[Intel-gfx] xf86-video-intel: src/i830_memory.c

[Jesse Barnes]

On Tue, 14 Apr 2009 08:22:03 -0400
Ben Gamari <bgamari.foss at gmail.com> wrote:

> On Tue, Apr 14, 2009 at 09:47:45AM +0200, Arkadiusz Miskiewicz wrote:
> > On Monday 13 of April 2009, Jesse Barnes wrote:
> > >  src/i830_memory.c |    2 --
> > >  1 file changed, 2 deletions(-)
> > >
> > > New commits:
> > > commit 9b615a52671aacf34666f90ecfff98651ce6afe2
> > > Author: Li Peng <peng.li at intel.com>
> > > Date:   Fri Apr 10 14:39:35 2009 +0800
> > >
> > >     Turn on front buffer tiling in KMS.
> > >
> > >     This code disabled front buffer tiling in KMS. Turn it on
> > > since kernel handles all tiling now, this improves performance of
> > > x11perf -aa10text from 97k to 286k on my 945GME.
> > >
> > >     Should help with #20761, if not totally fix it.
> > 
> > This commit breaks things for me on GM45.  Entire screen is
> > corrupted from beginning of kdm/X session. Reverting to before this
> > commit and everything is fine.
> >
> 
> Yep, same situation here on GM965. Supposedly the tiled allocation
> succeeded,
> 
> > (**) intel(0): Tiling enabled
> > (==) intel(0): VideoRam: -1 KB
> > (II) intel(0): Attempting memory allocation with tiled buffers.
> > (II) intel(0): Tiled allocation successful.
> 
> Yet the screen is full of corruption goodness. I've just forced tiling
> off in Xorg.conf for the moment and things are fine again.
> Everything's up to date so I can't imagine what the problem might be.
> Thanks!

Looks like something broke in the kernel or we're not calling the GTT
map function... I don't see any fences get allocated at fault time.
Debugging now.

-- 
Jesse Barnes, Intel Open Source Technology Center