From: Jack Wang <jinpu.wang@ionos.com>
To: song@kernel.org, yukuai3@huawei.com, linux-raid@vger.kernel.org
Cc: Florian-Ewald Mueller <florian-ewald.mueller@ionos.com>,
Paul Menzel <pmenzel@molgen.mpg.de>
Subject: [PATCHv3] md: Replace md_thread's wait queue with the swait variant
Date: Thu, 7 Mar 2024 13:08:35 +0100 [thread overview]
Message-ID: <20240307120835.87390-1-jinpu.wang@ionos.com> (raw)
From: Florian-Ewald Mueller <florian-ewald.mueller@ionos.com>
Replace md_thread's wait_event()/wake_up() related calls with their
simple swait~ variants to improve performance as well as memory and
stack footprint.
Use the IDLE state for the worker thread put to sleep instead of
misusing the INTERRUPTIBLE state combined with flushing signals
just for not contributing to system's cpu load-average stats.
Also check for sleeping thread before attempting its wake_up in
md_wakeup_thread() for avoiding unnecessary spinlock contention.
With this patch (backported) on a kernel 6.1, the IOPS improved
by around 4% with raid1 and/or raid5, in high IO load scenarios.
Fixes: 93588e2284b6 ("[PATCH] md: make md threads interruptible again")
Signed-off-by: Florian-Ewald Mueller <florian-ewald.mueller@ionos.com>
Signed-off-by: Jack Wang <jinpu.wang@ionos.com>
Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de>
---
v3: add reviewed by from Paul, add fixes tag.
v2: fix a type error for thread
drivers/md/md.c | 23 +++++++----------------
drivers/md/md.h | 2 +-
2 files changed, 8 insertions(+), 17 deletions(-)
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 48ae2b1cb57a..14d12ee4b06b 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -7970,22 +7970,12 @@ static int md_thread(void *arg)
* many dirty RAID5 blocks.
*/
- allow_signal(SIGKILL);
while (!kthread_should_stop()) {
- /* We need to wait INTERRUPTIBLE so that
- * we don't add to the load-average.
- * That means we need to be sure no signals are
- * pending
- */
- if (signal_pending(current))
- flush_signals(current);
-
- wait_event_interruptible_timeout
- (thread->wqueue,
- test_bit(THREAD_WAKEUP, &thread->flags)
- || kthread_should_stop() || kthread_should_park(),
- thread->timeout);
+ swait_event_idle_timeout_exclusive(thread->wqueue,
+ test_bit(THREAD_WAKEUP, &thread->flags) ||
+ kthread_should_stop() || kthread_should_park(),
+ thread->timeout);
clear_bit(THREAD_WAKEUP, &thread->flags);
if (kthread_should_park())
@@ -8017,7 +8007,8 @@ void md_wakeup_thread(struct md_thread __rcu *thread)
if (t) {
pr_debug("md: waking up MD thread %s.\n", t->tsk->comm);
set_bit(THREAD_WAKEUP, &t->flags);
- wake_up(&t->wqueue);
+ if (swq_has_sleeper(&t->wqueue))
+ swake_up_one(&t->wqueue);
}
rcu_read_unlock();
}
@@ -8032,7 +8023,7 @@ struct md_thread *md_register_thread(void (*run) (struct md_thread *),
if (!thread)
return NULL;
- init_waitqueue_head(&thread->wqueue);
+ init_swait_queue_head(&thread->wqueue);
thread->run = run;
thread->mddev = mddev;
diff --git a/drivers/md/md.h b/drivers/md/md.h
index b2076a165c10..1dc01bc5c038 100644
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -716,7 +716,7 @@ static inline void sysfs_unlink_rdev(struct mddev *mddev, struct md_rdev *rdev)
struct md_thread {
void (*run) (struct md_thread *thread);
struct mddev *mddev;
- wait_queue_head_t wqueue;
+ struct swait_queue_head wqueue;
unsigned long flags;
struct task_struct *tsk;
unsigned long timeout;
--
2.34.1
next reply other threads:[~2024-03-07 12:08 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-03-07 12:08 Jack Wang [this message]
2024-03-26 5:22 ` [PATCHv3] md: Replace md_thread's wait queue with the swait variant Jinpu Wang
2024-03-26 14:09 ` Yu Kuai
2024-03-26 15:51 ` Florian-Ewald Müller
2024-03-26 19:49 ` Florian-Ewald Müller
2024-03-27 1:07 ` Yu Kuai
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=20240307120835.87390-1-jinpu.wang@ionos.com \
--to=jinpu.wang@ionos.com \
--cc=florian-ewald.mueller@ionos.com \
--cc=linux-raid@vger.kernel.org \
--cc=pmenzel@molgen.mpg.de \
--cc=song@kernel.org \
--cc=yukuai3@huawei.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 read-only IMAP folder(s) and NNTP newsgroup(s).