Re: [PATCH v15 00/26] nfs/nfsd: add support for LOCALIO

[Anna Schumaker <anna.schumaker@oracle.com>]

Hi Mike,

On 9/10/24 3:14 PM, Mike Snitzer wrote:
> On Tue, Sep 10, 2024 at 12:45:11PM -0400, Mike Snitzer wrote:
>> On Fri, Sep 06, 2024 at 04:34:18PM -0400, Mike Snitzer wrote:
>>> On Fri, Sep 06, 2024 at 03:31:41PM -0400, Anna Schumaker wrote:
>>>> Hi Mike,
>>>>
>>>> I've been running tests on localio this afternoon after finishing up going through v15 of the patches (I was most of the way through when you posted v16, so I haven't updated yet!). Cthon tests passed on all NFS versions, and xfstests passed on NFS v4.x. However, I saw this crash from xfstests with NFS v3:
>>>>
>>>> [ 1502.440896] run fstests generic/633 at 2024-09-06 14:04:17
>>>> [ 1502.694356] process 'vfstest' launched '/dev/fd/4/file1' with NULL argv: empty string added
>>>> [ 1502.699514] Oops: general protection fault, probably for non-canonical address 0x6c616e69665f6140: 0000 [#1] PREEMPT SMP NOPTI
>>>> [ 1502.700970] CPU: 3 UID: 0 PID: 513 Comm: nfsd Not tainted 6.11.0-rc6-g0c79a48cd64d-dirty+ #42323 70d41673e6cbf8e3437eb227e0a9c3c46ed3b289
>>>> [ 1502.702506] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS unknown 2/2/2022
>>>> [ 1502.703593] RIP: 0010:nfsd_cache_lookup+0x2b3/0x840 [nfsd]
> 
> <snip>
> 
>>>>
>>>> Please let me know if there are any other details you need about my setup to help debug this!
>>>
>>> Hmm, I haven't seen this issue, my runs of xfstests with LOCALIO
>>> enabled look solid:
>>> https://evilpiepirate.org/~testdashboard/ci?user=snitzer&branch=snitm-nfs-next&test=^fs.nfs.fstests.generic.633$
>>>
>>> And I know Chuck has been testing xfstests and more with the patches
>>> applied but LOCALIO disabled in his kernel config.
>>>
>>> The stack seems to indicate nfsd is just handling a request (so it
>>> isn't using LOCALIO, at least not for this op).
>>>
>>> Probably best if you do try v16.  v15 has issues v16 addressed.  If
>>> you can reproduce with v16 please share your kernel .config and
>>> xfstests config. 
>>>
>>> Note that I've only really tested my changes against v6.11-rc4.  But I
>>> can rebase on v6.11-rc6 if you find v16 still fails for you.
>>
>> Hi Anna,
>>
>> Just checking back, how is LOCALIO for you at this point?  Anything
>> you're continuing to see as an issue or need from me?
> 
> In case it helps, I did just rebase LOCALIO (v16 + 1 fix) ontop of
> cel/nfsd-next (v6.11-rc6 based), and I've pushed the result here:
> https://git.kernel.org/pub/scm/linux/kernel/git/snitzer/linux.git/log/?h=nfs-localio-for-next

I'm seeing the same hang on generic/525 with your latest branch.

Anna

> 
> ktest is running xfstests against it (LOCALIO enabled and in use):
> https://evilpiepirate.org/~testdashboard/ci?user=snitzer&branch=snitm-nfs-next
> 
> And Chuck's kdevops testing should test it tomorrow morning.