[ubuntu-x] Two bugs with proposed patches and long discussions

[Bryce Harrington]

On Thu, May 07, 2009 at 12:32:52PM +0200, Martin Olsson wrote:
> Also, What to do about 965/Gx45 missing DRI on >2048 screen width?
> Timo had a patch in jaunty fixing this but it was dropped because it was (incorrectly?)
> suspected the patch was related to the intel freezes.

No, I think we want to stick with having this disabled, for several
reasons.  First, it had been proposed for mesa 7.4 but declined due to
concerns it could expose problems, so it seems to carry high risks with
it that it would probably be best to avoid, and wait for the "real"
solution that went upstream.  Second, there were testers who found it
helped a little; perhaps they were mistaken, but we don't know that for
certain.  Third, this is a feature change since the 2048 limit has
always been there; since it cannot be claimed to be a regression, fixing
it in jaunty is likely not to be possible.

> However, then we ended up doing that hardcoded virtalsize anyway
> right? It feels pretty scary to poke virtualsize related bits in
> jaunty so should we just tell them politely to wait for karmic and
> mesa 7.[5|6]? (patch is already commited upstream)
> 
> https://bugs.launchpad.net/ubuntu/+source/mesa/+bug/146298

It does seem scary, and I think waiting for karmic on this is the right
thing to do.  Meanwhile I see someone is maintaining a PPA that
re-enables this:

  https://edge.launchpad.net/~cavedon/+archive/ppa

The PPA is probably the best solution we're going to have for Jaunty.

Bryce