From: Ian Rogers <irogers@google.com>
To: Namhyung Kim <namhyung@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>,
Jiri Olsa <jolsa@kernel.org>, Ingo Molnar <mingo@kernel.org>,
Peter Zijlstra <peterz@infradead.org>,
LKML <linux-kernel@vger.kernel.org>,
Adrian Hunter <adrian.hunter@intel.com>,
linux-perf-users@vger.kernel.org,
Kan Liang <kan.liang@linux.intel.com>,
Leo Yan <leo.yan@linaro.org>, Andi Kleen <ak@linux.intel.com>,
Athira Rajeev <atrajeev@linux.vnet.ibm.com>,
James Clark <james.clark@arm.com>,
Xing Zhengjun <zhengjun.xing@linux.intel.com>
Subject: Re: [PATCH 09/19] perf stat: Aggregate events using evsel->stats->aggr
Date: Mon, 10 Oct 2022 16:11:07 -0700 [thread overview]
Message-ID: <CAP-5=fVW31CH6vhw4zECP-DZbxq8c2V=+C68e5c20LD-xxYbBA@mail.gmail.com> (raw)
In-Reply-To: <20221010053600.272854-10-namhyung@kernel.org>
On Sun, Oct 9, 2022 at 10:36 PM Namhyung Kim <namhyung@kernel.org> wrote:
>
> Add a logic to aggregate counter values to the new evsel->stats->aggr.
> This is not used yet so shadow stats are not updated. But later patch
> will convert the existing code to use it.
>
> With that, we don't need to handle AGGR_GLOBAL specially anymore. It
> can use the same logic with counts, prev_counts and aggr_counts.
>
> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
> ---
> tools/perf/builtin-stat.c | 3 --
> tools/perf/util/evsel.c | 9 +---
> .../scripting-engines/trace-event-python.c | 6 ---
> tools/perf/util/stat.c | 46 ++++++++++++++++---
> 4 files changed, 41 insertions(+), 23 deletions(-)
>
> diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
> index eaddafbd7ff2..139e35ed68d3 100644
> --- a/tools/perf/builtin-stat.c
> +++ b/tools/perf/builtin-stat.c
> @@ -963,9 +963,6 @@ static int __run_perf_stat(int argc, const char **argv, int run_idx)
> init_stats(&walltime_nsecs_stats);
> update_stats(&walltime_nsecs_stats, t1 - t0);
>
> - if (stat_config.aggr_mode == AGGR_GLOBAL)
> - evlist__save_aggr_prev_raw_counts(evsel_list);
> -
> evlist__copy_prev_raw_counts(evsel_list);
> evlist__reset_prev_raw_counts(evsel_list);
> perf_stat__reset_shadow_per_stat(&rt_stat);
> diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
> index a6ea91c72659..a1fcb3166149 100644
> --- a/tools/perf/util/evsel.c
> +++ b/tools/perf/util/evsel.c
> @@ -1526,13 +1526,8 @@ void evsel__compute_deltas(struct evsel *evsel, int cpu_map_idx, int thread,
> if (!evsel->prev_raw_counts)
> return;
>
> - if (cpu_map_idx == -1) {
> - tmp = evsel->prev_raw_counts->aggr;
> - evsel->prev_raw_counts->aggr = *count;
> - } else {
> - tmp = *perf_counts(evsel->prev_raw_counts, cpu_map_idx, thread);
> - *perf_counts(evsel->prev_raw_counts, cpu_map_idx, thread) = *count;
> - }
> + tmp = *perf_counts(evsel->prev_raw_counts, cpu_map_idx, thread);
> + *perf_counts(evsel->prev_raw_counts, cpu_map_idx, thread) = *count;
>
> count->val = count->val - tmp.val;
> count->ena = count->ena - tmp.ena;
> diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c
> index 1f2040f36d4e..7bc8559dce6a 100644
> --- a/tools/perf/util/scripting-engines/trace-event-python.c
> +++ b/tools/perf/util/scripting-engines/trace-event-python.c
> @@ -1653,12 +1653,6 @@ static void python_process_stat(struct perf_stat_config *config,
> struct perf_cpu_map *cpus = counter->core.cpus;
> int cpu, thread;
>
> - if (config->aggr_mode == AGGR_GLOBAL) {
> - process_stat(counter, (struct perf_cpu){ .cpu = -1 }, -1, tstamp,
> - &counter->counts->aggr);
> - return;
> - }
> -
> for (thread = 0; thread < threads->nr; thread++) {
> for (cpu = 0; cpu < perf_cpu_map__nr(cpus); cpu++) {
> process_stat(counter, perf_cpu_map__cpu(cpus, cpu),
> diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c
> index 374149628507..99874254809d 100644
> --- a/tools/perf/util/stat.c
> +++ b/tools/perf/util/stat.c
> @@ -387,6 +387,7 @@ process_counter_values(struct perf_stat_config *config, struct evsel *evsel,
> struct perf_counts_values *count)
> {
> struct perf_counts_values *aggr = &evsel->counts->aggr;
> + struct perf_stat_evsel *ps = evsel->stats;
> static struct perf_counts_values zero;
> bool skip = false;
>
> @@ -398,6 +399,44 @@ process_counter_values(struct perf_stat_config *config, struct evsel *evsel,
> if (skip)
> count = &zero;
>
> + if (!evsel->snapshot)
> + evsel__compute_deltas(evsel, cpu_map_idx, thread, count);
> + perf_counts_values__scale(count, config->scale, NULL);
> +
> + if (ps->aggr) {
> + struct perf_cpu cpu = perf_cpu_map__cpu(evsel->core.cpus, cpu_map_idx);
> + struct aggr_cpu_id aggr_id = config->aggr_get_id(config, cpu);
> + struct perf_stat_aggr *ps_aggr;
> + int i;
> +
> + for (i = 0; i < ps->nr_aggr; i++) {
Would it be cleaner to have a helper function here that returns i or
ps_aggr for the first CPU being aggregated into? That would avoid the
continue/break.
> + if (!aggr_cpu_id__equal(&aggr_id, &config->aggr_map->map[i]))
> + continue;
> +
> + ps_aggr = &ps->aggr[i];
> + ps_aggr->nr++;
> +
> + /*
> + * When any result is bad, make them all to give
> + * consistent output in interval mode.
> + */
> + if (count->ena == 0 || count->run == 0 ||
> + evsel->counts->scaled == -1) {
> + ps_aggr->counts.val = 0;
> + ps_aggr->counts.ena = 0;
> + ps_aggr->counts.run = 0;
> + ps_aggr->failed = true;
> + }
> +
> + if (!ps_aggr->failed) {
> + ps_aggr->counts.val += count->val;
> + ps_aggr->counts.ena += count->ena;
> + ps_aggr->counts.run += count->run;
> + }
> + break;
> + }
> + }
> +
> switch (config->aggr_mode) {
> case AGGR_THREAD:
> case AGGR_CORE:
> @@ -405,9 +444,6 @@ process_counter_values(struct perf_stat_config *config, struct evsel *evsel,
> case AGGR_SOCKET:
> case AGGR_NODE:
> case AGGR_NONE:
> - if (!evsel->snapshot)
> - evsel__compute_deltas(evsel, cpu_map_idx, thread, count);
> - perf_counts_values__scale(count, config->scale, NULL);
> if ((config->aggr_mode == AGGR_NONE) && (!evsel->percore)) {
> perf_stat__update_shadow_stats(evsel, count->val,
> cpu_map_idx, &rt_stat);
> @@ -469,10 +505,6 @@ int perf_stat_process_counter(struct perf_stat_config *config,
> if (config->aggr_mode != AGGR_GLOBAL)
> return 0;
>
> - if (!counter->snapshot)
> - evsel__compute_deltas(counter, -1, -1, aggr);
> - perf_counts_values__scale(aggr, config->scale, &counter->counts->scaled);
> -
It isn't clear to me how this relates to the patch.
Thanks,
Ian
> update_stats(&ps->res_stats, *count);
>
> if (verbose > 0) {
> --
> 2.38.0.rc1.362.ged0d419d3c-goog
>
next prev parent reply other threads:[~2022-10-10 23:13 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-10-10 5:35 [RFC/PATCHSET 00/19] perf stat: Cleanup counter aggregation (v1) Namhyung Kim
2022-10-10 5:35 ` [PATCH 01/19] perf tools: Save evsel->pmu in parse_events() Namhyung Kim
2022-10-10 22:21 ` Ian Rogers
2022-10-10 5:35 ` [PATCH 02/19] perf tools: Use pmu info in evsel__is_hybrid() Namhyung Kim
2022-10-10 22:31 ` Ian Rogers
2022-10-11 5:10 ` Namhyung Kim
2022-10-10 5:35 ` [PATCH 03/19] perf stat: Use evsel__is_hybrid() more Namhyung Kim
2022-10-10 22:32 ` Ian Rogers
2022-10-10 5:35 ` [PATCH 04/19] perf stat: Add aggr id for global mode Namhyung Kim
2022-10-10 22:46 ` Ian Rogers
2022-10-11 23:08 ` Namhyung Kim
2022-10-12 10:55 ` Jiri Olsa
2022-10-12 16:31 ` Namhyung Kim
2022-10-10 5:35 ` [PATCH 05/19] perf stat: Add cpu aggr id for no aggregation mode Namhyung Kim
2022-10-10 22:49 ` Ian Rogers
2022-10-12 10:40 ` Jiri Olsa
2022-10-12 16:27 ` Namhyung Kim
2022-10-10 5:35 ` [PATCH 06/19] perf stat: Add 'needs_sort' argument to cpu_aggr_map__new() Namhyung Kim
2022-10-10 22:53 ` Ian Rogers
2022-10-11 23:32 ` Namhyung Kim
2022-10-10 5:35 ` [PATCH 07/19] perf stat: Add struct perf_stat_aggr to perf_stat_evsel Namhyung Kim
2022-10-10 23:00 ` Ian Rogers
2022-10-11 23:37 ` Namhyung Kim
2022-10-10 5:35 ` [PATCH 08/19] perf stat: Allocate evsel->stats->aggr properly Namhyung Kim
2022-10-10 23:03 ` Ian Rogers
2022-10-11 23:38 ` Namhyung Kim
2022-10-10 5:35 ` [PATCH 09/19] perf stat: Aggregate events using evsel->stats->aggr Namhyung Kim
2022-10-10 23:11 ` Ian Rogers [this message]
2022-10-11 23:44 ` Namhyung Kim
2022-10-10 5:35 ` [PATCH 10/19] perf stat: Aggregate per-thread stats " Namhyung Kim
2022-10-10 23:17 ` Ian Rogers
2022-10-11 23:46 ` Namhyung Kim
2022-10-10 5:35 ` [PATCH 11/19] perf stat: Allocate aggr counts for recorded data Namhyung Kim
2022-10-10 23:18 ` Ian Rogers
2022-10-10 5:35 ` [PATCH 12/19] perf stat: Reset aggr counts for each interval Namhyung Kim
2022-10-10 23:20 ` Ian Rogers
2022-10-11 23:48 ` Namhyung Kim
2022-10-10 5:35 ` [PATCH 13/19] perf stat: Split process_counters() Namhyung Kim
2022-10-10 23:21 ` Ian Rogers
2022-10-10 5:35 ` [PATCH 14/19] perf stat: Add perf_stat_merge_counters() Namhyung Kim
2022-10-10 23:31 ` Ian Rogers
2022-10-11 23:55 ` Namhyung Kim
2022-10-10 5:35 ` [PATCH 15/19] perf stat: Add perf_stat_process_percore() Namhyung Kim
2022-10-10 23:32 ` Ian Rogers
2022-10-11 23:59 ` Namhyung Kim
2022-10-10 5:35 ` [PATCH 16/19] perf stat: Add perf_stat_process_shadow_stats() Namhyung Kim
2022-10-10 23:36 ` Ian Rogers
2022-10-10 5:35 ` [PATCH 17/19] perf stat: Display event stats using aggr counts Namhyung Kim
2022-10-10 23:38 ` Ian Rogers
2022-10-10 5:35 ` [PATCH 18/19] perf stat: Display percore events properly Namhyung Kim
2022-10-10 23:39 ` Ian Rogers
2022-10-10 5:36 ` [PATCH 19/19] perf stat: Remove unused perf_counts.aggr field Namhyung Kim
2022-10-10 23:40 ` Ian Rogers
2022-10-12 8:41 ` Jiri Olsa
2022-10-12 16:26 ` Namhyung Kim
2022-10-13 20:56 ` [PATCH] perf stat: Init aggr_map when reporting per-process stat Namhyung Kim
2022-10-11 0:25 ` [RFC/PATCHSET 00/19] perf stat: Cleanup counter aggregation (v1) Andi Kleen
2022-10-11 5:38 ` Namhyung Kim
2022-10-11 6:13 ` Ian Rogers
2022-10-12 3:55 ` Namhyung Kim
2022-10-11 11:57 ` Andi Kleen
2022-10-12 3:58 ` Namhyung Kim
-- strict thread matches above, loose matches on Subject: below --
2022-10-14 6:15 [PATCHSET 00/19] perf stat: Cleanup counter aggregation (v2) Namhyung Kim
2022-10-14 6:15 ` [PATCH 09/19] perf stat: Aggregate events using evsel->stats->aggr Namhyung Kim
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CAP-5=fVW31CH6vhw4zECP-DZbxq8c2V=+C68e5c20LD-xxYbBA@mail.gmail.com' \
--to=irogers@google.com \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=ak@linux.intel.com \
--cc=atrajeev@linux.vnet.ibm.com \
--cc=james.clark@arm.com \
--cc=jolsa@kernel.org \
--cc=kan.liang@linux.intel.com \
--cc=leo.yan@linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.org \
--cc=zhengjun.xing@linux.intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).