Siu Kwan Lam

@seanlaw, seeing py-spy has re-ignited my interest in getting backtraces from Numba. I have a side-project that uses libunwind (so no root-privilege requirement) to get stack frames: https://github.com/numba/stacktrace. I just...

> Is there value to designing a new library that wrapped around this one and provided the same diagnostics but in a different way? If so, what thoughts do people...

Are you planning to implement it into nbconvert?

I am not really spending time on this project at the moment. So, if anyone would like to work on this or take over this project, let me know.

Thanks @rocky for the detailed insights. > The control-flow to higher-level construct used in numba-rvsdg, while it is more conventional and is general purpose, does not make use of these...

I have a python bytecode based reproducer for this: ```python from numba_rvsdg.core.datastructures.byte_flow import ByteFlow from numba_rvsdg.rendering.rendering import ByteFlowRenderer def foo(x, y, z): if x: while z: print("A") z = 0...

The refactoring is ok but i'm seeing miscompilation in testing [#9012](https://github.com/numba/numba/pull/9012) once i switch to this patch. I will test again once this PR is stable.

I am changing the mockasm code so no branches have the same jump target on both side. However, the `assert len(diff) == 1` can trip on group like: YAML: ```yaml...

Tracking: https://github.com/numba/numba-rvsdg/blob/371b3dfa23d72ad93de1a60cced38ce8c45b1141/numba_rvsdg/tests/test_mock_asm.py#L270-L273

Tracking https://github.com/numba/numba-rvsdg/blob/371b3dfa23d72ad93de1a60cced38ce8c45b1141/numba_rvsdg/tests/test_mock_asm.py#L264-L268