GROMACS host requirement is breaking macOS CI

[lohedges]

I'm not 100% sure if this is the cause, but I'm seeing repeated failures in the environment setup stage for my sire-emle build on macOS. In the log I see:

/Users/runner/miniconda3/envs/sire_build/conda-bld/sire_1714379289268/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehol/etc/conda/activate.d/gromacs_activate.sh: line 22: /Users/runner/miniconda3/envs/sire_build/conda-bld/sire_1714379289268/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehol/bin.SSE2/GMXRC: No such file or directory
Traceback (most recent call last):
source tree in: /Users/runner/miniconda3/envs/sire_build/conda-bld/sire_1714379289268/work
  File "/Users/runner/miniconda3/envs/sire_build/bin/conda-mambabuild", line 10, in <module>
export PREFIX=/Users/runner/miniconda3/envs/sire_build/conda-bld/sire_1714379289268/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehol
export BUILD_PREFIX=/Users/runner/miniconda3/envs/sire_build/conda-bld/sire_1714379289268/_build_env
export SRC_DIR=/Users/runner/miniconda3/envs/sire_build/conda-bld/sire_1714379289268/work
    sys.exit(main())
  File "/Users/runner/miniconda3/envs/sire_build/lib/python3.10/site-packages/boa/cli/mambabuild.py", line 301, in main
    call_conda_build(action, config)
  File "/Users/runner/miniconda3/envs/sire_build/lib/python3.10/site-packages/boa/cli/mambabuild.py", line 273, in call_conda_build
    result = api.build(
  File "/Users/runner/miniconda3/envs/sire_build/lib/python3.10/site-packages/conda_build/api.py", line 250, in build
    return build_tree(
  File "/Users/runner/miniconda3/envs/sire_build/lib/python3.10/site-packages/conda_build/build.py", line 3762, in build_tree
    packages_from_this = build(
  File "/Users/runner/miniconda3/envs/sire_build/lib/python3.10/site-packages/conda_build/build.py", line 2634, in build
    utils.check_call_env(
  File "/Users/runner/miniconda3/envs/sire_build/lib/python3.10/site-packages/conda_build/utils.py", line 408, in check_call_env
    return _func_defaulting_env_to_os_environ("call", *popenargs, **kwargs)
  File "/Users/runner/miniconda3/envs/sire_build/lib/python3.10/site-packages/conda_build/utils.py", line 384, in _func_defaulting_env_to_os_environ
    raise subprocess.CalledProcessError(proc.returncode, _args)
subprocess.CalledProcessError: Command '['/bin/bash', '-o', 'errexit', '/Users/runner/miniconda3/envs/sire_build/conda-bld/sire_1714379289268/work/conda_build.sh']' returned non-zero exit status 1.

In particular, not the:

... /bin.SSE2/GMXRC: No such file or directory

The versions of GROMACS barely differs to the last working build:

works:

    gromacs:                       2023.4-nompi_h9dceb8e_102            conda-forge

fails:

    gromacs:                       2023.4-nompi_hf1d43a3_102              conda-forge

I've removed gromacs from the requirements and it works completely fine. I'm really puzzled by this since there haven't been any recent merges on their feedstock. (Last was March 28th.)

[lohedges]

Okay, this is because we are now using macOS ARM runners. According to the table the only intel-based macOS-14 runner is macos-14-large. @chryswoods: We should probably update our CI accordingly.

[chryswoods]

Cool - I knew the arm runners were coming but hadn't realised they are now here. I'll update the workflows once I get some time (likely later this week)

[lohedges]

One nice thing is that the macOS arm64 runners are really fast. It takes just shy of an hour for a build.

[lohedges]

Closing since the problematic GROMACS conda package has been rebuilt.