Using OpenGL with CUDA

Accelerated Computing CUDA CUDA Programming and Performance
1

Hi:

I am trying to compile a simple CUDA program on my local cluster. The program uses OpenGL APIs through

//#include <GL/glext.h>
//#include <GL/glx.h>
//#include <GL/glew.h>

When I compile the program using

nvcc -O2 -arch=sm_50 -o out my_julia_cpu.cu

I got the following :

/usr/include/GL/glext.h(111): error: identifier “GLenum” is undefined

/usr/include/GL/glext.h(111): error: identifier “GLuint” is undefined

/usr/include/GL/glext.h(111): error: identifier “GLuint” is undefined

/usr/include/GL/glext.h(111): error: identifier “GLsizei” is undefined

/usr/include/GL/glext.h(111): error: identifier “GLenum” is undefined

/usr/include/GL/glext.h(112): error: identifier “GLenum” is undefined

/usr/include/GL/glext.h(112): error: identifier “GLint” is undefined

/usr/include/GL/glext.h(112): error: identifier “GLint” is undefined

/usr/include/GL/glext.h(112): error: identifier “GLsizei” is undefined

/usr/include/GL/glext.h(112): error: identifier “GLsizei” is undefined

/usr/include/GL/glext.h(112): error: identifier “GLsizei” is undefined

/usr/include/GL/glext.h(112): error: identifier “GLint” is undefined

/usr/include/GL/glext.h(112): error: identifier “GLenum” is undefined

/usr/include/GL/glext.h(112): error: identifier “GLenum” is undefined

/usr/include/GL/glext.h(113): error: identifier “GLenum” is undefined

/usr/include/GL/glext.h(113): error: identifier “GLint” is undefined

/usr/include/GL/glext.h(113): error: identifier “GLint” is undefined

/usr/include/GL/glext.h(113): error: identifier “GLint” is undefined

How can I fix it?

Thanks,

malik

2

you’ll need to learn how to compile an OpenGL code. You need to specify additional libraries with:

-lGL -lGLX -lGLEW

etc. That may not be the exact set of libraries in the exact order, but with a bit of research you can figure it out.