Re: [PATCH] raid10: avoid spin_lock from fastpath from raid10_unplug()
From: Song Liu <song@kernel.org>
Date: 2023-06-20 17:38:42
Also in:
lkml
On Sat, Jun 17, 2023 at 11:26 PM Yu Kuai [off-list ref] wrote:
From: Yu Kuai <redacted>
Commit 0c0be98bbe67 ("md/raid10: prevent unnecessary calls to wake_up()
in fast path") missed one place, for example, while testing with:
fio -direct=1 -rw=write/randwrite -iodepth=1 ...
Then plug and unplug will be called for each io, then wake_up() from
raid10_unplug() will cause lock contention as well.
Avoid this contention by using wake_up_barrier() instead of wake_up(),
where spin_lock is not held while waitqueue is empty.
By the way, in this scenario, each blk_plug_cb() will be allocated and
freed for each io, this seems need to be optimized as well.
Reported-and-tested-by: Ali Gholami Rudi <redacted>
Link: https://lore.kernel.org/all/20231606122233@laper.mirepesht/ (local)I think ./scripts/checkpatch.pl would recommend using "Closes" instead of "Link" after "Reported-by". Please follow this pattern. Thanks, Song
quoted hunk ↗ jump to hunk
Signed-off-by: Yu Kuai <redacted> --- drivers/md/raid10.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index d0de8c9fb3cf..fbaaa5e05edc 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c@@ -1118,7 +1118,7 @@ static void raid10_unplug(struct blk_plug_cb *cb, bool from_schedule) spin_lock_irq(&conf->device_lock); bio_list_merge(&conf->pending_bio_list, &plug->pending); spin_unlock_irq(&conf->device_lock); - wake_up(&conf->wait_barrier); + wake_up_barrier(conf); md_wakeup_thread(mddev->thread); kfree(plug); return;@@ -1127,7 +1127,7 @@ static void raid10_unplug(struct blk_plug_cb *cb, bool from_schedule) /* we aren't scheduling, so we can do the write-out directly. */ bio = bio_list_get(&plug->pending); raid1_prepare_flush_writes(mddev->bitmap); - wake_up(&conf->wait_barrier); + wake_up_barrier(conf); while (bio) { /* submit pending writes */ struct bio *next = bio->bi_next; --2.39.2