[libcxx-commits] [libcxx] [libcxx] Use `aligned_alloc` for testing instead of `posix_memalign` (PR #101748)

[Louis Dionne via libcxx-commits]

https://github.com/ldionne commented:

I like this patch, I think it makes sense to try to rely on standard facilities more often. This makes it easier to run the test suite on alternative platforms by removing the (often incorrect) assumption that they provide `posix_memalign`.

I have a comment about rounding without which I think the patch is incorrect. But with that comment addressed, I'd be happy with the change.

https://github.com/llvm/llvm-project/pull/101748