I'm off-line as of this afternoon, back in a week, I hope everyone has a great
Holidays!
-Ted
__
Protecting essential freedoms is always a matter of restricting the actions
that would deny them. Remember, your freedom to swing yo
Hmm. Sounds like some mis-edits got put into the h files in CVS.
Alas, I won't have time to look at them until the weekend (work-based crunch
time got me going w/too many hours on the PC right now). I'll get those tracked
down Sat/Sun of no one else does first.
-Ted
_
I don't know who introduced it, but AFAIK it is used in two ways in the FX
driver right now (one of which is Windows specific):
1) Determines what pixel format the 3Dfx color buffer is configured to
2) Determines whether pixel format conversion is done for the 3Dfx-in-a-Window
hack on Windows.
I added the gloss demo file to the Win32 build set.
Based upon the various discussion, I changed over a few things both in gl.h and
in the nmake.mak/nmake.mif to default to Warning Level 4 - tons of warnings
generated now. ;^(
Anyway, I'll try to do a build and get the warning list posted somew
>I believe Ted Jump changed APIENTRY to GLAPIENTRY. Ted, why was
>that done?
Since each DLL component of the mesa package (Mesa itself "as" OpenGL32.DLL,
GLU32.DLL and GLUT32.DLL) need to have their and-only-their functions marked as
__declspec(dllexport) when building the respective DLL.
Othe
>Here's what they kinda look like now (from memory):
MSVC 6 SP3 definitely does NOT like varargs macros.
;^(
I tried about as many combinations as I could think of, no go.
Sorry.
___
Mesa-dev maillist - [EMAIL PROTECTED]
http://lists.mesa3d.or
Gareth,
I don't know how your macros are implemented, but variable length may not be at
all supported on MSVC variants, in fact it may not even treat it as a warning
but treat it as a compile-time error. ;^(
Is your var-arg macros in VSS that I can check them out?
-Ted
__
I've been building the mesa CVS stuff every couple days (mesa-for-cpu
rendering, mesa-for-3dfx, glu, glut, and all the samples/demos/book programs)
with MSVC 6 SP3. Now, I've only got MSVC set for Warning Level 3, not 4,
because the code base does generate a large amount of warnings under WL4
(co
>Does anyone know if src-glut is supposed to be understood/recognized by autoconf
>- I don't see any mention of it in configure.in, for instance.
It would be great if src-glut was also handled by autoconf, and a very
"complete" thing too, IMHO. But, I've no idea the details of dealing with
autoco
Keith,
I had to do some minor edits to get the latest building completely under MSVC
on Windblows, things seem good and I do not think I have broken anything for
other environments - but could you check?
Eero, I've also been able to simplify the handling of the
gl/glu-GetProcAddressEXT stuff in
>the book/anti.c demo is incredibly coloured and the transparent help
>screen in the 3dfx/demos/tunnel.c demo is not transparent.
I can't speak to the book/anti.c demo as it seems to run fine on Windows,
however the 3dfx/demos/tunnel.c demo has non-transparent help on Windows also -
but the othe
There were some minor changes to some source files required to get the Win32
build going cleanly. Please make sure I didn't blow anything for anyone else.
;^)
GUI project files to be forthcoming, just going to take a while to check and
validate.
I also added ESC key handling to the various book/
There shouldn't be "two" win32 directories - remember that Windblows doesn't
differentiate files/dirs base case of lettering, so all of the fines in both
"Win32/" and "win32/" should be in one hierarchy. The creation of these two
separate directories was probably a typing mistake (likely on my pa
I can't seem to get a cvs client connection.
Anyone else?
-Ted
___
Mesa-dev maillist - [EMAIL PROTECTED]
http://lists.mesa3d.org/mailman/listinfo/mesa-dev
The wmesa.c file on my system has ONE less backslash, marked below, which makes
sense. With the backslashes as follows the INNER_LOOP is getting defined as
part of the SETUP_CODE macro, without the backslash SETUP_CODE and INNER_LOOP
get defined as separate macros.
CVS as of 7/16/99 has it w/o
>I wish we wouldn't have to change glut.h in any way from Mark's stock
>version. Is that possible?
I think it's my brain-f@rt, it was glutint.h not glut.h that I modified (along
with the few places in .c files that used __cdecl instead of CDECL).
-Ted
>and GLAPIENTRY is defined in gl/gl to __stdcall
>but atexit really really needs a __cdecl function.
Boy, that was screwing up a couple things.
I'm CVS-ing a modifed glut.h that only does Brian's defs when NOT compiling
under _WIN32 - Brian I'm assuming you added those for compatability with ot
I got more cleanups/updates to the project files for building Windows posted,
along with some minor edits to various sources - these edits *should* not
affect other builds but I haven't been able to test that yet.
I am having a problem building GLUT with the current sources, and I haven't
figured
>in varray.c there are functions with "GLAPI void GLAPIENTRY" declaration.
Eero, what's the error you get with "GLAPI void GLAPIENTRY" declarations?
-Ted
___
Mesa-dev maillist - [EMAIL PROTECTED]
http://lists.mesa3d.org/mailman/listinfo/mesa-dev
While checking into why there are so many build warnings on Windows, I ran into
the warning that 'out' is used without being initialized.
'out' is defined as a stack-based GLvector4f in the function expanded by
USER_CLIPTEST. It looks to be used as some sort of early-out checking code but
I can't
I've successfully gotten mesa/mesa-on-3dfx/glu/glut and all the test proggys
built and executing via MSVC 6 SP3 on Win NT 4.
I've also made minor changes to fire.c/ipers.c/glbpaltx.c in the 3dfx/demos
directory to allow clean building on Windblows. Should not affect other builds
as it's only a mi
>On Tue, 8 Jun 1999, Holger Waechtler wrote:
>
>dlopen is really easy to understand:
Based on your explanation of uinx dl___() calls and my fairly extensive
knowledge of the Windblows counterparts I'd say it would take less than an hour
to hack a dlopen() interface on top of the Win32 API calls
>> of the "NASM ready" file, along with NASM. Oh, question, I haven't looked
>> recently but how large is NASM, how much would it add to the size of the d/l?
>Weighing in at around a meg, last I checked.
Okay, so, we'll need to make sure the distribution has the .obj file for all
NASM built code
>I think it is actually faster and safer to you to do these changes by
>editing instead of copying my files.
Okay. Thanks for the info.
>Regarding the NASM issue I don't think that manual translation
>is a good idea. You would get into trouble when the sources
>are updated for what ever reason.
>I assume there is a champion for the Win32 version (Theodore?).
Champion may be a bit of a strong word, but yes I'm still on board and about
80% done updating the Win32 build files.
I'm figuring I'll have everything building in about a week.
>If not, I could handle the win32 updates, assuming
>Secondly, and perhaps also related to this cheaper-than-cheap card I
>bought, do people see much of a speed up with the mesa demos like
>isosurf, etc. with a Voodoo2 vs. Voodoo1? I don't and I don't know
>why...
>
>Have I bought a pup?
Probably not. The triangle throughput via Glide 2 to V1 o
>Hope, the test_3dnow program will work now on a PII CPU, it checks
I don't have it converted to an asm format which works with gcc (it's in the
works as I have time), but I have CPUID source which properly Idents all
Pentium-and-later chip variants and properly checks for AMD, Cyrix, etc.,
spec
>I'm tempted to alias it to mesa-dev so those of us who develop/maintain
>the code can see the reports and take care of them.
I'd think that aliasing it to mesa-dev would be a good idea, more eyes on it
you know. Also, one of us may have already done a fix but hadn't had a chance
to post it.
-T
Rebuilt w/o "conditional is constant" warnings, and with some fixes included.
Revised log files now available at:
http://www.tertius.com/projects/library/index.html#mesa-3d
Thanks Brian and Gareth!
-Ted
__
Protecting
>I'll disable that warning in my copy now if you haven't committed that fix
>already.
Committed ...
__
Protecting essential freedoms is always a matter of restricting the actions
that would deny them. Remember, your freedom to
>Ted, if you could disable this warning:
> warning C4127: conditional expression is constant
Done.
>by preprocessor code templates. I don't see a way to change the code to
>silence this kind of warning. We use the preprocessor to systematically
>generate a lot of code.
I'll take a look
>Ted, I'm in at work at the moment on a Windows box, so I'll look at soem of
>the compiler warnings while I'm at it.
Cool, much appreciated, not going to have too much time to look at the code
myself until Sunday (working 6-day weeks right now). ;^(
I'll be posting new warning logs in the morni
http://www.tertius.com/projects/library/index.html#mesa-3d
There are four files, one for each of the DLL files generated by the Win32
hierarchy project files (fxmesa, mesa, glu, glut - all as DLL files).
The various programs all also generate warnings, however I don't feel like we
shoul
>And finally Ted, I just built Mesa on Win32 and got several thousand
>warnings. This does seems a little high... Are we still trying to fix
>this?
Yeah, that's about right. ;^(
I'll be working on it, but it's a slow/mechanical process ...
I wasn't able to get the warning logs up on the www
34 matches
Mail list logo
