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 #208

Closed regro-cf-autotick-bot closed 2 years ago

regro-cf-autotick-bot commented 2 years 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/autotick-bot/actions/runs/2442127476, please use this URL for debugging.

conda-forge-linter commented 2 years 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.

mkitti commented 2 years ago

We get better aarch64 support in 1.8. Stop.

ngam commented 2 years ago

We get better aarch64 support in 1.8. Stop.

This was automatically triggered. It gets rebased and recreated whenever the base branch changes. One way to stop it from doing this is to insert an empty commit, that stops the bot from recreating the PR.

Note, @mkitti, if you are interested in this, there are two ways to build the aarch64 and ppc64le variants in conda-forge. One is native build (on Travis CI like here) and one by emulation on Azure. Once the 1.8 release lands, we can discuss which path to take. Travis has a one-hour limit (though the CI is more powerful than Azure), Azure has a six-hour limit, though emulation makes it painfully slow...

mkitti commented 2 years ago

We might need to kill the testing for Travis then...

ngam commented 2 years ago

drafted a PR for rc1 going into the dev branch if you'd like to test stuff, #209

mkitti commented 2 years ago

As a side note, I do have a Mac Studio dev machine now, which I can use for testing purposes. I'm a bit busy at the moment, but I would be interested how I might be able to improve the situation on the M1.

ngam commented 2 years ago

As a side note, I do have a Mac Studio dev machine now, which I can use for testing purposes. I'm a bit busy at the moment, but I would be interested how I might be able to improve the situation on the M1.

Good news! I hope you're enjoying it :)

There are, broadly speaking, two things one can do:

  1. If you have Rosetta installed, you can effectively have a(n almost fully) functional x86_64 conda installation that acts like it is an osx-64 machine. This way, one could do the emulation we have here on the CI and see how far you can get --- but this is essentially the same as everything we can do here, so not sure how useful this option is...
  2. We can switch to native building of osx-arm64. This would mean we have to get some core member to copy the artifacts from a personal anaconda channel to the conda-forge's main channel. We can discuss this more later. Once you have time, and only after v1.8 comes out, we can try to tinker with this. For the native osx-arm64 build, I would guess the only problem is the dependencies (e.g. openblas-ilp64); otherwise, the build should be straightforward.

You're aware that they no longer ship the 1.7.x for M1, right? There seem to be some more problems to tighten. But I think it is basically 99% functional or so. Also, the homebrew people build natively on M1s these days (the rc)...

ngam commented 2 years ago

(I got very curious about Julia because I wanted to use it for a package I was developing --- but I kind of got nowhere; I still would like to make it happen, but my focus is on a Python-centric implementation for now. I also eventually would like to try implementing some of my research in Julia, but then again... I would need to find some room to do that between the constant demands of academic research...)

(I've been lurking around the discourse website for a while trying to get enough inspiration :) )

regro-cf-autotick-bot commented 2 years ago

I see that this PR has conflicts, and I'm the only committer. I'm going to close this PR and will make another one as appropriate. This was generated by https://github.com/regro/autotick-bot/actions/runs/2913090766