multi_reduce_core.cuh File Reference

#include <color_spinor_field_order.h>
#include <blas_helper.cuh>
#include <cub_helper.cuh>
#include <texture.h>

Include dependency graph for multi_reduce_core.cuh:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Classes
struct	quda::blas::MultiReduceArg< NXZ, ReduceType, SpinorX, SpinorY, SpinorZ, SpinorW, Reducer >
	Parameter struct for generic multi-blas kernel. More...
struct	quda::blas::coeff_array< T >
struct	quda::blas::MultiReduceFunctor< NXZ, ReduceType, Float2, FloatN >
struct	quda::blas::Dot< NXZ, ReduceType, Float2, FloatN >
struct	quda::blas::Cdot< NXZ, ReduceType, Float2, FloatN >
struct	quda::blas::CdotCopy< NXZ, ReduceType, Float2, FloatN >

Namespaces
	quda
	quda::blas

Macros
#define	BLAS_SPINOR
#define	MAX_MATRIX_SIZE   4096

Functions
template<int block_size, typename ReduceType , typename FloatN , int M, int NXZ, typename Arg >
__global__ void	quda::blas::multiReduceKernel (Arg arg_)
template<typename ReduceType >
__device__ __host__ void	quda::blas::dot_ (ReduceType &sum, const double2 &a, const double2 &b)
template<typename ReduceType >
__device__ __host__ void	quda::blas::dot_ (ReduceType &sum, const float2 &a, const float2 &b)
template<typename ReduceType >
__device__ __host__ void	quda::blas::dot_ (ReduceType &sum, const float4 &a, const float4 &b)
template<typename ReduceType >
__device__ __host__ void	quda::blas::cdot_ (ReduceType &sum, const double2 &a, const double2 &b)
template<typename ReduceType >
__device__ __host__ void	quda::blas::cdot_ (ReduceType &sum, const float2 &a, const float2 &b)
template<typename ReduceType >
__device__ __host__ void	quda::blas::cdot_ (ReduceType &sum, const float4 &a, const float4 &b)

Variables
static __constant__ signed char	quda::blas::Amatrix_d [MAX_MATRIX_SIZE]
static __constant__ signed char	quda::blas::Bmatrix_d [MAX_MATRIX_SIZE]
static __constant__ signed char	quda::blas::Cmatrix_d [MAX_MATRIX_SIZE]
static signed char *	quda::blas::Amatrix_h
static signed char *	quda::blas::Bmatrix_h
static signed char *	quda::blas::Cmatrix_h
static __constant__ signed char	quda::blas::arg_buffer [MAX_MATRIX_SIZE]

Macro Definition Documentation

BLAS_SPINOR

#define BLAS_SPINOR

Definition at line 15 of file multi_reduce_core.cuh.

MAX_MATRIX_SIZE

#define MAX_MATRIX_SIZE   4096

Definition at line 19 of file multi_reduce_core.cuh.