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

[Michel Dänzer]

On Sat, 2007-06-09 at 20:01 +1000, Hamish Moffatt wrote:
> On Fri, Jun 08, 2007 at 04:35:17PM +0200, Michel Dänzer wrote:
> > On Fri, 2007-06-08 at 21:18 +1000, Hamish Moffatt wrote:
> > > On Fri, Jun 08, 2007 at 08:23:05AM +0200, Michel Dänzer wrote:
> > > > Yes, this is because the X server DRI layer now counts DRI windows
> > > > according to visibility, to allow e.g. page flipping to work with
> > > > multiple windows as long as only one of them is visible. So the intel
> > > > driver can't tell the difference between a window being invisible and it
> > > > not existing at all and disables the vblank interrupts.
> > > 
> > > How is "3D in use" defined?
> > 
> > With current xserver master: One or more DRI windows is visible.
> > 
> > With xserver 1.3 or older: One or more DRI windows exists.
> > 
> > > MythTV uses DRM vertical blanking sync with Xv output; I wouldn't say it
> > > uses 3D by any stretch of the imagination.
> > 
> > Doesn't it use GL for the menus? If so, the above conditions are
> > probably satisfied with direct rendering or AIGLX.
> 
> Yes it does, although those won't be visible at the same time Myth wants
> to use DRI (and/or OpenGL) vblank sync. Is that a problem?

Could be, if it's using different windows for the menus and video
playback.

Anyway, as I pointed out in another thread, if you don't care for
disabling the vblank interrupts, just disable the xf86-video-intel code
that does it... Maybe this should be turned into an option (that
defaults to on with laptops and off otherwise?) until the proper DRM
solution is in place.


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