dgemm_vbatch.h File Reference

#include <cuda_runtime.h>

Include dependency graph for dgemm_vbatch.h:

Go to the source code of this file.

Functions
template<typename T >
void	gemm_nn_vbatch (int max_m, int max_n, int max_k, const int *m_d, const int *n_d, const int *k_d, const T *const *A_array_d, const int *lda_d, const T *const *B_array_d, const int *ldb_d, double **C_array_d, const int *ldc_d, int batchCount, cudaStream_t stream, const T *alpha=nullptr)
template<typename T >
void	gemm_tn_vbatch (int max_m, int max_n, int max_k, const int *m_d, const int *n_d, const int *k_d, const T *const *A_array_d, const int *lda_d, const T *const *B_array_d, const int *ldb_d, double **C_array_d, const int *ldc_d, int batchCount, cudaStream_t stream, const T *alpha=nullptr)
void	dgemm_nn_vbatch (int max_m, int max_n, int max_k, const int *m_d, const int *n_d, const int *k_d, const double *const *A_array_d, const int *lda_d, const double *const *B_array_d, const int *ldb_d, double **C_array_d, const int *ldc_d, int batchCount, cudaStream_t stream, const double *alpha=nullptr)
void	dgemm_tn_vbatch (int max_m, int max_n, int max_k, const int *m_d, const int *n_d, const int *k_d, const double *const *A_array_d, const int *lda_d, const double *const *B_array_d, const int *ldb_d, double **C_array_d, const int *ldc_d, int batchCount, cudaStream_t stream, const double *alpha=nullptr)

Function Documentation

dgemm_nn_vbatch()

void dgemm_nn_vbatch ( int  max_m, int  max_n, int  max_k, const int *  m_d, const int *  n_d, const int *  k_d, const double *const *  A_array_d, const int *  lda_d, const double *const *  B_array_d, const int *  ldb_d, double **  C_array_d, const int *  ldc_d, int  batchCount, cudaStream_t  stream, const double *  alpha = nullptr  )

inline

dgemm_tn_vbatch()

void dgemm_tn_vbatch ( int  max_m, int  max_n, int  max_k, const int *  m_d, const int *  n_d, const int *  k_d, const double *const *  A_array_d, const int *  lda_d, const double *const *  B_array_d, const int *  ldb_d, double **  C_array_d, const int *  ldc_d, int  batchCount, cudaStream_t  stream, const double *  alpha = nullptr  )

inline

gemm_nn_vbatch()

template<typename T >

void gemm_nn_vbatch	(	int	max_m,
		int	max_n,
		int	max_k,
		const int *	m_d,
		const int *	n_d,
		const int *	k_d,
		const T *const *	A_array_d,
		const int *	lda_d,
		const T *const *	B_array_d,
		const int *	ldb_d,
		double **	C_array_d,
		const int *	ldc_d,
		int	batchCount,
		cudaStream_t	stream,
		const T *	alpha = nullptr
	)

gemm_tn_vbatch()

template<typename T >

void gemm_tn_vbatch	(	int	max_m,
		int	max_n,
		int	max_k,
		const int *	m_d,
		const int *	n_d,
		const int *	k_d,
		const T *const *	A_array_d,
		const int *	lda_d,
		const T *const *	B_array_d,
		const int *	ldb_d,
		double **	C_array_d,
		const int *	ldc_d,
		int	batchCount,
		cudaStream_t	stream,
		const T *	alpha = nullptr
	)