[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CABPqkBSTK+Z2TAxjYb3ysyq0iz+sA+O-M2-TQz9QBvwK6g8SOA@mail.gmail.com>
Date: Fri, 23 Jul 2021 09:07:35 -0700
From: Stephane Eranian <eranian@...gle.com>
To: Namhyung Kim <namhyung@...nel.org>
Cc: Arnaldo Carvalho de Melo <acme@...nel.org>,
Jiri Olsa <jolsa@...hat.com>, Ingo Molnar <mingo@...nel.org>,
Peter Zijlstra <peterz@...radead.org>,
LKML <linux-kernel@...r.kernel.org>,
Andi Kleen <ak@...ux.intel.com>,
Ian Rogers <irogers@...gle.com>,
Adrian Hunter <adrian.hunter@...el.com>,
Jin Yao <yao.jin@...ux.intel.com>
Subject: Re: [PATCH 2/2] perf record: Add --synth option
On Thu, Jul 22, 2021 at 7:40 PM Namhyung Kim <namhyung@...nel.org> wrote:
>
> Add an option to control synthesize behavior.
>
> --synth <no|all|task|mmap|cgroup>
> Fine-tune event synthesis: default=all
>
> This can be useful when we know it doesn't need some synthesis like
> in a specific usecase and/or when using pipe:
>
> $ perf record -a --all-cgroups --synth cgroup -o- sleep 1 | \
> > perf report -i- -s cgroup
>
> Signed-off-by: Namhyung Kim <namhyung@...nel.org>
> ---
> tools/perf/Documentation/perf-record.txt | 4 ++
> tools/perf/builtin-record.c | 48 +++++++++++++++++++-----
> tools/perf/util/record.h | 1 +
> tools/perf/util/synthetic-events.c | 28 ++++++++++++++
> tools/perf/util/synthetic-events.h | 12 ++++++
> 5 files changed, 84 insertions(+), 9 deletions(-)
>
> diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
> index d71bac847936..421a7b93c4b5 100644
> --- a/tools/perf/Documentation/perf-record.txt
> +++ b/tools/perf/Documentation/perf-record.txt
> @@ -596,6 +596,10 @@ options.
> 'perf record --dry-run -e' can act as a BPF script compiler if llvm.dump-obj
> in config file is set to true.
>
> +--synth=TYPE::
> +Collect and synthesize given type of events. Available types are:
> +task, mmap, cgroup, all, no
> +
I would explain what each option disables in the synthesis. It is not
clear in your description
or your explain in the commit msg.
>
> --tail-synthesize::
> Instead of collecting non-sample events (for example, fork, comm, mmap) at
> the beginning of record, collect them during finalizing an output file.
> diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
> index 535da4dfd8d3..caaada3594ce 100644
> --- a/tools/perf/builtin-record.c
> +++ b/tools/perf/builtin-record.c
> @@ -1254,6 +1254,7 @@ static int record__synthesize_workload(struct record *rec, bool tail)
> {
> int err;
> struct perf_thread_map *thread_map;
> + bool needs_mmap = rec->opts.synth & PERF_SYNTH_MMAP;
>
> if (rec->opts.tail_synthesize != tail)
> return 0;
> @@ -1265,7 +1266,7 @@ static int record__synthesize_workload(struct record *rec, bool tail)
> err = perf_event__synthesize_thread_map(&rec->tool, thread_map,
> process_synthesized_event,
> &rec->session->machines.host,
> - true,
> + needs_mmap,
> rec->opts.sample_address);
> perf_thread_map__put(thread_map);
> return err;
> @@ -1500,20 +1501,26 @@ static int record__synthesize(struct record *rec, bool tail)
> if (err < 0)
> pr_warning("Couldn't synthesize bpf events.\n");
>
> - err = perf_event__synthesize_cgroups(tool, process_synthesized_event,
> - machine);
> - if (err < 0)
> - pr_warning("Couldn't synthesize cgroup events.\n");
> + if (rec->opts.synth & PERF_SYNTH_CGROUP) {
> + err = perf_event__synthesize_cgroups(tool, process_synthesized_event,
> + machine);
> + if (err < 0)
> + pr_warning("Couldn't synthesize cgroup events.\n");
> + }
>
> if (rec->opts.nr_threads_synthesize > 1) {
> perf_set_multithreaded();
> f = process_locked_synthesized_event;
> }
>
> - err = __machine__synthesize_threads(machine, tool, &opts->target,
> - rec->evlist->core.threads,
> - f, true, opts->sample_address,
> - rec->opts.nr_threads_synthesize);
> + if (rec->opts.synth & PERF_SYNTH_TASK) {
> + bool needs_mmap = rec->opts.synth & PERF_SYNTH_MMAP;
> +
> + err = __machine__synthesize_threads(machine, tool, &opts->target,
> + rec->evlist->core.threads,
> + f, needs_mmap, opts->sample_address,
> + rec->opts.nr_threads_synthesize);
> + }
>
> if (rec->opts.nr_threads_synthesize > 1)
> perf_set_singlethreaded();
> @@ -2422,6 +2429,26 @@ static int process_timestamp_boundary(struct perf_tool *tool,
> return 0;
> }
>
> +static int parse_record_synth_option(const struct option *opt,
> + const char *str,
> + int unset __maybe_unused)
> +{
> + struct record_opts *opts = opt->value;
> + char *p = strdup(str);
> +
> + if (p == NULL)
> + return -1;
> +
> + opts->synth = parse_synth_opt(p);
> + free(p);
> +
> + if (opts->synth < 0) {
> + pr_err("Invalid synth option: %s\n", str);
> + return -1;
> + }
> + return 0;
> +}
> +
> /*
> * XXX Ideally would be local to cmd_record() and passed to a record__new
> * because we need to have access to it in record__exit, that is called
> @@ -2447,6 +2474,7 @@ static struct record record = {
> .nr_threads_synthesize = 1,
> .ctl_fd = -1,
> .ctl_fd_ack = -1,
> + .synth = PERF_SYNTH_ALL,
> },
> .tool = {
> .sample = process_sample_event,
> @@ -2662,6 +2690,8 @@ static struct option __record_options[] = {
> "\t\t\t Optionally send control command completion ('ack\\n') to ack-fd descriptor.\n"
> "\t\t\t Alternatively, ctl-fifo / ack-fifo will be opened and used as ctl-fd / ack-fd.",
> parse_control_option),
> + OPT_CALLBACK(0, "synth", &record.opts, "no|all|task|mmap|cgroup",
> + "Fine-tune event synthesis: default=all", parse_record_synth_option),
> OPT_END()
> };
>
> diff --git a/tools/perf/util/record.h b/tools/perf/util/record.h
> index 68f471d9a88b..ef6c2715fdd9 100644
> --- a/tools/perf/util/record.h
> +++ b/tools/perf/util/record.h
> @@ -77,6 +77,7 @@ struct record_opts {
> int ctl_fd;
> int ctl_fd_ack;
> bool ctl_fd_close;
> + int synth;
> };
>
> extern const char * const *record_usage;
> diff --git a/tools/perf/util/synthetic-events.c b/tools/perf/util/synthetic-events.c
> index 566e0859fcfb..9485d0532b9c 100644
> --- a/tools/perf/util/synthetic-events.c
> +++ b/tools/perf/util/synthetic-events.c
> @@ -2186,3 +2186,31 @@ int perf_event__synthesize_features(struct perf_tool *tool, struct perf_session
> free(ff.buf);
> return ret;
> }
> +
> +int parse_synth_opt(char *synth)
> +{
> + char *p, *q;
> + int ret = 0;
> +
> + if (synth == NULL)
> + return -1;
> +
> + for (q = synth; (p = strsep(&q, ",")); p = q) {
> + if (!strcasecmp(p, "no") || !strcasecmp(p, "none"))
> + return 0;
> +
> + if (!strcasecmp(p, "all"))
> + return PERF_SYNTH_ALL;
> +
> + if (!strcasecmp(p, "task"))
> + ret |= PERF_SYNTH_TASK;
> + else if (!strcasecmp(p, "mmap"))
> + ret |= PERF_SYNTH_TASK | PERF_SYNTH_MMAP;
> + else if (!strcasecmp(p, "cgroup"))
> + ret |= PERF_SYNTH_CGROUP;
> + else
> + return -1;
> + }
> +
> + return ret;
> +}
> diff --git a/tools/perf/util/synthetic-events.h b/tools/perf/util/synthetic-events.h
> index 61bbdb3b64df..913803506345 100644
> --- a/tools/perf/util/synthetic-events.h
> +++ b/tools/perf/util/synthetic-events.h
> @@ -26,6 +26,18 @@ struct target;
>
> union perf_event;
>
> +enum perf_record_synth {
> + PERF_SYNTH_TASK = 1 << 0,
> + PERF_SYNTH_MMAP = 1 << 1,
> + PERF_SYNTH_CGROUP = 1 << 2,
> +
> + /* last element */
> + PERF_SYNTH_MAX = 1 << 3,
> +};
> +#define PERF_SYNTH_ALL (PERF_SYNTH_MAX - 1)
> +
> +int parse_synth_opt(char *str);
> +
> typedef int (*perf_event__handler_t)(struct perf_tool *tool, union perf_event *event,
> struct perf_sample *sample, struct machine *machine);
>
> --
> 2.32.0.432.gabb21c7263-goog
>
Powered by blists - more mailing lists