From: john.p.donnelly@oracle.com
To: Waiman Long <longman@redhat.com>,
chenguanyou <chenguanyou9338@gmail.com>,
gregkh@linuxfoundation.org
Cc: dave@stgolabs.net, hdanton@sina.com,
linux-kernel@vger.kernel.org, mazhenhua@xiaomi.com,
mingo@redhat.com, peterz@infradead.org, quic_aiquny@quicinc.com,
will@kernel.org, sashal@kernel.org
Subject: Re: [PATCH v5] locking/rwsem: Make handoff bit handling more consistent
Date: Tue, 26 Apr 2022 16:22:05 -0500 [thread overview]
Message-ID: <c0a068e7-18e3-87df-676c-e8270cd732b6@oracle.com> (raw)
In-Reply-To: <020aef66-6911-77e7-fd1a-25506dfcd3df@redhat.com>
On 4/26/22 3:21 PM, Waiman Long wrote:
> On 4/20/22 09:55, john.p.donnelly@oracle.com wrote:
>> On 4/12/22 11:28 AM, john.p.donnelly@oracle.com wrote:
>>> On 4/11/22 4:07 PM, Waiman Long wrote:
>>>>
>>>> On 4/11/22 17:03, john.p.donnelly@oracle.com wrote:
>>>>>
>>>>>>>
>>>>>>> I have reached out to Waiman and he suggested this for our next
>>>>>>> test pass:
>>>>>>>
>>>>>>>
>>>>>>> 1ee326196c6658 locking/rwsem: Always try to wake waiters in
>>>>>>> out_nolock path
>>>>>>
>>>>>> Does this commit help to avoid the lockup problem?
>>>>>>
>>>>>> Commit 1ee326196c6658 fixes a potential missed wakeup problem when
>>>>>> a reader first in the wait queue is interrupted out without
>>>>>> acquiring the lock. It is actually not a fix for commit
>>>>>> d257cc8cb8d5. However, this commit changes the out_nolock path
>>>>>> behavior of writers by leaving the handoff bit set when the wait
>>>>>> queue isn't empty. That likely makes the missed wakeup problem
>>>>>> easier to reproduce.
>>>>>>
>>>>>> Cheers,
>>>>>> Longman
>>>>>>
>>>>>
>>>>> Hi,
>>>>>
>>>>>
>>>>> We are testing now
>>>>>
>>>>> ETA for fio soak test completion is ~15hr from now.
>>>>>
>>>>> I wanted to share the stack traces for future reference + occurrences.
>>>>>
>>>> I am looking forward to your testing results tomorrow.
>>>>
>>>> Cheers,
>>>> Longman
>>>>
>>> Hi
>>>
>>> Our 24hr fio soak test with :
>>>
>>> 1ee326196c6658 locking/rwsem: Always try to wake waiters in
>>> out_nolock path
>>>
>>>
>>> applied to 5.15.30 passed.
>>>
>>> I suggest you append 1ee326196c6658 with :
>>>
>>>
>>> cc: stable
>>>
>>> Fixes: d257cc8cb8d5 ("locking/rwsem: Make handoff bit handling
>>> more consistent")
>>>
>>>
>>> I'll leave the implementation details up to the core maintainers how
>>> to do that ;-)
>>>
>>> ...
>>>
>>> Thank you
>>>
>>> John.
>>
>> Hi ,
>>
>>
>> We have observed another panic with :
>>
>> 1ee326196c6658 locking/rwsem: Always try to wake waiters in out_nolock
>> path
>>
>> Applied to 5.15.30 :
>>
>>
> Sorry for the late reply as I was busy with other important tasks.
>
> When you said panic, you mean a system hang, not an actual panic. Right?
Hi ,
Our setups turn on all the panic on-hung-task , on-opps, all those
various features:
./sys/kernel/hardlockup_panic
./sys/kernel/hung_task_panic
./sys/kernel/max_rcu_stall_to_panic
./sys/kernel/panic
./sys/kernel/panic_on_io_nmi
./sys/kernel/panic_on_oops
./sys/kernel/panic_on_rcu_stall
./sys/kernel/panic_on_unrecovered_nmi
./sys/kernel/panic_on_warn
./sys/kernel/panic_print
./sys/kernel/softlockup_panic
./sys/kernel/unknown_nmi_panic
The machine is unusable when this occurs.
>
>
>> PID: 3789 TASK: ffff900fc409b300 CPU: 29 COMMAND: "dio/dm-0"
>> #0 [fffffe00006bce50] crash_nmi_callback at ffffffff97c772c3
>> #1 [fffffe00006bce58] nmi_handle at ffffffff97c40778
>> #2 [fffffe00006bcea0] default_do_nmi at ffffffff988161e2
>> #3 [fffffe00006bcec8] exc_nmi at ffffffff9881648d
>> #4 [fffffe00006bcef0] end_repeat_nmi at ffffffff98a0153b
>> [exception RIP: _raw_spin_lock_irq+35]
>> RIP: ffffffff98827333 RSP: ffffa9320917fc78 RFLAGS: 00000046
>> RAX: 0000000000000000 RBX: ffff900fc409b300 RCX: 0000000000000000
>> RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
>> RBP: ffffa9320917fd20 R8: 0000000000000000 R9: 0000000000000000
>> R10: 0000000000000000 R11: 0000000000000000 R12: ffff90006259546c
>> R13: ffffa9320917fcb0 R14: ffff900062595458 R15: 0000000000000000
>> ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018
>> --- <NMI exception stack> ---
>> #5 [ffffa9320917fc78] _raw_spin_lock_irq at ffffffff98827333
>> #6 [ffffa9320917fc78] rwsem_down_write_slowpath at ffffffff97d25d49
>> #7 [ffffa9320917fd28] ext4_map_blocks at ffffffffc104b6dc [ext4]
>> #8 [ffffa9320917fd98] ext4_convert_unwritten_extents at
>> ffffffffc10369e0 [ext4]
>> #9 [ffffa9320917fdf0] ext4_dio_write_end_io at ffffffffc103b2aa [ext4]
>> #10 [ffffa9320917fe18] iomap_dio_complete at ffffffff98013f45
>> #11 [ffffa9320917fe48] iomap_dio_complete_work at ffffffff98014047
>> #12 [ffffa9320917fe60] process_one_work at ffffffff97cd9191
>> #13 [ffffa9320917fea8] rescuer_thread at ffffffff97cd991b
>> #14 [ffffa9320917ff10] kthread at ffffffff97ce11f7
>> #15 [ffffa9320917ff50] ret_from_fork at ffffffff97c04cf2
>> crash>
>>
>>
>> The failure is observed running "fio test suite" as a 24 hour soak
>> test on an LVM composed of four NVME devices, Intel 72 core server.
>> The test cycles through a variety of file-system types.
>>
>>
>> This kernel has these commits
>>
>> 1ee326196c6658 locking/rwsem: Always try to wake waiters in
>> out_nolock path
>>
>> d257cc8cb8d5 ("locking/rwsem: Make handoff bit handling more consistent")
>>
>> In earlier testing I had reverted d257cc8cb8d5 and did not observe
>> said panics. I still feel d257cc8cb8d5 is still the root cause.
>
> So it is possible that 1ee326196c6658 does not completely eliminate the
> missed wakeup situation.
>
> Regards,
> Longman
>
next prev parent reply other threads:[~2022-04-26 21:22 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-11-16 1:29 [PATCH v5] locking/rwsem: Make handoff bit handling more consistent Waiman Long
2021-11-16 2:52 ` Aiqun(Maria) Yu
2021-11-16 9:14 ` Peter Zijlstra
2021-11-16 9:24 ` Peter Zijlstra
2021-11-16 14:52 ` Waiman Long
2021-11-17 13:36 ` Peter Zijlstra
2021-11-23 8:53 ` [tip: locking/urgent] " tip-bot2 for Waiman Long
2022-02-14 15:47 ` Re:[PATCH v5] " chenguanyou
2022-02-14 16:01 ` [PATCH " Greg KH
2022-04-11 18:26 ` john.p.donnelly
2022-04-11 18:40 ` Waiman Long
2022-04-11 21:03 ` john.p.donnelly
2022-04-11 21:07 ` Waiman Long
2022-04-12 16:28 ` john.p.donnelly
2022-04-12 17:04 ` Waiman Long
2022-04-14 10:48 ` Greg KH
2022-04-14 15:18 ` Waiman Long
2022-04-14 15:42 ` Greg KH
2022-04-14 15:44 ` Waiman Long
2022-04-20 13:55 ` john.p.donnelly
2022-04-26 20:21 ` Waiman Long
2022-04-26 21:22 ` john.p.donnelly [this message]
2022-02-14 16:22 ` chenguanyou
2022-02-15 7:41 ` [PATCH " Greg KH
2022-02-16 16:30 ` Waiman Long
2022-02-17 15:41 ` chenguanyou
2022-03-14 8:07 ` [PATCH " Greg KH
2022-03-22 2:49 ` chenguanyou
2022-03-24 12:51 ` [PATCH " Greg KH
2022-07-19 0:27 ` Doug Anderson
[not found] ` <20220719104104.1634-1-hdanton@sina.com>
2022-07-19 15:30 ` Doug Anderson
[not found] ` <20220722115510.2101-1-hdanton@sina.com>
2022-07-22 14:02 ` Doug Anderson
[not found] ` <20220723001713.2156-1-hdanton@sina.com>
2022-08-05 17:14 ` Doug Anderson
2022-08-05 19:02 ` Waiman Long
2022-08-05 19:16 ` Doug Anderson
2022-08-30 16:18 ` Doug Anderson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=c0a068e7-18e3-87df-676c-e8270cd732b6@oracle.com \
--to=john.p.donnelly@oracle.com \
--cc=chenguanyou9338@gmail.com \
--cc=dave@stgolabs.net \
--cc=gregkh@linuxfoundation.org \
--cc=hdanton@sina.com \
--cc=linux-kernel@vger.kernel.org \
--cc=longman@redhat.com \
--cc=mazhenhua@xiaomi.com \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=quic_aiquny@quicinc.com \
--cc=sashal@kernel.org \
--cc=will@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).