Re: [RFC][PATCH 5/5] sched: Reduce ttwu rq->lock contention

[Oleg Nesterov <oleg@redhat.com>]

On 12/16, Peter Zijlstra wrote:
>
> It does the state and on_rq checks first, if we find on_rq,

The problem is, somehow we should check both on_rq and state
at the same time,

> +try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
>  {
> -	int cpu, orig_cpu, this_cpu, success = 0;
> +	int cpu, load, ret = 0;
>  	unsigned long flags;
> -	unsigned long en_flags = ENQUEUE_WAKEUP;
> -	struct rq *rq;
>
> -	this_cpu = get_cpu();
> +	smp_mb();

Yes, we need the full mb(). without subsequent spin_lock(), wmb()
can't act as a smp_store_load_barrier() (which we don't have).

> +	if (p->se.on_rq && ttwu_force(p, state, wake_flags))
> +		return 1;

	----- WINDOW -----

> +	for (;;) {
> +		unsigned int task_state = p->state;
> +
> +		if (!(task_state & state))
> +			goto out;
> +
> +		load = task_contributes_to_load(p);
> +
> +		if (cmpxchg(&p->state, task_state, TASK_WAKING) == task_state)
> +			break;

Suppose that we have a task T sleeping in TASK_INTERRUPTIBLE state,
and this cpu does try_to_wake_up(TASK_INTERRUPTIBLE). on_rq == false.
try_to_wake_up() starts the "for (;;)" loop.

However, in the WINDOW above, it is possible that somebody else wakes
it up, and then this task changes its state to TASK_INTERRUPTIBLE again.

Then we set ->state = TASK_WAKING, but this (still running) T restores
TASK_RUNNING after us.

Oleg.