Messages in this thread Patch in this message | | | Subject | Re: PostgreSQL pgbench performance regression in 2.6.23+ | From | Peter Zijlstra <> | Date | Thu, 22 May 2008 14:24:41 +0200 |
| |
On Thu, 2008-05-22 at 14:09 +0200, Mike Galbraith wrote: > On Thu, 2008-05-22 at 13:44 +0200, Peter Zijlstra wrote: > > > Humm,.. how to fix this.. we'd need to somehow detect the 1:n nature of > > its operation - I'm sure there are other scenarios that could benefit > > from this. > > Maybe simple (minded): cache waker's last non-interrupt context wakee, > if the wakee != cached, ignore SYNC_WAKEUP unless sync was requested at > call time?
Yeah, something like so - or perhaps like you say cache the wakee.
I picked the wake_affine() condition, because I think that is the biggest factor in this behaviour. You could of course also disable all of sync.
diff --git a/include/linux/sched.h b/include/linux/sched.h index c86c5c5..856c2a8 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -950,6 +950,8 @@ struct sched_entity { u64 last_wakeup; u64 avg_overlap; + struct sched_entity *waker; + #ifdef CONFIG_SCHEDSTATS u64 wait_start; u64 wait_max; diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 894a702..8971044 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1036,7 +1036,8 @@ wake_affine(struct rq *rq, struct sched_domain *this_sd, struct rq *this_rq, * a reasonable amount of time then attract this newly * woken task: */ - if (sync && curr->sched_class == &fair_sched_class) { + if (sync && curr->sched_class == &fair_sched_class && + p->se.waker == curr->se->waker) { if (curr->se.avg_overlap < sysctl_sched_migration_cost && p->se.avg_overlap < sysctl_sched_migration_cost) return 1; @@ -1210,6 +1211,7 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p) if (unlikely(se == pse)) return; + se->waker = pse; cfs_rq_of(pse)->next = pse; /*
| |