md/raid5: be careful not to resize_stripes too big.

When a RAID5 is reshaping, conf->raid_disks is increased
before mddev->delta_disks becomes zero.
This can result in check_reshape calling resize_stripes with a
number that is too large.  This particularly happens
when md_check_recovery calls ->check_reshape().

If we use ->previous_raid_disks, we don't risk this.

Signed-off-by: NeilBrown <neilb@suse.de>
This commit is contained in:
NeilBrown 2012-10-11 14:24:13 +11:00
parent db07d85ef6
commit e56108d65f

View File

@ -5871,7 +5871,8 @@ static int check_reshape(struct mddev *mddev)
if (!check_stripe_cache(mddev))
return -ENOSPC;
return resize_stripes(conf, conf->raid_disks + mddev->delta_disks);
return resize_stripes(conf, (conf->previous_raid_disks
+ mddev->delta_disks));
}
static int raid5_start_reshape(struct mddev *mddev)