cvs commit: src/sys/pci agp_amd64.c

[Adam Gregoire]

On Fri, 2005-04-08 at 19:18 -0700, Eric Anholt wrote: 
> On Fri, 2005-04-08 at 22:07 -0400, Adam Gregoire wrote:
> > On Fri, 2005-04-08 at 18:28 -0700, Eric Anholt wrote:
> > > On Fri, 2005-04-08 at 21:16 -0400, Adam Gregoire wrote:
> > 
> > > 
> > > Options "BusType" "PCI" in your Device section of xorg.conf will avoid
> > > all AGP stuff and tell you a little more about where to point the
> > > finger.
> > > 
> > 
> > Same result, I lose console and X locks.
> 
> OK, as long as you don't have agp loaded at the time of doing this, you
> can rule out agp for sure (otherwise, I would still say that
> misconfiguration could cause issues.  I've seen agp do weird things.).
> 
I recompiled for radeondrm and agp to be modules, set agp_load=yes
in /boot/loader.conf, followed your suggestion and was able to have DRI
running with Option "BusType" "PCI". With this in mind, and the failing
of apgtest as previously mentioned, would you think this is indeed gart
related?

from agptest:
Testing data integrity (1st pass): failed on first pass!
Testing data integrity (2nd pass): failed on second pass!

-- 
Adam Gregoire <ebola at psychoholics.org>