AW: OpenGL, -mno-cygwin AND GLUT (1/2)

[Andre Bleau]

>On Fri, Feb 16, 2001 at 05:32:47PM -0500, Earnie Boyd wrote:
> >Andre Bleau wrote:
> >>
> >>
> >> If you have the latest gcc for Cygwin, that is 2.95.2-7 (you can check 
> with
> >> gcc -v), and you compile programs using glut with -mno-cygwin, you must
> >> also use -mwin32.
> >>
> >
> >If that is true then it would be a package bug.  Hmm... after reviewing
> >the specs file I can see that indeed the /usr/include/w32api directory
> >will not get searched even for -mno-cygwin when it should.  The
> >workaround is as you state.
>
>Yep.  I've had this fixed in my sources for a while.  This is one reason
>why I haven't made -7 the official version.
>
>cgf

Here is some experimental version of gl.h, glu.h and glut.h that should 
solve  compilation problems with -mno-cygwin, hopefuly not creating 
problems without -mno-cygwin. I also included a very simple helloGlut.c 
program. I tested it with gcc 2.95.2-6 and gcc 2.95.2-7. Can anyone please 
try the new .h with their opengl/glu/glut programs ? If it works, I will 
make a new release soon.

I splited my post in 2 parts, to satisfy the mailing list limits. Here is 
gl.h; glu.h and glut.h will follow with helloGlut.c .

For compiling with -mno-cygwin, add the flag -I/usr/include, and, if using 
gcc-2.95.2-7, -mwin32 .

Christopher, what will be the status of _WIN32 in futur releases of gcc for 
Cygwin ? Is it going to be defined for -mno-cygwin or not ? This is giving 
me headaches, trying to write the .h files so that will compile correctly 
in Windows and UNIX, with gcc and other compilers, with and without 
-mno-cygwin...