From: NeilBrown <neilb@suse.com>
Date: Fri, 17 Jul 2015 02:06:02 +0000 (+1000)
Subject: md: be careful when testing resync_max against curr_resync_completed.
X-Git-Url: http://git.lede-project.org./?a=commitdiff_plain;h=c5e19d906a658f27fa858b09a95d9551b1a69bd0;p=openwrt%2Fstaging%2Fblogic.git

md: be careful when testing resync_max against curr_resync_completed.

While it generally shouldn't happen, it is not impossible for
curr_resync_completed to exceed resync_max.
This can particularly happen when reshaping RAID5 - the current
status isn't copied to curr_resync_completed promptly, so when it
is, it can exceed resync_max.
This happens when the reshape is 'frozen', resync_max is set low,
and reshape is re-enabled.

Taking a difference between two unsigned numbers is always dangerous
anyway, so add a test to behave correctly if
   curr_resync_completed > resync_max

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

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 5b62a3d49e12..b326cd26b027 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -7834,7 +7834,8 @@ void md_do_sync(struct md_thread *thread)
 		      > (max_sectors >> 4)) ||
 		     time_after_eq(jiffies, update_time + UPDATE_FREQUENCY) ||
 		     (j - mddev->curr_resync_completed)*2
-		     >= mddev->resync_max - mddev->curr_resync_completed
+		     >= mddev->resync_max - mddev->curr_resync_completed ||
+		     mddev->curr_resync_completed > mddev->resync_max
 			    )) {
 			/* time to update curr_resync_completed */
 			wait_event(mddev->recovery_wait,
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 19bbdbe1a52f..1c27aa10f89c 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -5542,7 +5542,8 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, int *sk
 	sector_nr += reshape_sectors;
 	retn = reshape_sectors;
 finish:
-	if ((sector_nr - mddev->curr_resync_completed) * 2
+	if (mddev->curr_resync_completed > mddev->resync_max ||
+	    (sector_nr - mddev->curr_resync_completed) * 2
 	    >= mddev->resync_max - mddev->curr_resync_completed) {
 		/* Cannot proceed until we've updated the superblock... */
 		wait_event(conf->wait_for_overlap,