Installing HILA on WSL

This section covers how to set up HILA on Windows using WSL, with specific instructions for CUDA configuration.

Table of contents

1. WSL Setup
2. HILA Dependencies
3. CUDA Installation
4. Environment Configuration

WSL Setup

HILA is fully compatible with Windows Subsystem for Linux (WSL). If you already have WSL installed with Ubuntu, you can skip to HILA Dependencies. For new WSL installations, follow your preferred WSL setup method.

HILA Dependencies

Once you have a working WSL Ubuntu installation, you'll need to install the required dependencies. First, update your system:

sudo apt update && sudo apt upgrade

Install LLVM (e.g. version 15):

export LLVM_VERSION=15
sudo apt-get -y install clang-$LLVM_VERSION libclang-$LLVM_VERSION-dev

Install other required dependencies:

sudo apt install build-essential libopenmpi-dev libfftw3-dev libomp-dev

Clone the HILA repository:

git clone https://github.com/CFT-HY/HILA

CUDA Installation

To use HILA with CUDA on WSL, do the following:

1. Have the NVIDIA drivers installed on Windows.
2. Install CUDA Toolkit (without drivers) in WSL:

   # Add CUDA repository configuration
   wget https://developer.download.nvidia.com/compute/cuda/repos/wsl-ubuntu/x86_64/cuda-wsl-ubuntu.pin
   sudo mv cuda-wsl-ubuntu.pin /etc/apt/preferences.d/cuda-repository-pin-600
    
   # Download and install CUDA repository package
   wget https://developer.download.nvidia.com/compute/cuda/12.8.0/local_installers/cuda-repo-wsl-ubuntu-12-8-local_12.8.0-1_amd64.deb
   sudo dpkg -i cuda-repo-wsl-ubuntu-12-8-local_12.8.0-1_amd64.deb
    
   # Copy CUDA GPG key
   sudo cp /var/cuda-repo-wsl-ubuntu-12-8-local/cuda-*-keyring.gpg /usr/share/keyrings/
    
   # Update package list and install CUDA toolkit
   sudo apt-get update
   sudo apt-get -y install cuda-toolkit-12-8

Note: If you are missing certain libraries after installation (e.g. cufft), run the following commands:

sudo dpkg --configure -a
sudo apt-get install cuda-libraries-dev-12-8

Afterwards, you can remove the downloaded repository package to free up space:

rm cuda-repo-wsl-ubuntu-12-8-local_12.8.0-1_amd64.deb

Environment Configuration

After installing CUDA, you need to configure your environment:

1. Add CUDA to your system PATH:

   nano ~/.bashrc

2. Add the following lines to the end of the file:

   # CUDA Toolkit paths
   export PATH=/usr/local/cuda-12.8/bin${PATH:+:${PATH}}
   export LD_LIBRARY_PATH=/usr/local/cuda-12.8/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

3. Save the file (Ctrl+O, Enter, Ctrl+X) and reload your bash configuration:

   source ~/.bashrc

4. Verify the installation by running:

   nvcc --version

This should display the CUDA compiler version if everything is set up correctly.

You can now compile HILA applications with CUDA by defining ARCH := cuda in the Makefile.