linux-perf-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Adrian Hunter <adrian.hunter@intel.com>
To: Leo Yan <leo.yan@arm.com>, Peter Zijlstra <peterz@infradead.org>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Namhyung Kim <namhyung@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>, Jiri Olsa <jolsa@kernel.org>,
	Ian Rogers <irogers@google.com>,
	"Liang, Kan" <kan.liang@linux.intel.com>,
	Suzuki K Poulose <suzuki.poulose@arm.com>,
	Mike Leach <mike.leach@linaro.org>,
	James Clark <james.clark@linaro.org>,
	John Garry <john.g.garry@oracle.com>,
	Will Deacon <will@kernel.org>,
	Yicong Yang <yangyicong@hisilicon.com>,
	Jonathan Cameron <jonathan.cameron@huawei.com>,
	linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org, coresight@lists.linaro.org
Subject: Re: [PATCH v6 4/8] perf auxtrace: Introduce auxtrace_record__validate_events()
Date: Tue, 3 Sep 2024 18:26:54 +0300	[thread overview]
Message-ID: <3f03541e-6dab-472f-bad9-4cdc0c0dc061@intel.com> (raw)
In-Reply-To: <20240823113306.2310957-5-leo.yan@arm.com>

On 23/08/24 14:33, Leo Yan wrote:
> A prerequisite for multiple AUX events is that the AUX events cannot
> overlap CPU maps. The reason is that every CPU has only one AUX trace
> buffer and maps it to an unique buffer index for CPU and system tracing
> mode.
> 
> To prevent the case of CPU maps overlapping occurring within multiple
> AUX events, the auxtrace_record__validate_events() function is
> introduced. It iterates through all AUX events and returns failure if
> it detects CPU maps overlapping.
> 
> Signed-off-by: Leo Yan <leo.yan@arm.com>
> ---
>  tools/perf/builtin-record.c |  4 +++
>  tools/perf/util/auxtrace.c  | 64 +++++++++++++++++++++++++++++++++++++
>  tools/perf/util/auxtrace.h  |  7 ++++
>  3 files changed, 75 insertions(+)
> 
> diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
> index adbaf80b398c..2c618efba97d 100644
> --- a/tools/perf/builtin-record.c
> +++ b/tools/perf/builtin-record.c
> @@ -862,6 +862,10 @@ static int record__auxtrace_init(struct record *rec)
>  
>  	auxtrace_regroup_aux_output(rec->evlist);
>  
> +	err = auxtrace_validate_events(rec->evlist);
> +	if (err)
> +		return err;
> +
>  	return auxtrace_parse_filters(rec->evlist);
>  }
>  
> diff --git a/tools/perf/util/auxtrace.c b/tools/perf/util/auxtrace.c
> index ca8682966fae..87e4f21b6edf 100644
> --- a/tools/perf/util/auxtrace.c
> +++ b/tools/perf/util/auxtrace.c
> @@ -2828,6 +2828,70 @@ int auxtrace_parse_filters(struct evlist *evlist)
>  	return 0;
>  }
>  
> +int auxtrace_validate_events(struct evlist *evlist)

'auxtrace_validate_aux_events' would better indicate that it is
looking only at AUX area events.

> +{
> +	struct evsel *evsel;
> +	struct perf_cpu_map *cpu_map = NULL;
> +	struct perf_cpu_map *cpu_map_intersect = NULL;
> +	struct perf_cpu_map *cpu_map_merged = NULL;
> +	int ret = 0;
> +
> +	if (!evlist)
> +		return 0;

Elsewhere we assume it is not NULL, might as well here too.

> +
> +	/*
> +	 * Currently the tool only supports multiple AUX events without
> +	 * overlapping CPU maps and every CPU has its unique AUX buffer
> +	 * for CPU or system mode tracing.
> +	 *
> +	 * Returns failure if detects CPU maps overlapping.
> +	 */
> +	evlist__for_each_entry(evlist, evsel) {
> +		if (!evsel__is_aux_event(evsel))
> +			continue;
> +
> +		if (perf_cpu_map__is_empty(evsel->pmu->cpus))
> +			continue;

Unless perf_cpu_map__intersect() is broken, the empty check
should not be needed.

Shouldn't we be looking at evsel->cpus ?

Possibly need to consider the perf_cpu_map__has_any_cpu() case?
e.g.
		if (cpu_map && (perf_cpu_map__has_any_cpu(evsel->cpus) || 
				perf_cpu_map__has_any_cpu(cpu_map)) {
			ret = -EINVAL;
			break;
		}

> +
> +		cpu_map_intersect = perf_cpu_map__intersect(cpu_map, evsel->pmu->cpus);
> +		if (cpu_map_intersect) {
> +			perf_cpu_map__put(cpu_map_intersect);
> +			pr_err("Doesn't support AUX events with overlapping CPU masks\n");
> +			ret = -EINVAL;
> +			break;
> +		}
> +		perf_cpu_map__put(cpu_map_intersect);

Maybe add a helper:

static bool perf_cpu_map__do_maps_intersect(struct perf_cpu_map *a, struct perf_cpu_map *b)
{
	struct perf_cpu_map *intersection = perf_cpu_map__intersect(a, b);
	bool ret = !perf_cpu_map__is_empty(intersection);

	perf_cpu_map__put(intersection);

	return ret;
}

> +
> +		cpu_map_merged = perf_cpu_map__merge(cpu_map, evsel->pmu->cpus);
> +		if (!cpu_map_merged) {
> +			ret = -ENOMEM;
> +			break;
> +		}
> +
> +		/* Update the CPU maps after merging */
> +		perf_cpu_map__put(cpu_map);
> +		cpu_map = cpu_map_merged;

perf_cpu_map__merge() is a bit tricky - see its comments.  This
should probably all just be:

		cpu_map = perf_cpu_map__merge(cpu_map, evsel->pmu->cpus);


> +	}
> +
> +	if (!ret)
> +		goto out;

Could we put the error path last i.e.

	perf_cpu_map__put(cpu_map);

	if (ret)
		goto out_err;

	return 0;

out_err:
> +
> +	/* If fails, dump CPU maps for debugging */
> +	evlist__for_each_entry(evlist, evsel) {
> +		char buf[200];
> +
> +		if (!evsel__is_aux_event(evsel))
> +			continue;
> +
> +		cpu_map__snprint(evsel->pmu->cpus, buf, sizeof(buf));
> +		pr_debug("AUX event [%s]'s cpu map is: %s\n", evsel->pmu->name, buf);

Could probably use cpu_map__fprintf(pmu->cpus, debug_file()) and
not need buf.

> +	}
> +
> +out:
> +	perf_cpu_map__put(cpu_map);
> +	return ret;
> +}
> +
>  int auxtrace__process_event(struct perf_session *session, union perf_event *event,
>  			    struct perf_sample *sample, const struct perf_tool *tool)
>  {
> diff --git a/tools/perf/util/auxtrace.h b/tools/perf/util/auxtrace.h
> index a1895a4f530b..67a74ad0c383 100644
> --- a/tools/perf/util/auxtrace.h
> +++ b/tools/perf/util/auxtrace.h
> @@ -636,6 +636,7 @@ void addr_filters__exit(struct addr_filters *filts);
>  int addr_filters__parse_bare_filter(struct addr_filters *filts,
>  				    const char *filter);
>  int auxtrace_parse_filters(struct evlist *evlist);
> +int auxtrace_validate_events(struct evlist *evlist);
>  
>  int auxtrace__process_event(struct perf_session *session, union perf_event *event,
>  			    struct perf_sample *sample, const struct perf_tool *tool);
> @@ -875,6 +876,12 @@ int auxtrace_parse_filters(struct evlist *evlist __maybe_unused)
>  	return 0;
>  }
>  
> +static inline
> +int auxtrace_validate_events(struct evlist *evlist __maybe_unused)
> +{
> +	return 0;
> +}
> +
>  int auxtrace_mmap__mmap(struct auxtrace_mmap *mm,
>  			struct auxtrace_mmap_params *mp,
>  			void *userpg, int fd);


  reply	other threads:[~2024-09-03 15:27 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-08-23 11:32 [PATCH v6 0/8] perf auxtrace: Support multiple AUX events Leo Yan
2024-08-23 11:32 ` [PATCH v6 1/8] perf/core: Allow multiple AUX PMU events with the same module Leo Yan
2024-08-23 11:40   ` Leo Yan
2024-09-03 10:06   ` Adrian Hunter
2024-09-04 19:35     ` Leo Yan
2024-09-05  7:13       ` Adrian Hunter
2024-08-23 11:33 ` [PATCH v6 2/8] perf auxtrace: Use evsel__is_aux_event() for checking AUX event Leo Yan
2024-08-23 11:33 ` [PATCH v6 3/8] perf auxtrace: Remove unused 'pmu' pointer from struct auxtrace_record Leo Yan
2024-08-23 11:33 ` [PATCH v6 4/8] perf auxtrace: Introduce auxtrace_record__validate_events() Leo Yan
2024-09-03 15:26   ` Adrian Hunter [this message]
2024-09-04 21:13     ` Leo Yan
2024-09-05  6:44       ` Adrian Hunter
2024-08-23 11:33 ` [PATCH v6 5/8] perf auxtrace: Refactor evlist__enable_event_idx() Leo Yan
2024-09-03 18:39   ` Adrian Hunter
2024-08-23 11:33 ` [PATCH v6 6/8] perf auxtrace: Bails out after finding the event for the map index Leo Yan
2024-09-03 18:41   ` Adrian Hunter
2024-08-23 11:33 ` [PATCH v6 7/8] perf auxtrace: Iterate all AUX events when finish reading Leo Yan
2024-08-23 11:33 ` [PATCH v6 8/8] perf arm-spe: Support multiple events in arm_spe_evsel_is_auxtrace() Leo Yan

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=3f03541e-6dab-472f-bad9-4cdc0c0dc061@intel.com \
    --to=adrian.hunter@intel.com \
    --cc=acme@kernel.org \
    --cc=coresight@lists.linaro.org \
    --cc=irogers@google.com \
    --cc=james.clark@linaro.org \
    --cc=john.g.garry@oracle.com \
    --cc=jolsa@kernel.org \
    --cc=jonathan.cameron@huawei.com \
    --cc=kan.liang@linux.intel.com \
    --cc=leo.yan@arm.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=mike.leach@linaro.org \
    --cc=namhyung@kernel.org \
    --cc=peterz@infradead.org \
    --cc=suzuki.poulose@arm.com \
    --cc=will@kernel.org \
    --cc=yangyicong@hisilicon.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).