Skip to content
This repository was archived by the owner on Oct 6, 2025. It is now read-only.

Comments

standalone: use nvidia runtime if using GPU#57

Merged
xenoscopic merged 1 commit intomainfrom
nvidia-runtime
May 13, 2025
Merged

standalone: use nvidia runtime if using GPU#57
xenoscopic merged 1 commit intomainfrom
nvidia-runtime

Conversation

@xenoscopic
Copy link
Contributor

We need the NVIDIA CDI to run the standalone GPU container.

Signed-off-by: Jacob Howard <jacob.howard@docker.com>
},
}
hostConfig.Runtime = "nvidia"
hostConfig.DeviceRequests = []container.DeviceRequest{{Count: -1, Capabilities: [][]string{{"gpu"}}}}
Copy link
Contributor

Choose a reason for hiding this comment

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

We don't need the Driver anymore in the DeviceRequest?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

It looks like the NVIDIA CDI handles it. Tested with cloud engine.

@xenoscopic xenoscopic merged commit bb895f3 into main May 13, 2025
4 checks passed
@xenoscopic xenoscopic deleted the nvidia-runtime branch May 13, 2025 23:28
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants