Build with CUDA 12.6

[jeongseok-meta]

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.

[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.

[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 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 pinnnings.
• 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/11185517647.}

[jeongseok-meta]

@hmaarrfk I lost from the first step. It seems we can only use CUDA versions listed here, but it doesn't have 12.2.

[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.

[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/11185822661.}

[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/11186641785.}

[jeongseok-meta]

@conda-forge-admin, please rerender

[hmaarrfk]

Suggested change

	- 12.0 # [linux64]
	- 12.2 # [linux64]

[hmaarrfk]

you might need to replicate the rest of everything too.

[jeongseok-meta]

Rerendering with 12.2 seems to fail per:

If there is a recipe/conda-build-config.yaml file in the feedstock make sure that it is compatible with the current global pinnnings.

[hmaarrfk]

you might need to replicate the rest of everything too.

[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.

[jeongseok-meta]

@conda-forge-admin, please rerender

[hmaarrfk]

👏🏾

[hmaarrfk]

"error": "User not authorized for the requested runners, user not present in 'users' list."

set it to azure for now, and see how far you get ;)

[hmaarrfk]

You might also want to rebase onto 22bf055

[hmaarrfk]

Suggested change

	linux_64: github_actions
	linux_64: azure

[jeongseok-meta]

@conda-forge-admin, please rerender

[jeongseok-meta]

@conda-forge-admin, please restart ci

[hmaarrfk]

@conda-forge-admin please rerender

[hmaarrfk]

the "yaml" files control the CI files, so you always have to rerender after changing them.

[hmaarrfk]

i'm speachless...

[hmaarrfk]

Packages (-cuda126.conda):

I actually cannot use this selector when uploading. 'Labels' work well. could you add the cfep03 to the new packages please?

[jeongseok-meta]

Do you want the label for cuda126 or including all the other pytorch 2.5.1?

[hmaarrfk]

just the new packages.

[hmaarrfk]

i think if you label all i can use the --skip-existing flag to not recopy the files over

[jeongseok-meta]

Done!

[hmaarrfk]

Sorry shoudln't there be a aarch cuda 12.6 package as well?

anaconda copy --from-label cfep03 --to-label main --to-owner conda-forge jslee02/libtorch/2.5.1
anaconda copy --from-label cfep03 --to-label main --to-owner conda-forge jslee02/pytorch/2.5.1
# anaconda copy --from-label cfep03 --to-label main --to-owner conda-forge jslee02/pytorch-cpu/2.5.1
anaconda copy --from-label cfep03 --to-label main --to-owner conda-forge jslee02/pytorch-gpu/2.5.1

[jeongseok-meta]

No, I intended to add only linux64 (not aarch) to move faster. But if I knew we could update this quickly, I would add them here too. How would you like to proceed? We could add them in this PR or in a follow-up. Either way works for me.

[hmaarrfk]

lets do it here without any force push. I want the OSX builds to finish building on main.

[hmaarrfk]

The cuda 12.6 builds have been uploaded already for linux 64. give them a test drive... you know just in case.

[hmaarrfk]

you can probably delete everything from your channel too for space.....

[jeongseok-meta]

lets do it here without any force push. I want the OSX builds to finish building on main.

Let me build locally for aarch64 + cuda126 without push a commit, as I expect the OSX build will be retriggered anyway, and then push the change for aarch64 after the current OSX build is done.

[jeongseok-meta]

• Deleted all the packages from my channel
• Initiated aarch64 + cuda126 build locally
• Will push the changes once the current OSX builds are done

[hmaarrfk]

you really don't need to worry about this branch, it was mostly about the 'main' branch where the packages get uploaded.

[jslee02]

Building aarch64 + cuda126 is done!

Build log: log_files.zip

Packages (only new ones are uploaded):

• https://anaconda.org/jslee02/libtorch
• https://anaconda.org/jslee02/pytorch
• https://anaconda.org/jslee02/pytorch-gpu

[hmaarrfk]

I'm not sure if this is important but things I observd:

  -- Unable to find cublas_v2.h in either "/home/conda/feedstock_root/build_artifacts/libtorch_1730670943607/_build_env/targets/sbsa-linux/include" or "/home/conda/feedstock_root/build_artifacts/libtorch_1730670943607/_build_env/math_libs/include"

[hmaarrfk]

anaconda copy --from-label main --to-label main --to-owner conda-forge jslee02/libtorch/2.5.1
anaconda copy --from-label main --to-label main --to-owner conda-forge jslee02/pytorch/2.5.1
anaconda copy --from-label main --to-label main --to-owner conda-forge jslee02/pytorch-gpu/2.5.1

[hmaarrfk]

Please feel free to issue the command:
https://conda-forge.org/docs/maintainer/infrastructure/#conda-forge-admin-please-add-user-username

with the appropriate username !

[jeongseok-meta]

@conda-forge-admin, please add user @jeongseok-meta

[hmaarrfk]

you might need to do this on an issue.