build: add aarch64 and ppc64le

[lgray]

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-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.

[lgray]

@conda-forge-admin, please rerender

[github-actions]

Hi! This is the friendly conda-forge automerge bot!

I considered the following status checks when analyzing this PR:

• linter: passed
• travis: failed
• azure: passed

Thus the PR was not passing and not merged.

[lgray]

@matthewfeickert it seems to only be failing for some combinations, please have a look!

[matthewfeickert]

Oh man. Travis CI error logs. A blast from the past that's giving me some mental and emotional damage. 😬

The build matrix failures are confusing too as they seem to only be CPython version based. :?

I'll have to dig through and also learn what

"if [[ ${PLATFORM} =~ .*linux.* ]]; then CONDA_FORGE_DOCKER_RUN_ARGS="--network=host --security-opt=seccomp=unconfined" ./.scripts/run_docker_build.sh; fi" exited with 137.

is all about.

[lgray]

@conda-forge-admin, please rerender

[github-actions]

Hi! This is the friendly conda-forge automerge bot!

Commits were made to this PR after the automerge label was added. For security reasons, I have disabled automerge by removing the automerge label. Please add the automerge label again (or ask a maintainer to do so) if you'd like to enable automerge again!

[lgray]

@matthewfeickert may have found an easier way through - just use azure instead ^_^

[lgray]

While this might take a while since it's emulated, I imagine it is far better than trying to setup this build system for cross compilation.

[matthewfeickert]

While this might take a while since it's emulated, I imagine it is far better than trying to setup this build system for cross compilation.

@lgray yeah if we can avoid using Travis CI I would be very happy. I assume there's no way to use Cirrus CI as a provider? If not, as long as this stays under 2 hours of build time I think that's okay (we can also turn things off as needed if we need to debug in a PR).

[matthewfeickert]

@lgray can you link to the docs that you read that cover providers? I'm trying to assemeble some conda-forge bookmarks for myself for the future. 😬

[lgray]

https://conda-forge.org/docs/maintainer/conda_forge_yml/#provider

[lgray]

Huh, the builds in azure keep getting to:

libtool: compile:  $BUILD_PREFIX/bin/aarch64-conda-linux-gnu-c++ -DHAVE_CONFIG_H -I. -I../include/fastjet -DNDEBUG -D_FORTIFY_SOURCE=2 -O2 -isystem $PREFIX/include -g -Woverloaded-virtual -frounding-math -I./../include -fvisibility-inlines-hidden -fmessage-length=0 -ftree-vectorize -fPIC -fstack-protector-strong -fno-plt -O3 -pipe -isystem $PREFIX/include -fdebug-prefix-map=$SRC_DIR=/usr/local/src/conda/fastjet-3.4.0.6 -fdebug-prefix-map=$PREFIX=/usr/local/src/conda-prefix -O3 -Bstatic -Bdynamic -std=c++17 -MT libfastjet_la-RangeDefinition.lo -MD -MP -MF .deps/libfastjet_la-RangeDefinition.Tpo -c RangeDefinition.cc -o libfastjet_la-RangeDefinition.o >/dev/null 2>&1

and then completing wreck the node they're on for some reason.

it is consistently that file on every single build.

[lgray]

@conda-forge-admin, please rerender

[github-actions]

Hi! This is the friendly conda-forge automerge bot!

Commits were made to this PR after the automerge label was added. For security reasons, I have disabled automerge by removing the automerge label. Please add the automerge label again (or ask a maintainer to do so) if you'd like to enable automerge again!

[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/fastjet-feedstock/actions/runs/8024264721.

[lgray]

@conda-forge-admin, please rerender

[matthewfeickert]

Could also try using Circle CI: https://circleci.com/docs/using-arm/

edit: You already did that. Never mind.

But it is strange that it keeps failing at that one spot.

[lgray]

@conda-forge-admin, please update circle.

[matthewfeickert]

@lgray have you actually succeeded in building the fastjet wheel on aarch_64 Linux machine before? If not, I can try to build on an RPi and see where things break down.

[lgray]

Oh it works on cirrus no problem and we use it to build wheels quickly. The builds were completing on travis in some cases just fine as well. It really seems to be a worker memory issue in some versions of python.

ppc64le was also succeeding for certain python versions, just not all of them.

[lgray]

@conda-forge-admin, please add circle

[matthewfeickert]

Oh it works on cirrus no problem and we use it to build wheels quickly

Oh right. I forgot that I added the PR for that. scikit-hep/fastjet#168 🤪 Time to step away from the keyboard and make a coffee.

Thanks for steering this and getting these builds going in the right direction.

[lgray]

@chrisburr does conda-forge no longer actually use circle-ci or do I need to go through some extra hoops I don't know about?

[lgray]

I have a feeling I need to first merge this and then it'll work.

[lgray]

I'm gonna yolo this one for now. It won't screw up the x86 builds that are already there.