Btrfs: add a cond_resched() into the worker loop

If we have a constant stream of end_io completions or crc work,
we can hit softlockup messages from the async helper threads.  This
adds a cond_resched() into the loop to avoid them.

Signed-off-by: Chris Mason <chris.mason@oracle.com>

diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index 7ec1409..203e849 100644 (file)
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -331,7 +331,7 @@ again:
                        run_ordered_completions(worker->workers, work);
 
                        check_pending_worker_creates(worker);
-
+                       cond_resched();
                }
 
                spin_lock_irq(&worker->lock);