simple_fft_block_shared.cu

#include <stdio.h>
#include <iostream>
#include <vector>

#include <cuda_runtime_api.h>
#include <cufftdx.hpp>

#include "block_io.hpp"
#include "common.hpp"

template<class FFT>
__launch_bounds__(FFT::max_threads_per_block) __global__ void block_fft_kernel(typename FFT::value_type* data) {
    using complex_type = typename FFT::value_type;

    extern __shared__  unsigned char shared_mem[];

    auto this_block_data = data + cufftdx::size_of<FFT>::value * (FFT::ffts_per_block/FFT::implicit_type_batching) * blockIdx.x;

    example::io<FFT>::load_to_smem(this_block_data, shared_mem);

    FFT().execute(reinterpret_cast<void*>(shared_mem));

    example::io<FFT>::store_from_smem(shared_mem, this_block_data);
}

// In this example a one-dimensional complex-to-complex transform is performed by a CUDA block.
//
// One block is run, it calculates two 128-point C2C float precision FFTs.
// Data is generated on host, copied to device buffer, and then results are copied back to host.
template<unsigned int Arch>
void simple_block_fft() {
    using namespace cufftdx;

    // FFT is defined, its: size, type, direction, precision. Block() operator informs that FFT
    // will be executed on block level. Shared memory is required for co-operation between threads.
    // Additionally,
    using FFT          = decltype(Block() + Size<128>() + Type<fft_type::c2c>() + Direction<fft_direction::forward>() +
                         Precision<float>() + ElementsPerThread<8>() + FFTsPerBlock<2>() + SM<Arch>());
    using complex_type = typename FFT::value_type;

    // Allocate managed memory for input/output
    complex_type* data;
    auto          size       = FFT::ffts_per_block * cufftdx::size_of<FFT>::value;
    auto          size_bytes = size * sizeof(complex_type);
    CUDA_CHECK_AND_EXIT(cudaMallocManaged(&data, size_bytes));
    for (size_t i = 0; i < size; i++) {
        data[i] = complex_type {float(i), -float(i)};
    }

    // Shared memory must fit input data and must be big enough to run FFT
    auto shared_memory_size = std::max((unsigned int)FFT::shared_memory_size, (unsigned int)size_bytes);

    std::cout << "input [1st FFT]:\n";
    for (size_t i = 0; i < cufftdx::size_of<FFT>::value; i++) {
        std::cout << data[i].x << " " << data[i].y << std::endl;
    }

    // Increase max shared memory if needed
    CUDA_CHECK_AND_EXIT(cudaFuncSetAttribute(
        block_fft_kernel<FFT>,
        cudaFuncAttributeMaxDynamicSharedMemorySize,
        shared_memory_size));

    // Invokes kernel with FFT::block_dim threads in CUDA block
    block_fft_kernel<FFT><<<1, FFT::block_dim, shared_memory_size>>>(data);
    CUDA_CHECK_AND_EXIT(cudaPeekAtLastError());
    CUDA_CHECK_AND_EXIT(cudaDeviceSynchronize());

    std::cout << "output [1st FFT]:\n";
    for (size_t i = 0; i < cufftdx::size_of<FFT>::value; i++) {
        std::cout << data[i].x << " " << data[i].y << std::endl;
    }

    CUDA_CHECK_AND_EXIT(cudaFree(data));
    std::cout << "Success" << std::endl;
}

template<unsigned int Arch>
struct simple_block_fft_functor {
    void operator()() { return simple_block_fft<Arch>(); }
};

int main(int, char**) {
    return example::sm_runner<simple_block_fft_functor>();
}