[PATCH] perf/core: Use local64_try_cmpxchg in perf_swevent_set_period

[PATCH] perf/core: Use local64_try_cmpxchg in perf_swevent_set_period

[Uros Bizjak]

Use local64_try_cmpxchg instead of local64_cmpxchg (*ptr, old, new) == old
in perf_swevent_set_period.  x86 CMPXCHG instruction returns success in ZF
flag, so this change saves a compare after cmpxchg (and related move
instruction in front of cmpxchg).

Also, try_cmpxchg implicitly assigns old *ptr value to "old" when cmpxchg
fails. There is no need to re-read the value in the loop.

No functional change intended.

Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
---
 kernel/events/core.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 78ae7b6f90fd..f84e2640ea2f 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -9595,16 +9595,16 @@ u64 perf_swevent_set_period(struct perf_event *event)
 
 	hwc->last_period = hwc->sample_period;
 
-again:
-	old = val = local64_read(&hwc->period_left);
-	if (val < 0)
-		return 0;
+	old = local64_read(&hwc->period_left);
+	do {
+		val = old;
+		if (val < 0)
+			return 0;
 
-	nr = div64_u64(period + val, period);
-	offset = nr * period;
-	val -= offset;
-	if (local64_cmpxchg(&hwc->period_left, old, val) != old)
-		goto again;
+		nr = div64_u64(period + val, period);
+		offset = nr * period;
+		val -= offset;
+	} while (!local64_try_cmpxchg(&hwc->period_left, &old, val));
 
 	return nr;
 }
-- 
2.41.0

[tip: perf/core] perf/core: Use local64_try_cmpxchg in perf_swevent_set_period

[tip-bot2 for Uros Bizjak]

The following commit has been merged into the perf/core branch of tip:

Commit-ID:     28fd85a10a2a73658c6e26056d9e093de06b5a22
Gitweb:        https://git.kernel.org/tip/28fd85a10a2a73658c6e26056d9e093de06b5a22
Author:        Uros Bizjak <ubizjak@gmail.com>
AuthorDate:    Sat, 08 Jul 2023 10:10:57 +02:00
Committer:     Peter Zijlstra <peterz@infradead.org>
CommitterDate: Mon, 10 Jul 2023 09:52:35 +02:00

perf/core: Use local64_try_cmpxchg in perf_swevent_set_period

Use local64_try_cmpxchg instead of local64_cmpxchg (*ptr, old, new) == old
in perf_swevent_set_period.  x86 CMPXCHG instruction returns success in ZF
flag, so this change saves a compare after cmpxchg (and related move
instruction in front of cmpxchg).

Also, try_cmpxchg implicitly assigns old *ptr value to "old" when cmpxchg
fails. There is no need to re-read the value in the loop.

No functional change intended.

Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20230708081129.45915-1-ubizjak@gmail.com
---
 kernel/events/core.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 78ae7b6..f84e264 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -9595,16 +9595,16 @@ u64 perf_swevent_set_period(struct perf_event *event)
 
 	hwc->last_period = hwc->sample_period;
 
-again:
-	old = val = local64_read(&hwc->period_left);
-	if (val < 0)
-		return 0;
+	old = local64_read(&hwc->period_left);
+	do {
+		val = old;
+		if (val < 0)
+			return 0;
 
-	nr = div64_u64(period + val, period);
-	offset = nr * period;
-	val -= offset;
-	if (local64_cmpxchg(&hwc->period_left, old, val) != old)
-		goto again;
+		nr = div64_u64(period + val, period);
+		offset = nr * period;
+		val -= offset;
+	} while (!local64_try_cmpxchg(&hwc->period_left, &old, val));
 
 	return nr;
 }