Skip to content

[CI][CUDA] Move away from cuda12.4, Add cuda12.6 eager CI tests#148602

Closed
tinglvv wants to merge 6 commits intopytorch:mainfrom
tinglvv:cu126-eager-benchmark
Closed

[CI][CUDA] Move away from cuda12.4, Add cuda12.6 eager CI tests#148602
tinglvv wants to merge 6 commits intopytorch:mainfrom
tinglvv:cu126-eager-benchmark

Conversation

@tinglvv
Copy link
Copy Markdown
Collaborator

@tinglvv tinglvv commented Mar 5, 2025 •
edited
Loading

#145570

breaking #140793 into eager and inductor benchmarks to unblock

cc @atalman @malfet @nWEIdia @ptrblck

@tinglvv tinglvv requested review from a team and jeffdaily as code owners March 5, 2025 21:04
@pytorch-bot
Copy link
Copy Markdown

pytorch-bot bot commented Mar 5, 2025 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/148602

Note: Links to docs will display an error until the docs builds have been completed.

❌ 2 New Failures, 14 Pending

As of commit 69d5251 with merge base 98458e5 (image):

NEW FAILURES - The following jobs have failed:

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@pytorch-bot pytorch-bot bot added the topic: not user facing topic category label Mar 5, 2025
@nWEIdia nWEIdia added the keep-going Don't stop on first failure, keep running tests until the end label Mar 5, 2025
nWEIdia
nWEIdia requested changes Mar 5, 2025
View reviewed changes
Copy link
Copy Markdown
Collaborator

@nWEIdia nWEIdia left a comment

Choose a reason for hiding this comment

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

there is also "slow" job.
https://github.com/pytorch/pytorch/blob/c9edd37ffb8c824db7da4e353a783b9738d71cca/.github/workflows/slow.yml

nWEIdia
nWEIdia reviewed Mar 5, 2025
View reviewed changes
@nWEIdia nWEIdia changed the title Move away from 12.4, Add 12.6 eager CI tests [CI][CUDA] Move away from cuda12.4, Add cuda12.6 eager CI tests Mar 5, 2025
@tinglvv
Copy link
Copy Markdown
Collaborator Author

tinglvv commented Mar 5, 2025

Failure #2 [internal] load metadata for docker.io/nvidia/cuda:12.4-devel-ubuntu20.04
#2 ERROR: docker.io/nvidia/cuda:12.4-devel-ubuntu20.04: not found, missed to replace this test. let me modify.

@tinglvv tinglvv mentioned this pull request Mar 6, 2025
Closed
26 tasks
atalman
atalman reviewed Mar 6, 2025
View reviewed changes
@atalman
Copy link
Copy Markdown
Contributor

atalman commented Mar 6, 2025

Failure #2 [internal] load metadata for docker.io/nvidia/cuda:12.4-devel-ubuntu20.04 #2 ERROR: docker.io/nvidia/cuda:12.4-devel-ubuntu20.04: not found, missed to replace this test. let me modify.

This looks like transient error with our infra with calculate-docker-image

@atalman atalman added the ciflow/trunk Trigger trunk jobs on your pull request label Mar 6, 2025
janeyx99
janeyx99 reviewed Mar 6, 2025
View reviewed changes
@atalman atalman mentioned this pull request Mar 6, 2025
Open
@tinglvv
Copy link
Copy Markdown
Collaborator Author

tinglvv commented Mar 6, 2025

Thanks Andrey for skipping the test, waiting for the green signal from the test to merge.

@colesbury colesbury added the triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module label Mar 6, 2025
@atalman atalman requested a review from malfet March 6, 2025 21:08
atalman
atalman approved these changes Mar 6, 2025
View reviewed changes
Copy link
Copy Markdown
Contributor

@atalman atalman left a comment

Choose a reason for hiding this comment

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

lgtm

@malfet malfet added ciflow/periodic Trigger jobs ran periodically on master (periodic.yml) on the PR ciflow/slow labels Mar 7, 2025
@malfet
Copy link
Copy Markdown
Contributor

malfet commented Mar 7, 2025

If you are modifying slow and periodic workflows, you should tests those as well

malfet
malfet reviewed Mar 7, 2025
View reviewed changes
.ci/docker/build.sh
;;
pytorch-linux-jammy-py3.12-triton-cpu)
CUDA_VERSION=12.4
CUDA_VERSION=12.6
Copy link
Copy Markdown
Contributor

@malfet malfet Mar 7, 2025

Choose a reason for hiding this comment

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

This sounds like an oxymoron: why CPU builds needs CUDA (and what this config is for to begin with)

.ci/docker/build.sh
;;
pytorch-linux-jammy-py3.12-halide)
CUDA_VERSION=12.4
CUDA_VERSION=12.6
Copy link
Copy Markdown
Contributor

@malfet malfet Mar 7, 2025

Choose a reason for hiding this comment

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

Halide is part of inductor, so not sure why you are modifying it here.

malfet
malfet approved these changes Mar 7, 2025
View reviewed changes
Copy link
Copy Markdown
Contributor

@malfet malfet left a comment

Choose a reason for hiding this comment

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

Let's test it in prod

@malfet
Copy link
Copy Markdown
Contributor

malfet commented Mar 7, 2025

@pytorchbot merge -f "Spartaa aa aa!"

@pytorchmergebot
Copy link
Copy Markdown
Collaborator

pytorchmergebot commented Mar 7, 2025

Merge started

Your change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use -f as last resort and instead consider -i/--ignore-current to continue the merge ignoring current failures. This will allow currently pending tests to finish and report signal before the merge.

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging
Check the merge workflow status
here

@anumol-baby anumol-baby mentioned this pull request Jun 17, 2025
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@janeyx99 janeyx99 janeyx99 left review comments

@nWEIdia nWEIdia nWEIdia requested changes

@malfet malfet malfet approved these changes

@atalman atalman atalman approved these changes

@jeffdaily jeffdaily Awaiting requested review from jeffdaily jeffdaily is a code owner

Assignees

No one assigned

Labels

ciflow/periodic Trigger jobs ran periodically on master (periodic.yml) on the PR ciflow/slow ciflow/trunk Trigger trunk jobs on your pull request keep-going Don't stop on first failure, keep running tests until the end Merged open source topic: not user facing topic category triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

8 participants

@tinglvv @atalman @malfet @pytorchmergebot @janeyx99 @nWEIdia @colesbury @pytorchbot