LinearAlgebra::CUDAWrappers::kernel Namespace Reference

Classes
struct	Binop_Addition
struct	Binop_Max
struct	Binop_Min
struct	Binop_Subtraction
struct	DotProduct
struct	ElemSum
struct	L1Norm
struct	LInfty

Functions
template<typename Number >
__global__ void	vec_scale (Number *val, const Number a, const size_type N)
template<typename Number , template< typename > class Binop>
__global__ void	vector_bin_op (Number *v1, const Number *v2, const size_type N)
template<typename Number , template< typename > class Binop>
__global__ void	masked_vector_bin_op (const unsigned int *mask, Number *v1, const Number *v2, const size_type N)
template<typename Number , typename Operation >
__global__ void	reduction (Number *result, const Number *v, const size_type N)
template<typename Number , typename Operation >
__global__ void	double_vector_reduction (Number *result, const Number *v1, const Number *v2, const size_type N)
template<typename Number >
__global__ void	vec_add (Number *val, const Number a, const size_type N)
template<typename Number >
__global__ void	add_aV (Number *val, const Number a, const Number *V_val, const size_type N)
template<typename Number >
__global__ void	add_aVbW (Number *val, const Number a, const Number *V_val, const Number b, const Number *W_val, const size_type N)
template<typename Number >
__global__ void	sadd (const Number s, Number *val, const Number a, const Number *V_val, const size_type N)
template<typename Number >
__global__ void	sadd (const Number s, Number *val, const Number a, const Number *V_val, const Number b, const Number *W_val, const size_type N)
template<typename Number >
__global__ void	scale (Number *val, const Number *V_val, const size_type N)
template<typename Number >
__global__ void	equ (Number *val, const Number a, const Number *V_val, const size_type N)
template<typename Number >
__global__ void	equ (Number *val, const Number a, const Number *V_val, const Number b, const Number *W_val, const size_type N)
template<typename Number >
__global__ void	add_and_dot (Number *res, Number *v1, const Number *v2, const Number *v3, const Number a, const size_type N)
template<typename Number >
__global__ void	set (Number *val, const Number s, const size_type N)
template<typename Number , typename IndexType >
__global__ void	set_permutated (const IndexType *indices, Number *val, const Number *v, const IndexType N)
template<typename Number , typename IndexType >
__global__ void	gather (Number *val, const IndexType *indices, const Number *v, const IndexType N)
template<typename Number >
__global__ void	add_permutated (const size_type *indices, Number *val, const Number *v, const size_type N)

Detailed Description

Namespace containing the CUDA kernels.