Received-SPF: pass (google.com: domain of linux-kernel+bounces-203739-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Precedence: bulk
MIME-Version: 1.0
References: <20240605160848.4116061-1-kan.liang@linux.intel.com> <CAP-5=fV+-ytA2st17Ar-jQ5xYqrWtxnF2TcADKrC5WoPyKz4wQ@mail.gmail.com>
In-Reply-To: <CAP-5=fV+-ytA2st17Ar-jQ5xYqrWtxnF2TcADKrC5WoPyKz4wQ@mail.gmail.com>
From: Namhyung Kim <namhyung@kernel.org>
Date: Thu, 6 Jun 2024 00:34:17 -0700
Message-ID: <CAM9d7cjuHYDMvcq10ZD=3LSmia4WcvAzsme89B-odHYBAZzWYg@mail.gmail.com>
Subject: Re: [PATCH] perf stat: Fix the hard-coded metrics calculation on the hybrid
To: Ian Rogers <irogers@google.com>
Cc: kan.liang@linux.intel.com, acme@kernel.org, jolsa@kernel.org, 
	adrian.hunter@intel.com, linux-perf-users@vger.kernel.org, 
	linux-kernel@vger.kernel.org, "Khalil, Amiri" <amiri.khalil@intel.com>, 
	stable@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Wed, Jun 5, 2024 at 10:21=E2=80=AFAM Ian Rogers <irogers@google.com> wro=
te:
>
> On Wed, Jun 5, 2024 at 9:10=E2=80=AFAM <kan.liang@linux.intel.com> wrote:
> >
> > From: Kan Liang <kan.liang@linux.intel.com>
> >
> > The hard-coded metrics is wrongly calculated on the hybrid machine.
> >
> > $ perf stat -e cycles,instructions -a sleep 1
> >
> >  Performance counter stats for 'system wide':
> >
> >         18,205,487      cpu_atom/cycles/
> >          9,733,603      cpu_core/cycles/
> >          9,423,111      cpu_atom/instructions/     #  0.52  insn per cy=
cle
> >          4,268,965      cpu_core/instructions/     #  0.23  insn per cy=
cle
> >
> > The insn per cycle for cpu_core should be 4,268,965 / 9,733,603 =3D 0.4=
4.
> >
> > When finding the metric events, the find_stat() doesn't take the PMU
> > type into account. The cpu_atom/cycles/ is wrongly used to calculate
> > the IPC of the cpu_core.
> >
> > Fixes: 0a57b910807a ("perf stat: Use counts rather than saved_value")
> > Reported-by: "Khalil, Amiri" <amiri.khalil@intel.com>
> > Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
>
> Reviewed-by: Ian Rogers <irogers@google.com>
>
> Thanks,
> Ian
>
> > Cc: stable@vger.kernel.org
> > ---
> >  tools/perf/util/stat-shadow.c | 4 ++++
> >  1 file changed, 4 insertions(+)
> >
> > diff --git a/tools/perf/util/stat-shadow.c b/tools/perf/util/stat-shado=
w.c
> > index 3466aa952442..4d0edc061f1a 100644
> > --- a/tools/perf/util/stat-shadow.c
> > +++ b/tools/perf/util/stat-shadow.c
> > @@ -176,6 +176,10 @@ static double find_stat(const struct evsel *evsel,=
 int aggr_idx, enum stat_type
> >                 if (type !=3D evsel__stat_type(cur))
> >                         continue;
> >
> > +               /* Ignore if not the PMU we're looking for. */
> > +               if (evsel->pmu !=3D cur->pmu)
> > +                       continue;

Hmm.. Don't some metrics need events from different PMU?
Like cycles per sec or branch instructions per sec..

Thanks,
Namhyung


> > +
> >                 aggr =3D &cur->stats->aggr[aggr_idx];
> >                 if (type =3D=3D STAT_NSECS)
> >                         return aggr->counts.val;
> > --
> > 2.35.1
> >