On Mon, Jul 20, 2015 at 05:42:38AM +0200, Mike Galbraith wrote:
> On Mon, 2015-07-20 at 09:34 +0900, Byungchul Park wrote:
>
> > yes, i also think that a preemption granularity has little meaning, atually
> > because of tick granularity.
>
> See HR_TICK. It's not cheap though, why it's default
On Mon, 2015-07-20 at 09:34 +0900, Byungchul Park wrote:
> yes, i also think that a preemption granularity has little meaning, atually
> because of tick granularity.
See HR_TICK. It's not cheap though, why it's default off.
-Mike
--
To unsubscribe from this list: send the line "unsubs
On Sun, Jul 19, 2015 at 01:57:14PM +0200, Mike Galbraith wrote:
> On Sun, 2015-07-19 at 18:11 +0900, byungchul.p...@lge.com wrote:
>
> > @@ -3226,6 +3226,12 @@ check_preempt_tick(struct cfs_rq *cfs_rq, struct
> > sched_entity *curr)
> > struct sched_entity *se;
> > s64 delta;
> >
> > +
On Sun, Jul 19, 2015 at 01:15:09PM +0200, Peter Zijlstra wrote:
> On Sun, Jul 19, 2015 at 06:11:00PM +0900, byungchul.p...@lge.com wrote:
> > doesn't it make sense?
>
> No, people have already given you all kinds of reasons why this isn't a
i feel sorry. but all kinds?. i got only a reason, that
On Sun, 2015-07-19 at 18:11 +0900, byungchul.p...@lge.com wrote:
> @@ -3226,6 +3226,12 @@ check_preempt_tick(struct cfs_rq *cfs_rq, struct
> sched_entity *curr)
> struct sched_entity *se;
> s64 delta;
>
> + /*
> + * Ensure that a task executes at least for sysctl_sched_min_
On Sun, Jul 19, 2015 at 06:11:00PM +0900, byungchul.p...@lge.com wrote:
> doesn't it make sense?
No, people have already given you all kinds of reasons why this isn't a
good change. It is also a very invasive change and you have no
performance numbers one way or another.
I'm going to fully ignore
[PATCH v3] sched: modify how to compute a slice and check a
preemptability
make cfs scheduler use rq level nr_running to compute a period in the case
of CONFIG_FAIR_GROUP_SCHED. using local cfs's nr_running to get period is
very weird. for example, imagine cgroup structure below.
root(
7 matches
Mail list logo
