Graham Markall comments

Results 776 comments of


                                            Graham Markall

numba with @jit silently handles float-to-integer conversion in slice indices

Many thanks for the report - I can reproduce this behaviour.

No implementation for `bool(OptionalType(int64))`

Note from triage meeting: https://github.com/numba/numba/issues/9977 may fix the "`types.Function` thing" edited: linked to this issue itself by mistake before.

`raise` and `assert` do not work correctly in CUDA device functions

Additionally, `assert` has the exact same problem.

[CI TEST] Test with Opaque Pointers

gpuci run tests

[CI TEST] Test with Opaque Pointers

gpuci run tests

[CI TEST] Test with Opaque Pointers

gpuci run tests

Migrate gpuCI to RAPIDS Ops Github Actions + self-hosted runners

Closing this as gpuCI is gone, and https://github.com/NVIDIA/numba-cuda uses the NVIDIA GHA runners.

Add support for opaque pointers

I note that this works with LLVM 15 but not 14. Given the imminent move to LLVM 15, I think it's not worth fixing for 14.

Add support for opaque pointers

Note from triage discussion - for function types, round tripping the type of arguments should still be able to work (correct me if this is wrong, @sklam)

Add support for opaque pointers

@rj-jesus Thanks for this, I think it looks good - I'll update the CI configuration and docs.