[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20231013073739.GA12118@noisy.programming.kicks-ass.net>
Date: Fri, 13 Oct 2023 09:37:39 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Abel Wu <wuyun.abel@...edance.com>
Cc: mingo@...nel.org, vincent.guittot@...aro.org,
linux-kernel@...r.kernel.org, juri.lelli@...hat.com,
dietmar.eggemann@....com, rostedt@...dmis.org, bsegall@...gle.com,
mgorman@...e.de, bristot@...hat.com, corbet@....net,
qyousef@...alina.io, chris.hyser@...cle.com,
patrick.bellasi@...bug.net, pjt@...gle.com, pavel@....cz,
qperret@...gle.com, tim.c.chen@...ux.intel.com, joshdon@...gle.com,
timj@....org, kprateek.nayak@....com, yu.c.chen@...el.com,
youssefesmat@...omium.org, joel@...lfernandes.org, efault@....de,
tglx@...utronix.de
Subject: Re: Re: [PATCH 03/15] sched/fair: Add lag based placement
On Thu, Oct 12, 2023 at 03:04:47PM +0800, Abel Wu wrote:
> On 10/11/23 9:24 PM, Peter Zijlstra Wrote:
> > > > + * we should inflate the lag before placement such that the
> > > > + * effective lag after placement comes out right.
> > > > + *
> > > > + * As such, invert the above relation for vl'_i to get the vl_i
> > > > + * we need to use such that the lag after placement is the lag
> > > > + * we computed before dequeue.
> > > > + *
> > > > + * vl'_i = vl_i - w_i*vl_i / (W + w_i)
> > > > + * = ((W + w_i)*vl_i - w_i*vl_i) / (W + w_i)
> > > > + *
> > > > + * (W + w_i)*vl'_i = (W + w_i)*vl_i - w_i*vl_i
> > > > + * = W*vl_i
> > > > + *
> > > > + * vl_i = (W + w_i)*vl'_i / W
> >
> > And then we obtain the scale factor: (W + w_i)/W, which is >1, right?
>
> Yeah, I see. But the scale factor is only for the to-be-placed entity.
> Say there is an entity k on the tree:
>
> vl_k = V - v_k
>
> adding the to-be-placed entity i affects this by:
>
> define delta := w_i*vl_i / (W + w_i)
>
> vl'_k = V' - v_k
> = V - delta - (V - vl_k)
> = vl_k - delta
>
> hence for any entity on the tree, its lag is offsetted by @delta. So
> I wonder if we should simply do offsetting rather than scaling.
I don't see the point, the result is the same and computing delta seems
numerically less stable.
Powered by blists - more mailing lists