Re: [patch] don't preempt not TASK_RUNNING tasks

[Peter Zijlstra <peterz@infradead.org>]

On Fri, 2009-03-20 at 10:43 +0100, Miklos Szeredi wrote:
> Ingo,
> 
> I tested this one, and I think it makes sense in any case as an
> optimization.  It should also be good for -stable kernels.
> 
> Does it look OK?

The idea is good, but there is a risk of preemption latencies here. Some
code paths aren't real quick between setting ->state != TASK_RUNNING and
calling schedule.

[ Both quick: as in O(1) and few instructions ]

So if we're going to do this, we'd need to audit all such code paths --
and there be lots.

The first line of attack for this problem is making wait_task_inactive()
sucks less, which shouldn't be too hard, that unconditional 1 jiffy
sleep is simply retarded.

> Index: linux.git/kernel/sched.c
> ===================================================================
> --- linux.git.orig/kernel/sched.c	2009-03-20 09:40:47.000000000 +0100
> +++ linux.git/kernel/sched.c	2009-03-20 10:28:56.000000000 +0100
> @@ -4632,6 +4632,10 @@ asmlinkage void __sched preempt_schedule
>  	if (likely(ti->preempt_count || irqs_disabled()))
>  		return;
>  
> +	/* No point in preempting we are just about to go to sleep. */
> +	if (current->state != TASK_RUNNING)
> +		return;
> +
>  	do {
>  		add_preempt_count(PREEMPT_ACTIVE);
>  		schedule();