RandR 1.3 additions?

[Jerome Glisse]

On 7/17/07, Alex Deucher <alexdeucher at gmail.com> wrote:
> On 7/17/07, Jerome Glisse <j.glisse at gmail.com> wrote:
> > On 7/17/07, Nicolas Trangez <eikke at eikke.com> wrote:
> > > On Tue, 2007-07-17 at 03:44 +0200, Jerome Glisse wrote:
> > > > Talking of power management i have been wondering if it wouldn't be
> > > > nice
> > > > to have some kind of infrastructure to scale down GPU & VRAM clock
> > > > (on card which can) anytime we don't have activity. The tricky point
> > > > is that
> > > > is bit hard to find out what we can call activity maybe we should just
> > > > provide
> > > > infrastructure and let a user program/daemon choose what is inactivity
> > > > (no
> > > > input in last minute, or nothing got refreshed, maybe better
> > > > information is how
> > > > many change happen to the currently displayed framebuffer things like
> > > > damage
> > > > extension might provide such information in more or less accurate
> > > > way).
> > > Providing infrastructure would be great, policy would go into something
> > > like gnome-volume-manager I guess, which could e.g. go into low-power
> > > mode when the session screensaver starts, or when the display goes into
> > > DPMS mode, when the battery reaches 10%,... This fits pretty well into
> > > HAL's power management methods.
> > >
> > > Nicolas
> >
> > I think we could save lot power by downclocking gpu & ram anytimes there
> > is not much activity on screen this why i have though about damage extension
> > to provide information on how much the screen change. If there is only the
> > cursor or slow refresh things you don't want to have gpu going at 400Mhz,
> > 200MHz should be more than enough. The things is that i would like to avoid
> > using a daemon which query damage extension in order to set gpu clock.
> > I guess it would require to add some new extension which report to a daemon
> > anytime there is a major change in graphics activity, we might also want to
> > avoid downclocking and upclocking the gpu every second.
> >
> > Anyway i guess for a first step providing infrastructure to change GPU clock
> > and VRAM clock (or any other power saving things the card can offer) would
> > be the first step. Then we could think to add more cleverness either by
> > adding new things to damage or by adding a new extension which can take
> > advantage of damage information. Btw i don't know much about damage
> > extension so maybe i am wrong when thinking that it could provide
> > usefull information on how much graphics activity we got.
>
> This is starting to move into the chips control domain.  Many newer
> GPUs already scale their clocks, voltage, etc. or toggle special power
> modes automatically without user space intervention.
>
> Alex
>

AFAIK for r500 family even though there is scaling you still can save
more power by down clocking GPU & VRAM. So i think that what
current GPU does is still a bit limited. That said it seems newer chipset
will have more advanced management that might make such
things useless.

best,
Jerome Glisse