Bug #5714 (was Re: xserver: Branch 'master')

[Eric Anholt]

On Fri, 2007-07-27 at 20:10 +0100, Alan Hourihane wrote:
> On Fri, 2007-07-27 at 10:05 -0700, Eric Anholt wrote:
> >  hw/xfree86/dri/dri.c       |   70 +++++++++++++++++++++++++--------------------
> >  hw/xfree86/dri/dri.h       |    8 ++++-
> >  hw/xfree86/dri/dristruct.h |    1 
> >  3 files changed, 47 insertions(+), 32 deletions(-)
> > 
> > New commits:
> > diff-tree 50cb6c7e4419e067c1f080d1de940811d21fc725 (from cec793ef7a6dac9fa2a6538683e363a72672cde9)
> > Author: Kristian Hgsberg <krh at redhat.com>
> > Date:   Fri Jun 15 15:29:00 2007 -0400
> > 
> >     Don't map the front buffer in libdri if the ddx driver doesn't set the size.
> >     
> >     This lets drivers map the front buffer themselves
> >     by setting dontMapFramebuffer.
> 
> Eric,
> 
> Did you see Kristian's work a little while ago on this ??
> 
> Kristian - what's the state of play ?

Yeah, this is actually cherry-picked out of his development tree.

krh and airlied have been working on exa-ttm.  I'm working on some
kernel interface cleanups that have been discussed, and buffer
allocation by the 2d driver which is passed to the clients so that I can
make the fake buffer type go away.

Also, note that this change moves the server context creation from
DRIFinishScreenInit to DRIScreenInit.  It slipped in with krh's exa-ttm
work, and slipped by my review.  I don't think it should ever hurt, and
helps if drivers try to do anything with the lock between DRIScreenInit
and DRIFinishScreenInit.

As far as the #5714 patches, though, I don't know what the state of the
remainder is.

-- 
Eric Anholt                             anholt at FreeBSD.org
eric at anholt.net                         eric.anholt at intel.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 187 bytes
Desc: This is a digitally signed message part
URL: <http://lists.x.org/archives/xorg/attachments/20070727/cd4c0cd9/attachment.pgp>