arm (cross-compile) builds, mlx v0.0.9, py312

[ngam]

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.

closes #1
closes #3
closes #6
closes #7
closes #9

[conda-forge-webservices]

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) and found it was in an excellent condition.

[ngam]

@conda-forge-admin, please rerender.

[github-actions]

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 latest rerendering GitHub actions workflow run for errors. You can also ping conda-forge/core for further assistance or try re-rendering locally.

This message was generated by GitHub actions workflow run https://github.com/conda-forge/mlx-feedstock/actions/runs/7283892070.

[ngam]

@xhochy green like heaven 🤑

[xhochy]

@conda-forge-admin please rerender

[xhochy]

Tests sadly fail locally on an M1 and an M3 with

import: 'mlx'
import: 'mlx'
+ pip check
No broken requirements found.
+ python -m unittest discover python/tests
libc++abi: terminating due to uncaught exception of type std::runtime_error: Failed to load device
/Users/uwe/mambaforge/conda-bld/mlx_1703158166871/test_tmp/run_test.sh: line 8: 53907 Abort trap: 6           python -m unittest discover python/tests

[ngam]

Interesting, the only other thing I can think of is what you suggested (the very latest SDK). Btw, here's the page with SDKs offered inside the vmImages: https://github.com/actions/runner-images/blob/main/images/macos/macos-13-Readme.md

Let's try SDK macosx14.2?

[ngam]

We can also play around with the settings following their note: https://ml-explore.github.io/mlx/build/html/install.html

Note

If you have multiple Xcode installations and wish to use a specific one while building, you can do so by adding the following environment variable before building export DEVELOPER_DIR="/path/to/Xcode.app/Contents/Developer/"

[github-actions]

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/mlx-feedstock/actions/runs/7312890249.

[ngam]

Current status: as is,

• ✅ builds and runs successfully without cf compilers (i.e., with apple's arm64 compilers) using both native and cross compiling (note that the compilers are arm64, so it is not really cross compiling). Also, python -m unittest discover python/tests passes
• ❌❌ does not run when built with conda-forge's clang compilers (not native and not cross compiling); that is, it builds fine, but doesn't actually work
• ❌ does not run when built with apple's x86_64 clang; that is, it builds fine, but doesn't actually work

For the latter two, the error is the same: libc++abi: terminating due to uncaught exception of type std::runtime_error: Failed to load device

[ngam]

Our options:

• keep trying to build in the conda-forge ecosystem (using cf compilers and cross compiling)
• upload packages built on arm64 manually

[xhochy]

@conda-forge-admin please rerender

[github-actions]

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/mlx-feedstock/actions/runs/7452400041.

[ngam]

@xhochy this now works; could you please test? 4772b90 (download artifacts and tested locally; no more device load issues)

We will delay until they have a new release, but we likely are going to hit another block with the gguf (ml-explore/mlx#350) or maybe you have ideas how to incorporate the lib correctly?

Anyway, please review when you have a moment (ignore the git_url, etc. for now)

[ngam]

the py312 build from 78df303:

(test_mlx) ~/mlx$ 
(test_mlx) ~/mlx$ python -m unittest discover python/tests
...............s............................................ssss...............................................................................................................................................................................
----------------------------------------------------------------------
Ran 234 tests in 9.561s

OK (skipped=5)
(test_mlx) ~/mlx$ 
(test_mlx) ~/mlx$ micromamba list | grep mlx 
List of packages in environment: "/Users/ngam/.micromamba/envs/test_mlx"
  mlx              0.0.9       py312hc9e2cf5_0       mlx-feedstock_conda_artifacts_20240111
(test_mlx) ~/mlx$ 

[xhochy]

Thank you and awesome work @ngam ! Also feel free to add you as a maintainer.