Commit 95d68593 authored by Vincent Guittot's avatar Vincent Guittot Committed by Peter Zijlstra

sched/pelt: Sync util/runnable_sum with PELT window when propagating

update_tg_cfs_*() propagate the impact of the attach/detach of an entity
down into the cfs_rq hierarchy and must keep the sync with the current pelt
window.

Even if we can't sync child cfs_rq and its group se, we can sync the group
se and its parent cfs_rq with current position in the PELT window. In fact,
we must keep them sync in order to stay also synced with others entities
and group entities that are already attached to the cfs_rq.
Signed-off-by: default avatarVincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: default avatarPeter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200506155301.14288-1-vincent.guittot@linaro.org
parent 12aa2587
......@@ -3441,52 +3441,46 @@ static inline void
update_tg_cfs_util(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq *gcfs_rq)
{
long delta = gcfs_rq->avg.util_avg - se->avg.util_avg;
/*
* cfs_rq->avg.period_contrib can be used for both cfs_rq and se.
* See ___update_load_avg() for details.
*/
u32 divider = LOAD_AVG_MAX - 1024 + cfs_rq->avg.period_contrib;
/* Nothing to update */
if (!delta)
return;
/*
* The relation between sum and avg is:
*
* LOAD_AVG_MAX - 1024 + sa->period_contrib
*
* however, the PELT windows are not aligned between grq and gse.
*/
/* Set new sched_entity's utilization */
se->avg.util_avg = gcfs_rq->avg.util_avg;
se->avg.util_sum = se->avg.util_avg * LOAD_AVG_MAX;
se->avg.util_sum = se->avg.util_avg * divider;
/* Update parent cfs_rq utilization */
add_positive(&cfs_rq->avg.util_avg, delta);
cfs_rq->avg.util_sum = cfs_rq->avg.util_avg * LOAD_AVG_MAX;
cfs_rq->avg.util_sum = cfs_rq->avg.util_avg * divider;
}
static inline void
update_tg_cfs_runnable(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq *gcfs_rq)
{
long delta = gcfs_rq->avg.runnable_avg - se->avg.runnable_avg;
/*
* cfs_rq->avg.period_contrib can be used for both cfs_rq and se.
* See ___update_load_avg() for details.
*/
u32 divider = LOAD_AVG_MAX - 1024 + cfs_rq->avg.period_contrib;
/* Nothing to update */
if (!delta)
return;
/*
* The relation between sum and avg is:
*
* LOAD_AVG_MAX - 1024 + sa->period_contrib
*
* however, the PELT windows are not aligned between grq and gse.
*/
/* Set new sched_entity's runnable */
se->avg.runnable_avg = gcfs_rq->avg.runnable_avg;
se->avg.runnable_sum = se->avg.runnable_avg * LOAD_AVG_MAX;
se->avg.runnable_sum = se->avg.runnable_avg * divider;
/* Update parent cfs_rq runnable */
add_positive(&cfs_rq->avg.runnable_avg, delta);
cfs_rq->avg.runnable_sum = cfs_rq->avg.runnable_avg * LOAD_AVG_MAX;
cfs_rq->avg.runnable_sum = cfs_rq->avg.runnable_avg * divider;
}
static inline void
......@@ -3496,19 +3490,26 @@ update_tg_cfs_load(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq
unsigned long load_avg;
u64 load_sum = 0;
s64 delta_sum;
u32 divider;
if (!runnable_sum)
return;
gcfs_rq->prop_runnable_sum = 0;
/*
* cfs_rq->avg.period_contrib can be used for both cfs_rq and se.
* See ___update_load_avg() for details.
*/
divider = LOAD_AVG_MAX - 1024 + cfs_rq->avg.period_contrib;
if (runnable_sum >= 0) {
/*
* Add runnable; clip at LOAD_AVG_MAX. Reflects that until
* the CPU is saturated running == runnable.
*/
runnable_sum += se->avg.load_sum;
runnable_sum = min(runnable_sum, (long)LOAD_AVG_MAX);
runnable_sum = min_t(long, runnable_sum, divider);
} else {
/*
* Estimate the new unweighted runnable_sum of the gcfs_rq by
......@@ -3533,7 +3534,7 @@ update_tg_cfs_load(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq
runnable_sum = max(runnable_sum, running_sum);
load_sum = (s64)se_weight(se) * runnable_sum;
load_avg = div_s64(load_sum, LOAD_AVG_MAX);
load_avg = div_s64(load_sum, divider);
delta_sum = load_sum - (s64)se_weight(se) * se->avg.load_sum;
delta_avg = load_avg - se->avg.load_avg;
......@@ -3697,6 +3698,10 @@ update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq)
*/
static void attach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se)
{
/*
* cfs_rq->avg.period_contrib can be used for both cfs_rq and se.
* See ___update_load_avg() for details.
*/
u32 divider = LOAD_AVG_MAX - 1024 + cfs_rq->avg.period_contrib;
/*
......
......@@ -237,6 +237,30 @@ ___update_load_sum(u64 now, struct sched_avg *sa,
return 1;
}
/*
* When syncing *_avg with *_sum, we must take into account the current
* position in the PELT segment otherwise the remaining part of the segment
* will be considered as idle time whereas it's not yet elapsed and this will
* generate unwanted oscillation in the range [1002..1024[.
*
* The max value of *_sum varies with the position in the time segment and is
* equals to :
*
* LOAD_AVG_MAX*y + sa->period_contrib
*
* which can be simplified into:
*
* LOAD_AVG_MAX - 1024 + sa->period_contrib
*
* because LOAD_AVG_MAX*y == LOAD_AVG_MAX-1024
*
* The same care must be taken when a sched entity is added, updated or
* removed from a cfs_rq and we need to update sched_avg. Scheduler entities
* and the cfs rq, to which they are attached, have the same position in the
* time segment because they use the same clock. This means that we can use
* the period_contrib of cfs_rq when updating the sched_avg of a sched_entity
* if it's more convenient.
*/
static __always_inline void
___update_load_avg(struct sched_avg *sa, unsigned long load)
{
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment