Torrance Hodgson

(It's not obvious in the docs that examples/utils.jl is importing this.

At your suggestion, I've made the following change: ```diff diff --git a/src/fft.cu b/src/fft.cu index eeace96..26cd458 100644 --- a/src/fft.cu +++ b/src/fft.cu @@ -44,7 +44,7 @@ #include "ArrayIndexer.cuh" #include #include -#include +#include...

A little more investigation, and it turns out the entire callback that performs that fftshift isn't running. What's incredible is that if I add an empty print statement to `post_fftshift`...

> I'm also getting a 'cuda/stream.hpp(85): error: namespace "cuda::std" has no member "runtime_error"' error there as well Yes, I got that too and had to make it an absolute import....

@TorreZuk Thank you! HIPIFY complained there was no suitable equivalent and I clearly didn't spend long enough verifying that. If I can hijack my own issue (!), what about a...

Hi @amcamd > Can you say what application is using this CUDA_C_8I datatype? Real 8 bit integers are used in machine learning, what is the use case for complex 8...

> Does this work on CUDA? If so, I can take a look at how they do it and try to mirror their implementation. @jpsamaroo In fact you're right, my...

This might be related to this issue: https://github.com/ROCm/hipFFT/issues/91

Here's a patch if y'all AMD folks want to ensure this is fixed (?): ```diff diff --git a/include/hip/nvidia_detail/nvidia_hip_runtime_api.h b/include/hip/nvidia_detail/nvidia_hip_runtime_api.h index 0c492b7c..7dbfedd5 100644 --- a/include/hip/nvidia_detail/nvidia_hip_runtime_api.h +++ b/include/hip/nvidia_detail/nvidia_hip_runtime_api.h @@ -3465,7 +3473,7 @@...

Huh. I didn't even know there was a vertical split option. Had to consult the Help to work out how to do this. Thanks for the heads up!