Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add GPU support to minikube. #2936

Merged
merged 3 commits into from
Jul 9, 2018
Merged

Conversation

rohitagarwal003
Copy link
Member

@rohitagarwal003 rohitagarwal003 commented Jun 26, 2018

Fixes #2115.

Testing this PR locally:

See the instructions added to docs/gpu.md in this PR. Even if you don't satisfy the prerequisites mentioned there, it would be useful to run these and check if the error messages returned are useful.

# From your minikube checkout
git fetch origin pull/2936/head:gpu-support
git checkout gpu-support

# Build the docker machine plugin
make out/docker-machine-driver-kvm2
cp out/docker-machine-driver-kvm2 /usr/local/bin/docker-machine-driver-kvm2

# Build minikube
make out/minikube

# Run minikube with GPUs
out/minikube start --vm-driver kvm2 --gpu --logtostderr

@k8s-ci-robot k8s-ci-robot added cncf-cla: yes Indicates the PR's author has signed the CNCF CLA. size/L Denotes a PR that changes 100-499 lines, ignoring generated files. labels Jun 26, 2018
@rohitagarwal003
Copy link
Member Author

/assign @vishh

@minikube-bot
Copy link
Collaborator

Can one of the admins verify this patch?

@rohitagarwal003 rohitagarwal003 force-pushed the gpu-support branch 2 times, most recently from b65beec to 3dc19ac Compare June 26, 2018 17:20
@vishh
Copy link
Contributor

vishh commented Jun 27, 2018

This PR overall lgtm. I'd recommend having some of the users wanting this feature in #2115 to try this out as well.

@vishh
Copy link
Contributor

vishh commented Jun 27, 2018

What would be the testing and maintenance story for this feature? If we need community help with maintenance, should we mark it as alpha/experimental until we can demonstrate sustained usage and maintenance?

@k8s-ci-robot k8s-ci-robot added size/XL Denotes a PR that changes 500-999 lines, ignoring generated files. and removed size/L Denotes a PR that changes 100-499 lines, ignoring generated files. labels Jun 27, 2018
@rohitagarwal003
Copy link
Member Author

/assign @dlorenc @r2d4 @aaron-prindle

@rohitagarwal003
Copy link
Member Author

This PR overall lgtm. I'd recommend having some of the users wanting this feature in #2115 to try this out as well.

Commented on that issue.

What would be the testing and maintenance story for this feature? If we need community help with maintenance, should we mark it as alpha/experimental until we can demonstrate sustained usage and maintenance?

The doc and help text say experimental. I would love to e2e test this but since GCE's nested virtualization doesn't have a virtual IOMMU, so we can't test this on GCE. If people have other ideas, I am all ears.

@k8s-ci-robot
Copy link
Contributor

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: mindprince
To fully approve this pull request, please assign additional approvers.
We suggest the following additional approver: aaron-prindle

Assign the PR to them by writing /assign @aaron-prindle in a comment when ready.

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

@dlorenc
Copy link
Contributor

dlorenc commented Jun 27, 2018

Thanks! The PR looks good to me. I'd like to think about the testing for another day or so and see if we can come up with something.

@rohitagarwal003
Copy link
Member Author

Ping.

@dlorenc
Copy link
Contributor

dlorenc commented Jul 4, 2018

@minikube-bot OK to test

@dlorenc dlorenc merged commit db90809 into kubernetes:master Jul 9, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
cncf-cla: yes Indicates the PR's author has signed the CNCF CLA. size/XL Denotes a PR that changes 500-999 lines, ignoring generated files.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

7 participants