685s
> >  #
> >
> > The reason is a lookup for the server for debug symbols as shown in
> >  # cat /etc/debuginfod/elfutils.urls
> >  https://debuginfod.fedoraproject.org/
> >  #
> > This lookup is done for every symbol/sample, so about 3500 lookups
> > will take place.
> > To omit this lookup, which is not needed, unset environment variable
> > DEBUGINFOD_URLS=''.
> >
> >  Output after:
> >  # time ./perf test 108
> >  108: Test java symbol                  : Ok
> >
> >  real   0m6.242s
> >  user   0m4.982s
> >  sys    0m3.243s
> >  #
> >
> > Signed-off-by: Thomas Richter <tmricht@...ux.ibm.com>
> > Cc: leo.yan@...aro.org
> > Cc: irogers@...gle.com
> > ---
> >  tools/perf/tests/shell/test_java_symbol.sh | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/tools/perf/tests/shell/test_java_symbol.sh b/tools/perf/tests/shell/test_java_symbol.sh
> > index 90cea8811926..499539d1c479 100755
> > --- a/tools/perf/tests/shell/test_java_symbol.sh
> > +++ b/tools/perf/tests/shell/test_java_symbol.sh
> > @@ -56,7 +56,7 @@ if [ $? -ne 0 ]; then
> >         exit 1
> >  fi
> >
> > -if ! perf inject -i $PERF_DATA -o $PERF_INJ_DATA -j; then
> > +if ! DEBUGINFOD_URLS='' perf inject -i $PERF_DATA -o $PERF_INJ_DATA -j; then
> 
> Jiri, is this the best way to workaround a debuginfod slowness?

I noticed this delay, It is good that someone worked on rootcausing this
excessive delay, thanks Thomas.

I'll test it now.

- Arnaldo