ViennaCL - The Vienna Computing Library  1.5.1
Namespaces | Functions
viennacl/linalg/cuda/matrix_operations_row.hpp File Reference

Implementations of row-major dense matrix related operations, including matrix-vector products, using CUDA. More...

Go to the source code of this file.

Namespaces

namespace  viennacl
 

Main namespace in ViennaCL. Holds all the basic types such as vector, matrix, etc. and defines operations upon them.


namespace  viennacl::linalg
 

Provides all linear algebra operations which are not covered by operator overloads.


namespace  viennacl::linalg::cuda
 

Holds all CUDA compute kernels used by ViennaCL.


Functions

template<typename T >
__global__ void am_row_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, T fac2, unsigned int options2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void am_row_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *fac2, unsigned int options2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void ambm_row_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, T fac2, unsigned int options2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2, T fac3, unsigned int options3, const T *C, unsigned int C_start1, unsigned int C_start2, unsigned int C_inc1, unsigned int C_inc2, unsigned int C_internal_size1, unsigned int C_internal_size2)
template<typename T >
__global__ void ambm_row_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, T fac2, unsigned int options2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2, const T *fac3, unsigned int options3, const T *C, unsigned int C_start1, unsigned int C_start2, unsigned int C_inc1, unsigned int C_inc2, unsigned int C_internal_size1, unsigned int C_internal_size2)
template<typename T >
__global__ void ambm_row_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *fac2, unsigned int options2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2, T fac3, unsigned int options3, const T *C, unsigned int C_start1, unsigned int C_start2, unsigned int C_inc1, unsigned int C_inc2, unsigned int C_internal_size1, unsigned int C_internal_size2)
template<typename T >
__global__ void ambm_row_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *fac2, unsigned int options2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2, const T *fac3, unsigned int options3, const T *C, unsigned int C_start1, unsigned int C_start2, unsigned int C_inc1, unsigned int C_inc2, unsigned int C_internal_size1, unsigned int C_internal_size2)
template<typename T >
__global__ void ambm_m_row_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, T fac2, unsigned int options2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2, T fac3, unsigned int options3, const T *C, unsigned int C_start1, unsigned int C_start2, unsigned int C_inc1, unsigned int C_inc2, unsigned int C_internal_size1, unsigned int C_internal_size2)
template<typename T >
__global__ void ambm_m_row_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, T fac2, unsigned int options2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2, const T *fac3, unsigned int options3, const T *C, unsigned int C_start1, unsigned int C_start2, unsigned int C_inc1, unsigned int C_inc2, unsigned int C_internal_size1, unsigned int C_internal_size2)
template<typename T >
__global__ void ambm_m_row_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *fac2, unsigned int options2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2, T fac3, unsigned int options3, const T *C, unsigned int C_start1, unsigned int C_start2, unsigned int C_inc1, unsigned int C_inc2, unsigned int C_internal_size1, unsigned int C_internal_size2)
template<typename T >
__global__ void ambm_m_row_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *fac2, unsigned int options2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2, const T *fac3, unsigned int options3, const T *C, unsigned int C_start1, unsigned int C_start2, unsigned int C_inc1, unsigned int C_inc2, unsigned int C_internal_size1, unsigned int C_internal_size2)
template<typename T >
__global__ void matrix_row_assign_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, T alpha)
template<typename T >
__global__ void matrix_row_diagonal_assign_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, T alpha)
template<typename T >
__global__ void element_op_row_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2, const T *C, unsigned int C_start1, unsigned int C_start2, unsigned int C_inc1, unsigned int C_inc2, unsigned int C_internal_size1, unsigned int C_internal_size2, unsigned int op_type)
template<typename T >
__global__ void element_op_int_row_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2, const T *C, unsigned int C_start1, unsigned int C_start2, unsigned int C_inc1, unsigned int C_inc2, unsigned int C_internal_size1, unsigned int C_internal_size2, unsigned int op_type)
template<typename T >
__global__ void matrix_row_element_abs_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_acos_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_asin_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_atan_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_ceil_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_cos_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_cosh_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_exp_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_fabs_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_floor_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_log_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_log10_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_sin_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_sinh_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_sqrt_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_tan_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void matrix_row_element_tanh_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *B, unsigned int B_start1, unsigned int B_start2, unsigned int B_inc1, unsigned int B_inc2, unsigned int B_internal_size1, unsigned int B_internal_size2)
template<typename T >
__global__ void vec_mul_row_kernel (const T *A, unsigned int A_row_start, unsigned int A_col_start, unsigned int A_row_inc, unsigned int A_col_inc, unsigned int A_row_size, unsigned int A_col_size, unsigned int A_internal_rows, unsigned int A_internal_cols, const T *v, unsigned int v_start, unsigned int v_inc, unsigned int v_size, T *result, unsigned int result_start, unsigned int result_inc, unsigned int result_size)
template<typename T >
__global__ void trans_vec_mul_row_kernel (const T *A, unsigned int A_row_start, unsigned int A_col_start, unsigned int A_row_inc, unsigned int A_col_inc, unsigned int A_row_size, unsigned int A_col_size, unsigned int A_internal_rows, unsigned int A_internal_cols, const T *v, unsigned int v_start, unsigned int v_inc, unsigned int v_size, T *result, unsigned int result_start, unsigned int result_inc, unsigned int result_size)
template<typename T >
__global__ void scaled_rank1_update_row_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, T val, unsigned int options2, const T *vec1, unsigned int start1, unsigned int inc1, unsigned int size1, const T *vec2, unsigned int start2, unsigned int inc2, unsigned int size2)
template<typename T >
__global__ void scaled_rank1_update_row_kernel (T *A, unsigned int A_start1, unsigned int A_start2, unsigned int A_inc1, unsigned int A_inc2, unsigned int A_size1, unsigned int A_size2, unsigned int A_internal_size1, unsigned int A_internal_size2, const T *val, unsigned int options2, const T *vec1, unsigned int start1, unsigned int inc1, unsigned int size1, const T *vec2, unsigned int start2, unsigned int inc2, unsigned int size2)

Detailed Description

Implementations of row-major dense matrix related operations, including matrix-vector products, using CUDA.