btrfs: always choose work from prio_head first
diff mbox

Message ID 20140208221843.GA12866@localhost.localdomain
State Accepted
Headers show

Commit Message

Stanislaw Gruszka Feb. 8, 2014, 10:18 p.m. UTC
In case we do not refill, we can overwrite cur pointer from prio_head
by one from not prioritized head, what looks as something that was
not intended.

This change make we always take works from prio_head first until it's
not empty.

Signed-off-by: Stanislaw Gruszka <stf_xl@wp.pl>
---
I found this by reading code, not sure if change is correct.
Patch is only compile tested.

 fs/btrfs/async-thread.c |    9 +++++----
 1 files changed, 5 insertions(+), 4 deletions(-)

Patch
diff mbox

diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index c1e0b0c..0b78bf2 100644
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -262,18 +262,19 @@  static struct btrfs_work *get_next_work(struct btrfs_worker_thread *worker,
 	struct btrfs_work *work = NULL;
 	struct list_head *cur = NULL;
 
-	if (!list_empty(prio_head))
+	if (!list_empty(prio_head)) {
 		cur = prio_head->next;
+		goto out;
+	}
 
 	smp_mb();
 	if (!list_empty(&worker->prio_pending))
 		goto refill;
 
-	if (!list_empty(head))
+	if (!list_empty(head)) {
 		cur = head->next;
-
-	if (cur)
 		goto out;
+	}
 
 refill:
 	spin_lock_irq(&worker->lock);