lkml.org 
[lkml]   [2014]   [Jun]   [23]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH v2] sched: Fast idling of CPU when system is partially loaded
From
Date
On Mon, 2014-06-23 at 20:50 +0200, Peter Zijlstra wrote:
> On Mon, Jun 23, 2014 at 09:40:45AM -0700, Tim Chen wrote:
>
> > > @@ -5886,7 +5886,7 @@ static inline void update_sg_lb_stats(st
> > >
> > > sgs->group_load += load;
> > > sgs->sum_nr_running += rq->nr_running;
> > > - if (overload && rq->nr_running > 1)
> > > + if (rq->nr_running > 1)
> > > *overload = true;
> > > #ifdef CONFIG_NUMA_BALANCING
> > > sgs->nr_numa_running += rq->nr_numa_running;
>
> > With this change, we'll be returning the overload indicator
> > that we don't use for non-root domains, which will be
> > extra work in sg_lb_stats as it loops through each rq checking
> > the nr_running to update the indicator. I was hoping to avoid
> > that if possible.
>
> What extra work? We already load nr_running and overloaded is on-stack
> and should be quite dirty already due to that.

Okay then. I'll need to modify the v3 patch with the following bits:

- /* only need to update overload indicator for root domain */
- if (!env->sd->parent && rq->nr_running > 1)
+ if (rq->nr_running > 1)
*overload = true;

Just sent v3 patch out before your email reached me.

Tim



\
 
 \ /
  Last update: 2014-06-23 21:41    [W:0.048 / U:0.212 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site