Ubuntu and CUDA on laptop with 2x GTX980m GPUs

I purchased a Clevo P375SM-A from Metabox with 2 x NVIDIA GTX980m GPUs.

These were the steps I followed to get Ubuntu and CUDA up and running:

1. Install Ubuntu 14.04 amd64 version. This is a universal 64 bit version for intel and amd 64 bit cpus
2. GTX 980m needs a very new nvidia driver - not in normal repository so:
3. sudo add-apt-repository ppa:xorg-edgers/ppa
4. sudo apt-get update
5. sudo apt-get install nvidia-346
6. reboot or restart lightdm
7. download cuda6.5 64bit and install (but not drivers since we already have those)
8. The only remaining wrinkle is that you have to run a cuda app as root before you can run one as non-root. This seems to be related to needing nvidia-uvm kernel module installed and the /dev/nvidia-uvm node created.
9. Set up binary and library paths needed to access nvcc, nvvp etc. as well as libraries. I added a couple of lines to the end of my .bashrc
   1. export PATH=/usr/local/cuda/bin:$PATH
   2. export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH

The CuBLAS matrix multiply example in 6.5 gives ~1TFlop on a single GTX980m.