Re: Stopping the tick on a fully loaded system

[Rafael J. Wysocki]

On Thu, Jul 27, 2023 at 9:59 AM Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
>
> On Wed, Jul 26, 2023 at 06:14:32PM +0200, Peter Zijlstra wrote:
> > On Wed, Jul 26, 2023 at 05:53:46PM +0200, Rafael J. Wysocki wrote:
> >
> > > > > That means we don't track nearly enough data to reliably tell anything
> > > > > about disabling the tick or not. We should have at least one bucket
> > > > > beyond TICK_NSEC for this.
> > > >
> > > > Quite likely.
> > >
> > > So the reasoning here was that those additional bins would not be
> > > necessary for idle state selection, but the problem of whether or not
> > > to stop the tick is kind of separate from the idle state selection
> > > problem if the target residency values for all of the idle states are
> > > relatively short.  And so it should be addressed separately which
> > > currently it is not.  Admittedly, this is a mistake.
> >
> > Right, the C state buckets are enough to pick a state, but not to handle
> > the tick thing.
> >
> > The below hack boots on my ivb-ep with extra (disabled) states. Now let
> > me go hack up teo to make use of that.
> >
> > name          residency
> >
> > POLL          0
> > C1              1
> > C1E             80
> > C3              156
> > C6              300
> > TICK            1000
> > POST-TICK       2000
> >
>
> Ah, so last night (or rather, somewhat realy today) I realized I has the
> buckets wrong.
>
> We don't have buckets to the left, but buckets to the right, so the
> above would give:
>
> 0: [0,1)
> 1: [1,80)
> 2: [80,156)
> 3: [156,300)
> 4: [300,1000)
> 5: [1000,2000)
> 6: [2000,...)
>
> Which also means I can ditch the whole POST-TICK bucket. Let me get
> breakfast and try all this again.

Note that this is about the cases between the target residency of the
deepest enabled idle state and the tick period and from the
Anna-Maria's results it looks that's about 0,1% of the total number.