md: allow metadata update while suspending.

commit 35bfc52187f6df8779d0f1cebdb52b7f797baf4e upstream.

There are various deadlocks that can occur
when a thread holds reconfig_mutex and calls
->quiesce(mddev, 1).
As some write request block waiting for
metadata to be updated (e.g. to record device
failure), and as the md thread updates the metadata
while the reconfig mutex is held, holding the mutex
can stop write requests completing, and this prevents
->quiesce(mddev, 1) from completing.

->quiesce() is now usually called from mddev_suspend(),
and it is always called with reconfig_mutex held.  So
at this time it is safe for the thread to update metadata
without explicitly taking the lock.

So add 2 new flags, one which says the unlocked updates is
allowed, and one which ways it is happening.  Then allow it
while the quiesce completes, and then wait for it to finish.

Reported-and-tested-by: Xiao Ni <xni@redhat.com>
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
Signed-off-by: Jack Wang <jinpu.wang@profitbricks.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

2 files changed, 20 insertions, 0 deletions

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 3d1650db2064..b74cca273e38 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -364,8 +364,12 @@ void mddev_suspend(struct mddev *mddev)
 		return;
 	synchronize_rcu();
 	wake_up(&mddev->sb_wait);
+	set_bit(MD_ALLOW_SB_UPDATE, &mddev->flags);
+	smp_mb__after_atomic();
 	wait_event(mddev->sb_wait, atomic_read(&mddev->active_io) == 0);
 	mddev->pers->quiesce(mddev, 1);
+	clear_bit_unlock(MD_ALLOW_SB_UPDATE, &mddev->flags);
+	wait_event(mddev->sb_wait, !test_bit(MD_UPDATING_SB, &mddev->flags));
 
 	del_timer_sync(&mddev->safemode_timer);
 }
@@ -8882,6 +8886,16 @@ void md_check_recovery(struct mddev *mddev)
 	unlock:
 		wake_up(&mddev->sb_wait);
 		mddev_unlock(mddev);
+	} else if (test_bit(MD_ALLOW_SB_UPDATE, &mddev->flags) && mddev->sb_flags) {
+		/* Write superblock - thread that called mddev_suspend()
+		 * holds reconfig_mutex for us.
+		 */
+		set_bit(MD_UPDATING_SB, &mddev->flags);
+		smp_mb__after_atomic();
+		if (test_bit(MD_ALLOW_SB_UPDATE, &mddev->flags))
+			md_update_sb(mddev, 0);
+		clear_bit_unlock(MD_UPDATING_SB, &mddev->flags);
+		wake_up(&mddev->sb_wait);
 	}
 }
 EXPORT_SYMBOL(md_check_recovery);
diff --git a/drivers/md/md.h b/drivers/md/md.h
index 9b0a896890ef..37c19b7b5df9 100644
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -237,6 +237,12 @@ enum mddev_flags {
 				 */
 	MD_HAS_PPL,		/* The raid array has PPL feature set */
 	MD_HAS_MULTIPLE_PPLS,	/* The raid array has multiple PPLs feature set */
+	MD_ALLOW_SB_UPDATE,	/* md_check_recovery is allowed to update
+				 * the metadata without taking reconfig_mutex.
+				 */
+	MD_UPDATING_SB,		/* md_check_recovery is updating the metadata
+				 * without explicitly holding reconfig_mutex.
+				 */
 };
 
 enum mddev_sb_flags {