conda-forge / julia-feedstock

A conda-smithy repository for julia.
BSD 3-Clause "New" or "Revised" License
23 stars 32 forks source link

Arch Migrator #262

Open regro-cf-autotick-bot opened 11 months ago

regro-cf-autotick-bot commented 11 months ago

This feedstock is being rebuilt as part of the aarch64/ppc64le migration.

Feel free to merge the PR if CI is all green, but please don't close it without reaching out the the ARM migrators first at @conda-forge/arm-arch.

If this PR was opened in error or needs to be updated please add the bot-rerun label to this PR. The bot will close this PR and schedule another one. If you do not have permissions to add this label, you can use the phrase code>@<space/conda-forge-admin, please rerun bot in a PR comment to have the conda-forge-admin add it for you.

This PR was created by the regro-cf-autotick-bot. The regro-cf-autotick-bot is a service to automatically track the dependency graph, migrate packages, and propose package version updates for conda-forge. Feel free to drop us a line if there are any issues! This PR was generated by https://github.com/regro/cf-scripts/actions/runs/6806434710, please use this URL for debugging.

conda-forge-webservices[bot] commented 11 months ago

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.

mfansler commented 11 months ago

This seems pertinent, RE: Travis CI jobs aren't being generated here (they should be): Gitter Thread

In case the link is a hassle, here's the content:

FYI if your feedstock uses Travis CI and you are not seeing Travis CI builds, please make sure the commit author (yourself or another contributor) does a relogin to Travis CI and then restart CI. This should remedy the issue

Please reach out in the thread if you have questions or issues. Thank you! 🙏

ngam commented 11 months ago

I will look into it later today. I think there may be issues beyond just compiling natively, but we shall see...

ngam commented 11 months ago
_maybe_raise_for_conda_build
        raise exc
    conda_libmamba_solver.conda_build_exceptions.ExplainedDependencyNeedsBuildingError: Unsatisfiable dependencies for platform linux-aarch64: {MatchSpec("openblas-ilp64")}
    Encountered problems while solving:
      - nothing provides requested openblas-ilp64

    Could not solve for environment specs
ngam commented 11 months ago

Seems like we have two distinct issues now:

  1. Tests failing and testing taking more resources --- we could likely fix this by adjusting our tests on all platforms
  2. Missing openblas-ilp64 on aarch64 --- I believe we discussed this before and I don't remember what happened

I can probably push this forward, but my general use of Julia is close to zero these days and I am likely out of the loop on what's happened in julialand.

Hi @mfansler, what exactly is your need (aarch or ppc)? Are you interested in helping to get these issues resolved?

mfansler commented 11 months ago

@schlichtanders would be the one to ask - he has the interest in Arch builds. I was mostly just querying about the status of the PR to communicate this to others.

schlichtanders commented 11 months ago

I am running stuff on Arm64 with CondaPkg and wanted to use r-juliacall. That depends on julia and because of this it is not installable on Arm64 out of the box. I now created a workaround for myself - a custom build of r-juliacall. But in general, it would be great if julia-dependent conda packages would just install everywhere out of the box.

mkitti commented 11 months ago

We have a few choices here:

  1. Work on Julia dependencies in the conda-forge ecosystem to better accomodate Julia's needs
  2. Compromise conda-forge principles and download binaries built elsewhere (e.g. juliaup)
  3. Create a distinct julia conda channel that just download binaries built elsewhere
  4. Don't use conda to obtain julia
schlichtanders commented 11 months ago

For me all options are fine (my motivation is very practical - being able to install the conda package r-juliacall out of the box.)