V7.3.14

[mattip]

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.

Build for newly released PyPy 7.3.14

[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.

[h-vetinari]

Look what santa brought! :)

On OSX:

memprotect.c:16:70: error: use of undeclared identifier 'MAP_ANONYMOUS'
    new_ptr = mmap(NULL, size, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
                                                                     ^
1 error generated.

For that, you'll need to either bump the deployment target, or do something like https://github.com/conda-forge/abseil-cpp-feedstock/blob/main/recipe/patches/0003-add-missing-osx-workaround.patch.

[mattip]

I wonder why we don't see this on the buildbot. As far as I can tell (using xcodebuild -version) the xcode is up to date there, but clang --version shows Apple clang version 13.1.6 where in the failed build I see x86_64-apple-darwin13.4.0-clang -O3 -fomit-frame-pointer -mmacosx-version-min=10.9 -arch x86_64 .... Confusing.

As for what to do about the failure, what is the conda-forge standard for -mmacosx-version-min ? I think CPython is still on 10.5, so I wouldn't want to get too far from that. I guess I will add the patch in your link.

[h-vetinari]

As for what to do about the failure, what is the conda-forge standard for -mmacosx-version-min?

Currently 10.9, but will soon™ switch 10.13 (pending infrastructure work).

I think CPython is still on 10.5, so I wouldn't want to get too far from that.

As described in that issue, I personally think it's utterly pointless to spend dev time on anything below 10.13. Several foundational projects (chrome, libcxx, +/- all google projects etc.) started dropping support for 10.12 and below, aside from the fact that it has been EOL'd by Apple for over 4 years already.

To the degree that numbers are available, usage numbers¹ support this (at least IMO), not least because Apple pushes people to upgrade aggressively.

Footnotes

1. modulo the fact that every browser fakes the Macos version past 10.15 in the user agent string, rendering web-based stats after that effectively useless ↩

[mattip]

The error is again in darwin:

/Users/runner/miniforge3/conda-bld/pypy3.9_1703500807357/work/pypy3/rpython/../pypy/module/_hpy_universal/_vendored/hpy/trace/src/trace_internal.h:64:12: \
    error: call to undeclared function 'clock_gettime'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
    return clock_gettime(CLOCK_MONOTONIC_RAW, t);
           ^

but man clock_gettime on macos says it should be available from time.h, and #include <time.h> is before this. What is going on?

[mattip]

Ahh, the man page also says

HISTORY
     These functions first appeared in Mac OSX 10.12

So I guess the easiest path will be to update the minimum macos version to 10.13

[mattip]

@h-vetinari the hint about conda_build_config.yaml seems to have done the magic. CI is passing. Would anyone care to review?

[h-vetinari]

LGTM!

[mattip]

Hmm. Maybe we should wait for a resolution of pypy/pypy#4816