[Intel-gfx] Modesetting bug on 945GSE (Atom N270) with Chris Wilson drm-intel-next kernel

[Daniel Vetter]

On Fri, Nov 19, 2010 at 09:28:24PM +0000, Chris Wilson wrote:
> On Fri, 19 Nov 2010 17:44:49 +0000, Simon Farnsworth <simon.farnsworth at onelan.co.uk> wrote:
> > I'm seeing an interesting bug on 945GSE, and I have absolutely no idea where 
> > to start looking. I'm trying to deal with performance issues on this platform, 
> > so I'm updating to master of as many bits as possible, to see if they're 
> > better.
> 
> Do you use a lot of fences? If so the pipelined fencing may be of interest
> and also the relaxed fencing rules for i945. It would be good to get
> some feedback on those.

I might be seeing something similar on my i945gme. I'm trying to get
x-tiling (using tiling bits) to work in the i915 gallium driver. It seems
to work rather well (given the state the driver is in, generally), safe
for something very odd: When I relocate textures with NEED_FENCE (still
using tiling bits in the command stream), the fbo texture renders all
black in fbo_firecube. Relocating without NEED_FENCE works flawless.
Latest -linus seems to be affected, too.

This is just fyi, I have currently no clue what's going on and I'm still
investigating.
-Daniel
-- 
Daniel Vetter
Mail: daniel at ffwll.ch
Mobile: +41 (0)79 365 57 48