Home Home > GIT Browse
diff options
authorNeilBrown <neilb@suse.de>2014-01-22 11:45:03 +1100
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>2014-01-29 05:06:19 -0800
commite4e80e0bec9e03b6e323857e9bb2820f5bbe6287 (patch)
parent34fa7e82ec661cf54f21b1677d2e05173e5a812c (diff)
md/raid5: close recently introduced race in stripe_head management.
commit 7da9d450ab2843bf1db378c156acc6304dbc1c2b upstream. As release_stripe and __release_stripe decrement ->count and then manipulate ->lru both under ->device_lock, it is important that get_active_stripe() increments ->count and clears ->lru also under ->device_lock. However we currently list_del_init ->lru under the lock, but increment the ->count outside the lock. This can lead to races and list corruption. So move the atomic_inc(&sh->count) up inside the ->device_lock protected region. Note that we still increment ->count without device lock in the case where get_free_stripe() was called, and in fact don't take ->device_lock at all in that path. This is safe because if the stripe_head can be found by get_free_stripe, then the hash lock assures us the no-one else could possibly be calling release_stripe() at the same time. Fixes: 566c09c53455d7c4f1130928ef8071da1a24ea65 Reported-and-tested-by: Ian Kumlien <ian.kumlien@gmail.com> Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
1 files changed, 4 insertions, 4 deletions
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 3088d3af5a89..03f82ab87d9e 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -675,8 +675,10 @@ get_active_stripe(struct r5conf *conf, sector_t sector,
|| !conf->inactive_blocked),
*(conf->hash_locks + hash));
conf->inactive_blocked = 0;
- } else
+ } else {
init_stripe(sh, sector, previous);
+ atomic_inc(&sh->count);
+ }
} else {
if (atomic_read(&sh->count)) {
@@ -695,13 +697,11 @@ get_active_stripe(struct r5conf *conf, sector_t sector,
sh->group = NULL;
+ atomic_inc(&sh->count);
} while (sh == NULL);
- if (sh)
- atomic_inc(&sh->count);
spin_unlock_irq(conf->hash_locks + hash);
return sh;