Waiting for vertical refresh on Intel 965 and 945 chips in recent git

[Michel Dänzer]

On Fri, 2007-06-08 at 09:29 +0100, Simon Farnsworth wrote:
> Michel Dänzer wrote:
> > On Thu, 2007-06-07 at 18:00 +0100, Simon Farnsworth wrote:
> >> Is there any way for me to wait for VBlank that doesn't depend on having
> >> a window at the top of the stack?
> > 
> > I'm afraid not without code changes.
> > 
> >> We're willing to custom patch anything to make this work (from the
> >> kernel upwards).
> > 
> > I guess the simplest solution for you right now would be to disable the
> > code in the intel driver that disables the vblank interrupts.
> > 
> Any idea where I should look in the intel driver? Even just knowing
> whether it's the DRI driver or the X driver I need to hack would be helpful.

In xf86-video-intel, the I830DRITransitionTo[23]d functions in src/i830_dri.c.


> > There's been some discussion on and off the dri-devel list about a
> > generic solution for the DRM to only enable the vblank interrupts when
> > needed and keep them disabled otherwise. If you're interested in working
> > on that, we can discuss the current ideas there.
> > 
> I'm not likely to have time to work on this soon, but I've subscribed to
> dri-devel so that I can join in next time the discussion comes round (or
> indeed, start the discussion myself when I've got the time to implement
> something).

Excellent.


-- 
Earthling Michel Dänzer           |          http://tungstengraphics.com
Libre software enthusiast         |          Debian, X and DRI developer