Re: [PATCH v1] kthread/smpboot: Serialize kthread parking against wakeup

From: "Kohli, Gaurav" <gkohli@codeaurora.org>
To: Peter Zijlstra <peterz@infradead.org>
Cc: tglx@linutronix.de, mpe@ellerman.id.au, mingo@kernel.org,
	bigeasy@linutronix.de, linux-kernel@vger.kernel.org,
	linux-arm-msm@vger.kernel.org,
	Neeraj Upadhyay <neeraju@codeaurora.org>,
	Will Deacon <will.deacon@arm.com>,
	Oleg Nesterov <oleg@redhat.com>
Subject: Re: [PATCH v1] kthread/smpboot: Serialize kthread parking against wakeup
Date: Tue, 5 Jun 2018 16:43:45 +0530	[thread overview]
Message-ID: <ba12bc49-33c0-b4ee-05cf-f84124bd66cc@codeaurora.org> (raw)
In-Reply-To: <55221a5b-dd52-3359-f582-86830dd9f205@codeaurora.org>

Hi Peter,

As last mentioned on mail, we are still seeing issue with the latest 
approach and below is the susceptible race as mentioned earlier..
controller Thread                               CPUHP Thread
takedown_cpu
kthread_park
kthread_parkme
Set KTHREAD_SHOULD_PARK
                                                 smpboot_thread_fn
                                                 set Task interruptible

wake_up_process
  if (!(p->state & state))
                 goto out;

                                                 Kthread_parkme
                                                 SET TASK_PARKED
                                                 schedule
                                                 raw_spin_lock(&rq->lock)
ttwu_remote
waiting for __task_rq_lock
                                                 context_switch

                                                 finish_lock_switch

                                                 Case TASK_PARKED
                                                 kthread_park_complete

SET Running

So it seems issue is still their with the latest mentioned fix
kthread, sched/wait: Fix kthread_parkme() completion issue.

Regards
Gaurav

On 5/7/2018 4:53 PM, Kohli, Gaurav wrote:
> Corrected the formatting, Sorry for spam.
> 
> 
>>
>> HI Peter,
>>
>> We have tested with new patch and still seeing same issue, in this 
>> dumps we don't have debug traces, but seems there still exist race 
>> from code review , Can you please check it once:
>>
>> Controller Thread                               CPUHP Thread
>> takedown_cpu
>> kthread_park
>> kthread_parkme
>> Set KTHREAD_SHOULD_PARK
>>                                                 smpboot_thread_fn
>>                                                 set Task interruptible
>>
>>
>> wake_up_process
>>
>>                                                 Kthread_parkme
>>                                                 SET TASK_PARKED
>>                                                 schedule
>>                                                 raw_spin_lock(&rq->lock)
>>
>>                                                 context_switch
>>
>>                                                 finish_lock_switch
>>
>>
>>
>>                                                 Case TASK_PARKED
>>                                                 kthread_park_complete
>>
>>
>> SET TASK_INTERRUPTIBLE
>>
>>
>> And also seeing the same warning during unpark of cpuhp from controller:
>>   if (!wait_task_inactive(p, state)) {
>>                  WARN_ON(1);
>>                  return;
>>          }
>> 325.065893] [<ffffff8920ed0200>] kthread_unpark+0x80/0xd8
>> [  325.065902] [<ffffff8920eab754>] bringup_cpu+0xa0/0x12c
>> [  325.065910] [<ffffff8920eaae90>] cpuhp_invoke_callback+0xb4/0x5c8
>> [  325.065917] [<ffffff8920eabd98>] cpuhp_up_callbacks+0x3c/0x154
>> [  325.065924] [<ffffff8920ead220>] _cpu_up+0x134/0x208
>> [  325.065931] [<ffffff8920ead45c>] do_cpu_up+0x168/0x1a0
>> [  325.065938] [<ffffff8920ead4b8>] cpu_up+0x24/0x30
>> [  325.065948] [<ffffff89215b1408>] cpu_subsys_online+0x20/0x2c
>> [  325.065956] [<ffffff89215aac64>] device_online+0x70/0xb4
>> [  325.065962] [<ffffff89215aad78>] online_store+0xd0/0xdc
>> [  325.065971] [<ffffff89215a7424>] dev_attr_store+0x40/0x54
>> [  325.065982] [<ffffff89210d8a98>] sysfs_kf_write+0x5c/0x74
>> [  325.065988] [<ffffff89210d7b9c>] kernfs_fop_write+0xcc/0x1ec
>> [  325.065999] [<ffffff8921049288>] vfs_write+0xb4/0x1d0
>> [  325.066006] [<ffffff892104a858>] SyS_write+0x60/0xc0
>> [  325.066014] [<ffffff8920e83770>] el0_svc_naked+0x24/0x28
>>
>>
>> And after this same crash occured:
>> [  325.521307] [<ffffff8920ed4aac>] smpboot_thread_fn+0x26c/0x2c8
>> [  325.527295] [<ffffff8920ecfb24>] kthread+0xf4/0x108
>>
>> I will put more debug ftraces to check what is going on exactly.
>>
>> Regards
>> Gaurav
>>
>>
>>
>>
> 

-- 
Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, 
Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project.