Re: [PATCH v2 1/2] perf evsel: Create counts for collecting summary data

[Jiri Olsa <jolsa@redhat.com>]

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@linux.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