Message-ID: <CAP-5=fVfYR2xCi=ZiT-xmcRzq2y6L6gZg8i_DWMPzhzWyeLsgQ@mail.gmail.com>
Date:   Mon, 27 Nov 2023 14:23:57 -0800
From:   Ian Rogers <irogers@...gle.com>
To:     Arnaldo Carvalho de Melo <acme@...nel.org>
Cc:     Peter Zijlstra <peterz@...radead.org>,
        Ingo Molnar <mingo@...hat.com>,
        Mark Rutland <mark.rutland@....com>,
        Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        Jiri Olsa <jolsa@...nel.org>,
        Namhyung Kim <namhyung@...nel.org>,
        Adrian Hunter <adrian.hunter@...el.com>,
        Nick Terrell <terrelln@...com>,
        Kan Liang <kan.liang@...ux.intel.com>,
        Andi Kleen <ak@...ux.intel.com>,
        Kajol Jain <kjain@...ux.ibm.com>,
        Athira Rajeev <atrajeev@...ux.vnet.ibm.com>,
        Huacai Chen <chenhuacai@...nel.org>,
        Masami Hiramatsu <mhiramat@...nel.org>,
        Vincent Whitchurch <vincent.whitchurch@...s.com>,
        "Steinar H. Gunderson" <sesse@...gle.com>,
        Liam Howlett <liam.howlett@...cle.com>,
        Miguel Ojeda <ojeda@...nel.org>,
        Colin Ian King <colin.i.king@...il.com>,
        Dmitrii Dolgov <9erthalion6@...il.com>,
        Yang Jihong <yangjihong1@...wei.com>,
        Ming Wang <wangming01@...ngson.cn>,
        James Clark <james.clark@....com>,
        K Prateek Nayak <kprateek.nayak@....com>,
        Sean Christopherson <seanjc@...gle.com>,
        Leo Yan <leo.yan@...aro.org>,
        Ravi Bangoria <ravi.bangoria@....com>,
        German Gomez <german.gomez@....com>,
        Changbin Du <changbin.du@...wei.com>,
        Paolo Bonzini <pbonzini@...hat.com>, Li Dong <lidong@...o.com>,
        Sandipan Das <sandipan.das@....com>,
        liuwenyu <liuwenyu7@...wei.com>, linux-kernel@...r.kernel.org,
        linux-perf-users@...r.kernel.org
Subject: Re: [PATCH v4 10/53] perf record: Be lazier in allocating lost
 samples buffer

On Mon, Nov 27, 2023 at 2:03 PM Arnaldo Carvalho de Melo
<acme@...nel.org> wrote:
>
> Em Thu, Nov 02, 2023 at 10:56:52AM -0700, Ian Rogers escreveu:
> > Wait until a lost sample occurs to allocate the lost samples buffer,
> > often the buffer isn't necessary. This saves a 64kb allocation and
> > 5.3kb of peak memory consumption.
> >
> > Signed-off-by: Ian Rogers <irogers@...gle.com>
> > ---
> >  tools/perf/builtin-record.c | 29 +++++++++++++++++++----------
> >  1 file changed, 19 insertions(+), 10 deletions(-)
> >
> > diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
> > index 9b4f3805ca92..b6c8c1371b39 100644
> > --- a/tools/perf/builtin-record.c
> > +++ b/tools/perf/builtin-record.c
> > @@ -1924,21 +1924,13 @@ static void __record__save_lost_samples(struct record *rec, struct evsel *evsel,
> >  static void record__read_lost_samples(struct record *rec)
> >  {
> >       struct perf_session *session = rec->session;
> > -     struct perf_record_lost_samples *lost;
> > +     struct perf_record_lost_samples *lost = NULL;
> >       struct evsel *evsel;
> >
> >       /* there was an error during record__open */
> >       if (session->evlist == NULL)
> >               return;
> >
> > -     lost = zalloc(PERF_SAMPLE_MAX_SIZE);
> > -     if (lost == NULL) {
> > -             pr_debug("Memory allocation failed\n");
> > -             return;
> > -     }
>
> Shouldn't we take the time here and instead improve this error message
> and then propagate the error?
>
> For instance, we may want to still get some perf.data file without these
> records but inform the user at this point how many records were lost
> (count.lost)?

Sounds like a follow-up, the messages here are just moving the
existing message and the point of the patch is to postpone/avoid a
memory allocation when possible.

Thanks,
Ian

> - Arnaldo
>
> > -
> > -     lost->header.type = PERF_RECORD_LOST_SAMPLES;
> > -
> >       evlist__for_each_entry(session->evlist, evsel) {
> >               struct xyarray *xy = evsel->core.sample_id;
> >               u64 lost_count;
> > @@ -1961,6 +1953,14 @@ static void record__read_lost_samples(struct record *rec)
> >                               }
> >
> >                               if (count.lost) {
> > +                                     if (!lost) {
> > +                                             lost = zalloc(PERF_SAMPLE_MAX_SIZE);
> > +                                             if (!lost) {
> > +                                                     pr_debug("Memory allocation failed\n");
> > +                                                     return;
> > +                                             }
> > +                                             lost->header.type = PERF_RECORD_LOST_SAMPLES;
> > +                                     }
> >                                       __record__save_lost_samples(rec, evsel, lost,
> >                                                                   x, y, count.lost, 0);
> >                               }
> > @@ -1968,9 +1968,18 @@ static void record__read_lost_samples(struct record *rec)
> >               }
> >
> >               lost_count = perf_bpf_filter__lost_count(evsel);
> > -             if (lost_count)
> > +             if (lost_count) {
> > +                     if (!lost) {
> > +                             lost = zalloc(PERF_SAMPLE_MAX_SIZE);
> > +                             if (!lost) {
> > +                                     pr_debug("Memory allocation failed\n");
> > +                                     return;
> > +                             }
> > +                             lost->header.type = PERF_RECORD_LOST_SAMPLES;
> > +                     }
> >                       __record__save_lost_samples(rec, evsel, lost, 0, 0, lost_count,
> >                                                   PERF_RECORD_MISC_LOST_SAMPLES_BPF);
> > +             }
> >       }
> >  out:
> >       free(lost);
> > --
> > 2.42.0.869.gea05f2083d-goog
> >
>
> --
>
> - Arnaldo

Powered by blists - more mailing lists