Re: [RFC][PATCH 08/17] sched: Drop the rq argument to sched_class::select_task_rq()

[Yong Zhang]

On Mon, Jan 3, 2011 at 11:49 PM, Oleg Nesterov <oleg@xxxxxxxxxx> wrote:
>
> Ah, sorry for the confusion, I only meant sched_exec() case.
> set_cpus_allowed_ptr() does need need_migrate_task(), of course.
>
>
> As for set_cpus_allowed_ptr()->need_migrate_task() path, I have another
> question,
>
> Â Â Â Âstatic bool need_migrate_task(struct task_struct *p)
> Â Â Â Â{
> Â Â Â Â Â Â Â Â/*
> Â Â Â Â Â Â Â Â * If the task is not on a runqueue (and not running), then
> Â Â Â Â Â Â Â Â * the next wake-up will properly place the task.
> Â Â Â Â Â Â Â Â */
> Â Â Â Â Â Â Â Âsmp_rmb(); /* finish_lock_switch() */
> Â Â Â Â Â Â Â Âreturn p->on_rq || p->on_cpu;
> Â Â Â Â}
>
> I don't understand this smp_rmb(). Yes, finish_lock_switch() does
> wmb() before it clears ->on_cpu, but how these 2 barriers can pair?
>
> In fact, I am completely confused. I do not understand why do we
> check task_running() at all. If we see on_rq == 0 && on_cpu == 1,
> then this task is going to clear its on_cpu soon, once it finishes
> context_switch().
>
> Probably, this check was needed before, try_to_wake_up() could
> activate the task_running() task without migrating. But, at first
> glance, this is no longer possible after this series?

Yeah, task_running() is not needed after patch 13 which
may be the suitable place to poke :)

Thanks,
Yong

-- 
Only stand for myself
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/