On Thu, Feb 20, 2020 at 07:13:02PM +0800, Zheng Bin wrote:
> We recently used fuzz(hydra) to test XFS and automatically generate
> tmp.img(XFS v5 format, but some metadata is wrong)
>
> xfs_repair information(just one AG):
> agf_freeblks 0, counted 3224 in ag 0
> agf_longest 536874136, counted 3224 in ag 0
> sb_fdblocks 613, counted 3228
>
> Test as follows:
> mount tmp.img tmpdir
> cp file1M tmpdir
> sync
>
> In 4.19-stable, sync will stuck, the reason is:
> xfs_mountfs
> xfs_check_summary_counts
> if ((!xfs_sb_version_haslazysbcount(&mp->m_sb) ||
> XFS_LAST_UNMOUNT_WAS_CLEAN(mp)) &&
> !xfs_fs_has_sickness(mp, XFS_SICK_FS_COUNTERS))
> return 0; -->just return, incore sb_fdblocks still be 613
> xfs_initialize_perag_data
>
> cp file1M tmpdir -->ok(write file to pagecache)
> sync -->stuck(write pagecache to disk)
> xfs_map_blocks
> xfs_iomap_write_allocate
> while (count_fsb != 0) {
> nimaps = 0;
> while (nimaps == 0) { --> endless loop
> nimaps = 1;
> xfs_bmapi_write(..., &nimaps) --> nimaps becomes 0 again
> xfs_bmapi_write
> xfs_bmap_alloc
> xfs_bmap_btalloc
> xfs_alloc_vextent
> xfs_alloc_fix_freelist
> xfs_alloc_space_available -->fail(agf_freeblks is 0)
>
> In linux-next, sync not stuck, cause commit c2b3164320b5 ("xfs:
> use the latest extent at writeback delalloc conversion time") remove
> the above while, dmesg is as follows:
> [ 55.250114] XFS (loop0): page discard on page ffffea0008bc7380, inode 0x1b0c, offset 0.
>
> Users do not know why this page is discard, the better soultion is:
> 1. Like xfs_repair, make sure sb_fdblocks is equal to counted
> (xfs_initialize_perag_data did this, who is not called at this mount)
> 2. Add agf verify, if fail, will tell users to repair
>
> This patch use the second soultion.
>
> Signed-off-by: Zheng Bin <zhengbin13@xxxxxxxxxx>
> Signed-off-by: Ren Xudong <renxudong1@xxxxxxxxxx>
> ---
> v1->v2: modify comment, add more agf verify
> fs/xfs/libxfs/xfs_alloc.c | 17 +++++++++++++++++
> 1 file changed, 17 insertions(+)
>
> diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
> index d8053bc..5faed42 100644
> --- a/fs/xfs/libxfs/xfs_alloc.c
> +++ b/fs/xfs/libxfs/xfs_alloc.c
> @@ -2839,6 +2839,7 @@ xfs_agf_verify(
> {
> struct xfs_mount *mp = bp->b_mount;
> struct xfs_agf *agf = XFS_BUF_TO_AGF(bp);
> + int i;
>
> if (xfs_sb_version_hascrc(&mp->m_sb)) {
> if (!uuid_equal(&agf->agf_uuid, &mp->m_sb.sb_meta_uuid))
> @@ -2858,6 +2859,22 @@ xfs_agf_verify(
> be32_to_cpu(agf->agf_flcount) <= xfs_agfl_size(mp)))
> return __this_address;
>
> + if (be32_to_cpu(agf->agf_length) > mp->m_sb.sb_dblocks ||
> + be32_to_cpu(agf->agf_btreeblks) > be32_to_cpu(agf->agf_length) ||
Isn't this already covered later on?
> + be32_to_cpu(agf->agf_rmap_blocks) > be32_to_cpu(agf->agf_length) ||
> + be32_to_cpu(agf->agf_refcount_blocks) > be32_to_cpu(agf->agf_length) ||
Do these fields need checking when the corresponding feature isn't
enabled?
> + be32_to_cpu(agf->agf_spare2) != 0)
If, for some reason, these unused "spare" fields are *not* zero, won't
this cause mount failures on existing filesystems?
> + return __this_address;
Please try to check only one agf field per if clause, because we use the
logged __this_address to figure out which field triggered the corruption
error.
> +
> + for (i = 0; i < ARRAY_SIZE(agf->agf_spare64); i++)
> + if (be64_to_cpu(agf->agf_spare64[i]) != 0)
> + return __this_address;
memchr_inv if you leave in the spare check.
> +
> + if (be32_to_cpu(agf->agf_freeblks) < be32_to_cpu(agf->agf_longest) ||
> + be32_to_cpu(agf->agf_freeblks) > be32_to_cpu(agf->agf_length) ||
Already covered in this function.
--D
> + be32_to_cpu(agf->agf_freeblks) > mp->m_sb.sb_fdblocks)
> + return __this_address;
> +
> if (be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]) < 1 ||
> be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]) < 1 ||
> be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]) > XFS_BTREE_MAXLEVELS ||
> --
> 2.7.4
>
