Btrfs: fix wrong sync_writers decrement in btrfs_file_aio_write()

If the checks at the beginning of btrfs_file_aio_write() fail, we needn't
decrease ->sync_writers, because we have not increased it. Fix it.
Signed-off-by: NMiao Xie <miaox@cn.fujitsu.com>
Signed-off-by: NJosef Bacik <jbacik@fusionio.com>

fs/btrfs/file.c

@@ -1595,9 +1595,10 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
 		if (err < 0 && num_written > 0)
 			num_written = err;
 	}
-out:
+
 	if (sync)
 		atomic_dec(&BTRFS_I(inode)->sync_writers);
+out:
 	sb_end_write(inode->i_sb);
 	current->backing_dev_info = NULL;
 	return num_written ? num_written : err;