On 24/06/2025 15:07, Christoph Hellwig wrote:
On Tue, Jun 17, 2025 at 01:09:43PM +0100, John Garry wrote:
diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c
index 8af83bd161f9..91647a43e1b2 100644
--- a/fs/xfs/xfs_buf.c
+++ b/fs/xfs/xfs_buf.c
@@ -1744,8 +1744,7 @@ xfs_configure_buftarg(
*/
if (bdev_can_atomic_write(btp->bt_bdev))
xfs_configure_buftarg_atomic_writes(btp);
-
- return sync_blockdev(btp->bt_bdev);
+ return 0;
we only ever return 0 now, so we can get rid of the return code
The call to bdev_validate_blocksize above the diff context can still
return an error.
Right, but I think that my comment would be correct after the next
patch. I just noticed this when I applied your series, but commented on
the incorrect patch.
But I also did mention that maybe we can get rid of
xfs_configure_buftarg() since it is effectively a wrapper now. The
ASSERT (call in that function) looks pointless to me.
