Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1422036

Re: [PATCH 2/3] sched: Unloop sched avg decaying

From Peter Zijlstra <peterz@infradead.org>
Newsgroups linux.kernel
Subject Re: [PATCH 2/3] sched: Unloop sched avg decaying
Date 2016-06-14 18:00 +0200
Message-ID <rJYUV-6QY-3@gated-at.bofh.it> (permalink)
References <rJYrT-6Cf-3@gated-at.bofh.it> <rJYrT-6Cf-1@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


On Tue, Jun 14, 2016 at 05:28:01PM +0200, Frederic Weisbecker wrote:
> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> index 385c947..0c0578a 100644
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -666,17 +666,17 @@ bool sched_can_stop_tick(struct rq *rq)
>  void sched_avg_update(struct rq *rq)
>  {
>  	s64 period = sched_avg_period();
> +	s64 delta;
> +	u64 rem;
> +	int pending;
>  
> +	delta = (s64)(rq_clock(rq) - rq->age_stamp);
> +	if (delta <= period)
> +		return;
> +
> +	pending = div64_u64_rem(delta, period, &rem);
> +	rq->age_stamp += delta - rem;
> +	rq->rt_avg >>= pending;
>  }

Blergh, and now do the profile on machine that doesn't have major
transistor count dedicated to divisions.

Why not add the division to the nohz exit path only?

Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread


Thread

[PATCH 2/3] sched: Unloop sched avg decaying Frederic Weisbecker <fweisbec@gmail.com> - 2016-06-14 17:30 +0200
  Re: [PATCH 2/3] sched: Unloop sched avg decaying Peter Zijlstra <peterz@infradead.org> - 2016-06-14 18:00 +0200

csiph-web