Re: [PATCH -tip V3 0/8] workqueue: break affinity initiatively

[Lai Jiangshan <jiangshanlai@gmail.com>]

On Wed, Jan 13, 2021 at 7:11 PM Peter Zijlstra <peterz@infradead.org> wrote:
>
> On Tue, Jan 12, 2021 at 11:38:12PM +0800, Lai Jiangshan wrote:
>
> > But the hard problem is "how to suppress the warning of
> > online&!active in __set_cpus_allowed_ptr()" for late spawned
> > unbound workers during hotplug.
>
> I cannot see create_worker() go bad like that.
>
> The thing is, it uses:
>
>   kthread_bind_mask(, pool->attr->cpumask)
>   worker_attach_to_pool()
>     set_cpus_allowed_ptr(, pool->attr->cpumask)
>
> which means set_cpus_allowed_ptr() must be a NOP, because the affinity
> is already set by kthread_bind_mask(). Further, the first wakeup of that
> worker will then hit:
>
>   select_task_rq()
>     is_cpu_allowed()
>       is_per_cpu_kthread() -- false
>     select_fallback_rq()
>
>
> So normally that really isn't a problem. I can only see a tiny hole
> there, where someone changes the cpumask between kthread_bind_mask() and
> set_cpus_allowed_ptr(). AFAICT that can be fixed in two ways:
>
>  - add wq_pool_mutex around things in create_worker(), or
>  - move the set_cpus_allowed_ptr() out of worker_attach_to_pool() and
>    into rescuer_thread().
>
> Which then brings us to rescuer_thread...  If we manage to trigger the
> rescuer during hotplug, then yes, I think that can go wobbly.

Oh, I forgot set_cpus_allowed_ptr() is NOP when combined with
kthread_bind_mask()(create_worker()).

So the problem becomes "how to suppress the warning of online&!active in
__set_cpus_allowed_ptr()" for late *attached unbound rescuer* workers
during hotplug.


>
> Let me consider that a bit more while I try and make sense of that splat
> Paul reported.
>
> ---
>
> diff --git a/kernel/workqueue.c b/kernel/workqueue.c
> index ec0771e4a3fb..fe05308dc472 100644
> --- a/kernel/workqueue.c
> +++ b/kernel/workqueue.c
> @@ -1844,15 +1844,19 @@ static struct worker *alloc_worker(int node)
>   * cpu-[un]hotplugs.
>   */
>  static void worker_attach_to_pool(struct worker *worker,
> -                                  struct worker_pool *pool)
> +                                 struct worker_pool *pool,
> +                                 bool set_affinity)
>  {
>         mutex_lock(&wq_pool_attach_mutex);
>
> -       /*
> -        * set_cpus_allowed_ptr() will fail if the cpumask doesn't have any
> -        * online CPUs.  It'll be re-applied when any of the CPUs come up.
> -        */
> -       set_cpus_allowed_ptr(worker->task, pool->attrs->cpumask);
> +       if (set_affinity) {
> +               /*
> +                * set_cpus_allowed_ptr() will fail if the cpumask doesn't have
> +                * any online CPUs.  It'll be re-applied when any of the CPUs
> +                * come up.
> +                */
> +               set_cpus_allowed_ptr(worker->task, pool->attrs->cpumask);
> +       }
>
>         /*
>          * The wq_pool_attach_mutex ensures %POOL_DISASSOCIATED remains
> @@ -1944,7 +1948,7 @@ static struct worker *create_worker(struct worker_pool *pool)
>         kthread_bind_mask(worker->task, pool->attrs->cpumask);
>
>         /* successful, attach the worker to the pool */
> -       worker_attach_to_pool(worker, pool);
> +       worker_attach_to_pool(worker, pool, false);
>
>         /* start the newly created worker */
>         raw_spin_lock_irq(&pool->lock);
> @@ -2509,7 +2513,11 @@ static int rescuer_thread(void *__rescuer)
>
>                 raw_spin_unlock_irq(&wq_mayday_lock);
>
> -               worker_attach_to_pool(rescuer, pool);
> +               /*
> +                * XXX can go splat when running during hot-un-plug and
> +                * the pool affinity is wobbly.
> +                */
> +               worker_attach_to_pool(rescuer, pool, true);
>
>                 raw_spin_lock_irq(&pool->lock);
>