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

e2e: add GPU test #1136

Merged
merged 1 commit into from
Jan 23, 2025
Merged

e2e: add GPU test #1136

merged 1 commit into from
Jan 23, 2025

Conversation

msanft
Copy link
Contributor

@msanft msanft commented Jan 10, 2025

This adds an E2E test for GPU use on Contrast.
It currently runs on the GPU-enabled bare-metal SNP runner.

The test currently only verifies that the GPU is available via
nvidia-smi, which also verifies that driver and CUDA work correctly.

@msanft msanft added the no changelog PRs not listed in the release notes label Jan 10, 2025
@msanft msanft added this to the v1.4.0 milestone Jan 10, 2025
@msanft msanft force-pushed the msanft/gpu-e2e-test branch from f1fcfd7 to e92e97d Compare January 10, 2025 12:55
@msanft msanft force-pushed the msanft/gpu-runtime-class branch from 2218242 to 5bfe77f Compare January 10, 2025 15:21
@msanft msanft force-pushed the msanft/gpu-e2e-test branch from e92e97d to b24fc44 Compare January 10, 2025 15:23
e2e/gpu/testdata/gpu-pod.yaml Outdated Show resolved Hide resolved
.github/workflows/e2e_on_pull_request.yml Outdated Show resolved Hide resolved
flake.nix Outdated Show resolved Hide resolved
@msanft msanft force-pushed the msanft/gpu-runtime-class branch from 5bfe77f to fa64242 Compare January 13, 2025 07:41
@msanft msanft force-pushed the msanft/gpu-e2e-test branch 4 times, most recently from f7384f3 to 1b54d8c Compare January 13, 2025 08:15
@msanft msanft force-pushed the msanft/gpu-runtime-class branch from fa64242 to 8674ee5 Compare January 13, 2025 08:30
@msanft msanft force-pushed the msanft/gpu-e2e-test branch from 1b54d8c to 437b977 Compare January 13, 2025 09:43
@msanft msanft force-pushed the msanft/gpu-runtime-class branch 2 times, most recently from a59e508 to 2c70c4e Compare January 14, 2025 08:05
@msanft msanft force-pushed the msanft/gpu-e2e-test branch 2 times, most recently from 896901f to 9ee8c10 Compare January 14, 2025 08:18
@msanft msanft force-pushed the msanft/gpu-runtime-class branch from 2c70c4e to 1f69236 Compare January 14, 2025 08:26
@msanft msanft force-pushed the msanft/gpu-e2e-test branch 4 times, most recently from 60a66f5 to b8b5b31 Compare January 16, 2025 09:59
@msanft msanft force-pushed the msanft/gpu-runtime-class branch from 1f69236 to acf8e0a Compare January 17, 2025 15:17
@msanft msanft force-pushed the msanft/gpu-e2e-test branch from b8b5b31 to e91c280 Compare January 17, 2025 15:18
@msanft msanft force-pushed the msanft/gpu-runtime-class branch from acf8e0a to fe591c1 Compare January 17, 2025 16:37
@msanft msanft force-pushed the msanft/gpu-e2e-test branch from e91c280 to 3bd65be Compare January 17, 2025 16:38
@msanft msanft force-pushed the msanft/gpu-runtime-class branch from fe591c1 to eb8de6c Compare January 20, 2025 10:03
@msanft msanft force-pushed the msanft/gpu-e2e-test branch from 3bd65be to 3592c49 Compare January 20, 2025 10:03
.github/workflows/e2e_manual.yml Show resolved Hide resolved
@msanft msanft force-pushed the msanft/gpu-runtime-class branch 2 times, most recently from 64bba4f to 3a5d4bc Compare January 20, 2025 12:27
@msanft msanft force-pushed the msanft/gpu-e2e-test branch 3 times, most recently from a0c1f64 to 721a4c6 Compare January 20, 2025 12:32
@msanft msanft force-pushed the msanft/gpu-runtime-class branch 3 times, most recently from 9afca0a to 0a1f4fa Compare January 22, 2025 09:38
@msanft msanft force-pushed the msanft/gpu-e2e-test branch 2 times, most recently from c7ebadd to c201276 Compare January 22, 2025 09:39
@msanft msanft force-pushed the msanft/gpu-runtime-class branch from 0a1f4fa to b5f9a13 Compare January 22, 2025 15:20
@msanft msanft force-pushed the msanft/gpu-e2e-test branch from c201276 to 4ed3505 Compare January 22, 2025 15:23
@msanft msanft force-pushed the msanft/gpu-runtime-class branch from b5f9a13 to 5255d32 Compare January 22, 2025 15:24
@msanft msanft force-pushed the msanft/gpu-e2e-test branch from 4ed3505 to 1dc1300 Compare January 22, 2025 15:25
- name: K3s-QEMU-SNP-GPU
runner: SNP
self-hosted: true
test-name: [servicemesh, openssl, policy, workloadsecret, volumestatefulset, gpu]
Copy link
Member

Choose a reason for hiding this comment

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

Do we really need to run all those tests for GPU?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

I would like to also run tests that exercise service mesh functionality, etc. on the GPU runtime class.

.github/workflows/e2e_nightly.yml Outdated Show resolved Hide resolved
packages/by-name/contrast/package.nix Outdated Show resolved Hide resolved
packages/by-name/contrast/package.nix Outdated Show resolved Hide resolved
@katexochen
Copy link
Member

commit message should be e2e: add GPU test

@msanft msanft force-pushed the msanft/gpu-e2e-test branch from 1dc1300 to 776afff Compare January 23, 2025 07:17
This adds an E2E test for GPU use on Contrast.
It currently runs on the GPU-enabled bare-metal SNP runner.

The test currently only verifies that the GPU is available via
nvidia-smi, which also verifies that driver and CUDA work correctly.
@msanft msanft force-pushed the msanft/gpu-e2e-test branch from 776afff to 4fbe39a Compare January 23, 2025 07:54
@msanft msanft requested a review from katexochen January 23, 2025 07:55
@msanft msanft changed the title Add GPU E2E Test e2e: add GPU test Jan 23, 2025
@msanft msanft merged commit b664359 into msanft/gpu-runtime-class Jan 23, 2025
12 checks passed
@msanft msanft deleted the msanft/gpu-e2e-test branch January 23, 2025 12:13
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
no changelog PRs not listed in the release notes
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants