[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <93e36dac-47d4-1e2b-6cb1-0092a5371810@arm.com>
Date: Thu, 28 Apr 2022 17:14:09 +0100
From: James Clark <james.clark@....com>
To: Jeremy Linton <jeremy.linton@....com>,
linux-perf-users@...r.kernel.org,
"acme@...nel.org" <acme@...nel.org>
Cc: peterz@...radead.org, mingo@...hat.com, mark.rutland@....com,
alexander.shishkin@...ux.intel.com, jolsa@...nel.org,
namhyung@...nel.org, irogers@...gle.com, leo.yan@...aro.org,
linux-kernel@...r.kernel.org, Leo Yan <leo.yan@...aro.org>
Subject: Re: [PATCH] perf tests: Fix coresight `perf test` failure.
On 28/04/2022 16:19, Jeremy Linton wrote:
> Currently the `perf test` always fails the coresight test like:
>
> 89: Check Arm CoreSight trace data recording and synthesized samples: FAILED!
>
> That is because the test_arm_coresight.sh is attempting to SIGINT the
> parent but is using $$ rather than $PPID and it sigint's itself when
> run under the perf test framework. Since this is done in a trap clause
> it ends up returning a non zero return. Since $PPID is a bash ism and
> not all distros are linking /bin/sh to bash, the alternative
> parent pid lookups are uglier than just dropping the kill, and its
> not strictly needed, lets pick the simple solution and drop the sigint.
>
> Fixes: 133fe2e617e4 ("perf tests: Improve temp file cleanup in test_arm_coresight.sh")
> Cc: James Clark <james.clark@....com>
> Signed-off-by: Jeremy Linton <jeremy.linton@....com>
> ---
> tools/perf/tests/shell/test_arm_coresight.sh | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/tools/perf/tests/shell/test_arm_coresight.sh b/tools/perf/tests/shell/test_arm_coresight.sh
> index 6de53b7ef5ff..e4cb4f1806ff 100755
> --- a/tools/perf/tests/shell/test_arm_coresight.sh
> +++ b/tools/perf/tests/shell/test_arm_coresight.sh
> @@ -29,7 +29,6 @@ cleanup_files()
> rm -f ${file}
> rm -f "${perfdata}.old"
> trap - exit term int
> - kill -2 $$
> exit $glb_err
> }
>
Reviewed-by: James Clark <james.clark@....com>
Powered by blists - more mailing lists