I have done a make distcheck on the LibGGIMISC library and it works
fine. Since I have all the targets on my system, it is not needed
to put in the EXTRA_DIST changes (but I notice someone's been up
to something). I've enabled the fbdev target build and added code
in configure.in to try to determine the version of fb.h.
This release will be 2:1:0 == 2.0.1, to get something out the door.
I'm sorry I said on IRC I thought it was 0.4.1, I apparently was
lacking caffeine and was mistaken, as I checked the current .debs and
it has always been synced with the LibGGI release number up to now (duh.)
I'd prefer to start now with this liibrary and keep the releases in
sync with the libtool versions. Christoph: I think I got everyplace
that you put the 0.4.1 version including the rpm spec, but please
do check.
The next release I will add some API functions for vsync behavior
on frame flips and on palette set, and a DGA target. Really this will
not add much functionality since the DGA target automatically does
wait for vretrace when you call ggiSetOrigin/ggiSetDisplayFrame if the
XF86DGA driver supports such a thing, but it will be allow you to check
to see if the target behaves in this way so you'll be able to tell
that those functions do this rather than working blind. I'll also add
support for these into the fbdev target, though the only fbdev driver
I have found that actually implements this is the special patches for
atyfb on directfb. But, at least that will get things working when
the kernel folks do catch up. That release will also use LibGAlloc.
Since there is an API change it will be 3:0:1 == 3.1.0.
Down the road, there will be added support for character cell data in
textmodes, overscan border colors, and various things having to do with
special KGI features for choosing/configuring the output sinks on cards
like the all-in-wonder (once we have a KGI all-in-wonder driver, that is).
I have the "make distcheck" tarball, if I should upload it somewhere,
let me know.
--
Brian
