linux-perf-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Namhyung Kim <namhyung@kernel.org>
To: Arnaldo Carvalho de Melo <acme@kernel.org>, Jiri Olsa <jolsa@kernel.org>
Cc: Ian Rogers <irogers@google.com>,
	Adrian Hunter <adrian.hunter@intel.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@kernel.org>,
	LKML <linux-kernel@vger.kernel.org>,
	linux-perf-users@vger.kernel.org, Song Liu <song@kernel.org>,
	Hao Luo <haoluo@google.com>,
	bpf@vger.kernel.org, Juri Lelli <juri.lelli@redhat.com>
Subject: [PATCH 4/7] perf lock contention: Add data failure stat
Date: Thu,  6 Apr 2023 14:06:08 -0700	[thread overview]
Message-ID: <20230406210611.1622492-5-namhyung@kernel.org> (raw)
In-Reply-To: <20230406210611.1622492-1-namhyung@kernel.org>

It's possible to fail to update the data when the lock_stat map is full.
We should check that case and show the number at the end.

  $ sudo ./perf lock con -ablv -E3 -- ./perf bench sched messaging
  ...
   contended   total wait     max wait     avg wait            address   symbol

        6157    208.48 ms     69.29 us     33.86 us   ffff934c001c1f00    (spinlock)
        4030     72.04 ms     61.84 us     17.88 us   ffff934c000415c0    (spinlock)
        3201     50.30 ms     47.73 us     15.71 us   ffff934c2eead850    (spinlock)

  === output for debug ===

  bad: 0, total: 13388
  bad rate: 0.00 %
  histogram of failure reasons
         task: 0
        stack: 0
         time: 0
         data: 0      <----- added

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
---
 tools/perf/builtin-lock.c                      | 4 +++-
 tools/perf/util/bpf_lock_contention.c          | 1 +
 tools/perf/util/bpf_skel/lock_contention.bpf.c | 4 +++-
 tools/perf/util/lock-contention.h              | 1 +
 4 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/tools/perf/builtin-lock.c b/tools/perf/builtin-lock.c
index 9b92c7a5aefb..01b318d6c80a 100644
--- a/tools/perf/builtin-lock.c
+++ b/tools/perf/builtin-lock.c
@@ -1626,7 +1626,7 @@ static void sort_contention_result(void)
 static void print_bpf_events(int total, struct lock_contention_fails *fails)
 {
 	/* Output for debug, this have to be removed */
-	int broken = fails->task + fails->stack + fails->time;
+	int broken = fails->task + fails->stack + fails->time + fails->data;
 
 	if (quiet || total == 0 || (broken == 0 && verbose <= 0))
 		return;
@@ -1640,7 +1640,9 @@ static void print_bpf_events(int total, struct lock_contention_fails *fails)
 	pr_info(" %10s: %d\n", "task", fails->task);
 	pr_info(" %10s: %d\n", "stack", fails->stack);
 	pr_info(" %10s: %d\n", "time", fails->time);
+	pr_info(" %10s: %d\n", "data", fails->data);
 }
+
 static void print_contention_result(struct lock_contention *con)
 {
 	struct lock_stat *st;
diff --git a/tools/perf/util/bpf_lock_contention.c b/tools/perf/util/bpf_lock_contention.c
index 8a5d0eb44189..0071058ac3d2 100644
--- a/tools/perf/util/bpf_lock_contention.c
+++ b/tools/perf/util/bpf_lock_contention.c
@@ -262,6 +262,7 @@ int lock_contention_read(struct lock_contention *con)
 	con->fails.task = skel->bss->task_fail;
 	con->fails.stack = skel->bss->stack_fail;
 	con->fails.time = skel->bss->time_fail;
+	con->fails.data = skel->bss->data_fail;
 
 	stack_trace = zalloc(stack_size);
 	if (stack_trace == NULL)
diff --git a/tools/perf/util/bpf_skel/lock_contention.bpf.c b/tools/perf/util/bpf_skel/lock_contention.bpf.c
index f9d2d792ccc8..cb87c98e5340 100644
--- a/tools/perf/util/bpf_skel/lock_contention.bpf.c
+++ b/tools/perf/util/bpf_skel/lock_contention.bpf.c
@@ -124,6 +124,7 @@ int aggr_mode;
 int task_fail;
 int stack_fail;
 int time_fail;
+int data_fail;
 
 static inline int can_record(u64 *ctx)
 {
@@ -380,7 +381,8 @@ int contention_end(u64 *ctx)
 		if (aggr_mode == LOCK_AGGR_ADDR)
 			first.flags |= check_lock_type(pelem->lock, pelem->flags);
 
-		bpf_map_update_elem(&lock_stat, &key, &first, BPF_NOEXIST);
+		if (bpf_map_update_elem(&lock_stat, &key, &first, BPF_NOEXIST) < 0)
+			__sync_fetch_and_add(&data_fail, 1);
 		bpf_map_delete_elem(&tstamp, &pid);
 		return 0;
 	}
diff --git a/tools/perf/util/lock-contention.h b/tools/perf/util/lock-contention.h
index 10c28302420c..3ed1cad370fc 100644
--- a/tools/perf/util/lock-contention.h
+++ b/tools/perf/util/lock-contention.h
@@ -126,6 +126,7 @@ struct lock_contention_fails {
 	int task;
 	int stack;
 	int time;
+	int data;
 };
 
 struct lock_contention {
-- 
2.40.0.577.gac1e443424-goog


  parent reply	other threads:[~2023-04-06 21:06 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-04-06 21:06 [PATCHSET 0/7] perf lock contention: Improve performance if map is full (v1) Namhyung Kim
2023-04-06 21:06 ` [PATCH 1/7] perf lock contention: Simplify parse_lock_type() Namhyung Kim
2023-04-06 21:06 ` [PATCH 2/7] perf lock contention: Use -M for --map-nr-entries Namhyung Kim
2023-04-06 21:06 ` [PATCH 3/7] perf lock contention: Update default map size to 16384 Namhyung Kim
2023-04-06 21:06 ` Namhyung Kim [this message]
2023-04-06 21:06 ` [PATCH 5/7] perf lock contention: Update total/bad stats for hidden entries Namhyung Kim
2023-04-06 21:06 ` [PATCH 6/7] perf lock contention: Revise needs_callstack() condition Namhyung Kim
2023-04-06 21:06 ` [PATCH 7/7] perf lock contention: Do not try to update if hash map is full Namhyung Kim
2023-04-07  0:35 ` [PATCHSET 0/7] perf lock contention: Improve performance if map is full (v1) Ian Rogers
2023-04-07  0:41 ` Arnaldo Carvalho de Melo

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=20230406210611.1622492-5-namhyung@kernel.org \
    --to=namhyung@kernel.org \
    --cc=acme@kernel.org \
    --cc=adrian.hunter@intel.com \
    --cc=bpf@vger.kernel.org \
    --cc=haoluo@google.com \
    --cc=irogers@google.com \
    --cc=jolsa@kernel.org \
    --cc=juri.lelli@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=peterz@infradead.org \
    --cc=song@kernel.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;
as well as URLs for NNTP newsgroup(s).