Try compiling cuda version on linux-aarch64

[diegoferigo]

Checklist

• Used a personal fork of the feedstock to propose changes
• Bumped the build number (if the version is unchanged)
• Reset the build number to 0 (if the version changed)
• Re-rendered with the latest conda-smithy (Use the phrase @conda-forge-admin, please rerender in a comment in this PR for automated rerendering)
• Ensured the license file is being packaged.

Worth trying, from #168 (comment).

[conda-forge-admin]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe/meta.yaml) and found it was in an excellent condition.

I do have some suggestions for making it better though...

For recipe/meta.yaml:

• ℹ️ The recipe is not parsable by parser conda-souschef (grayskull). This parser is not currently used by conda-forge, but may be in the future. We are collecting information to see which recipes are compatible with grayskull.
• ℹ️ The recipe is not parsable by parser conda-recipe-manager. The recipe can only be automatically migrated to the new v1 format if it is parseable by conda-recipe-manager.

_{This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/22623678598. Examine the logs at this URL for more detail.}

[diegoferigo-rai]

@conda-forge-admin, please rerender

[diegoferigo]

@conda-forge-admin, please rerender

[conda-forge-admin]

Hi! This is the friendly automated conda-forge-linting service.

I wanted to let you know that I linted all conda-recipes in your PR (recipe/meta.yaml) and found some lint.

Here's what I've got...

For recipe/meta.yaml:

• ❌ The feedstock has no .ci_support files and thus will not build any packages.

For recipe/meta.yaml:

• ℹ️ The recipe is not parsable by parser conda-souschef (grayskull). This parser is not currently used by conda-forge, but may be in the future. We are collecting information to see which recipes are compatible with grayskull.
• ℹ️ The recipe is not parsable by parser conda-recipe-manager. The recipe can only be automatically migrated to the new v1 format if it is parseable by conda-recipe-manager.

_{This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/22623918551. Examine the logs at this URL for more detail.}

[diegoferigo]

@conda-forge-admin, please rerender

[conda-forge-admin]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe/meta.yaml) and found it was in an excellent condition.

I do have some suggestions for making it better though...

For recipe/meta.yaml:

• ℹ️ The recipe is not parsable by parser conda-souschef (grayskull). This parser is not currently used by conda-forge, but may be in the future. We are collecting information to see which recipes are compatible with grayskull.
• ℹ️ The recipe is not parsable by parser conda-recipe-manager. The recipe can only be automatically migrated to the new v1 format if it is parseable by conda-recipe-manager.

_{This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/22663521834. Examine the logs at this URL for more detail.}

[diegoferigo]

@conda-forge-admin, please rerender

[diegoferigo]

@conda-forge-admin, please rerender

[conda-forge-admin]

Hi! This is the friendly automated conda-forge-webservice.

I tried to rerender for you but ran into some issues. Please check the output logs of the GitHub Actions workflow below for more details. You can also ping conda-forge/core (using the @ notation) for further assistance or you can try rerendering locally.

The following suggestions might help debug any issues:

• Is the recipe/{meta.yaml,recipe.yaml} file valid?
• If there is a recipe/conda-build-config.yaml file in the feedstock make sure that it is compatible with the current global pinnings.
• Is the fork used for this PR on an organization or user GitHub account? Automated rerendering via the webservices admin bot only works for user GitHub accounts.

_{This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/22662563688. Examine the logs at this URL for more detail.}

[diegoferigo]

@conda-forge-admin, please rerender

[conda-forge-admin]

Hi! This is the friendly automated conda-forge-webservice.

I tried to rerender for you but ran into some issues. Please check the output logs of the GitHub Actions workflow below for more details. You can also ping conda-forge/core (using the @ notation) for further assistance or you can try rerendering locally.

The following suggestions might help debug any issues:

• Is the recipe/{meta.yaml,recipe.yaml} file valid?
• If there is a recipe/conda-build-config.yaml file in the feedstock make sure that it is compatible with the current global pinnings.
• Is the fork used for this PR on an organization or user GitHub account? Automated rerendering via the webservices admin bot only works for user GitHub accounts.

_{This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/22663207477. Examine the logs at this URL for more detail.}

[diegoferigo]

@conda-forge-admin, please rerender

[diegoferigo]

@conda-forge-admin, please rerender

[conda-forge-admin]

Hi! This is the friendly automated conda-forge-linting service.

I wanted to let you know that I linted all conda-recipes in your PR (recipe/meta.yaml) and found some lint.

Here's what I've got...

For recipe/meta.yaml:

• ❌ The feedstock has no .ci_support files and thus will not build any packages.

For recipe/meta.yaml:

• ℹ️ The recipe is not parsable by parser conda-souschef (grayskull). This parser is not currently used by conda-forge, but may be in the future. We are collecting information to see which recipes are compatible with grayskull.
• ℹ️ The recipe is not parsable by parser conda-recipe-manager. The recipe can only be automatically migrated to the new v1 format if it is parseable by conda-recipe-manager.

_{This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/22664480430. Examine the logs at this URL for more detail.}

[diegoferigo]

@conda-forge-admin, please rerender

[conda-forge-admin]

Hi! This is the friendly automated conda-forge-webservice.

I tried to rerender for you, but it looks like there was nothing to do.

_{This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/22663706426. Examine the logs at this URL for more detail.}

[diegoferigo]

@conda-forge-admin, please rerender

[conda-forge-admin]

Hi! This is the friendly automated conda-forge-webservice.

I tried to rerender for you, but it looks like there was nothing to do.

_{This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/22664074781. Examine the logs at this URL for more detail.}

[diegoferigo]

@conda-forge-admin, please rerender

[conda-forge-admin]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe/meta.yaml) and found it was in an excellent condition.

I do have some suggestions for making it better though...

For recipe/meta.yaml:

• ℹ️ The recipe is not parsable by parser conda-souschef (grayskull). This parser is not currently used by conda-forge, but may be in the future. We are collecting information to see which recipes are compatible with grayskull.
• ℹ️ The recipe is not parsable by parser conda-recipe-manager. The recipe can only be automatically migrated to the new v1 format if it is parseable by conda-recipe-manager.

_{This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/22664588996. Examine the logs at this URL for more detail.}

[diegoferigo]

The Azure runners hit their 6-hour limit, but there were no failures before that point.

After a few iterations (I'm still getting familiar with this setup), I managed to add support for building this recipe locally with Pixi. I currently have a Jetson Thor on my desk, might be worth giving it a try.

Note1: it seems there's a bug in this combination with pixi as install tool and old meta.yaml recipe. In order to build locally, I had to modify the pixi.toml task to pass an extra -c conda-forge option to conda-build. FYI @conda-forge/core

Note2: solving the environment with conda-build is painfully slow, at least locally. It takes ~20 minutes before it starts building... @conda-forge/onnxruntime any plan to port the recipe to rattler-build?
Update: second run was much faster, ~5 minutes.

[diegoferigo]

@cbourjau I successfully built the CUDA13 version with Python 3.12 on linux-aarch64 (Nvidia Jetson Thor in 3h), good job on the previous PR! I think you can enable back arm cuda builds in the next minor release 🎉

[flferretti]

Thanks a lot for taking care of this @diegoferigo! Is this ready for review?

[diegoferigo]

Thanks a lot for taking care of this @diegoferigo! Is this ready for review?

It is ready in the sense that it served its purpose: checking if linux-aarch64 with GPU support compiles. The only relevant commit of this PR is 3cecf6d, and maybe it can be folded directly into either #174 or #171, cc @cbourjau.

[flferretti]

Thanks a lot for taking care of this @diegoferigo! Is this ready for review?

It is ready in the sense that it served its purpose: checking if linux-aarch64 with GPU support compiles. The only relevant commit of this PR is 3cecf6d, and maybe it can be folded directly into either #174 or #171, cc @cbourjau.

Alright, thank you! Have you perhaps tried to run an inference? I've built this recipe locally and I'm getting CUDAExecutionProvider available on Blackwell GB10, but model execution fails with cudaErrorNoKernelImageForDevice. I wasn't expecting this since it should be compiled for sm120, that is theoretically compatible with sm121

EDIT: onnxruntime is not compiled for sm_121. Would it be possible to add it to CUDA_ARCH_LIST?

CC @giotherobot

[cbourjau]

I was hoping to make faster progress with the "megabuilds" that would significantly reduce the size of the build matrix, but it has proven difficult. In the meantime I'm afraid that a new problem has come up: conda-forge/.cirun#174 . Even if everything were in order and ready to build right now, we wouldn't be able to finish it by Monday. I'm afraid we won't be making new onnxruntime releases until we get an alternative to those runners :(

EDIT: onnxruntime is not compiled for sm_121. Would it be possible to add it to CUDA_ARCH_LIST?

Should be possible.

[flferretti]

No worries @cbourjau, thanks a lot for your work!