[PATCH 4.4 36/48] btrfs: fix error handling when run_delayed_extent_op fails

From: Greg Kroah-Hartman
Date: Wed Jan 18 2017 - 05:51:00 EST


4.4-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jeff Mahoney <jeffm@xxxxxxxx>

commit aa7c8da35d1905d80e840d075f07d26ec90144b5 upstream.

In __btrfs_run_delayed_refs, the error path when run_delayed_extent_op
fails sets locked_ref->processing = 0 but doesn't re-increment
delayed_refs->num_heads_ready. As a result, we end up triggering
the WARN_ON in btrfs_select_ref_head.

Fixes: d7df2c796d7 (Btrfs: attach delayed ref updates to delayed ref heads)
Reported-by: Jon Nelson <jnelson-suse@xxxxxxxxxxx>
Signed-off-by: Jeff Mahoney <jeffm@xxxxxxxx>
Reviewed-by: Liu Bo <bo.li.liu@xxxxxxxxxx>
Signed-off-by: David Sterba <dsterba@xxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
fs/btrfs/extent-tree.c | 3 +++
1 file changed, 3 insertions(+)

--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -2570,7 +2570,10 @@ static noinline int __btrfs_run_delayed_
*/
if (must_insert_reserved)
locked_ref->must_insert_reserved = 1;
+ spin_lock(&delayed_refs->lock);
locked_ref->processing = 0;
+ delayed_refs->num_heads_ready++;
+ spin_unlock(&delayed_refs->lock);
btrfs_debug(fs_info, "run_delayed_extent_op returned %d", ret);
btrfs_delayed_ref_unlock(locked_ref);
return ret;