An AMD GPU runtime and development toolkit. It installs ROCm from the official AMD APT repository, typically alongside a language SDK for GPU-accelerated workloads.
A minimal workshop:
# workshop.yaml
name: gpu-workload
base: [email protected]
sdks:
- name: uv
channel: 0.9/stable
- name: rocm
channel: 7.1/stable
actions:
check-gpu: |
rocminfoThis demonstrates GPU access verification inside a workshop with both a language SDK and the ROCm companion runtime.
The SDK installs the complete ROCm GPU compute development and system toolset from the AMD APT repository and automatically configures environment paths.
To build a project in the workshop:
git clone [email protected]:ROCm/rocm-examples.git
workshop launch
workshop shell
makeOnce the workshop is ready:
workshop shell
rocminfoThis lists detected AMD GPUs and their capabilities. For a monitoring view:
rocm-smi- Interface:
gpu - Purpose: Grants access to AMD GPU hardware on the host.
This SDK doesn't define any slots.
- ROCm community: ROCm GitHub
- Workshop forum: Discourse
- Please review our Code of Conduct before participating.
All contributions, including code, documentation updates, and issue reports, are welcome!
- See
CONTRIBUTING.mdfor guidelines. - Open issues or pull requests on the official repository.
Copyright 2025 Canonical Ltd.
This program is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License version 2.1 (LGPLv2.1) as published by the Free Software Foundation.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details.
ROCm is licensed under the MIT License.