[Intel-gfx] [PATCH] agp: Fix stolen memory counting on G4X.

[Zhenyu Wang]

On 2008.10.15 00:01:20 -0700, Eric Anholt wrote:
> On Wed, 2008-10-15 at 13:25 +0800, Zhenyu Wang wrote:
> > On 2008.10.14 11:38:18 -0700, Eric Anholt wrote:
> > > On the GM45, the amount of stolen memory mapped to the GTT was underestimated,
> > > even though we had 508KB more available since the GTT doesn't take from
> > > stolen memory. On the non-GM45 G4X, we overestimated how much stolen was
> > > mapped to the GTT by 4KB, resulting in GPU page faults when that page was
> > > accessed.
> > > 
> > > This update requires a corresponding update to xf86-video-intel to work
> > > correctly.
> > 
> > How should we handle compability when pushing this upstream? 2.6.26/27 user
> > with upstream intel video driver... (my origin G33 stolen memory fix got
> > reverted because of this, although I think that revert is wrong. people has
> > bug so upgrade shouldn't be a problem...)
> 
> Servers fail massively without both fixes, so it's only a regression in
> a very strange sense.  I'll be adamant about getting this in, it's not
> optional.
> 

So do the G33 stolen size fix to me. Would you queue that in drm-intel-next?
You may just revert that revert commit, or Brandon Philips has also sent patch
today.

-- 
Open Source Technology Center, Intel ltd.

$gpg --keyserver wwwkeys.pgp.net --recv-keys 4D781827
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: Digital signature
URL: <http://lists.freedesktop.org/archives/intel-gfx/attachments/20081015/a630b02a/attachment.sig>