intel-batchbuffer problem (was: Re: DRI2 direct rendering)

[Steven J Newbury]

On Sat, 2008-04-05 at 04:25 +0100, Steven J Newbury wrote: 
> > 
> > There seems to be no working vblank interrupt, could this be related?
> > My kernel log gets filled with: "trying to get vblank count for disabled
> > pipe 0"
> > 
The above error log message is coming from the DRM, specifically
i915_drm.c. The i915_get_pipe() function is returning the wrong pipe on
my system ( pipe 0, where it should be pipe 1).  The Xorg.0.log shows:

(II) intel(0):   Pipe A is off 
(II) intel(0):   Display plane A is now disabled and connected to pipe A.
(II) intel(0):   Pipe B is on
(II) intel(0):   Display plane B is now enabled and connected to pipe B.
(II) intel(0):   Output VGA is connected to pipe none
(II) intel(0):   Output LVDS is connected to pipe B
(II) intel(0):   Output TMDS-1 is connected to pipe none
(II) intel(0):   Output TV is connected to pipe none

Making i915_get_pipe() always return 1 stopped the log messages but
hasn't addressed the stability problem.  I have noticed though, that the
issue I described with composited windows "flickering" only occcurs with
semi-transparent windows (or transparent GNOME terminals).

Finally, while glxgears appears to work fine, other GL apps only get
updated when their window is moved, maybe vblank interrupts still aren't
working..

> I'm going to be away skiing for the next week but I'm attaching my
> Xorg.log file.  The register debug info looks like it might be
> indicating my problem.  Hope it helps.

I'm back now, I'll do any testing as time allows.