In xfs_bmap_process_allocated_extent(), for when we found that we could not
provide the requested length completely, the mapping is moved so that we
can provide as much as possible for the original request.

For forcealign, this would mean ignoring alignment guaranteed, so don't do
this.

Reviewed-by: "Darrick J. Wong" <djwong@kernel.org>
Signed-off-by: John Garry <john.g.garry@oracle.com>
---
 fs/xfs/libxfs/xfs_bmap.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 3ab2cecf09d2..a9e98f83e57e 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -3491,11 +3491,15 @@ xfs_bmap_process_allocated_extent(
 	 * original request as possible.  Free space is apparently
 	 * very fragmented so we're unlikely to be able to satisfy the
 	 * hints anyway.
+	 * However, for an inode with forcealign, continue with the
+	 * found offset as we need to honour the alignment hint.
 	 */
-	if (ap->length <= orig_length)
-		ap->offset = orig_offset;
-	else if (ap->offset + ap->length < orig_offset + orig_length)
-		ap->offset = orig_offset + orig_length - ap->length;
+	if (!xfs_inode_has_forcealign(ap->ip)) {
+		if (ap->length <= orig_length)
+			ap->offset = orig_offset;
+		else if (ap->offset + ap->length < orig_offset + orig_length)
+			ap->offset = orig_offset + orig_length - ap->length;
+	}
 	xfs_bmap_alloc_account(ap);
 }
 
-- 
2.31.1