Im trying to run a code from: https://www.wiki.ed.ac.uk/display/ecdfwiki/Use+multiple+GPU+devices+with+OpenMP+and+CUDA, the cudaOpenMP. The program is really simple yet i cant run it.
I have compiled it for my system, and i have run the samples from the NVIDIA CUDA SDK succesfully and i dont see much difference between those and this program here.
I have noted that if i try to run the program as a user it fails, but when i run it as root, it works. The program was compiled as user yet it fails.
Also i saw that if i run any program as root then switch to user and run the cudaOpenMP, it runs too.