Re: [PATCH] sched: Skip double execution of pick_next_task_fair

2014-04-24 Thread Peter Zijlstra
On Thu, Apr 24, 2014 at 10:06:07AM -0700, Tim Chen wrote: > Yes, this version is more concise. > > > > > Its a little more contained. > > > > --- a/kernel/sched/core.c > > +++ b/kernel/sched/core.c > > @@ -2636,8 +2636,14 @@ pick_next_task(struct rq *rq, struct tas > > if (likely(prev->sched

Re: [PATCH] sched: Skip double execution of pick_next_task_fair

2014-04-24 Thread Tim Chen
On Thu, 2014-04-24 at 12:00 +0200, Peter Zijlstra wrote: > On Wed, Apr 23, 2014 at 03:31:57PM -0700, Tim Chen wrote: > > The current code will call pick_next_task_fair a second time > > in the slow path if we did not pull any task in our first try. > > This is really unnecessary as we already know

Re: [PATCH] sched: Skip double execution of pick_next_task_fair

2014-04-24 Thread Peter Zijlstra
On Wed, Apr 23, 2014 at 03:31:57PM -0700, Tim Chen wrote: > The current code will call pick_next_task_fair a second time > in the slow path if we did not pull any task in our first try. > This is really unnecessary as we already know no task can > be pulled and it doubles the delay for the cpu to e

[PATCH] sched: Skip double execution of pick_next_task_fair

2014-04-23 Thread Tim Chen
The current code will call pick_next_task_fair a second time in the slow path if we did not pull any task in our first try. This is really unnecessary as we already know no task can be pulled and it doubles the delay for the cpu to enter idle. We instrumented some network workloads and that saw th