[Intel-gfx] [PATCH] drm/i915: add render standby support

[Jesse Barnes]

On Thu, 8 Oct 2009 12:55:00 -0400
Andrew Lutomirski <luto at mit.edu> wrote:

> On Thu, Oct 8, 2009 at 12:46 PM, Jesse Barnes
> <jbarnes at virtuousgeek.org> wrote:
> > On Thu, 8 Oct 2009 12:04:32 +0200
> > Erik Andrén <erik.andren at gmail.com> wrote:
> >
> >> 2009/10/7 Jesse Barnes <jbarnes at virtuousgeek.org>:
> >> > Render standy allows the GPU to power down the render unit when
> >> > idle. In order for this to work, it needs a page of graphics
> >> > memory to save state.  This patch allocates that page and
> >> > enables the feature on supported chipsets.
> >> >
> >> > Signed-off-by: Jesse Barnes <jbarnes at virtuousgeek.org>
> >> > ---
> >>
> >> Do you have any figures on how much power this saves?
> >
> > I haven't done any measurements yet.  I had to do this as part of
> > some other work and figured I'd get it out there after some light
> > testing.
> 
> For fun, I compiled and ran it on my laptop.  I didn't get any
> noticeable savings over 2.6.32-rc3 (with a hack added in to
> force-enable render_reclock_avail and lvds_downclock_avail on my
> GM45), but nothing went wrong either.
> 
> While I'm here, silly question: what's supposed to enable
> lvds_downclock_avail?

Not silly at all. :)  That code can't be enabled until we figure out a
reliable way of getting secondary (reduced) refresh timings for the
builtin panel.  We have some info on it but haven't tested it against a
wide variety of VBIOSes yet.

-- 
Jesse Barnes, Intel Open Source Technology Center