[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20161121111243.GK3102@twins.programming.kicks-ass.net>
Date: Mon, 21 Nov 2016 12:12:43 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: Viresh Kumar <viresh.kumar@...aro.org>
Cc: Rafael Wysocki <rjw@...ysocki.net>, Ingo Molnar <mingo@...hat.com>,
linaro-kernel@...ts.linaro.org, linux-pm@...r.kernel.org,
linux-kernel@...r.kernel.org,
Vincent Guittot <vincent.guittot@...aro.org>,
Juri Lelli <Juri.Lelli@....com>,
Robin Randhawa <robin.randhawa@....com>,
Steve Muckle <smuckle.linux@...il.com>, tkjos@...gle.com,
Morten Rasmussen <morten.rasmussen@....com>
Subject: Re: [PATCH] cpufreq: schedutil: add up/down frequency transition
rate limits
On Mon, Nov 21, 2016 at 04:18:00PM +0530, Viresh Kumar wrote:
> On 21-11-16, 11:19, Peter Zijlstra wrote:
> > Urgh...
> >
> >
> > So no tunables and rate limits here at all please.
> >
> > During LPC we discussed the rampup and decay issues and decided that we
> > should very much first address them by playing with the PELT stuff.
> > Morton was going to play with capping the decay on the util signal. This
> > should greatly improve the ramp-up scenario and cure some other wobbles.
> >
> > The decay can be set by changing the over-all pelt decay, if so desired.
> >
> > Also, there was the idea of; once the above ideas have all been
> > explored; tying the freq ram rate to the power curve.
> >
> > So NAK on everything tunable here.
>
> Okay, as I told you on IRC, we already have a tunable: rate_limit_us for the
> schedutil governor which defines the minimum time before which the governor
> wouldn't try to update the frequency again. Perhaps 10-20 ms is the ideal value
> for that everyone is using.
>
> So eventually that should also die and we should get inputs from PELT stuff ?
I think it should be replaced by a value provided by the driver. It
makes sense to have a rate-limit in so far as that it doesn't make sense
to try and program the hardware faster than it can actually change
frequencies and/or have a programming cost amortization. And this very
clearly is a driver specific thing.
It however doesn't make sense to me to fudge with this in order to
achieve ramp up/down differences.
Powered by blists - more mailing lists