[PATCH v2] xfs: sanitize da node pad field on write

[Yuto Ohnuki <ytohnuki@amazon.com>]

The DA node block header (xfs_da3_node_hdr) contains a __pad32 field
that should always be zero. Prior to commit a45086e27dfa21 ("xfs:
validate metadata LSNs against log on v5 superblocks"),
xfs_da3_node_create() did not zero the full header, so __pad32 could
contain stale data on older filesystems.

Clear this field in the write verifier to ensure it is corrected
whenever the block is next written to disk.

Suggested-by: "Darrick J. Wong" <djwong@kernel.org>
Reviewed-by: "Darrick J. Wong" <djwong@kernel.org>
Signed-off-by: Yuto Ohnuki <ytohnuki@amazon.com>
---
Changes in v2:
- Added comments explaining why __pad32 is cleared unconditionally.
- Link to v1: https://lore.kernel.org/all/20260421054311.25301-2-ytohnuki@amazon.com/
---
 fs/xfs/libxfs/xfs_da_btree.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/fs/xfs/libxfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c
index ad801b7bd2dd..4edbf4bb5c7f 100644
--- a/fs/xfs/libxfs/xfs_da_btree.c
+++ b/fs/xfs/libxfs/xfs_da_btree.c
@@ -318,6 +318,12 @@ xfs_da3_node_write_verify(
 	if (bip)
 		hdr3->info.lsn = cpu_to_be64(bip->bli_item.li_lsn);
 
+	/*
+	 * Old kernels may not have zeroed the padding field in the header,
+	 * so clear it unconditionally to sanitize stale data on disk.
+	 */
+	hdr3->__pad32 = 0;
+
 	xfs_buf_update_cksum(bp, XFS_DA3_NODE_CRC_OFF);
 }
 
-- 
2.50.1




Amazon Web Services EMEA SARL, 38 avenue John F. Kennedy, L-1855 Luxembourg, R.C.S. Luxembourg B186284

Amazon Web Services EMEA SARL, Irish Branch, One Burlington Plaza, Burlington Road, Dublin 4, Ireland, branch registration number 908705