Re: [PATCH 2/2] sched: Document Energy Aware Scheduling

["Rafael J. Wysocki" <rafael@kernel.org>]

On Fri, Jan 18, 2019 at 12:11 PM Quentin Perret <quentin.perret@arm.com> wrote:
>
> On Friday 18 Jan 2019 at 12:01:35 (+0100), Rafael J. Wysocki wrote:
> > On Fri, Jan 18, 2019 at 11:58 AM Rafael J. Wysocki <rafael@kernel.org> wrote:
> > >
> > > On Fri, Jan 18, 2019 at 11:34 AM Quentin Perret <quentin.perret@arm.com> wrote:
> > > >
> > > > Hi Rafael,
> > > >
> > > > On Friday 18 Jan 2019 at 10:57:08 (+0100), Rafael J. Wysocki wrote:
> > > > > On Fri, Jan 18, 2019 at 10:16 AM Quentin Perret <quentin.perret@arm.com> wrote:
> > > > > >
> > > > > > Hi Juri,
> > > > > >
> > > > > > On Thursday 17 Jan 2019 at 16:51:17 (+0100), Juri Lelli wrote:
> > > > > > > On 10/01/19 11:05, Quentin Perret wrote:
> > > > > > [...]
> > > > > > > > +The idea behind introducing an EM is to allow the scheduler to evaluate the
> > > > > > > > +implications of its decisions rather than blindly applying energy-saving
> > > > > > > > +techniques that may have positive effects only on some platforms. At the same
> > > > > > > > +time, the EM must be as simple as possible to minimize the scheduler latency
> > > > > > > > +impact.
> > > > > > > > +
> > > > > > > > +In short, EAS changes the way CFS tasks are assigned to CPUs. When it is time
> > > > > > >
> > > > > > > Not sure if we want to remark the fact that EAS is looking at CFS tasks
> > > > > > > only ATM.
> > > > > >
> > > > > > Oh, what's wrong about mentioning it ? I mean, it is a fact ATM ...
> > > > >
> > > > > But it won't hurt to mention that it may cover other scheduling
> > > > > classes in the future.  IOW, the scope limit is not fundamental.
> > > >
> > > > Agreed, I can do that.
> > > >
> > > > > > > > +for the scheduler to decide where a task should run (during wake-up), the EM
> > > > > > > > +is used to break the tie between several good CPU candidates and pick the one
> > > > > > > > +that is predicted to yield the best energy consumption without harming the
> > > > > > > > +system's throughput. The predictions made by EAS rely on specific elements of
> > > > > > > > +knowledge about the platform's topology, which include the 'capacity' of CPUs,
> > > > > > >
> > > > > > > Add a reference to DT bindings docs defining 'capacity' (or define it
> > > > > > > somewhere)?
> > > > > >
> > > > > > Right, I can mention this is defined in the next section. But are you
> > > > > > sure about the reference to the DT bindings ? They're arm-specific right ?
> > > > > > Maybe I can give that as an example or something ...
> > > > >
> > > > > Example sounds right.
> > > > >
> > > > > You also can point to the section below from here.
> > > >
> > > > Sounds good.
> > > >
> > > > > Side note: If the doc is in the .rst format (which Peter won't like
> > > > > I'm sure :-)), you can actually use cross-references in it and you get
> > > > > a translation to an HTML doc (hosted at kernel.org) for free and the
> > > > > cross-references become clickable links in that.
> > > >
> > > > Right, I personally don't mind the .rst format, but the existing files
> > > > in Documentation/power/ and Documentation/scheduler/ are good old txt
> > > > files so I just wanted to keep things consistent.
> > >
> > > In fact, Documentation/power/ is under a slow on-going transition to
> > > .rst (due to the benefits mentioned above).
> > >
> > > > I don't mind converting to rst if necessary :-)
> > >
> > > It is not necessary, but maybe worth considering.
> >
> > That said, as this is targeted at Documentation/scheduler/, being
> > consistent with the other material in there is probably more
> > important.
>
> Right. Patch 01/02 is targeted at Documentation/power/ though. So if
> that makes your life easier I can turn that one into a .rst file, no
> problem at all.

Yes, I'd prefer it that way.  And please put it into
Documentation/driver-api/pm/.