On 11/16/23 1:05 AM, Yang Xu (Fujitsu) wrote:
> Hi Zorro
>
>> As we've filter out the inode number, so the "inode numbers control" doesn't make sense anymore. It's fine for me to remove the "-d size=xxx" part (if it's still test passed).
> I have tested and pass. Will send a v2.
>
> Best Regards
> Yang Xu
>
> -----Original Message-----
> From: Zorro Lang <zlang@xxxxxxxxxx>
> Sent: Thursday, November 16, 2023 10:34 AM
> To: Xu, Yang/徐 杨 <xuyang2018.jy@xxxxxxxxxxx>
> Cc: Darrick J. Wong <djwong@xxxxxxxxxx>; fstests@xxxxxxxxxxxxxxx
> Subject: Re: [PATCH] xfs/263: increase data section size to 1024M
>
> On Wed, Nov 15, 2023 at 08:05:43AM -0800, Darrick J. Wong wrote:
>> On Wed, Nov 15, 2023 at 12:56:59AM -0500, Yang Xu wrote:
>>> On machine with using raid, this case will trigger the following
>>> error:
>>> ==== NO CRC ====
>>> +mkfs.xfs: small data volume, ignoring data volume stripe unit 512
>>> +and stripe width 512
>>> == Options: rw ==
>>> == Options: usrquota,rw ==
>>>
>>> mkfs.xfs generates this error since xfsprogs commit 42371fb36
>>> ("mkfs: ignore data blockdev stripe geometry for small filesystems").
>>> It disables automatic detection of stripe unit and width if the data
>>> device is less than 1GB.
>>>
>>> To slove false poistive, just increase data section size to 1G.
>>
>> Is there a particular reason why this test needs -d size= at all?
>>
>> There's a single comment about "Control size to control inode numbers"
>> but then filter_quota_state() seds the inode numbers to #XXX. So
>> perhaps that part of the mkfs argument isn't necessary anymore?
>
> As we've filter out the inode number, so the "inode numbers control" doesn't make sense anymore. It's fine for me to remove the "-d size=xxx" part (if it's still test passed).
Sorry for the late reply, but I agree that it's not needed now.
(I wonder if we should just filter the "small data volume, ignoring data volume
stripe" message in a common filter, but we don't need to tie up this fix, I
think it's fine to change the size for this test.)
Thanks,
-Eric
