DM RAID: Use safe version of rdev_for_each
authorJonathan Brassow <jbrassow@redhat.com>
Tue, 24 Apr 2012 00:23:13 +0000 (10:23 +1000)
committerNeilBrown <neilb@suse.de>
Tue, 24 Apr 2012 00:23:13 +0000 (10:23 +1000)
Fix segfault caused by using rdev_for_each instead of rdev_for_each_safe

Commit dafb20fa34320a472deb7442f25a0c086e0feb33 mistakenly replaced a safe
iterator with an unsafe one when making some macro changes.

Signed-off-by: Jonathan Brassow <jbrassow@redhat.com>
Signed-off-by: NeilBrown <neilb@suse.de>
drivers/md/dm-raid.c

index b0ba52459ed7381d4a802acb94e69df02f979149..68965e663248e2f0b0158aaf9a75f2583be2704c 100644 (file)
@@ -859,7 +859,7 @@ static int analyse_superblocks(struct dm_target *ti, struct raid_set *rs)
        int ret;
        unsigned redundancy = 0;
        struct raid_dev *dev;
-       struct md_rdev *rdev, *freshest;
+       struct md_rdev *rdev, *tmp, *freshest;
        struct mddev *mddev = &rs->md;
 
        switch (rs->raid_type->level) {
@@ -877,7 +877,7 @@ static int analyse_superblocks(struct dm_target *ti, struct raid_set *rs)
        }
 
        freshest = NULL;
-       rdev_for_each(rdev, mddev) {
+       rdev_for_each_safe(rdev, tmp, mddev) {
                if (!rdev->meta_bdev)
                        continue;