Received-SPF: pass (google.com: domain of linux-kernel+bounces-153957-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Precedence: bulk
MIME-Version: 1.0
References: <20240402214436.1409476-1-weilin.wang@intel.com> <20240402214436.1409476-6-weilin.wang@intel.com>
In-Reply-To: <20240402214436.1409476-6-weilin.wang@intel.com>
From: Namhyung Kim <namhyung@kernel.org>
Date: Mon, 22 Apr 2024 12:55:04 -0700
Message-ID: <CAM9d7ciKUYAdu97O3p5VRYGjXwvOv+4kgERXKM6mjr6OUbRQZA@mail.gmail.com>
Subject: Re: [RFC PATCH v7 5/6] perf stat: Add retire latency print functions
 to print out at the very end of print out
To: weilin.wang@intel.com
Cc: Ian Rogers <irogers@google.com>, Arnaldo Carvalho de Melo <acme@kernel.org>, 
	Peter Zijlstra <peterz@infradead.org>, Ingo Molnar <mingo@redhat.com>, 
	Alexander Shishkin <alexander.shishkin@linux.intel.com>, Jiri Olsa <jolsa@kernel.org>, 
	Adrian Hunter <adrian.hunter@intel.com>, Kan Liang <kan.liang@linux.intel.com>, 
	linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, 
	Perry Taylor <perry.taylor@intel.com>, Samantha Alt <samantha.alt@intel.com>, 
	Caleb Biggers <caleb.biggers@intel.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Tue, Apr 2, 2024 at 2:46=E2=80=AFPM <weilin.wang@intel.com> wrote:
>
> From: Weilin Wang <weilin.wang@intel.com>
>
> Add print out functions so that users could read retire latency values.
>
> Example output:
>
>  Performance counter stats for 'system wide':
>
>             25,717      MEM_INST_RETIRED.SPLIT_STORES    #      2.2 %  tm=
a_split_stores
>         28,365,080      CPU_CLK_UNHALTED.THREAD
>              24.00      MEM_INST_RETIRED.SPLIT_STORES:p  #       96      =
  4
>
>        2.054365083 seconds time elapsed
>
> This output of retire latency data is in format:
>  <val> <event-name> # <sum> <count>.

Nop, please follow the perf stat output format.  It'd be

  <sum> <event-name> # ...

Hmm.. maybe you don't need this at all as it'll have the event
anyway (probably without the 'p' modifier like in the example
above).

Then I think you can just add a metric value saying the average
latency in the comment area.

  Performance counter stats for 'system wide':

             25,717      MEM_INST_RETIRED.SPLIT_STORES    #      2.2 %
 tma_split_stores
                                                                     #
     24.0  average latency
       28,365,080      CPU_CLK_UNHALTED.THREAD

Thanks,
Namhyung

>
> Signed-off-by: Weilin Wang <weilin.wang@intel.com>
> Reviewed-by: Ian Rogers <irogers@google.com>