Re: add lowpower_idle sysctl

From: Dominik Brodowski
Date: Thu Mar 18 2004 - 04:07:19 EST


On Wed, Mar 17, 2004 at 10:40:31PM -0500, Zwane Mwaikambo wrote:
> On Wed, 17 Mar 2004, Andrew Morton wrote:
>
> > "Kenneth Chen" <kenneth.w.chen@xxxxxxxxx> wrote:
> > >
> > > > > Logically it means a sysctl entry in /proc/sys/kernel.
> > > > Yes, but the *meanings* of the different values of that sysctl need
> > > > to be defined, and documented. If lowpower_idle=42 has a totally
> > > > different meaning on different architectures then that's unfortunate
> > > > but understandable. But we should at least enumerate the different
> > > > values and try to get different architectures to honour `42' in the
> > > > same way.
> > >
> > > Writing to sysctl should be a bool, reading the value can be number of
> > > module currently disabled low power idle. I think the original intent
> > > is to use ref count for enabling/disabling. (granted, we copied the
> > > code from other arch).

I assume ia64 does idling using the ACPI processor.c driver? If so, couldn't
writing to /proc/acpi/processor/./power be an option?

> > OK, so why not give us:
> >
> > #define IDLE_HALT 0
> > #define IDLE_POLL 1
> > #define IDLE_SUPER_LOW_POWER_HALT 2
> >
> > and so forth (are there any others?).
> >
> > Set some system-wide integer via a sysctl and let the particular
> > architecture decide how best to implement the currently-selected idle mode?
>
> I'm wondering whether the setting of these magic numbers can't be done
> using cpufreq infrastructure.

I doubt it -- there's no ia64 cpufreq driver anyway, and cpufreq is about
frequency scaling and (sometimes) throttling, but not "idling". And
"idling" is a too different implementation anyways.

Dominik

Attachment: pgp00000.pgp
Description: PGP signature