The CUT functions are NVIDIA’s SDK provided. There are there in one of the libraries provided by the SDK. YOu need to specify those libraries to your Linker. For example, I am on windows and I have to specify “cutil32D.lib” in “C:\Program Files\NVIDIA Corporation\NVIDIA CUDA SDK\common\lib” directory as linker input. There must be a similar mechanism to do in linux as well. YOu may need to specify this in your make file, I believe. CHeck out,.