Alvaro Vazquez-Mayagoitia alvarovm

🏠

Working from home

alvarovm / clangoffload.md

Created June 30, 2022 22:50 — forked from anjohan/clangoffload.md

Build Clang with OpenMP Target Offloading for NVIDIA GPUs

Note to self

This works with Clang 12 on Manjaro, and maybe other Linux versions. CUDA 11.1 is installed with pacman in /opt/cuda.

git clone https://github.com/llvm/llvm-project.git

alvarovm / MACE_install.md

Last active February 27, 2024 17:35 — forked from jingsk/MACE_install.md

Install MACE with cuda-enabled Pytorch

Mace installation on ANL's Swing as of 1/30/24

Check resources available on Swing: NVIDIA A100 GPUs (8 GPUs per node) - 1/8 node allocated when requesting 1 gpu

MACE requires Pytorch2, which needs CUDA 11.8 or 11.7. Check required CUDA versions here

Create a Conda environment with Python 3.10

conda create -n "CUDA-torch-base" python=3.10.0