[ubuntu-x] Nouveau in Lucid (again)

Bryce Harrington bryce at canonical.com
Thu Jan 21 22:14:50 UTC 2010

On Tue, Jan 19, 2010 at 01:33:02PM +1100, Christopher James Halse Rogers wrote:
> Now that Alpha 2 is out of the way, I'd like to raise the status of
> nouveau in Lucid again.  As far as I can tell there's nobody on the
> kernel team responsible for driving nouveau in Lucid at the moment,
> and everyone's busy with other tasks.  I'm happy to help, but at this
> point I'm not sure what I can do to help.

Hi Chris,

I'm hoping Andy, Steve, or Tim can give some comment on this.  Thanks
for highlighting three options, hopefully they can indicate which way
we're going to go.
> As far as I'm aware, there hasn't been a decision as to which of the
> following paths will be taken for nouveau in Lucid:
> 1) Backporting the driver from staging in 2.6.33; this would be easier
> if we were picking up some of the drm changes in 2.6.33, and I'm not
> sure how much radeon or intel wants these changes.
> 2) Taking an oldish snapshot of the driver that works against our
> 2.6.32 drm; a branch using this approach is available at
> http://cooperteam.net/kernel-lucid-nouveau.git .  This would make
> picking up upstream fixes that much more difficult, but this is likely
> to be be difficult regardless, and the X team expects to use a PPA for
> this purpose anyway.
> 3) Adding nouveau+drm from 2.6.33 to linux-backports-modules; I'm not
> sure what's involved here, but could find out.

If I understand correctly, it sounds like #1 would be the best action
but would involve work to backport the kernel bits; #2 would be the
least effort but gives Lucid a driver that will be more difficult to
maintain than it would be with #1.

> What information do we need to determine a course of action?  I'd like
> to get a direction so I can start getting the DDX ready and such.

Andy - I seem to recall we had planned to make a go/no-go decision
around alpha-2 as to whether to keep full steam ahead on nouveau or bag
it and leave it to MM.  I think on the X side of things we are good to
go (I'm still concerned we'll have a lot of bugs, but we can deal); can
you give an official thumb's up or thumb's down for the kernel side of
things?  Once we are officially "go" on nouveau we can start flipping
some of the big switches.


More information about the kernel-team mailing list