The locking order for stuff is
__sb_start_write
ordered_mutex
but with sync() we don't do __sb_start_write for some strange reason, which
means that our iput in wait_ordered_extents could start a transaction which does
the __sb_start_write while we're holding the ordered_mutex. Fix this by using
delayed iput in sync. Thanks,
Reported-by: David Sterba <dsterba@xxxxxxx>
Signed-off-by: Josef Bacik <jbacik@xxxxxxxxxxxx>
---
fs/btrfs/super.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index bb60fe5..2a0bece 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -866,7 +866,7 @@ int btrfs_sync_fs(struct super_block *sb, int wait)
return 0;
}
- btrfs_wait_ordered_extents(root, 0);
+ btrfs_wait_ordered_extents(root, 1);
trans = btrfs_attach_transaction_barrier(root);
if (IS_ERR(trans)) {
--
1.7.7.6
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
