md/raid1: ensure device failure recorded before write request returns.

When a write to one of the legs of a RAID1 fails, the failure is
recorded in the metadata of the other leg(s) so that after a restart
the data on the failed drive wont be trusted even if that drive seems
to be working again  (maybe a cable was unplugged).

Similarly when we record a bad-block in response to a write failure,
we must not let the write complete until the bad-block update is safe.

Currently there is no interlock between the write request completing
and the metadata update.  So it is possible that the write will
complete, the app will confirm success in some way, and then the
machine will crash before the metadata update completes.

This is an extremely small hole for a racy to fit in, but it is
theoretically possible and so should be closed.

So:
 - set MD_CHANGE_PENDING when requesting a metadata update for a
   failed device, so we can know with certainty when it completes
 - queue requests that experienced an error on a new queue which
   is only processed after the metadata update completes
 - call raid_end_bio_io() on bios in that queue when the time comes.

Signed-off-by: NeilBrown <neilb@suse.com>

1 files changed, 28 insertions, 1 deletions

diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 742b50794dfd..3d9ca836247f 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1508,6 +1508,7 @@ static void error(struct mddev *mddev, struct md_rdev *rdev)
 	 */
 	set_bit(MD_RECOVERY_INTR, &mddev->recovery);
 	set_bit(MD_CHANGE_DEVS, &mddev->flags);
+	set_bit(MD_CHANGE_PENDING, &mddev->flags);
 	printk(KERN_ALERT
 	       "md/raid1:%s: Disk failure on %s, disabling device.\n"
 	       "md/raid1:%s: Operation continuing on %d devices.\n",
@@ -2289,6 +2290,7 @@ static void handle_sync_write_finished(struct r1conf *conf, struct r1bio *r1_bio
 static void handle_write_finished(struct r1conf *conf, struct r1bio *r1_bio)
 {
 	int m;
+	bool fail = false;
 	for (m = 0; m < conf->raid_disks * 2 ; m++)
 		if (r1_bio->bios[m] == IO_MADE_GOOD) {
 			struct md_rdev *rdev = conf->mirrors[m].rdev;
@@ -2301,6 +2303,7 @@ static void handle_write_finished(struct r1conf *conf, struct r1bio *r1_bio)
 			 * narrow down and record precise write
 			 * errors.
 			 */
+			fail = true;
 			if (!narrow_write_error(r1_bio, m)) {
 				md_error(conf->mddev,
 					 conf->mirrors[m].rdev);
@@ -2312,7 +2315,13 @@ static void handle_write_finished(struct r1conf *conf, struct r1bio *r1_bio)
 		}
 	if (test_bit(R1BIO_WriteError, &r1_bio->state))
 		close_write(r1_bio);
-	raid_end_bio_io(r1_bio);
+	if (fail) {
+		spin_lock_irq(&conf->device_lock);
+		list_add(&r1_bio->retry_list, &conf->bio_end_io_list);
+		spin_unlock_irq(&conf->device_lock);
+		md_wakeup_thread(conf->mddev->thread);
+	} else
+		raid_end_bio_io(r1_bio);
 }
 
 static void handle_read_error(struct r1conf *conf, struct r1bio *r1_bio)
@@ -2418,6 +2427,23 @@ static void raid1d(struct md_thread *thread)
 
 	md_check_recovery(mddev);
 
+	if (!list_empty_careful(&conf->bio_end_io_list) &&
+	    !test_bit(MD_CHANGE_PENDING, &mddev->flags)) {
+		LIST_HEAD(tmp);
+		spin_lock_irqsave(&conf->device_lock, flags);
+		if (!test_bit(MD_CHANGE_PENDING, &mddev->flags)) {
+			list_add(&tmp, &conf->bio_end_io_list);
+			list_del_init(&conf->bio_end_io_list);
+		}
+		spin_unlock_irqrestore(&conf->device_lock, flags);
+		while (!list_empty(&tmp)) {
+			r1_bio = list_first_entry(&conf->bio_end_io_list,
+						  struct r1bio, retry_list);
+			list_del(&r1_bio->retry_list);
+			raid_end_bio_io(r1_bio);
+		}
+	}
+
 	blk_start_plug(&plug);
 	for (;;) {
 
@@ -2819,6 +2845,7 @@ static struct r1conf *setup_conf(struct mddev *mddev)
 	conf->raid_disks = mddev->raid_disks;
 	conf->mddev = mddev;
 	INIT_LIST_HEAD(&conf->retry_list);
+	INIT_LIST_HEAD(&conf->bio_end_io_list);
 
 	spin_lock_init(&conf->resync_lock);
 	init_waitqueue_head(&conf->wait_barrier);