Re: [patch 13/16] sched: update_cfs_shares at period edge

["\"Jan H. Schönherr\"" <schnhrr@cs.tu-berlin.de>]

Am 23.08.2012 16:14, schrieb pjt@google.com:
> From: Paul Turner <pjt@google.com>
> 
> Now that our measurement intervals are small (~1ms) we can amortize the posting
> of update_shares() to be about each period overflow.  This is a large cost
> saving for frequently switching tasks.

[snip]

> @@ -1181,6 +1181,7 @@ static void update_cfs_rq_blocked_load(struct cfs_rq *cfs_rq, int force_update)
>  	}
>  
>  	__update_cfs_rq_tg_load_contrib(cfs_rq, force_update);
> +	update_cfs_shares(cfs_rq);
>  }

Here a call to update_cfs_shares() gets added. Doesn't that make the call to
update_cfs_shares() in __update_blocked_averages_cpu() superfluous?


Function pasted here for reference:

static void __update_blocked_averages_cpu(struct task_group *tg, int cpu)
{
	struct sched_entity *se = tg->se[cpu];
	struct cfs_rq *cfs_rq = tg->cfs_rq[cpu];

	/* throttled entities do not contribute to load */
	if (throttled_hierarchy(cfs_rq))
		return;

	update_cfs_rq_blocked_load(cfs_rq, 1);

	if (se) {
		update_entity_load_avg(se, 1);
		/*
		 * We can pivot on the runnable average decaying to zero for
		 * list removal since the parent average will always be >=
		 * child.
		 */
		if (se->avg.runnable_avg_sum)
			update_cfs_shares(cfs_rq);
		else
			list_del_leaf_cfs_rq(cfs_rq);
	} else {
		struct rq *rq = rq_of(cfs_rq);
		update_rq_runnable_avg(rq, rq->nr_running);
	}
}


Regards
Jan