md/raid10 - avoid reading from known bad blocks - part 3
authorNeilBrown <neilb@suse.de>
Thu, 28 Jul 2011 01:39:24 +0000 (11:39 +1000)
committerNeilBrown <neilb@suse.de>
Thu, 28 Jul 2011 01:39:24 +0000 (11:39 +1000)
When attempting to repair a read error, don't read from
devices with a known bad block.

As we are only reading PAGE_SIZE blocks, we don't try to
narrow down to smaller regions in the hope that only part of this
page is bad - it isn't worth the effort.

Signed-off-by: NeilBrown <neilb@suse.de>
drivers/md/raid10.c

index 37801d68a4cdb365193a6335e14e7266062fc67e..a5ecea2672b57dff49ed1b7aa0d616821a38f273 100644 (file)
@@ -1594,10 +1594,15 @@ static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio)
 
                rcu_read_lock();
                do {
+                       sector_t first_bad;
+                       int bad_sectors;
+
                        d = r10_bio->devs[sl].devnum;
                        rdev = rcu_dereference(conf->mirrors[d].rdev);
                        if (rdev &&
-                           test_bit(In_sync, &rdev->flags)) {
+                           test_bit(In_sync, &rdev->flags) &&
+                           is_badblock(rdev, r10_bio->devs[sl].addr + sect, s,
+                                       &first_bad, &bad_sectors) == 0) {
                                atomic_inc(&rdev->nr_pending);
                                rcu_read_unlock();
                                success = sync_page_io(rdev,