Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <1541587845-9150-1-git-send-email-eranian@google.com>
In-Reply-To: <1541587845-9150-1-git-send-email-eranian@google.com>
From:   Stephane Eranian <eranian@google.com>
Date:   Wed, 21 Nov 2018 14:33:23 -0800
Message-ID: <CABPqkBQrAFSJBJSGTPRaJu1_LJmurgYpF69RxWHht-+o8cvqFg@mail.gmail.com>
Subject: Re: [PATCH] perf stat: fix csv mode column output for non-cgroup events
To:     LKML <linux-kernel@vger.kernel.org>
Cc:     Arnaldo Carvalho de Melo <acme@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>, mingo@elte.hu,
        Jiri Olsa <jolsa@redhat.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Jiri,

Any update on this patch?
Thanks.
On Wed, Nov 7, 2018 at 2:50 AM Stephane Eranian <eranian@google.com> wrote:
>
> When using the -x option, perf stat prints csv-style output
> with one event per line.  For each event, it prints the count,
> the unit, the event name, the cgroup, and a bunch of other event
> specific fields (such as insn per cycles).
>
> When you use csv-style mode, you expect a normalized output where
> each event is printed with the same number of fields regardless of
> what it is so it can easily be imported into a spreadsheet or parsed.
> For instance, if an event does not have a unit, then print an empty field
> for it. Although this approach was implemented for the unit, it was not
> for the cgroup.  When mixing cgroup and non-cgroup events, then non-cgroup
> events would not show an empty field, instead the next field was printed,
> make columns not line up correctly.
>
> This patch fixes the cgroup output issues by forcing an empty field
> for non-cgroup events as soon as one event has cgroup.
>
> Before:
> <not counted> @ @cycles @foo    @ 0    @100.00@@
> 2531614       @ @cycles @6420922@100.00@    @
>
> foo cgroup lines up with time_running!
>
> After:
> <not counted> @ @cycles @foo @0       @100.00@@
> 2594834       @ @cycles @    @5287372 @100.00@@
>
> Fields line up.
>
> Signed-off-by: Stephane Eranian <eranian@google.com>
> ---
>  tools/perf/util/stat-display.c | 16 +++++++++++-----
>  1 file changed, 11 insertions(+), 5 deletions(-)
>
> diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c
> index e7b4c44ebb62..665ee374fc01 100644
> --- a/tools/perf/util/stat-display.c
> +++ b/tools/perf/util/stat-display.c
> @@ -59,6 +59,15 @@ static void print_noise(struct perf_stat_config *config,
>         print_noise_pct(config, stddev_stats(&ps->res_stats[0]), avg);
>  }
>
> +static void print_cgroup(struct perf_stat_config *config, struct perf_evsel *evsel)
> +{
> +       if (nr_cgroups) {
> +               const char *cgrp_name = evsel->cgrp ? evsel->cgrp->name  : "";
> +               fprintf(config->output, "%s%s", config->csv_sep, cgrp_name);
> +       }
> +}
> +
> +
>  static void aggr_printout(struct perf_stat_config *config,
>                           struct perf_evsel *evsel, int id, int nr)
>  {
> @@ -336,8 +345,7 @@ static void abs_printout(struct perf_stat_config *config,
>
>         fprintf(output, "%-*s", config->csv_output ? 0 : 25, perf_evsel__name(evsel));
>
> -       if (evsel->cgrp)
> -               fprintf(output, "%s%s", config->csv_sep, evsel->cgrp->name);
> +       print_cgroup(config, evsel);
>  }
>
>  static bool is_mixed_hw_group(struct perf_evsel *counter)
> @@ -431,9 +439,7 @@ static void printout(struct perf_stat_config *config, int id, int nr,
>                         config->csv_output ? 0 : -25,
>                         perf_evsel__name(counter));
>
> -               if (counter->cgrp)
> -                       fprintf(config->output, "%s%s",
> -                               config->csv_sep, counter->cgrp->name);
> +               print_cgroup(config, counter);
>
>                 if (!config->csv_output)
>                         pm(config, &os, NULL, NULL, "", 0);
> --
> 2.7.4
>