Axel Feldmann

Edge Partitioning From The Interface

6

Hi! I'm trying to understand how I would do edge partitioning from the C library interface. I'm very happy with the results I get from running the `edge_partitioning` executable, and...

Previously, `clearInputs()` is declared as: `bool clearInputs();` in `include/element.h`. However, in its definition in `src/element.cpp`, it does not actually have a return statement. This is undefined behavior and causes crashes...

[QST] Deadlock in producer consumer loop

7

Hi! I'm trying to understand how to use the PipelineTmaAsync utility, so I wrote a simple program where one warp group loads tiles and another one does nothing with them...

[BUG] segfault with `cute.gemm`

13

Hi! I'm trying to use `cute.gemm` with wgmma, and I'm getting a segfault that I don't quite know how to debug. I have the following simple code: ``` import torch...

[BUG] Illegal instruction on H100 TMA

6

Hi, I am trying to do a very simple TMA load on H100 and my code is producing an illegal instruction exception. Small reproducer: ``` import torch import cutlass import...