Re: [PATCH 4/7 v3] sched: propagate load during synchronous attach/detach

On 15 September 2016 at 16:43, Peter Zijlstra peterz@infradead.org wrote:

On Mon, Sep 12, 2016 at 09:47:49AM +0200, Vincent Guittot wrote:

...

+static inline void +update_tg_cfs_load(struct cfs_rq *cfs_rq, struct sched_entity *se) +{

• struct cfs_rq *gcfs_rq = group_cfs_rq(se);
  

• long delta, load = gcfs_rq->avg.load_avg;
  

• /* If the load of group cfs_rq is null, the load of the
  

•  * sched_entity will also be null so we can skip the formula
  

•  */
  

• if (load) {
  

•         long tg_load;
  

•         /* Get tg's load and ensure tg_load > 0 */
  

•         tg_load = atomic_long_read(&gcfs_rq->tg->load_avg) + 1;
  

•         /* Ensure tg_load >= load and updated with current load*/
  

•         tg_load -= gcfs_rq->tg_load_avg_contrib;
  

•         tg_load += load;
  

•         /* scale gcfs_rq's load into tg's shares*/
  

•         load *= scale_load_down(gcfs_rq->tg->shares);
  

•         load /= tg_load;
  

•         /*
  

•          * we need to compute a correction term in the case that the
  

•          * task group is consuming <1 cpu so that we would contribute
  

•          * the same load as a task of equal weight.
  

•         */
  

•         if (tg_load < scale_load_down(gcfs_rq->tg->shares)) {
  

•                 load *= tg_load;
  

•                 load /= scale_load_down(gcfs_rq->tg->shares);
  

•         }
  

Note that you're reversing the exact scaling you just applied.

Yes, Indeed

That is: shares tg_load load * ------- * ------- == load tg_load shares

...

• }
  

So something like:

    shares = scale_load_down(gcfs_rq->tg->shares);

    if (tg_load >= shares) {
            load *= shares;
            load /= tg_load;
    }

should be the same as the above and saves a bunch of math, no?

Yes