The re-vote on mandating inclusion of glext.h from gl.h is pretty
much where the original vote ended: hung with equal totals for choices
(A) and (C) (I think there's maybe 1 point in favor of (A) now that
Brett voted, but if Blythe votes it might well slide back). We all seem
to have taken a collective breather for a bit, although there is some
behind the scenes activity to try and conclude this (will let the people
involved say more when they're ready).
In the meantime, I've done a bunch of updates to the OpenGL
extension registry at
http://oss.sgi.com/projects/ogl-sample/registry/
- New ARB extension specifications from the March meeting have been
added.
- Added some enumerant values in extension specifications that didn't
include them, and added a few SGI extension specifications which had
been left out.
- The ".spec" files and generator scripts used to convert those into
e.g. header files are now available in the ogl-sample CVS repository
on oss.sgi.com, under main/doc/registry/specs/ . I would appreciate
fixes / additions, but please note that this *is* the extension
registry, and should not be modified by anyone except me, in the
interests of namespace control. Brian, I believe this has been
updated to be consistent with the dispatch offsets you ended up
using for XFree86 4.0, but there are now a whole bunch of new entry
points in the file which should likely have offsets assigned.
- A new version of the glext.h extension interface header, for
use both with the OpenGL ABI for Linux as well as other platforms,
is available at:
http://oss.sgi.com/projects/ogl-sample/ABI/glext.h
I've spent quite a bit of effort tracking down every possible (well
defined) extension I could, and some that aren't so well defined,
though I'm sure I've missed a few - please let me know. Preferably
in the form of patches to the enum.spec/gl.spec files.
- Following the notion I expressed I few weeks ago, and subject to
further discussion on just how to structure/mandate this, I've also
created separate headers for GLX (and for WGL, though that's not
subject to the Linux ABI group), in
http://oss.sgi.com/projects/ogl-sample/ABI/glxext.h
http://oss.sgi.com/projects/ogl-sample/ABI/wglext.h
- For testing purposes, versions of these headers that (a) remove the
protective #ifndef <extension-symbol>/#endif code and (b) include C
prototypes for extension functions, as well as typedefs for function
pointers obtained via wglGetProcAddress() or glXGetProcAddressARB(),
are available in:
http://oss.sgi.com/projects/ogl-sample/ABI/glext_proto.h
http://oss.sgi.com/projects/ogl-sample/ABI/glxext_proto.h
http://oss.sgi.com/projects/ogl-sample/ABI/wglext_proto.h
I request and encourage people to try these latter headers out
against their platform implementations and let me know of errors or
inconstencies. You're likely to see some issues as to whether some
parameters are declared 'const' or not, in cases where the extension
specification and the source code are not consistent for one reason
or another. Those are unlikely to be a problem in practice since the
#ifndef/#endif protections will not declare an extension if it was
already declared in gl.h, but there may be more serious problems
which need to be resolved.
Jon Leech
SGI
