work_aggregation_cuda_triad.cpp File Reference

#include <chrono>
#include <hpx/futures/future.hpp>
#include <hpx/async_cuda/cuda_executor.hpp>
#include <boost/program_options.hpp>
#include "cppuddle/memory_recycling/cuda_recycling_allocators.hpp"
#include "cppuddle/memory_recycling/util/cuda_recycling_device_buffer.hpp"
#include "cppuddle/kernel_aggregation/kernel_aggregation_interface.hpp"

Include dependency graph for work_aggregation_cuda_triad.cpp:

Functions
template<typename float_t>
__global__ void	__launch_bounds__ (1024, 2) triad_kernel(float_t *A
int	hpx_main (int argc, char *argv[])
int	main (int argc, char *argv[])

Variables
__global__ void const float_t *	B
__global__ void const float_t const float_t *	C
__global__ void const float_t const float_t const float_t	scalar
__global__ void const float_t const float_t const float_t const size_t	start_id
__global__ void const float_t const float_t const float_t const size_t const size_t	kernel_size
__global__ void const float_t const float_t const float_t const size_t const size_t const size_t	problem_size
	A [i] = B[i] + scalar * C[i]

Function Documentation

__launch_bounds__()

template<typename float_t>

__global__ void __launch_bounds__	(	1024	,
		2	)

hpx_main()

int hpx_main	(	int	argc,
		char *	argv[] )

main()

int main	(	int	argc,
		char *	argv[] )

Variable Documentation

A

A[i] = B[i] + scalar * C[i]

B

__global__ void const float_t* B

C

__global__ void const float_t const float_t* C

kernel_size

__global__ void const float_t const float_t const float_t const size_t const size_t kernel_size

problem_size

__global__ void const float_t const float_t const float_t const size_t const size_t const size_t problem_size

Initial value:

{
  const size_t i = start_id + blockIdx.x * blockDim.x + threadIdx.x

scalar

__global__ void const float_t const float_t const float_t scalar

start_id

__global__ void const float_t const float_t const float_t const size_t start_id