Graham Markall
Graham Markall
Thanks for the PR! I need to look over the code more closely, but I think the idea of these is good in general.
I've added this to the next milestone since it seems like it should be pretty easy to get in before the next release.
Many thanks for your contribution, @thomaspinckney3! This will be included in the next release of llvmlite (v0.44.0).
Similar error rendering numba-feedstock with NumPy 2 and the llvmlite_rc channel (at commit https://github.com/gmarkall/numba-feedstock/commit/fb24c5327608aa16f46ee48e13dd005310c5ab49): ``` $ conda-smithy rerender INFO:conda_smithy.configure_feedstock:__pycache__ rendering is skipped INFO:conda_smithy.configure_feedstock:README rendering is skipped INFO:conda_smithy.configure_feedstock:python312.yaml from feedstock is...
This seems very similar to #9938. @yojeep, if you use `parallel=True` on the outermost function decorator, does it provide a workaround for you?
Glad to hear the workaround works. The issue should be resolved properly in Numba with PR #9950.
I'm struggling to reproduce this with 0.61.0 - can you provide the output of `numba -s` please? Perhaps it's related to the Python version as well, and knowing your complete...
Thanks, I can hit the issue with Python 3.13 and Numba 0.61.0.
A couple of items from the triage meeting: * We should check for this occurring in literal_unroll and raise an exception * We should check / update the documentation to...
Accidentally closed the wrong PR here.