Nathan Goldbaum

I think originally the plan was to make a full set of multithreaded programming utilities available in the C API. However, if you look at the discussion around adding PyMutex...

> How is the ufunc cache used? Are entries ever removed or modified or are they only added? They're only ever added to the cache. It's in the hot path...

@robinvandernoord maybe it makes sense to split out the pyo3 update into its own PR? That would at least make the `main` branch buildable on Python 3.14.

You could also use a lock or some form of atomic reference counting. If another thread tries to read an object's state while it's being mutated you can then immediately...

Argh, sorry for commenting hastily. You're right, that's the wrong thing to point to. Probably better to point to this change I made to NumPy: https://github.com/numpy/numpy/pull/29394. See also [this test](https://github.com/numpy/numpy/blob/2640bec58c9072553e7381c999ce18a679da9ffe/numpy/_core/tests/test_multithreading.py#L309-L353)...

I tried setting up free-threaded wheels on my fork of llvmlite just now: https://github.com/ngoldbaum/llvmlite/pull/1 Currently the Mac builds fail because 3.14t isn't a valid python version to use when creating...

I think it should be possible to build PyLIEF from source without going through the pain of properly packaging it. I'm a little concerned about the nanobind leak warnings I...

I was able to get the mac builds to work by building PyLIEF from source, see https://github.com/ngoldbaum/llvmlite/actions/runs/18986468571?pr=1 and https://github.com/ngoldbaum/llvmlite/pull/1/files for the corresponding diff. It's not very fast: building PyLIEF is...

See https://github.com/numba/llvmlite/pull/1351 for an attempt at setting up cp314t wheel builds.