From: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
To: Namhyung Kim <namhyung.kim@lge.com>
Cc: Namhyung Kim <namhyung@gmail.com>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Paul Mackerras <paulus@samba.org>, Ingo Molnar <mingo@elte.hu>,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 11/11] perf record: Fix event grouping on forked task
Date: Mon, 13 Feb 2012 16:42:33 -0200 [thread overview]
Message-ID: <20120213184233.GK15955@infradead.org> (raw)
In-Reply-To: <1329118064-9412-12-git-send-email-namhyung.kim@lge.com>
Em Mon, Feb 13, 2012 at 04:27:43PM +0900, Namhyung Kim escreveu:
> When event group is enabled for forked task (i.e. no target task/cpu
> was specified) all events were disabled and marked ->enable_on_exec.
> However they wouldn't be counted at all since only group leader will
> be enabled on exec actually.
>
> In contrast to perf stat, perf record doesn't have a real problem
> as it enables all the event before proceeding. But it needs to be
> fixed anyway IMHO.
>
> Signed-off-by: Namhyung Kim <namhyung.kim@lge.com>
> ---
> tools/perf/builtin-record.c | 2 +-
> tools/perf/builtin-test.c | 2 +-
> tools/perf/util/evlist.c | 5 +++--
> tools/perf/util/evlist.h | 3 ++-
> tools/perf/util/evsel.c | 5 +++--
> tools/perf/util/evsel.h | 3 ++-
> 6 files changed, 12 insertions(+), 8 deletions(-)
>
> diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
> index fbf71edd31ba..a65b0a5a81ea 100644
> --- a/tools/perf/builtin-record.c
> +++ b/tools/perf/builtin-record.c
> @@ -183,7 +183,7 @@ static void perf_record__open(struct perf_record *rec)
>
> first = list_entry(evlist->entries.next, struct perf_evsel, node);
>
> - perf_evlist__config_attrs(evlist, opts);
> + perf_evlist__config_attrs(evlist, opts, first);
No need to pass the first entry in the evlist, since we can find it at
perf_evlist__config_attrs using:
list_entry(evlist->entries.next, struct perf_evsel, node);
That way you reduce the patch size by not touching all the callers of
perf_evlist__config_attrs.
- Arnaldo
> list_for_each_entry(pos, &evlist->entries, node) {
> struct perf_event_attr *attr = &pos->attr;
> diff --git a/tools/perf/builtin-test.c b/tools/perf/builtin-test.c
> index 91a6eba000f2..b72b12aedd29 100644
> --- a/tools/perf/builtin-test.c
> +++ b/tools/perf/builtin-test.c
> @@ -1083,7 +1083,7 @@ static int test__PERF_RECORD(void)
> evsel->attr.sample_type |= PERF_SAMPLE_CPU;
> evsel->attr.sample_type |= PERF_SAMPLE_TID;
> evsel->attr.sample_type |= PERF_SAMPLE_TIME;
> - perf_evlist__config_attrs(evlist, &opts);
> + perf_evlist__config_attrs(evlist, &opts, evsel);
>
> err = sched__get_first_possible_cpu(evlist->workload.pid, &cpu_mask,
> &cpu_mask_size);
> diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
> index aa0418034d82..7ab81906c1aa 100644
> --- a/tools/perf/util/evlist.c
> +++ b/tools/perf/util/evlist.c
> @@ -49,7 +49,8 @@ struct perf_evlist *perf_evlist__new(struct cpu_map *cpus,
> }
>
> void perf_evlist__config_attrs(struct perf_evlist *evlist,
> - struct perf_record_opts *opts)
> + struct perf_record_opts *opts,
> + struct perf_evsel *first)
> {
> struct perf_evsel *evsel;
>
> @@ -57,7 +58,7 @@ void perf_evlist__config_attrs(struct perf_evlist *evlist,
> opts->no_inherit = true;
>
> list_for_each_entry(evsel, &evlist->entries, node) {
> - perf_evsel__config(evsel, opts);
> + perf_evsel__config(evsel, opts, first);
>
> if (evlist->nr_entries > 1)
> evsel->attr.sample_type |= PERF_SAMPLE_ID;
> diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
> index 7e99ac513893..7ebb6419e8d0 100644
> --- a/tools/perf/util/evlist.h
> +++ b/tools/perf/util/evlist.h
> @@ -81,7 +81,8 @@ union perf_event *perf_evlist__mmap_read(struct perf_evlist *self, int idx);
> int perf_evlist__open(struct perf_evlist *evlist, bool group);
>
> void perf_evlist__config_attrs(struct perf_evlist *evlist,
> - struct perf_record_opts *opts);
> + struct perf_record_opts *opts,
> + struct perf_evsel *first);
>
> int perf_evlist__prepare_workload(struct perf_evlist *evlist,
> struct perf_record_opts *opts,
> diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
> index e7ddf27f240b..cb5c18b66d06 100644
> --- a/tools/perf/util/evsel.c
> +++ b/tools/perf/util/evsel.c
> @@ -63,7 +63,8 @@ struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr, int idx)
> return evsel;
> }
>
> -void perf_evsel__config(struct perf_evsel *evsel, struct perf_record_opts *opts)
> +void perf_evsel__config(struct perf_evsel *evsel, struct perf_record_opts *opts,
> + struct perf_evsel *first)
> {
> struct perf_event_attr *attr = &evsel->attr;
> int track = !evsel->idx; /* only the first counter needs these */
> @@ -130,7 +131,7 @@ void perf_evsel__config(struct perf_evsel *evsel, struct perf_record_opts *opts)
> attr->mmap = track;
> attr->comm = track;
>
> - if (no_target_maps(&opts->maps)) {
> + if (no_target_maps(&opts->maps) && (!opts->group || evsel == first)) {
> attr->disabled = 1;
> attr->enable_on_exec = 1;
> }
> diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
> index 326b8e4d5035..3158ca3d69a1 100644
> --- a/tools/perf/util/evsel.h
> +++ b/tools/perf/util/evsel.h
> @@ -80,7 +80,8 @@ void perf_evsel__exit(struct perf_evsel *evsel);
> void perf_evsel__delete(struct perf_evsel *evsel);
>
> void perf_evsel__config(struct perf_evsel *evsel,
> - struct perf_record_opts *opts);
> + struct perf_record_opts *opts,
> + struct perf_evsel *first);
>
> int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads);
> int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads);
> --
> 1.7.9
prev parent reply other threads:[~2012-02-13 18:42 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <1329118064-9412-1-git-send-email-namhyung.kim@lge.com>
2012-02-13 7:27 ` [PATCH 01/11] perf tools: Introduce struct perf_maps_opts Namhyung Kim
2012-02-13 7:44 ` [RFC PATCHSET] perf: Fix cpu/thread map and group event handling Namhyung Kim
2012-02-13 18:32 ` [PATCH 01/11] perf tools: Introduce struct perf_maps_opts Arnaldo Carvalho de Melo
2012-02-13 18:50 ` David Ahern
2012-02-13 19:05 ` Arnaldo Carvalho de Melo
2012-02-13 19:19 ` David Ahern
2012-02-13 20:12 ` Arnaldo Carvalho de Melo
2012-02-13 7:27 ` [PATCH 02/11] perf stat: Convert to perf_maps_opts Namhyung Kim
2012-02-13 18:33 ` Arnaldo Carvalho de Melo
2012-02-13 7:27 ` [PATCH 03/11] perf top: " Namhyung Kim
2012-02-13 7:27 ` [PATCH 04/11] perf tools: Introduce check_target_maps() helper Namhyung Kim
2012-02-13 18:36 ` Arnaldo Carvalho de Melo
2012-02-13 7:27 ` [PATCH 05/11] perf tools: Make perf_evlist__create_maps() take struct perf_maps_opts Namhyung Kim
2012-02-13 18:36 ` Arnaldo Carvalho de Melo
2012-02-13 7:27 ` [PATCH 06/11] perf tools: Check more combinations of PID/TID, UID and CPU switches Namhyung Kim
2012-02-13 7:27 ` [PATCH 07/11] perf tools: Fix creation of cpu map Namhyung Kim
2012-02-13 7:27 ` [PATCH 08/11] perf tools: Consolidate target task/cpu checking Namhyung Kim
2012-02-13 18:39 ` Arnaldo Carvalho de Melo
2012-02-13 7:27 ` [PATCH 09/11] perf stat: Use perf_evlist__create_maps Namhyung Kim
2012-02-13 18:40 ` Arnaldo Carvalho de Melo
2012-02-13 7:27 ` [PATCH 10/11] perf stat: Fix event grouping on forked task Namhyung Kim
2012-02-13 18:41 ` Arnaldo Carvalho de Melo
2012-02-14 1:20 ` [PATCH] " Namhyung Kim
2012-02-13 7:27 ` [PATCH 11/11] perf record: " Namhyung Kim
2012-02-13 18:42 ` Arnaldo Carvalho de Melo [this message]
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=20120213184233.GK15955@infradead.org \
--to=acme@ghostprotocols.net \
--cc=a.p.zijlstra@chello.nl \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=namhyung.kim@lge.com \
--cc=namhyung@gmail.com \
--cc=paulus@samba.org \
/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
Powered by JetHome