ld can't find libglut on RHEL 5.6

Accelerated Computing CUDA CUDA Programming and Performance
1

I’m currently compiling the example programs in the SDK for version 4.1. I currently get the following error:

/usr/bin/ld: cannot find -lglut
collect2: ld returned 1 exit status

However, I do have both 64-bit and 32-bit versions of freeglut rpm package installed:

[chemadm@rnuusprgpu01 C]$ rpm -qf /usr/lib64/libglut.so.3
freeglut-2.4.0-7.1.el5
[chemadm@rnuusprgpu01 C]$ rpm -qf /usr/lib/libglut.so.3
freeglut-2.4.0-7.1.el5

Do I need a newer version of freeglut? I don’t remember seeing this error with version 4.0 in the past. I setup my environment like so for the compile:

cd /dbs/apps/cuda-4.1/NVIDIA_GPU_Computing_SDK/C
setenv PATH /dbs/apps/cuda-4.1/cuda/bin:/usr/bin:/bin:/usr/sbin:/sbin
setenv LD_LIBRARY_PATH /dbs/apps/cuda-4.1/cuda/lib64:/usr/lib64:/lib64
unsetenv PYTHONPATH
unsetenv PYTHONHOME
make | tee make.out

So I do have /usr/lib64 in my LD_LIBRARY_PATH above. Anyone else seen this linker error with libglut before?

Thanks,
Philippe

2

Noone has run into this error? Tried searching but could not find it in the forum so far.

3

Maybe the doc http://developer.download.nvidia.com/compute/DevZone/docs/html/C/doc/CUDA_Getting_Started_Linux.pdf will help:
sudo apt-get install freeglut3-dev build-essential libx11-dev libxmu-dev libxi-dev libgl1-mesa-glx libglu1-mesa libglu1-mesa-dev
and further notes.

4

linking requires the dynlib symlink, which is packaged in the freeglut-devel rpm:

$ rpm -qil freeglut-devel|grep libglut

/usr/lib64/libglut.so

davide