Joe Rowell

Fix SMEM index for C in CuTe examples

2

The static check against `CSmemLayout` in all of the CuTe sgemm examples checks against `size` twice. This has only passed thus far because `size(CSmemLayout{}) == size(CSmemLayout{});`, but it isn't strictly...

_Atomic use on C++ compilers can cause issues

2

I've noticed that cysignals uses _Atomic where it is available (c.f 80738037). In C land, this is fine. However, for C++ this is more questionable, and I've recently run into...

Should these loops iterate from 1?

2

It seems like these loops: https://github.com/emp-toolkit/emp-ag2pc/blob/19751229d79deccfcf5f8d3a97a15d3dee4b917a/emp-ag2pc/fpre.h#L314 https://github.com/emp-toolkit/emp-ag2pc/blob/19751229d79deccfcf5f8d3a97a15d3dee4b917a/emp-ag2pc/fpre.h#L327 Iterate starting from 1. Is this intentional? Given that the whole array is sent later, this means that an uninitialised value gets written,...

Setting ```__debug``` in ```amortized_2pc.h``` is broken

1

If one sets ```__debug``` in ```amortized_2pc.h``` then the code doesn't compile. This is because variables like ```value``` aren't defined in the class anymore.

[BUG] cutlass-python does not work on H100 for CUDA 11.8

5

**Describe the bug** The following sample program fails with the latest version of cutlass: ``` import cutlass import numpy as np plan = cutlass.op.Gemm(element=np.float16, layout=cutlass.LayoutType.RowMajor) A, B, C, D =...

[BUG] Cutlass' static library defines fewer symbols than the shared library

5

**Describe the bug** I've recently been playing around with building cutlass as both a shared and static library. However, I've noticed that the static library often omits some symbols that...