Geforce 4 Ti 4200 (and GF 2 MX 400 too), Slackware 9

Hi,
I’m trying to run my opengl windows app under Wine. I’ve got the newest wine (18032003), Mesa-5.0, NVidia GLX 4191, NVidia Kernel 4191. I’ve compiled them, made wine -enable-opengl, and installed as instructed in NVidia readme (i suppose correctly).

On one computer - with Riva TNT 2 M64 OpenGL works with acceleration and looks great.
On two computers with GeForce-s (GF2 MX 400 and GF 4 Ti 4200) OpenGl doesn’t work at all. Even KDE 3.1 screensavers (the OpenGL ones) do not work.

I don’t know where can I find FAQ, HowTO, step by step instruction what to install. I don’t understand, why doing the same things on computers with RivaTNT 2 and GF doesn’t give the same result.

When I run (on both not working computers)
make check
in Mesa-5.0 src folder (to check if it works) it gives me such output (the errors, even though I’ve installed it on the Riva TNT 2 computer in the same way it works and shows no error). Errors are:

…/si-glu/.libs/libGLU.so: undefined reference to operator new[](unsigned)' ../si-glu/.libs/libGLU.so: undefined reference to vtable for __cxxabiv1::__si_class_type_info’
…/si-glu/.libs/libGLU.so: undefined reference to operator delete(void*)' ../si-glu/.libs/libGLU.so: undefined reference to __gxx_personality_v0’
…/si-glu/.libs/libGLU.so: undefined reference to __cxa_pure_virtual' ../si-glu/.libs/libGLU.so: undefined reference to vtable for __cxxabiv1::__class_type_info’
…/si-glu/.libs/libGLU.so: undefined reference to operator delete[](void*)' ../si-glu/.libs/libGLU.so: undefined reference to vtable for __cxxabiv1::__vmi_class_type_info’
…/si-glu/.libs/libGLU.so: undefined reference to operator new(unsigned)' collect2: ld returned 1 exit status make[2]: *** [aaindex] Error 1 make[2]: Leaving directory /usr/src/Mesa-5.0/book’
make[1]: *** [check-am] Error 2
make[1]: Leaving directory `/usr/src/Mesa-5.0/book’
make: *** [check-recursive] Error 1

What is the problem?

That is a g++ / gcc error. Somewhere ther is OOP code that is being compiled by gcc. Since GCC 3.1, gcc does not anymore can compile c++ (now they are separate stuff).

i do not know to much about this but my 2 cents worth…

why do you have MESA? the nvidia drivers take care of the ‘interface’ to opengl with the glx module. i was under the impression you had the nividia kernel module and glx module OR MESA with X Windows and DRI BUT NOT MESA and the nvidia modules.

tho, i could be wrong… :P~ just for the record, i have a GF 4 Ti4200/Slackware 9 running OPENGL HW accelerated just fine with the nvidia modules installed BUT NO mesa.

As luke123 stated, if you install the nVidia drivers you should not need MESA. nVidia and MESA both implement OpenGL, but MESA implements it in software.

Secondly, wine is a big hack. The point of wine is to provide a temporary solution until native apps are compiled for Linux. One of the biggest advantages about OpenGL is that it is cross-platform.

Abstract out all your win32 code and write some X11 code. Or pick a toolkit such as GLUT, GLFW, SDL, wxWindows, GLOW, QT, etc. Do a google search and you should be able to find more than enough info.

[This message has been edited by PK (edited 03-29-2003).]

This topic was automatically closed 183 days after the last reply. New replies are no longer allowed.