Re: [PATCH] xfs: correct statx's result_mask value

[Eric Sandeen]

On 1/7/19 11:15 PM, Su Yanjun <suyj.fnst@xxxxxxxxxxxxxx> wrote:
> 
> 
> On 1/8/2019 1:07 PM, Darrick J. Wong wrote:
>> On Tue, Jan 08, 2019 at 12:58:43PM +0800, Su Yanjun <suyj.fnst@xxxxxxxxxxxxxx> wrote:
>>>
>>> On 1/8/2019 2:04 AM, Eric Sandeen wrote:
>>>> On 1/7/19 11:52 AM, Darrick J. Wong wrote:
>>>>> On Mon, Jan 07, 2019 at 04:53:10AM -0500, Su Yanjun wrote:
>>>>>> For statx syscall, xfs return the wrong result_mask.
>>>>>>
>>>>>> Signed-off-by: Su Yanjun<suyj.fnst@xxxxxxxxxxxxxx>
>>>>>> ---
>>>>>> ÂÂ fs/xfs/xfs_iops.c | 3 +++
>>>>>> ÂÂ 1 file changed, 3 insertions(+)
>>>>>>
>>>>>> diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c
>>>>>> index f48ffd7..3811457 100644
>>>>>> --- a/fs/xfs/xfs_iops.c
>>>>>> +++ b/fs/xfs/xfs_iops.c
>>>>>> @@ -521,6 +521,9 @@ xfs_vn_getattr(
>>>>>> ÂÂÂÂÂÂÂÂÂÂÂÂÂÂ stat->btime.tv_nsec = ip->i_d.di_crtime.t_nsec;
>>>>>> ÂÂÂÂÂÂÂÂÂÂ }
>>>>>> ÂÂÂÂÂÂ }
>>>>>> +ÂÂÂ
>>>>>> +ÂÂÂ /* Only return mask that we care */
>>>>>> +ÂÂÂ stat->result_mask &= request_mask;
>>>>> Why not just:
>>>>>
>>>>> ÂÂÂÂstat->result_mask = STATX_BASIC_STATS;
>>>>>
>>>>> at the top of the function?
>>>>>
>>>>> I don't see the need to mask off result_mask at all, since we could some
>>>>> day elect to return more than what's in request_mask...
>>> When we run xfstests with nfs, the generic/423 case runs failed. So i review
>>> the nfs'
>>> nfs_getattr code it does validate the request_mask.
>>>
>>> Then i review the xfs' getattr code, it has no such check. Whatever
>>> request_mask
>>> ÂÂis set, the stat's result_mask always the 0x7ff.
>> Yes, statx can return more data than what userspace callers ask for:
>>
>>> Maybe it has Unclear semantics about statx's result_mask.
>> "A filesystem may also fill in fields that the caller didn't ask for if
>> it has values for them available and the information is available at no
>> extra cost. If this happens, the corresponding bits will be set in
>> stx_mask."
>>
>> --D
> 
> I get it, then the testcase generic/423 may need update in xfstests.
> Thanks for your reply.

Can you please share the details of the failure you're seeing when you run
it over nfs?

Thanks,
-=Eric