[PATCH 2/2] btrfs: Fix deadlock caused by missing memory barrier

[Nikolay Borisov <nborisov@suse.com>]

Commit 06297d8cefca ("btrfs: switch extent_buffer blocking_writers from atomic to int")
changed the type of blocking_writers but forgot to adjust relevant code
in btrfs_tree_unlock by converting the smp_mb__after_atomic to smp_mb.
This opened up the possibility of a deadlock due to re-ordering of
setting blocking_writers and checking/waking up the  waiter. This
particular lockup is explained in a comment above waitqueue_active()
function.

Fix it by converting the memory barrier to a full smp_mb, accounting
for the fact that blocking_writers is a simple integer.

Fixes: 06297d8cefca ("btrfs: switch extent_buffer blocking_writers from atomic to int")
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Tested-by: Johannes Thumshirn <jthumshirn@suse.com>
---
 fs/btrfs/locking.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c
index ceb23690b89b..5e867720df88 100644
--- a/fs/btrfs/locking.c
+++ b/fs/btrfs/locking.c
@@ -310,9 +310,12 @@ void btrfs_tree_unlock(struct extent_buffer *eb)
 	if (blockers) {
 		btrfs_assert_no_spinning_writers(eb);
 		eb->blocking_writers--;
-		/* Use the lighter barrier after atomic */
-		smp_mb__after_atomic();
-		cond_wake_up_nomb(&eb->write_lock_wq);
+		/*
+		 * We need to order modifying blocking_writers above with
+		 * actually waking up the sleepers to ensure they see the
+		 * updated value of blocking_writers
+		 */
+		cond_wake_up(&eb->write_lock_wq);
 	} else {
 		btrfs_assert_spinning_writers_put(eb);
 		write_unlock(&eb->lock);
-- 
2.17.1