Re: [RFC PATCH v3 1/3] workqueue: Hold wq_pool_mutex while affining tasks to wq_unbound_cpumask

[Valentin Schneider]

On 03/08/22 11:40, Lai Jiangshan wrote:
> On 2022/8/2 16:41, Valentin Schneider wrote:
>> When unbind_workers() reads wq_unbound_cpumask to set the affinity of
>> freshly-unbound kworkers, it only holds wq_pool_attach_mutex. This isn't
>> sufficient as wq_unbound_cpumask is only protected by wq_pool_mutex.
>>
>> This is made more obvious as of recent commit
>>
>>    46a4d679ef88 ("workqueue: Avoid a false warning in unbind_workers()")
>>
>> e.g.
>>
>> unbind_workers()                             workqueue_set_unbound_cpumask()
>>    kthread_set_per_cpu(p, -1);
>>    if (cpumask_intersects(wq_unbound_cpumask, cpu_active_mask))
>>                                             cpumask_copy(wq_unbound_cpumask, cpumask);
>>      WARN_ON_ONCE(set_cpus_allowed_ptr(worker->task, wq_unbound_cpumask) < 0);
>>
>> Make workqueue_offline_cpu() invoke unbind_workers() with wq_pool_mutex
>> held.
>
> I would prefer to protect wq_unbound_cpumask with wq_pool_attach_mutex.

That looks alright to me, do you want to push that separately as it's a
standalone patch, or should I carry it with this series?