Hi Alex,
On Mon, 2011-01-24 at 17:57 +0100, Alexander Thurgood wrote:
> Are there any moves afoot to support Wayland, the potential replacement
> for X on Linux ?
None yet. Now, of course X apps will still sit on top of wayland, so
there is not a huge urgency here. Also, IMHO the timeline for wayland is
perhaps over-hyped, if you talk to eg. Kristian (actually working on it)
vs. reading third hand guesses plastered over the media you become more
sanguine :-)
So - having said that, almost everything that wayland would want us to
do, is already a good thing to do to get better graphics performance.
There is a certain amount that we should get for 'free' from an
(inevitable) move from gtk2 to gtk3 (which has a native wayland
backend). That should cover most of the input and window event changes
(I hope).
Then of course there is rendering - and we need lots of work done
there, to move us to fully non-X-dependent rendering; I guess we should
do that via cairo implementation - but that might have a big knock-on
impact through the app.
If you want to start chipping away at that, I suggest that you talk
first to Thorsten; and then consider re-implementing the
vcl/inc/vcl/salgdi.hxx interface to target cairo directly. That in turn
will create some fun issues eg. 'XOR rendering' - which we need to chase
out of the code (there is an easy hack for this), and then try to work
out how we can deal with the vile meta-files that serialize this in
lots of files[1] ;-]
HTH,
Michael.
[1] - we prolly have to fall back to an in-memory surface for that,
unfortunately - and also (I guess) start to hand-code things; but XOR
rendering should be by far the exception not the rule around the place.
--
michael.me...@novell.com <><, Pseudo Engineer, itinerant idiot
_______________________________________________
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice
