xfs: validity check agbnos on the AGFL
authorDave Chinner <dchinner@redhat.com>
Mon, 5 Jun 2023 04:48:15 +0000 (14:48 +1000)
committerDave Chinner <david@fromorbit.com>
Mon, 5 Jun 2023 04:48:15 +0000 (14:48 +1000)
If the agfl or the indexing in the AGF has been corrupted, getting a
block form the AGFL could return an invalid block number. If this
happens, bad things happen. Check the agbno we pull off the AGFL
and return -EFSCORRUPTED if we find somethign bad.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Dave Chinner <david@fromorbit.com>
fs/xfs/libxfs/xfs_alloc.c

index fd3293a..643d178 100644 (file)
@@ -2780,6 +2780,9 @@ xfs_alloc_get_freelist(
         */
        agfl_bno = xfs_buf_to_agfl_bno(agflbp);
        bno = be32_to_cpu(agfl_bno[be32_to_cpu(agf->agf_flfirst)]);
+       if (XFS_IS_CORRUPT(tp->t_mountp, !xfs_verify_agbno(pag, bno)))
+               return -EFSCORRUPTED;
+
        be32_add_cpu(&agf->agf_flfirst, 1);
        xfs_trans_brelse(tp, agflbp);
        if (be32_to_cpu(agf->agf_flfirst) == xfs_agfl_size(mp))