Re: [PATCH 1/2] sched: normalize sleeper's vruntime during group change

[Mike Galbraith <efault@gmx.de>]

On Tue, 2010-09-28 at 23:46 -0700, Dima Zavin wrote:
> This adds a new callback, prep_move_task, to struct sched_class
> to give sched_fair the opportunity to adjust the task's vruntime
> just before setting its new group.
> 
> This allows us to properly normalize a sleeping task's vruntime
> when moving it between different cgroups.


> diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
> index db3f674..008fe57 100644
> --- a/kernel/sched_fair.c
> +++ b/kernel/sched_fair.c
> @@ -3827,10 +3827,23 @@ static void set_curr_task_fair(struct rq *rq)
>  static void moved_group_fair(struct task_struct *p, int on_rq)
>  {
>  	struct cfs_rq *cfs_rq = task_cfs_rq(p);
> +	struct sched_entity *se = &p->se;
>  
>  	update_curr(cfs_rq);
> -	if (!on_rq)
> +	if (!on_rq) {
> +		se->vruntime += cfs_rq->min_vruntime;
>  		place_entity(cfs_rq, &p->se, 1);
> +	}
> +}

I'm still missing the place_entity() rationale.  Why is a sleeping task
eligible for pain therapy that a runnable task is spared?  That state
can change a usec from now.

Nit: place_entity(cfs_rq, se, 1);

	-Mike