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 Gitlab CI/CD targets for gh200 #1546

Merged
merged 76 commits into from
Jul 4, 2024
Merged

Add Gitlab CI/CD targets for gh200 #1546

merged 76 commits into from
Jul 4, 2024

Conversation

edopao
Copy link
Contributor

@edopao edopao commented May 29, 2024

This PR adds the CI/CD configuration for GH200 nodes, while keeping the currrent configuration for x86_64+CUDA on PizDaint.

Some differences between GH200 (on Todi vCluster) and x86_64+CUDA (on PizDaint):

  • CUDA v11.2.2 / CUDA ARCH=60 on PizDaint vs. CUDA v12.4.1 / CUDA ARCH=90 on GH200
  • Support for Python 3.8, 3.9, 3.10, 3.11 on x86_64 Ubuntu base image, while Python 3.8 is not supported on the ARM base image for GH200
  • JAX dependency updated from v0.4.13 to v0.4.18 because this is the minimum version available on the ARM base image
  • A compiler is allowed to choose if char is signed or unsigned. The Python bindings in GT4Py cartesian rely on the signed representation, which was the default for the compiler on the x86_64 Ubuntu base image. This behavior is not the default on ARM base image, so we have to enforce it with the flag -fsigned-char

@edopao
Copy link
Contributor Author

edopao commented May 31, 2024

cscs-ci run default

@edopao
Copy link
Contributor Author

edopao commented Jul 2, 2024

cscs-ci run default

@edopao edopao marked this pull request as ready for review July 2, 2024 06:24
@edopao edopao requested a review from havogt July 2, 2024 06:29
Copy link
Contributor

@havogt havogt left a comment

Choose a reason for hiding this comment

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

I don't know how to solve the version constraints problem, but we will discuss today if we can drop 3.8.

src/gt4py/cartesian/backend/pyext_builder.py Outdated Show resolved Hide resolved
requirements-dev.txt Outdated Show resolved Hide resolved
ci/cscs-ci.yml Outdated Show resolved Hide resolved
@edopao
Copy link
Contributor Author

edopao commented Jul 2, 2024

cscs-ci run default

@edopao edopao requested a review from havogt July 2, 2024 19:06
ipython==8.12.3 # via -c constraints.txt, ipykernel
isort==5.13.2 # via -c constraints.txt, -r requirements-dev.in
jax[cpu]==0.4.13 # via -c constraints.txt, gt4py (pyproject.toml)
Copy link
Contributor

Choose a reason for hiding this comment

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

We are removing jax from requirements-dev.txt because it's build for python 3.8. For me that's fine as we anyway don't have jax developers atm. Once we drop 3.8 and 3.9 we will have it back. What do you think @egparedes ?

Copy link
Contributor

@havogt havogt left a comment

Choose a reason for hiding this comment

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

Discussed with Enrique. Looks good.

pyproject.toml Outdated Show resolved Hide resolved
edopao and others added 2 commits July 3, 2024 16:50
Co-authored-by: Hannes Vogt <hannes@havogt.de>
@edopao
Copy link
Contributor Author

edopao commented Jul 4, 2024

cscs-ci run default

@edopao edopao merged commit df9aec2 into GridTools:main Jul 4, 2024
55 checks passed
@edopao edopao deleted the ci-gh200 branch July 4, 2024 06:29
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants