[tip: perf/urgent] perf: Optimize perf_pmu_migrate_context()

From: tip-bot2 for Peter Zijlstra
Date: Thu Apr 06 2023 - 06:05:29 EST


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

Commit-ID: b168098912926236bbeebaf7795eb7aab76d2b45
Gitweb: https://git.kernel.org/tip/b168098912926236bbeebaf7795eb7aab76d2b45
Author: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
AuthorDate: Mon, 03 Apr 2023 11:08:58 +02:00
Committer: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
CommitterDate: Wed, 05 Apr 2023 09:58:46 +02:00

perf: Optimize perf_pmu_migrate_context()

Thomas reported that offlining CPUs spends a lot of time in
synchronize_rcu() as called from perf_pmu_migrate_context() even though
he's not actually using uncore events.

Turns out, the thing is unconditionally waiting for RCU, even if there's
no actual events to migrate.

Fixes: 0cda4c023132 ("perf: Introduce perf_pmu_migrate_context()")
Reported-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
Tested-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Reviewed-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Reviewed-by: Paul E. McKenney <paulmck@xxxxxxxxxx>
Link: https://lkml.kernel.org/r/20230403090858.GT4253@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
---
kernel/events/core.c | 12 +++++++-----
1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/kernel/events/core.c b/kernel/events/core.c
index fb3e436..115320f 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -12893,12 +12893,14 @@ void perf_pmu_migrate_context(struct pmu *pmu, int src_cpu, int dst_cpu)
__perf_pmu_remove(src_ctx, src_cpu, pmu, &src_ctx->pinned_groups, &events);
__perf_pmu_remove(src_ctx, src_cpu, pmu, &src_ctx->flexible_groups, &events);

- /*
- * Wait for the events to quiesce before re-instating them.
- */
- synchronize_rcu();
+ if (!list_empty(&events)) {
+ /*
+ * Wait for the events to quiesce before re-instating them.
+ */
+ synchronize_rcu();

- __perf_pmu_install(dst_ctx, dst_cpu, pmu, &events);
+ __perf_pmu_install(dst_ctx, dst_cpu, pmu, &events);
+ }

mutex_unlock(&dst_ctx->mutex);
mutex_unlock(&src_ctx->mutex);