lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:   Tue, 8 Nov 2022 17:26:41 -0800
From:   Ian Rogers <irogers@...gle.com>
To:     Namhyung Kim <namhyung@...nel.org>
Cc:     Arnaldo Carvalho de Melo <acme@...nel.org>,
        Jiri Olsa <jolsa@...nel.org>, Ingo Molnar <mingo@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Adrian Hunter <adrian.hunter@...el.com>,
        linux-perf-users@...r.kernel.org,
        Kan Liang <kan.liang@...ux.intel.com>,
        Zhengjun Xing <zhengjun.xing@...ux.intel.com>,
        James Clark <james.clark@....com>
Subject: Re: [PATCH 5/9] perf stat: Fix --metric-only --json output

On Mon, Nov 7, 2022 at 1:33 PM Namhyung Kim <namhyung@...nel.org> wrote:
>
> Currently it prints all metric headers for JSON output.  But actually it
> skips some metrics with valid_only_metric().  So the output looks like:
>
>   $ perf stat --metric-only --json true
>   {"unit" : "CPUs utilized", "unit" : "/sec", "unit" : "/sec", "unit" : "/sec", "unit" : "GHz", "unit" : "insn per cycle", "unit" : "/sec", "unit" : "branch-misses of all branches"}
>   {"metric-value" : "3.861"}{"metric-value" : "0.79"}{"metric-value" : "3.04"}
>
> As you can see there are 8 units in the header but only 3 metric-values
> are there.  It should skip the unused headers as well.  Also each unit
> should be printed as a separate object like metric values.
>
> With this patch:
>
>   $ perf stat --metric-only --json true
>   {"unit" : "GHz"}{"unit" : "insn per cycle"}{"unit" : "branch-misses of all branches"}
>   {"metric-value" : "4.166"}{"metric-value" : "0.73"}{"metric-value" : "2.96"}

It looks like the correct output would be to place the unit with the
metric-value, but perhaps that's something best done by deleting the
output code and starting again - the new output is wrong but better.
We could really do with functions like emit counter, emit metric and
have appropriate output for CSV, JSON, etc. A lot of the problems in
JSON and CSV outputting come from the unnatural way it is driven and
placing a large significance on newline characters. Putting the CSV
and JSON output to stdout also is problematic, as it naturally gets
interwoven with the monitored commands output. The dictionaries in the
new output should really be in a list, this is fixed in the test here:
https://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git/tree/tools/perf/tests/shell/lib/perf_json_output_lint.py?h=perf/core#n74

Thanks,
Ian

>
> Signed-off-by: Namhyung Kim <namhyung@...nel.org>
> ---
>  tools/perf/util/stat-display.c | 22 +++-------------------
>  1 file changed, 3 insertions(+), 19 deletions(-)
>
> diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c
> index 115477461224..515fb6db3d67 100644
> --- a/tools/perf/util/stat-display.c
> +++ b/tools/perf/util/stat-display.c
> @@ -430,12 +430,12 @@ static void print_metric_header(struct perf_stat_config *config,
>             os->evsel->priv != os->evsel->evlist->selected->priv)
>                 return;
>
> -       if (!valid_only_metric(unit) && !config->json_output)
> +       if (!valid_only_metric(unit))
>                 return;
>         unit = fixunit(tbuf, os->evsel, unit);
>
>         if (config->json_output)
> -               fprintf(os->fh, "\"unit\" : \"%s\"", unit);
> +               fprintf(os->fh, "{\"unit\" : \"%s\"}", unit);
>         else if (config->csv_output)
>                 fprintf(os->fh, "%s%s", unit, config->csv_sep);
>         else
> @@ -847,10 +847,6 @@ static void print_metric_headers(struct perf_stat_config *config,
>                 .new_line = new_line_metric,
>                 .force_header = true,
>         };
> -       bool first = true;
> -
> -       if (config->json_output && !config->interval)
> -               fprintf(config->output, "{");
>
>         if (prefix && !config->json_output)
>                 fprintf(config->output, "%s", prefix);
> @@ -871,18 +867,12 @@ static void print_metric_headers(struct perf_stat_config *config,
>         evlist__for_each_entry(evlist, counter) {
>                 os.evsel = counter;
>
> -               if (!first && config->json_output)
> -                       fprintf(config->output, ", ");
> -               first = false;
> -
>                 perf_stat__print_shadow_stats(config, counter, 0,
>                                               0,
>                                               &out,
>                                               &config->metric_events,
>                                               &rt_stat);
>         }
> -       if (config->json_output)
> -               fprintf(config->output, "}");
>         fputc('\n', config->output);
>  }
>
> @@ -954,14 +944,8 @@ static void print_interval(struct perf_stat_config *config,
>                 }
>         }
>
> -       if ((num_print_interval == 0 || config->interval_clear)
> -                        && metric_only && !config->json_output)
> +       if ((num_print_interval == 0 || config->interval_clear) && metric_only)
>                 print_metric_headers(config, evlist, " ", true);
> -       if ((num_print_interval == 0 || config->interval_clear)
> -                        && metric_only && config->json_output) {
> -               fprintf(output, "{");
> -               print_metric_headers(config, evlist, " ", true);
> -       }
>         if (++num_print_interval == 25)
>                 num_print_interval = 0;
>  }
> --
> 2.38.1.431.g37b22c650d-goog
>

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ