提交 3f883f5b 编写于 作者: D Darrick J. Wong

xfs: convert a few more directory asserts to corruption

Yet another round of playing whack-a-mole with directory code that
asserts on corrupt on-disk metadata when it really should be returning
-EFSCORRUPTED instead of ASSERTing.  Found by a xfs/391 crash while
lastbit fuzzing of ltail.bestcount.
Signed-off-by: NDarrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
上级 8241f7f9
......@@ -1415,7 +1415,8 @@ xfs_dir2_leaf_removename(
oldbest = be16_to_cpu(bf[0].length);
ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
bestsp = xfs_dir2_leaf_bests_p(ltp);
ASSERT(be16_to_cpu(bestsp[db]) == oldbest);
if (be16_to_cpu(bestsp[db]) != oldbest)
return -EFSCORRUPTED;
/*
* Mark the former data entry unused.
*/
......
......@@ -387,8 +387,9 @@ xfs_dir2_leaf_to_node(
dp->d_ops->free_hdr_from_disk(&freehdr, free);
leaf = lbp->b_addr;
ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
ASSERT(be32_to_cpu(ltp->bestcount) <=
(uint)dp->i_d.di_size / args->geo->blksize);
if (be32_to_cpu(ltp->bestcount) >
(uint)dp->i_d.di_size / args->geo->blksize)
return -EFSCORRUPTED;
/*
* Copy freespace entries from the leaf block to the new block.
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册