Highly Efficient FFT for Exascale: HeFFTe v2.3
|
Implementation of Cosine Transform pre-post processing methods using CUDA. More...
#include <heffte_backend_oneapi.h>
Static Public Member Functions | |
template<typename precision > | |
static void | pre_forward (sycl::queue &, int length, precision const input[], precision fft_signal[]) |
Pre-process in the forward transform. | |
template<typename precision > | |
static void | post_forward (sycl::queue &, int length, std::complex< precision > const fft_result[], precision result[]) |
Post-process in the forward transform. | |
template<typename precision > | |
static void | pre_backward (sycl::queue &, int length, precision const input[], std::complex< precision > fft_signal[]) |
Pre-process in the inverse transform. | |
template<typename precision > | |
static void | post_backward (sycl::queue &, int length, precision const fft_result[], precision result[]) |
Post-process in the inverse transform. | |
Implementation of Cosine Transform pre-post processing methods using CUDA.