sched: fixlet for group load balance
We should not only correct the increment for the initial group, but should be consistent and do so for all the groups we encounter. Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Ingo Molnar <mingo@elte.hu>
This commit is contained in:
parent
63e5c39859
commit
940959e939
|
@ -1027,7 +1027,6 @@ static long effective_load(struct task_group *tg, int cpu,
|
|||
long wl, long wg)
|
||||
{
|
||||
struct sched_entity *se = tg->se[cpu];
|
||||
long more_w;
|
||||
|
||||
if (!tg->parent)
|
||||
return wl;
|
||||
|
@ -1039,6 +1038,10 @@ static long effective_load(struct task_group *tg, int cpu,
|
|||
if (!wl && sched_feat(ASYM_EFF_LOAD))
|
||||
return wl;
|
||||
|
||||
for_each_sched_entity(se) {
|
||||
long S, rw, s, a, b;
|
||||
long more_w;
|
||||
|
||||
/*
|
||||
* Instead of using this increment, also add the difference
|
||||
* between when the shares were last updated and now.
|
||||
|
@ -1047,11 +1050,6 @@ static long effective_load(struct task_group *tg, int cpu,
|
|||
wl += more_w;
|
||||
wg += more_w;
|
||||
|
||||
for_each_sched_entity(se) {
|
||||
#define D(n) (likely(n) ? (n) : 1)
|
||||
|
||||
long S, rw, s, a, b;
|
||||
|
||||
S = se->my_q->tg->shares;
|
||||
s = se->my_q->shares;
|
||||
rw = se->my_q->rq_weight;
|
||||
|
@ -1059,7 +1057,11 @@ static long effective_load(struct task_group *tg, int cpu,
|
|||
a = S*(rw + wl);
|
||||
b = S*rw + s*wg;
|
||||
|
||||
wl = s*(a-b)/D(b);
|
||||
wl = s*(a-b);
|
||||
|
||||
if (likely(b))
|
||||
wl /= b;
|
||||
|
||||
/*
|
||||
* Assume the group is already running and will
|
||||
* thus already be accounted for in the weight.
|
||||
|
@ -1068,7 +1070,6 @@ static long effective_load(struct task_group *tg, int cpu,
|
|||
* alter the group weight.
|
||||
*/
|
||||
wg = 0;
|
||||
#undef D
|
||||
}
|
||||
|
||||
return wl;
|
||||
|
|
Loading…
Reference in New Issue