Siu Kwan Lam comments

Results 276 comments of


                                            Siu Kwan Lam

Migrate random `glue_lowering` to `overload` where easy

/azp run

Ambiguous overloads are allowed for calls to jitted functions inside other jitted functions

the problem seems to be at https://github.com/numba/numba/blob/main/numba/core/typing/context.py#L620 and `allows_ambiguous` is `True` when resolving `add()` in `add_caller()`.

Ambiguous overloads are allowed for calls to jitted functions inside other jitted functions

turning off the ambiguous overload leads to error like: ``` TypeError: Ambiguous overloading for (uint64, int64): (float64, float64) -> bool (complex128, complex128) -> bool ``` in the testsuite. So, `eq(uin64,...

Function compilation not being updated for specific types on 32-bit systems

Note: - check typeof logic in dispatching. seems like it's always picking intp.

Export NRT_MemInfo_new

`nrt_external.h` from https://github.com/numba/numba/pull/4554 should satisfy the usecase for this.

Export NRT_MemInfo_new

Is this still needed?

Drop pre-3.7 code path (CPU only)

> @sklam is this intended for 0.57? Sure. We already dropped py

Drop pre-3.7 code path (CPU only)

> This is comment, but it seems like it is a "dead comment" now: > > https://github.com/numba/numba/blob/main/numba/tests/test_withlifting.py#L1109-L1110 changed in https://github.com/numba/numba/pull/7920/commits/d2d76a663bcbb7d0df4eb0b81176f425362ee2ff > > This could be made more precise by using...

Drop pre-3.7 code path (CPU only)

@esc, can you approve this PR please?

CFuncs cannot use numba.carray when jitting is disabled

There are two different ideas here: 1. `cfunc` doesn't respect `NUMBA_DISABLE_JIT=1`. That's something we can fix by wrapping the function with `ctypes.CFUNCTYPE` so the interpreted python function is exposed as...