question about generic/473 on xfs filesystem

question about generic/473 on xfs filesystem

[Yang Xu]

Hi Eryu, Darrick

Currently, on lastest kernel, generic/473 fails ,as below:
# ./check generic/473
FSTYP         -- xfs (non-debug)
PLATFORM      -- 5.2.0-rc4+
MKFS_OPTIONS  -- -f -bsize=4096 /dev/sda11
MOUNT_OPTIONS -- -o context=system_u:object_r:root_t:s0 /dev/sda11 /mnt/xfstests/scratch
generic/473 2s ... - output mismatch (see /var/lib/xfstests/results//generic/473.out.bad)
    --- tests/generic/473.out   2019-06-12 19:30:37.219753498 +0800
    +++ /var/lib/xfstests/results//generic/473.out.bad  2019-06-12 20:34:12.499984113 +0800
    @@ -6,7 +6,7 @@
     1: [256..287]: hole
     Hole + Data
     0: [0..127]: hole
     -1: [128..255]: data
     +1: [128..135]: data
     Hole + Data + Hole
     0: [0..127]: hole

I have seen previous mail[1]that mentioned "new ranged fiemap test (generic/473) fails on XFS, because there's no 
final decision on the expected behavior yet and discussion has stalled somehow.". This case has been merged in 2017.12,
but this case also fails on lastest kernel, does it have any update or I miss something? 

more question:
xfs/107 and xfs/136 also fail on lastest kernel on my machine. 

[1]https://www.spinics.net/lists/fstests/msg08426.html

Re: question about generic/473 on xfs filesystem

[Darrick J. Wong]

On Wed, Jun 12, 2019 at 08:44:44PM +0800, Yang Xu wrote:
> Hi Eryu, Darrick
> 
> Currently, on lastest kernel, generic/473 fails ,as below:
> # ./check generic/473
> FSTYP         -- xfs (non-debug)
> PLATFORM      -- 5.2.0-rc4+
> MKFS_OPTIONS  -- -f -bsize=4096 /dev/sda11
> MOUNT_OPTIONS -- -o context=system_u:object_r:root_t:s0 /dev/sda11 /mnt/xfstests/scratch
> generic/473 2s ... - output mismatch (see /var/lib/xfstests/results//generic/473.out.bad)
>     --- tests/generic/473.out   2019-06-12 19:30:37.219753498 +0800
>     +++ /var/lib/xfstests/results//generic/473.out.bad  2019-06-12 20:34:12.499984113 +0800
>     @@ -6,7 +6,7 @@
>      1: [256..287]: hole
>      Hole + Data
>      0: [0..127]: hole
>      -1: [128..255]: data
>      +1: [128..135]: data
>      Hole + Data + Hole
>      0: [0..127]: hole
> 
> I have seen previous mail[1]that mentioned "new ranged fiemap test (generic/473) fails on XFS, because there's no 
> final decision on the expected behavior yet and discussion has stalled somehow.". This case has been merged in 2017.12,
> but this case also fails on lastest kernel, does it have any update or I miss something? 

Nope, you haven't missed anything.  The FIEMAP documentation says that
filesystems /can/ report more data than the range requested, but XFS
and ext4 still differ as to whether or not they will.

--D

> more question:
> xfs/107 and xfs/136 also fail on lastest kernel on my machine. 
> 
> [1]https://www.spinics.net/lists/fstests/msg08426.html
> 
> 
> 

Re: question about generic/473 on xfs filesystem

[Yang Xu]

on 2019/06/13 0:21, Darrick J. Wong wrote:

> On Wed, Jun 12, 2019 at 08:44:44PM +0800, Yang Xu wrote:
>> Hi Eryu, Darrick
>>
>> Currently, on lastest kernel, generic/473 fails ,as below:
>> # ./check generic/473
>> FSTYP         -- xfs (non-debug)
>> PLATFORM      -- 5.2.0-rc4+
>> MKFS_OPTIONS  -- -f -bsize=4096 /dev/sda11
>> MOUNT_OPTIONS -- -o context=system_u:object_r:root_t:s0 /dev/sda11 /mnt/xfstests/scratch
>> generic/473 2s ... - output mismatch (see /var/lib/xfstests/results//generic/473.out.bad)
>>      --- tests/generic/473.out   2019-06-12 19:30:37.219753498 +0800
>>      +++ /var/lib/xfstests/results//generic/473.out.bad  2019-06-12 20:34:12.499984113 +0800
>>      @@ -6,7 +6,7 @@
>>       1: [256..287]: hole
>>       Hole + Data
>>       0: [0..127]: hole
>>       -1: [128..255]: data
>>       +1: [128..135]: data
>>       Hole + Data + Hole
>>       0: [0..127]: hole
>>
>> I have seen previous mail[1]that mentioned "new ranged fiemap test (generic/473) fails on XFS, because there's no
>> final decision on the expected behavior yet and discussion has stalled somehow.". This case has been merged in 2017.12,
>> but this case also fails on lastest kernel, does it have any update or I miss something?
> Nope, you haven't missed anything.  The FIEMAP documentation says that
> filesystems /can/ report more data than the range requested, but XFS
> and ext4 still differ as to whether or not they will.
>
> --D
Hi Darrick

   Thanks. I got it.

>> more question:
>> xfs/107 and xfs/136 also fail on lastest kernel on my machine.
Hi Eryu, Darrick

These cases(xfs/107, xfs/136) are too old and they still fail on lastest kernel.
Can you give me more information about them?

xfs/107 fail information
+++ /var/lib/xfstests/results//xfs/107.out.bad      2019-06-13 14:33:03.345457763 +0800
     @@ -9,1542 +9,980 @@
      ### populate filesystem
      ### initial report
      [SCR_DEV] ([SCR_MNT]) Project:
     +   96168    #0
     +    4968    #3
     +    4344    #1
     +    3804    #2
     ...

xfs/136 fail information
   +++ /var/lib/xfstests/results//xfs/136.out.bad      2019-06-13 14:30:14.662466225 +0800
     @@ -1,5 +1,5 @@
      QA output created by 136
     -inum=67
     +inum=131
      core.format = 2 (extents)
      core.size = 0
      core.extsize = 0
     @@ -7,6 +7,7 @@
     ...

Thanks

>> [1]https://www.spinics.net/lists/fstests/msg08426.html
>>
>>
>>
>
>