Re: [RFC PATCH] workqueue: Fix rescuer task's name truncated

[Aaron Tomlin]

On Wed, Apr 24, 2024 at 11:12:24PM GMT, Wenchao Hao wrote:
> Hi Aaron, thanks a lot for your reply.

No problem.

> I think destroy_workqueue() may not race with wq_worker_comm(),
> wq_pool_attach_mutex is used to avoid race, below is my analysis.
> (Welcome to point out if my understand is incorrect)
> 
> t1 which call destroy_workqueue()     rescuer->task
> 
> destroy_workqueue()
>   kthread_stop(rescuer->task)
>                                       rescuer_thread()
> 	                                if (should_stop) {
> 	                                  __set_current_state(TASK_RUNNING);
> 	                                  set_pf_worker(false);
>                                             mutex_lock(&wq_pool_attach_mutex);
> 				            current->flags &= ~PF_WQ_WORKER;
> 				            mutex_unlock(&wq_pool_attach_mutex);
> 	                                  return 0;
> 	                                }
> 
>   kfree(rescuer)
> 
> wq_worker_comm() would acquire wq_pool_attach_mutex then check if task->flags
> is set PF_WQ_WORKER.
> If PF_WQ_WORKER is not set, wq_worker_comm() would not access this task's worker
> any more;
> If PF_WQ_WORKER is set, the wq_pool_attach_mutex is held durning access of task's
> worker.

Indeed. If I understand correctly then a use-after-free is theoretically
impossible due to the use of the 'wq_pool_attach_mutex' in the context of
rescuer_thread() and wq_pool_attach_mutex(), respectively.

> What confuse me mostly is why the origin logic only append worker's desc when
> worker is attached to a work pool.

I can only assume there was no intention to use the rescuer kworker's
description information for this purpose. Your patch looks fine to me now.




-- 
Aaron Tomlin