Ubuntu – how to completely remove cuda toolkit/driver from ubuntu 12.04

The problem with undefined references to the functions is because you don't link libcuda library using -lcuda. The guide at How can I get nVidia CUDA or OpenCL working on a laptop with nVidia discrete card/Intel Integrated Graphics? mentions that you should remove -lcuda at one place in common.mk, but that did not work for me. But, as the compiler couldn't find the library, it is important to add the directory, where the library is located, so that the corresponding line in common.mk (it is line 276 for me) looks like

    LIB += -L/usr/lib/nvidia-current -lcuda $(RENDERCHECKGLLIB) ${OPENGLLIB} $(PARAMGLLIB) ${LIB}

Similar correction had to be done in the Makefile of the src/deviceQuery folder, last line, changing

    LIB += -lcuda

to

    LIB += -L/usr/lib/nvidia-current -lcuda

Hope that helps!

I have exactely the same issue. I am trying to test my working setup with Ubuntu 16.04 and GTX1080 with CUDA 7.5 for parallel computing, and was hoping to easily get this done by using the packages provided in the repos. Unfortunately as noted above they are missing. It is possible to use the runfile install for installing only the samples. See e.g. for an older version:

https://stackoverflow.com/questions/27306724/ubuntu-14-04-how-to-install-cuda-6-5-without-installing-nvidia-driver

This works also for 7.5. Simply say "no" when prompted for driver and toolkit installation. You will also need to set the right toolkit libraries. This installs cuda samples into your home by default. However there seem to be an issue with libraries not being found as the package from the repo also puts them into a different location.

However some simple samples like matrixMul compile and run fine.

There might be a simple workaround by adding the right path, but probably installing at least cuda-toolkit together with samples from the runfile should make it work 'out-of-the-box' - have not tried it yet...