All of lore.kernel.org
 help / color / mirror / Atom feed
From: Arnaldo Carvalho de Melo <acme@kernel.org>
To: Disha Goel <disgoel@linux.ibm.com>
Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com>,
	jolsa@kernel.org, irogers@google.com, namhyung@kernel.org,
	linux-perf-users@vger.kernel.org, linuxppc-dev@lists.ozlabs.org,
	maddy@linux.ibm.com, kjain@linux.ibm.com,
	disgoel@linux.vnet.ibm.com
Subject: Re: [PATCH V2 2/2] tools/perf/tests: perf all metricgroups test fails when perf_event access is restricted
Date: Mon, 7 Aug 2023 18:19:30 -0300	[thread overview]
Message-ID: <ZNFf4iliPNnqlvbH@kernel.org> (raw)
In-Reply-To: <3c0eb25b-a51b-edba-82ea-22e6d2b07978@linux.ibm.com>

Em Mon, Aug 07, 2023 at 08:14:39PM +0530, Disha Goel escreveu:
> On 04/08/23 10:30 am, Athira Rajeev wrote:
> > Perf all metricgroups test fails as below when perf_event access
> > is restricted.
> > 
> >      ./perf test -v "perf all metricgroups test"
> >      Testing Memory_BW
> >      Error:
> >      Access to performance monitoring and observability operations is limited.
> >      Enforced MAC policy settings (SELinux) can limit access to performance
> >      access to performance monitoring and observability operations for processes
> >      without CAP_PERFMON, CAP_SYS_PTRACE or CAP_SYS_ADMIN Linux capability.
> > 
> >      test child finished with -1
> >      ---- end ----
> >      perf all metricgroups test: FAILED!
> > 
> > Fix the testcase to skip those metric events which needs perf_event access
> > explicitly. The exit code of the testcase is based on return code of
> > the perf stat command ( enabled by set -e option ). Hence save the
> > exit status in a variable and use that to decide success or fail for the
> > testcase.

I wonder if we shouldn't somehow check if the credentials needed to
performing a test shouldn't be checked before trying it. This way we
would check if the check that the tool or the kernel is doing is the
appropriate one.

I.e. the kernel refusal for doing something may be an error.

- Arnaldo

> > Signed-off-by: Athira Rajeev<atrajeev@linux.vnet.ibm.com>
> 
> With this patch applied(on power) perf metricgroups test works correctly when perf_event access is restricted.
> 
>  # ./perf test "perf all metricgroups test"
>  96: perf all metricgroups test                                      : Ok
> 
> Tested-by: Disha Goel<disgoel@linux.ibm.com>
> 
> > ---
> > Changelog:
> > v1 -> v2:
> >   Changed the condition to use "echo" and "grep" so it works on
> >   Posix shell as well.
> > 
> >   tools/perf/tests/shell/stat_all_metricgroups.sh | 14 +++++++++++---
> >   1 file changed, 11 insertions(+), 3 deletions(-)
> > 
> > diff --git a/tools/perf/tests/shell/stat_all_metricgroups.sh b/tools/perf/tests/shell/stat_all_metricgroups.sh
> > index cb35e488809a..eaa5e1172294 100755
> > --- a/tools/perf/tests/shell/stat_all_metricgroups.sh
> > +++ b/tools/perf/tests/shell/stat_all_metricgroups.sh
> > @@ -2,11 +2,19 @@
> >   # perf all metricgroups test
> >   # SPDX-License-Identifier: GPL-2.0
> > 
> > -set -e
> > -
> >   for m in $(perf list --raw-dump metricgroups); do
> >     echo "Testing $m"
> > -  perf stat -M "$m" -a true
> > +  result=$(perf stat -M "$m" -a true 2>&1)
> > +  rc=$?
> > +  # Skip if there is no access to perf_events monitoring
> > +  # Otherwise exit based on the return code of perf comamnd.
> > +  if echo "$result" | grep -q "Access to performance monitoring and observability operations is limited";
> > +  then
> > +      continue
> > +  else
> > +      [ $rc -ne 0 ] && exit $rc
> > +  fi
> > +
> >   done
> > 
> >   exit 0

-- 

- Arnaldo

WARNING: multiple messages have this Message-ID (diff)
From: Arnaldo Carvalho de Melo <acme@kernel.org>
To: Disha Goel <disgoel@linux.ibm.com>
Cc: irogers@google.com, Athira Rajeev <atrajeev@linux.vnet.ibm.com>,
	kjain@linux.ibm.com, linux-perf-users@vger.kernel.org,
	maddy@linux.ibm.com, jolsa@kernel.org, namhyung@kernel.org,
	disgoel@linux.vnet.ibm.com, linuxppc-dev@lists.ozlabs.org
Subject: Re: [PATCH V2 2/2] tools/perf/tests: perf all metricgroups test fails when perf_event access is restricted
Date: Mon, 7 Aug 2023 18:19:30 -0300	[thread overview]
Message-ID: <ZNFf4iliPNnqlvbH@kernel.org> (raw)
In-Reply-To: <3c0eb25b-a51b-edba-82ea-22e6d2b07978@linux.ibm.com>

Em Mon, Aug 07, 2023 at 08:14:39PM +0530, Disha Goel escreveu:
> On 04/08/23 10:30 am, Athira Rajeev wrote:
> > Perf all metricgroups test fails as below when perf_event access
> > is restricted.
> > 
> >      ./perf test -v "perf all metricgroups test"
> >      Testing Memory_BW
> >      Error:
> >      Access to performance monitoring and observability operations is limited.
> >      Enforced MAC policy settings (SELinux) can limit access to performance
> >      access to performance monitoring and observability operations for processes
> >      without CAP_PERFMON, CAP_SYS_PTRACE or CAP_SYS_ADMIN Linux capability.
> > 
> >      test child finished with -1
> >      ---- end ----
> >      perf all metricgroups test: FAILED!
> > 
> > Fix the testcase to skip those metric events which needs perf_event access
> > explicitly. The exit code of the testcase is based on return code of
> > the perf stat command ( enabled by set -e option ). Hence save the
> > exit status in a variable and use that to decide success or fail for the
> > testcase.

I wonder if we shouldn't somehow check if the credentials needed to
performing a test shouldn't be checked before trying it. This way we
would check if the check that the tool or the kernel is doing is the
appropriate one.

I.e. the kernel refusal for doing something may be an error.

- Arnaldo

> > Signed-off-by: Athira Rajeev<atrajeev@linux.vnet.ibm.com>
> 
> With this patch applied(on power) perf metricgroups test works correctly when perf_event access is restricted.
> 
>  # ./perf test "perf all metricgroups test"
>  96: perf all metricgroups test                                      : Ok
> 
> Tested-by: Disha Goel<disgoel@linux.ibm.com>
> 
> > ---
> > Changelog:
> > v1 -> v2:
> >   Changed the condition to use "echo" and "grep" so it works on
> >   Posix shell as well.
> > 
> >   tools/perf/tests/shell/stat_all_metricgroups.sh | 14 +++++++++++---
> >   1 file changed, 11 insertions(+), 3 deletions(-)
> > 
> > diff --git a/tools/perf/tests/shell/stat_all_metricgroups.sh b/tools/perf/tests/shell/stat_all_metricgroups.sh
> > index cb35e488809a..eaa5e1172294 100755
> > --- a/tools/perf/tests/shell/stat_all_metricgroups.sh
> > +++ b/tools/perf/tests/shell/stat_all_metricgroups.sh
> > @@ -2,11 +2,19 @@
> >   # perf all metricgroups test
> >   # SPDX-License-Identifier: GPL-2.0
> > 
> > -set -e
> > -
> >   for m in $(perf list --raw-dump metricgroups); do
> >     echo "Testing $m"
> > -  perf stat -M "$m" -a true
> > +  result=$(perf stat -M "$m" -a true 2>&1)
> > +  rc=$?
> > +  # Skip if there is no access to perf_events monitoring
> > +  # Otherwise exit based on the return code of perf comamnd.
> > +  if echo "$result" | grep -q "Access to performance monitoring and observability operations is limited";
> > +  then
> > +      continue
> > +  else
> > +      [ $rc -ne 0 ] && exit $rc
> > +  fi
> > +
> >   done
> > 
> >   exit 0

-- 

- Arnaldo

  reply	other threads:[~2023-08-07 21:19 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-04  5:00 [PATCH V2 1/2] tools/perf/tests: perf all metrics test fails when perf_event access is restricted Athira Rajeev
2023-08-04  5:00 ` Athira Rajeev
2023-08-04  5:00 ` [PATCH V2 2/2] tools/perf/tests: perf all metricgroups " Athira Rajeev
2023-08-04  5:00   ` Athira Rajeev
2023-08-07 14:44   ` Disha Goel
2023-08-07 21:19     ` Arnaldo Carvalho de Melo [this message]
2023-08-07 21:19       ` Arnaldo Carvalho de Melo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=ZNFf4iliPNnqlvbH@kernel.org \
    --to=acme@kernel.org \
    --cc=atrajeev@linux.vnet.ibm.com \
    --cc=disgoel@linux.ibm.com \
    --cc=disgoel@linux.vnet.ibm.com \
    --cc=irogers@google.com \
    --cc=jolsa@kernel.org \
    --cc=kjain@linux.ibm.com \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=maddy@linux.ibm.com \
    --cc=namhyung@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.