From: linan666@huaweicloud.com
To: song@kernel.org, yukuai3@huawei.com
Cc: linux-raid@vger.kernel.org, linux-kernel@vger.kernel.org,
linan666@huaweicloud.com, yi.zhang@huawei.com,
houtao1@huawei.com, yangerkun@huawei.com
Subject: [PATCH 1/2] md: change the return value type of md_write_start to void
Date: Sun, 26 May 2024 02:52:56 +0800 [thread overview]
Message-ID: <20240525185257.3896201-2-linan666@huaweicloud.com> (raw)
In-Reply-To: <20240525185257.3896201-1-linan666@huaweicloud.com>
From: Li Nan <linan122@huawei.com>
Commit cc27b0c78c79 ("md: fix deadlock between mddev_suspend() and
md_write_start()") aborted md_write_start() with false when mddev is
suspended, which fixed a deadlock if calling mddev_suspend() with
holding reconfig_mutex(). Since mddev_suspend() now includes
lockdep_assert_not_held(), it no longer holds the reconfig_mutex. This
makes previous abort unnecessary. Now, remove unnecessary abort and
change function return value to void.
Signed-off-by: Li Nan <linan122@huawei.com>
---
drivers/md/md.h | 2 +-
drivers/md/md.c | 14 ++++----------
drivers/md/raid1.c | 3 +--
drivers/md/raid10.c | 3 +--
drivers/md/raid5.c | 3 +--
5 files changed, 8 insertions(+), 17 deletions(-)
diff --git a/drivers/md/md.h b/drivers/md/md.h
index ca085ecad504..487582058f74 100644
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -785,7 +785,7 @@ extern void md_unregister_thread(struct mddev *mddev, struct md_thread __rcu **t
extern void md_wakeup_thread(struct md_thread __rcu *thread);
extern void md_check_recovery(struct mddev *mddev);
extern void md_reap_sync_thread(struct mddev *mddev);
-extern bool md_write_start(struct mddev *mddev, struct bio *bi);
+extern void md_write_start(struct mddev *mddev, struct bio *bi);
extern void md_write_inc(struct mddev *mddev, struct bio *bi);
extern void md_write_end(struct mddev *mddev);
extern void md_done_sync(struct mddev *mddev, int blocks, int ok);
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 509e5638cea1..14d6e615bcbb 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -8638,12 +8638,12 @@ EXPORT_SYMBOL(md_done_sync);
* A return value of 'false' means that the write wasn't recorded
* and cannot proceed as the array is being suspend.
*/
-bool md_write_start(struct mddev *mddev, struct bio *bi)
+void md_write_start(struct mddev *mddev, struct bio *bi)
{
int did_change = 0;
if (bio_data_dir(bi) != WRITE)
- return true;
+ return;
BUG_ON(mddev->ro == MD_RDONLY);
if (mddev->ro == MD_AUTO_READ) {
@@ -8676,15 +8676,9 @@ bool md_write_start(struct mddev *mddev, struct bio *bi)
if (did_change)
sysfs_notify_dirent_safe(mddev->sysfs_state);
if (!mddev->has_superblocks)
- return true;
+ return;
wait_event(mddev->sb_wait,
- !test_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags) ||
- is_md_suspended(mddev));
- if (test_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags)) {
- percpu_ref_put(&mddev->writes_pending);
- return false;
- }
- return true;
+ !test_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags));
}
EXPORT_SYMBOL(md_write_start);
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 7b8a71ca66dd..0d80ff471c73 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1687,8 +1687,7 @@ static bool raid1_make_request(struct mddev *mddev, struct bio *bio)
if (bio_data_dir(bio) == READ)
raid1_read_request(mddev, bio, sectors, NULL);
else {
- if (!md_write_start(mddev,bio))
- return false;
+ md_write_start(mddev,bio);
raid1_write_request(mddev, bio, sectors);
}
return true;
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index a4556d2e46bf..f8d7c02c6ed5 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -1836,8 +1836,7 @@ static bool raid10_make_request(struct mddev *mddev, struct bio *bio)
&& md_flush_request(mddev, bio))
return true;
- if (!md_write_start(mddev, bio))
- return false;
+ md_write_start(mddev, bio);
if (unlikely(bio_op(bio) == REQ_OP_DISCARD))
if (!raid10_handle_discard(mddev, bio))
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 2bd1ce9b3922..a84389311dd1 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -6078,8 +6078,7 @@ static bool raid5_make_request(struct mddev *mddev, struct bio * bi)
ctx.do_flush = bi->bi_opf & REQ_PREFLUSH;
}
- if (!md_write_start(mddev, bi))
- return false;
+ md_write_start(mddev, bi);
/*
* If array is degraded, better not do chunk aligned read because
* later we might have to read it again in order to reconstruct
--
2.39.2
next prev parent reply other threads:[~2024-05-25 11:00 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-05-25 18:52 [PATCH 0/2] md: flush deadlock bugfix linan666
2024-05-25 18:52 ` linan666 [this message]
2024-05-28 12:53 ` [PATCH 1/2] md: change the return value type of md_write_start to void Yu Kuai
2024-05-25 18:52 ` [PATCH 2/2] md: fix deadlock between mddev_suspend and flush bio linan666
2024-05-28 13:12 ` Yu Kuai
2024-08-17 22:49 ` Michael
2024-06-10 20:52 ` [PATCH 0/2] md: flush deadlock bugfix Song Liu
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=20240525185257.3896201-2-linan666@huaweicloud.com \
--to=linan666@huaweicloud.com \
--cc=houtao1@huawei.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-raid@vger.kernel.org \
--cc=song@kernel.org \
--cc=yangerkun@huawei.com \
--cc=yi.zhang@huawei.com \
--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 NNTP newsgroup(s).