[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200504235059.GG1916255@krava>
Date: Tue, 5 May 2020 01:50:59 +0200
From: Jiri Olsa <jolsa@...hat.com>
To: Jin Yao <yao.jin@...ux.intel.com>
Cc: acme@...nel.org, jolsa@...nel.org, peterz@...radead.org,
mingo@...hat.com, alexander.shishkin@...ux.intel.com,
Linux-kernel@...r.kernel.org, ak@...ux.intel.com,
kan.liang@...el.com, yao.jin@...el.com
Subject: Re: [PATCH v2 1/2] perf evsel: Create counts for collecting summary
data
On Sat, May 02, 2020 at 10:07:04AM +0800, Jin Yao wrote:
> It would be useful to support the overall statistics for perf-stat
> interval mode. For example, report the summary at the end of
> "perf-stat -I" output.
>
> But since perf-stat can support many aggregation modes, such as
> --per-thread, --per-socket, -M and etc, we need a solution which
> doesn't bring much complexity.
>
> The idea is to create new 'evsel->summary_counts' which sums up the
> counts delta per interval. Before reporting the summary, we copy the
> data from evsel->summary_counts to evsel->counts, and next we just
> follow current code.
>
> v2:
> ---
> Rebase to perf/core branch
>
> Signed-off-by: Jin Yao <yao.jin@...ux.intel.com>
> ---
> tools/perf/util/evsel.c | 10 ++++++++--
> tools/perf/util/evsel.h | 1 +
> tools/perf/util/stat.c | 20 ++++++++++++++++++++
> 3 files changed, 29 insertions(+), 2 deletions(-)
>
> diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
> index a75bcb95bf23..abc503dd6eda 100644
> --- a/tools/perf/util/evsel.c
> +++ b/tools/perf/util/evsel.c
> @@ -1280,22 +1280,28 @@ void evsel__delete(struct evsel *evsel)
> void evsel__compute_deltas(struct evsel *evsel, int cpu, int thread,
> struct perf_counts_values *count)
> {
> - struct perf_counts_values tmp;
> + struct perf_counts_values tmp, *summary;
>
> - if (!evsel->prev_raw_counts)
> + if (!evsel->prev_raw_counts || !evsel->summary_counts)
> return;
>
> if (cpu == -1) {
> tmp = evsel->prev_raw_counts->aggr;
> evsel->prev_raw_counts->aggr = *count;
> + summary = &evsel->summary_counts->aggr;
> } else {
> tmp = *perf_counts(evsel->prev_raw_counts, cpu, thread);
> *perf_counts(evsel->prev_raw_counts, cpu, thread) = *count;
> + summary = perf_counts(evsel->summary_counts, cpu, thread);
shouldn't this be enough?
perf_counts(evsel->summary_counts, cpu, thread) = *count
without the code below.. and similar for aggr case
however I still wonder if we should count this in
perf_stat_process_counter and only for interval mode
> }
>
> count->val = count->val - tmp.val;
> count->ena = count->ena - tmp.ena;
> count->run = count->run - tmp.run;
> +
> + summary->val += count->val;
> + summary->ena += count->ena;
> + summary->run += count->run;
> }
>
> void perf_counts_values__scale(struct perf_counts_values *count,
> diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
> index 783246bf8d0d..430639c99d04 100644
> --- a/tools/perf/util/evsel.h
> +++ b/tools/perf/util/evsel.h
> @@ -46,6 +46,7 @@ struct evsel {
> char *filter;
> struct perf_counts *counts;
> struct perf_counts *prev_raw_counts;
> + struct perf_counts *summary_counts;
'sum_counts' might be better
jirka
> int idx;
> unsigned long max_events;
> unsigned long nr_events_printed;
> diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c
> index 774468341851..c3fd008b4e84 100644
> --- a/tools/perf/util/stat.c
SNIP
Powered by blists - more mailing lists