Received-SPF: pass (google.com: domain of linux-kernel+bounces-145597-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Precedence: bulk
MIME-Version: 1.0
References: <20240415063626.453987-1-irogers@google.com>
In-Reply-To: <20240415063626.453987-1-irogers@google.com>
From: Ian Rogers <irogers@google.com>
Date: Mon, 15 Apr 2024 09:59:00 -0700
Message-ID: <CAP-5=fWhPTWiRjiUW7vRCetm26fYgOWfHzV4Qig3rSfe-x5VLQ@mail.gmail.com>
Subject: Re: [PATCH v1 0/9] Consistently prefer sysfs/json events
To: Peter Zijlstra <peterz@infradead.org>, Ingo Molnar <mingo@redhat.com>, 
	Arnaldo Carvalho de Melo <acme@kernel.org>, Namhyung Kim <namhyung@kernel.org>, 
	Mark Rutland <mark.rutland@arm.com>, 
	Alexander Shishkin <alexander.shishkin@linux.intel.com>, Jiri Olsa <jolsa@kernel.org>, 
	Ian Rogers <irogers@google.com>, Adrian Hunter <adrian.hunter@intel.com>, 
	Kan Liang <kan.liang@linux.intel.com>, James Clark <james.clark@arm.com>, 
	linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, 
	bpf@vger.kernel.org, Atish Patra <atishp@rivosinc.com>, linux-riscv@lists.infradead.org, 
	Beeman Strong <beeman@rivosinc.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Sun, Apr 14, 2024 at 11:36=E2=80=AFPM Ian Rogers <irogers@google.com> wr=
ote:
>
> As discussed in:
> https://lore.kernel.org/lkml/20240217005738.3744121-1-atishp@rivosinc.com=
/
> preferring sysfs/json events consistently (with or without a given
> PMU) will enable RISC-V's hope to customize legacy events in the perf
> tool.
>
> Some minor clean-up is performed on the way.

A side-effect of prioritizing sysfs/json events over legacy hardware
events is that the hard coded metric logic in stat-shadow fails:
https://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git/tr=
ee/tools/perf/util/stat-shadow.c?h=3Dperf-tools-next#n100
This is because the hard coded metrics assume that legacy events will
be used. Rather than make the hard coded metrics match sysfs/json I
think it is better to remove the hard coded metrics. This is because
the hard coded metrics lack checks on things like grouping that rather
than fix should be transitioned to json metrics. My preference for the
json metrics is to generate them using the python generation scripts
that are out for review.

Thanks,
Ian

> Ian Rogers (9):
>   perf parse-events: Factor out '<event_or_pmu>/.../' parsing
>   perf parse-events: Directly pass PMU to parse_events_add_pmu
>   perf parse-events: Avoid copying an empty list
>   perf pmu: Refactor perf_pmu__match
>   perf tests parse-events: Use branches rather than cache-references
>   perf parse-events: Legacy cache names on all PMUs and lower priority
>   perf parse-events: Handle PE_TERM_HW in name_or_raw
>   perf parse-events: Constify parse_events_add_numeric
>   perf parse-events: Prefer sysfs/json hardware events over legacy
>
>  tools/perf/tests/parse-events.c |   6 +-
>  tools/perf/util/parse-events.c  | 201 ++++++++++++++++++++++----------
>  tools/perf/util/parse-events.h  |  16 +--
>  tools/perf/util/parse-events.l  |  76 ++++++------
>  tools/perf/util/parse-events.y  | 166 +++++++++-----------------
>  tools/perf/util/pmu.c           |  27 +++--
>  tools/perf/util/pmu.h           |   2 +-
>  7 files changed, 262 insertions(+), 232 deletions(-)
>
> --
> 2.44.0.683.g7961c838ac-goog
>