1
0
Fork 0

md-cluster: fix wrong condition check in raid1_write_request

The check used here is to avoid conflict between write and
resync, however we used the wrong logic, it should be the
inverse of the checking inside "if".

Fixes: 589a1c4 ("Suspend writes in RAID1 if within range")
Signed-off-by: Guoqing Jiang <gqjiang@suse.com>
Signed-off-by: Shaohua Li <shli@fb.com>
hifive-unleashed-5.1
Guoqing Jiang 2017-09-29 09:16:43 +08:00 committed by Shaohua Li
parent 938b533d47
commit 385f4d7f94
1 changed files with 5 additions and 5 deletions

View File

@ -1325,12 +1325,12 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio,
sigset_t full, old;
prepare_to_wait(&conf->wait_barrier,
&w, TASK_INTERRUPTIBLE);
if (bio_end_sector(bio) <= mddev->suspend_lo ||
bio->bi_iter.bi_sector >= mddev->suspend_hi ||
(mddev_is_clustered(mddev) &&
if ((bio_end_sector(bio) <= mddev->suspend_lo ||
bio->bi_iter.bi_sector >= mddev->suspend_hi) &&
(!mddev_is_clustered(mddev) ||
!md_cluster_ops->area_resyncing(mddev, WRITE,
bio->bi_iter.bi_sector,
bio_end_sector(bio))))
bio->bi_iter.bi_sector,
bio_end_sector(bio))))
break;
sigfillset(&full);
sigprocmask(SIG_BLOCK, &full, &old);