Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.37 as permitted sender) client-ip=23.128.96.37;
MIME-Version: 1.0
References: <20230913125157.2790375-1-tmricht@linux.ibm.com>
In-Reply-To: <20230913125157.2790375-1-tmricht@linux.ibm.com>
From:   Ian Rogers <irogers@google.com>
Date:   Wed, 13 Sep 2023 09:31:29 -0700
Message-ID: <CAP-5=fUiHMRPVYhbQv-YM+EMKyBF6TEopea=PPX2thbtdmhGsg@mail.gmail.com>
Subject: Re: [PATCH] perf jevent: fix core dump on software events on s390
To:     Thomas Richter <tmricht@linux.ibm.com>
Cc:     linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
        acme@kernel.org, sumanthk@linux.ibm.com, dengler@linux.ibm.com,
        svens@linux.ibm.com, gor@linux.ibm.com, hca@linux.ibm.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Wed, Sep 13, 2023 at 5:52=E2=80=AFAM Thomas Richter <tmricht@linux.ibm.c=
om> wrote:
>
> Running commands such as
>  # ./perf stat -e cs -- true
>  Segmentation fault (core dumped)
>  # ./perf stat -e cpu-clock-- true
>  Segmentation fault (core dumped)
>  #
>
> dump core. This should not happen as these events are defined
> even when no hardware PMU is available.
> Debugging this reveals this call chain:
>
>   perf_pmus__find_by_type(type=3D1)
>   +--> pmu_read_sysfs(core_only=3Dfalse)
>        +--> perf_pmu__find2(dirfd=3D3, name=3D0x152a113 "software")
>             +--> perf_pmu__lookup(pmus=3D0x14f0568 <other_pmus>, dirfd=3D=
3,
>                                   lookup_name=3D0x152a113 "software")
>                  +--> perf_pmu__find_events_table (pmu=3D0x1532130)
>
> Now the pmu is "software" and it tries to find a proper table
> generated by the pmu-event generation process for s390:
>
>  # cd pmu-events/
>  # ./jevents.py  s390 all /root/linux/tools/perf/pmu-events/arch |\
>         grep -E '^const struct pmu_table_entry'
>  const struct pmu_table_entry pmu_events__cf_z10[] =3D {
>  const struct pmu_table_entry pmu_events__cf_z13[] =3D {
>  const struct pmu_table_entry pmu_metrics__cf_z13[] =3D {
>  const struct pmu_table_entry pmu_events__cf_z14[] =3D {
>  const struct pmu_table_entry pmu_metrics__cf_z14[] =3D {
>  const struct pmu_table_entry pmu_events__cf_z15[] =3D {
>  const struct pmu_table_entry pmu_metrics__cf_z15[] =3D {
>  const struct pmu_table_entry pmu_events__cf_z16[] =3D {
>  const struct pmu_table_entry pmu_metrics__cf_z16[] =3D {
>  const struct pmu_table_entry pmu_events__cf_z196[] =3D {
>  const struct pmu_table_entry pmu_events__cf_zec12[] =3D {
>  const struct pmu_table_entry pmu_metrics__cf_zec12[] =3D {
>  const struct pmu_table_entry pmu_events__test_soc_cpu[] =3D {
>  const struct pmu_table_entry pmu_metrics__test_soc_cpu[] =3D {
>  const struct pmu_table_entry pmu_events__test_soc_sys[] =3D {
>  #
>
> However event "software" is not listed, as can be seen in the
> generated const struct pmu_events_map pmu_events_map[].
> So in function perf_pmu__find_events_table(), the variable
> table is initialized to NULL, but never set to a proper
> value. The function scans all generated &pmu_events_map[]
> tables, but no table matches, because the tables are
> s390 CPU Measurement unit specific:
>
>   i =3D 0;
>   for (;;) {
>       const struct pmu_events_map *map =3D &pmu_events_map[i++];
>       if (!map->arch)
>            break;
>
>       --> the maps are there because the build generated them
>
>            if (!strcmp_cpuid_str(map->cpuid, cpuid)) {
>                 table =3D &map->event_table;
>                 break;
>            }
>       --> Since no matching CPU string the table var remains 0x0
>       }
>       free(cpuid);
>       if (!pmu)
>            return table;
>
>       --> The pmu is "software" so it exists and no return
>
>       --> and here perf dies because table is 0x0
>       for (i =3D 0; i < table->num_pmus; i++) {
>               ...
>       }
>       return NULL;
>
> Fix this and do not access the table variable. Instead return 0x0
> which is the same return code when the for-loop was not successful.
>
> Output after:
>  # ./perf stat -e cs -- true
>
>  Performance counter stats for 'true':
>
>                  0      cs
>
>        0.000853105 seconds time elapsed
>
>        0.000061000 seconds user
>        0.000827000 seconds sys
>
>  # ./perf stat -e cpu-clock -- true
>
>  Performance counter stats for 'true':
>
>               0.25 msec cpu-clock #    0.341 CPUs utilized
>
>        0.000728383 seconds time elapsed
>
>        0.000055000 seconds user
>        0.000706000 seconds sys
>
>  # ./perf stat -e cycles -- true
>
>  Performance counter stats for 'true':
>
>    <not supported>      cycles
>
>        0.000767298 seconds time elapsed
>
>        0.000055000 seconds user
>        0.000739000 seconds sys
>
>  #
>
> Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>

Reviewed-by: Ian Rogers <irogers@google.com>

Thanks!
Ian

> ---
>  tools/perf/pmu-events/jevents.py | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/tools/perf/pmu-events/jevents.py b/tools/perf/pmu-events/jev=
ents.py
> index a7e88332276d..72ba4a9239c6 100755
> --- a/tools/perf/pmu-events/jevents.py
> +++ b/tools/perf/pmu-events/jevents.py
> @@ -991,7 +991,7 @@ const struct pmu_events_table *perf_pmu__find_events_=
table(struct perf_pmu *pmu)
>                  }
>          }
>          free(cpuid);
> -        if (!pmu)
> +        if (!pmu || !table)
>                  return table;
>
>          for (i =3D 0; i < table->num_pmus; i++) {
> --
> 2.41.0
>