linux-perf-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: weilin.wang@intel.com
To: weilin.wang@intel.com, Namhyung Kim <namhyung@kernel.org>,
	Ian Rogers <irogers@google.com>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@redhat.com>,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Jiri Olsa <jolsa@kernel.org>,
	Adrian Hunter <adrian.hunter@intel.com>,
	Kan Liang <kan.liang@linux.intel.com>
Cc: linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org,
	Perry Taylor <perry.taylor@intel.com>,
	Samantha Alt <samantha.alt@intel.com>,
	Caleb Biggers <caleb.biggers@intel.com>
Subject: [RFC PATCH v5 4/6] perf stat: Create another thread for sample data processing
Date: Tue, 26 Mar 2024 16:28:57 -0400	[thread overview]
Message-ID: <20240326202859.960577-5-weilin.wang@intel.com> (raw)
In-Reply-To: <20240326202859.960577-1-weilin.wang@intel.com>

From: Weilin Wang <weilin.wang@intel.com>

Another thread is required to synchronize between perf stat and perf record
when we pass data through pipe.

Signed-off-by: Weilin Wang <weilin.wang@intel.com>
Reviewed-by: Ian Rogers <irogers@google.com>
---
 tools/perf/builtin-stat.c     | 51 +++++++++++++++++++++--------------
 tools/perf/util/stat-shadow.c |  4 +--
 2 files changed, 33 insertions(+), 22 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index fc887d9aaa66..14488cb0cfc8 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -802,9 +802,9 @@ static int process_feature_event(struct perf_session *session,
 	return 0;
 }
 
-static int __cmd_script(struct child_process *cmd __maybe_unused)
+static void *__cmd_script(void *arg __maybe_unused)
 {
-	int err = 0;
+	struct child_process *cmd = arg;
 	struct perf_session *session;
 	struct perf_data data = {
 		.mode = PERF_DATA_MODE_READ,
@@ -819,29 +819,15 @@ static int __cmd_script(struct child_process *cmd __maybe_unused)
 		.attr		 = perf_event__process_attr,
 		},
 	};
-	struct tpebs_event *e;
-
-	list_for_each_entry(e, &stat_config.tpebs_events, nd) {
-		struct tpebs_retire_lat *new = malloc(sizeof(struct tpebs_retire_lat));
 
-		if (!new)
-			return -1;
-		new->event.name = strdup(e->name);
-		new->event.tpebs_name = strdup(e->tpebs_name);
-		new->count = 0;
-		new->sum = 0;
-		list_add_tail(&new->event.nd, &stat_config.tpebs_results);
-	}
-
-	kill(cmd->pid, SIGTERM);
 	session = perf_session__new(&data, &script.tool);
 	if (IS_ERR(session))
-		return PTR_ERR(session);
+		return NULL;
 	script.session = session;
-	err = perf_session__process_events(session);
+	perf_session__process_events(session);
 	perf_session__delete(session);
 
-	return err;
+	return NULL;
 }
 
 static int __run_perf_stat(int argc, const char **argv, int run_idx)
@@ -861,16 +847,37 @@ static int __run_perf_stat(int argc, const char **argv, int run_idx)
 	int err;
 	bool second_pass = false;
 	struct child_process cmd;
+	pthread_t thread_script;
 
 	/* Prepare perf record for sampling event retire_latency before fork and
 	 * prepare workload */
 	if (stat_config.tpebs_event_size > 0) {
 		int ret;
+		struct tpebs_event *e;
 
 		pr_debug("perf stat pid = %d\n", getpid());
+		list_for_each_entry(e, &stat_config.tpebs_events, nd) {
+			struct tpebs_retire_lat *new = malloc(sizeof(struct tpebs_retire_lat));
+
+			if (!new)
+				return -1;
+			new->event.name = strdup(e->name);
+			new->event.tpebs_name = strdup(e->tpebs_name);
+			new->count = 0;
+			new->sum = 0;
+			list_add_tail(&new->event.nd, &stat_config.tpebs_results);
+		}
 		ret = prepare_perf_record(&cmd);
 		if (ret)
 			return ret;
+		if (pthread_create(&thread_script, NULL, __cmd_script, &cmd)) {
+			kill(cmd.pid, SIGTERM);
+			close(cmd.out);
+			pr_err("Could not create thread to process sample data.\n");
+			return -1;
+		}
+		/* Wait for perf record initialization a little bit.*/
+		sleep(2);
 	}
 
 	if (forks) {
@@ -1081,8 +1088,12 @@ static int __run_perf_stat(int argc, const char **argv, int run_idx)
 	if (stat_config.tpebs_event_size > 0) {
 		int ret;
 
-		ret = __cmd_script(&cmd);
+		kill(cmd.pid, SIGTERM);
+		pthread_join(thread_script, NULL);
 		close(cmd.out);
+		ret = finish_command(&cmd);
+		if (ret != -ERR_RUN_COMMAND_WAITPID_SIGNAL)
+			return ret;
 	}
 
 	if (stat_config.walltime_run_table)
diff --git a/tools/perf/util/stat-shadow.c b/tools/perf/util/stat-shadow.c
index 8fc3415de106..bc77e9e02892 100644
--- a/tools/perf/util/stat-shadow.c
+++ b/tools/perf/util/stat-shadow.c
@@ -361,8 +361,8 @@ static int prepare_retire_lat(struct expr_parse_ctx *pctx,
 	int ret = 0;
 	struct tpebs_retire_lat *t;
 
-	list_for_each_entry(t, retire_lats, nd) {
-		ret = expr__add_id_val(pctx, strdup(t->tpebs_name), t->val);
+	list_for_each_entry(t, retire_lats, event.nd) {
+		ret = expr__add_id_val(pctx, strdup(t->event.tpebs_name), t->val);
 		if (ret < 0)
 			return ret;
 	}
-- 
2.43.0


  parent reply	other threads:[~2024-03-26 20:29 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-03-26 20:28 [RFC PATCH v5 0/6] TPEBS counting mode support weilin.wang
2024-03-26 20:28 ` [RFC PATCH v5 1/6] perf stat: Parse and find tpebs events when parsing metrics to prepare for perf record sampling weilin.wang
2024-03-29  5:46   ` Namhyung Kim
2024-03-29  6:43     ` Wang, Weilin
2024-03-29  6:55       ` Namhyung Kim
2024-03-26 20:28 ` [RFC PATCH v5 2/6] perf stat: Fork and launch perf record when perf stat needs to get retire latency value for a metric weilin.wang
2024-03-29  6:43   ` Namhyung Kim
2024-03-29 17:25     ` Wang, Weilin
2024-03-26 20:28 ` [RFC PATCH v5 3/6] perf stat: Add retire latency values into the expr_parse_ctx to prepare for final metric calculation weilin.wang
2024-03-26 20:28 ` weilin.wang [this message]
2024-03-29  6:47   ` [RFC PATCH v5 4/6] perf stat: Create another thread for sample data processing Namhyung Kim
2024-03-26 20:28 ` [RFC PATCH v5 5/6] perf stat: Add retire latency print functions to print out at the very end of print out weilin.wang
2024-03-26 20:28 ` [RFC PATCH v5 6/6] perf vendor events intel: Add MTL metric json files weilin.wang
2024-03-28 16:36 ` [RFC PATCH v5 0/6] TPEBS counting mode support Ian Rogers
2024-03-28 19:43   ` Namhyung Kim
2024-03-29  0:12     ` Wang, Weilin

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=20240326202859.960577-5-weilin.wang@intel.com \
    --to=weilin.wang@intel.com \
    --cc=acme@kernel.org \
    --cc=adrian.hunter@intel.com \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=caleb.biggers@intel.com \
    --cc=irogers@google.com \
    --cc=jolsa@kernel.org \
    --cc=kan.liang@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=namhyung@kernel.org \
    --cc=perry.taylor@intel.com \
    --cc=peterz@infradead.org \
    --cc=samantha.alt@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).