Skip to content

CI build for HIP#178

Draft
peastman wants to merge 22 commits into
openmm:masterfrom
peastman:hip
Draft

CI build for HIP#178
peastman wants to merge 22 commits into
openmm:masterfrom
peastman:hip

Conversation

@peastman

Copy link
Copy Markdown
Member

No description provided.

Comment thread .github/workflows/CI.yml
run: |
wget https://repo.radeon.com/amdgpu-install/7.2.4/ubuntu/noble/amdgpu-install_7.2.4.70204-1_all.deb
sudo apt install ./amdgpu-install_7.2.4.70204-1_all.deb
sudo apt update

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think an additional line is needed here to install ROCm. The amdgpu-install_7.2.4.70204-1_all.deb package they provide just adds some repositories to APT but doesn't install the libraries themselves: https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/quick-start.html#install-rocm

Suggested change
sudo apt update
sudo apt update
sudo apt install rocm

@peastman

peastman commented Jul 7, 2026

Copy link
Copy Markdown
Member Author

Some progress. I have it building everything with HIP. It then gets to this line which just reports what plugins could or couldn't be loaded:

python -c "import openmm as mm; print('---Loaded---', *mm.pluginLoadedLibNames, '---Failed---', *mm.Platform.getPluginLoadFailures(), sep='\n')"

It prints the output (reference, CPU, and OpenCL all loaded, CUDA and HIP not loaded because of missing dependencies). And then it crashes with a memory corruption error: corrupted size vs. prev_size in fastbins. It hasn't actually run any HIP code. All it's done is try and fail to load the HIP plugins because it couldn't find libhiprtc.so.6. But something has still managed to corrupt memory.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants