Btrfs: be more polite in the async caching threads

The semaphore used by the async caching threads can prevent a
transaction commit, which can make the FS appear to stall.  This
releases the semaphore more often when a transaction commit is
in progress.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 2fe21fa..dc84dae 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -302,10 +302,11 @@
 			else if (ret)
 				break;
 
-			if (need_resched()) {
+			if (need_resched() ||
+			    btrfs_transaction_in_commit(fs_info)) {
 				btrfs_release_path(fs_info->extent_root, path);
 				up_read(&fs_info->extent_commit_sem);
-				cond_resched();
+				schedule_timeout(1);
 				goto again;
 			}