Because setup is always a pain

Here is an in depth breakdown of each step from chapter 2 if you are on windows. Feel free to copy and modify into your own notes. Note that the `wget` and `sudo sh cuda` lines are already slightly different from the course. ### WSL Setup: Open the search bar and type Windows Features Select turn windows features on and off Make sure the following are checked: * Hypervisor * Virtual Machine Platform * Windows Subsystem for Linux Open the command prompt as Administrator `wsl --install -d Ubuntu` `wsl.exe --install --no-distribution` restart as needed Open wsl by typing `wsl` then enter the password `sudo apt update && sudo apt upgrade` `sudo apt install wget curl git` `sudo apt install python3-pip` ### Install the CUDA toolkit: https://developer.nvidia.com/cuda-downloads Click through the options to install in Linux which for me were: * Linux * x86_64 * WSL-Ubuntu * 2.0 * runfile(local) `wget https://developer.download.nvidia.com/compute/cuda/12.6.1/local_installers/cuda_12.6.1_560.35.03_linux.run` `sudo sh cuda_12.6.1_560.35.03_linux.run` The environment vars still had to be installed so: `nano ~/.bashrc` add the lines somewhere in the file: `export PATH=/usr/local/cuda-12.6/bin${PATH:+:${PATH}}` `export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/usr/local/cuda/lib64` back in the terminal run `source ~/.bashrc` ### Test run `nvcc --version` and `nvidia-smi` Also create a test file `main.cu` ``` #include <iostream> using namespace std; int main() { cout << "hello world" >> endl; } ``` Compile back in the terminal with `nvcc -o main main.cu` Then run with `./main`