On 25.02.21 г. 9:53 ч., Wang Yugui wrote:
> Hi,
>
> xfstest btrfs/154 failed at kernel 5.4.100
>
> frequency: always
> kernel version: 5.4.100, other kernel version yet not tested.
> xfstest: https://github.com/kdave/xfstests.git
> btrfs-progs: 5.10.1
> but mkfs.btrfs default enable no-holes and free-space-tree.
>
> # ./check btrfs/154
> FSTYP -- btrfs
> PLATFORM -- Linux/x86_64 T7610 5.4.100-3.el7.x86_64 #1 SMP Thu Feb 25
> 13:19:45 CST 2021
> MKFS_OPTIONS -- /dev/sdb
> MOUNT_OPTIONS -- /dev/sdb /mnt/scratch
>
> btrfs/154 _check_dmesg: something found in dmesg (see
> /ssd/git/os/xfstests/results//btrfs/154.dmesg)
> - output mismatch (see /ssd/git/os/xfstests/results//btrfs/154.out.bad)
> --- tests/btrfs/154.out 2021-02-25 13:41:27.906590386 +0800
> +++ /ssd/git/os/xfstests/results//btrfs/154.out.bad 2021-02-25
> 15:45:52.182865707 +0800
> @@ -1,2 +1,6 @@
> QA output created by 154
> +Traceback (most recent call last):
> + File "/ssd/git/os/xfstests/src/btrfs_crc32c_forged_name.py", line 89,
> in <module>
> + os.rename(srcpath, dstpath)
> +OSError: [Errno 75] Value too large for defined data type
> Silence is golden
>
That's expected to fail, the commit for it landed in 5.11-rc2 actually
if you've taken the time to look into the test itself it even states the
name of the commit. Furthermore, the commit is tagged for stable 4.4 so
it will eventually land in all stable kernels.
