Messages in this thread | | | From | Vincent Guittot <> | Date | Tue, 21 Jan 2014 21:06:19 +0100 | Subject | Re: [PATCH] sched: fix sched_entity avg statistics update |
| |
On 21 January 2014 19:38, <bsegall@google.com> wrote: > Vincent Guittot <vincent.guittot@linaro.org> writes: > >> With the current implementation, the load average statistics of a sched entity >> change according to other activity on the CPU even if this activity is done >> between the running window of the sched entity and have no influence on the >> running duration of the task. >> >> When a task wakes up on the same CPU, we currently update last_runnable_update >> with the return of __synchronize_entity_decay without updating the >> runnable_avg_sum and runnable_avg_period accordingly. In fact, we have to sync >> the load_contrib of the se with the rq's blocked_load_contrib before removing >> it from the latter (with __synchronize_entity_decay) but we must keep >> last_runnable_update unchanged for updating runnable_avg_sum/period during the >> next update_entity_load_avg. > > ... Gah, that's correct, we had this right the first time. Could you do > this as a full revert of 282cf499f03ec1754b6c8c945c9674b02631fb0f (ie > remove the now inaccurate comment, or maybe replace it with a correct one).
Ok i'm going to remove comment as well and replace it with a new description
Vincent
>> >> Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org> >> --- >> kernel/sched/fair.c | 3 +-- >> 1 file changed, 1 insertion(+), 2 deletions(-) >> >> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c >> index e64b079..5b0ef90 100644 >> --- a/kernel/sched/fair.c >> +++ b/kernel/sched/fair.c >> @@ -2370,8 +2370,7 @@ static inline void enqueue_entity_load_avg(struct cfs_rq *cfs_rq, >> * would have made count negative); we must be careful to avoid >> * double-accounting blocked time after synchronizing decays. >> */ >> - se->avg.last_runnable_update += __synchronize_entity_decay(se) >> - << 20; >> + __synchronize_entity_decay(se); >> } >> >> /* migrated tasks did not contribute to our blocked load */
| |