Matrix multiply, dot product, BLAS

Matrix multiply: Device pointer interface
See also Linear Algebra Parameters: TRANSA Must be 'N' indicating non-transposed TRANSB Must be 'N' indicating non-transposed. M Rows of A and C N Columns of B and C K Columns of A and rows of B ALPHA Must be 1.0 [in] d_A Left matrix (A). [in] d_B Right matrix (B). [out] d_C Output matrix (C). BETA Must be 0.0 batch_A Number of tiles in matrix A batch_B Number of tiles in matrix A
afError	af_gemm_S (char TRANSA, char TRANSB, int M, int N, int K, float ALPHA, const float *d_A, int LDA, const float *d_B, int LDB, float BETA, float *d_C, int LDC, unsigned batch_A=0, unsigned batch_B=0)
	Single precision, Real matrix multiply.
afError	af_gemm_D (char TRANSA, char TRANSB, int M, int N, int K, double ALPHA, const double *d_A, int LDA, const double *d_B, int LDB, double BETA, double *d_C, int LDC, unsigned batch_A=0, unsigned batch_B=0)
	Double precision, Real matrix multiply.
afError	af_gemm_C (char TRANSA, char TRANSB, int M, int N, int K, cuComplex ALPHA, const cuComplex *d_A, int LDA, const cuComplex *d_B, int LDB, cuComplex BETA, cuComplex *d_C, int LDC, unsigned batch_A=0, unsigned batch_B=0)
	Single precision, Complex matrix multiply.
afError	af_gemm_Z (char TRANSA, char TRANSB, int M, int N, int K, cuDoubleComplex ALPHA, const cuDoubleComplex *d_A, int LDA, const cuDoubleComplex *d_B, int LDB, cuDoubleComplex BETA, cuDoubleComplex *d_C, int LDC, unsigned batch_A=0, unsigned batch_B=0)
	Double precision, Complex matrix multiply.
afError	af_matmul_SS (const float *d_A, const float *d_B, float *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
	Mixed precision, mixed complexity matrix multiply.
afError	af_matmul_DS (const double *d_A, const float *d_B, float *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
afError	af_matmul_SD (const float *d_A, const double *d_B, float *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
afError	af_matmul_DD (const double *d_A, const double *d_B, double *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
afError	af_matmul_CS (const cuComplex *d_A, const float *d_B, cuComplex *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
afError	af_matmul_ZS (const cuDoubleComplex *d_A, const float *d_B, cuComplex *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
afError	af_matmul_CD (const cuComplex *d_A, const double *d_B, cuComplex *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
afError	af_matmul_ZD (const cuDoubleComplex *d_A, const double *d_B, cuDoubleComplex *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
afError	af_matmul_SC (const float *d_A, const cuComplex *d_B, cuComplex *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
afError	af_matmul_DC (const double *d_A, const cuComplex *d_B, cuComplex *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
afError	af_matmul_SZ (const float *d_A, const cuDoubleComplex *d_B, cuComplex *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
afError	af_matmul_DZ (const double *d_A, const cuDoubleComplex *d_B, cuDoubleComplex *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
afError	af_matmul_CC (const cuComplex *d_A, const cuComplex *d_B, cuComplex *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
afError	af_matmul_ZC (const cuDoubleComplex *d_A, const cuComplex *d_B, cuComplex *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
afError	af_matmul_CZ (const cuComplex *d_A, const cuDoubleComplex *d_B, cuComplex *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
afError	af_matmul_ZZ (const cuDoubleComplex *d_A, const cuDoubleComplex *d_B, cuDoubleComplex *d_C, int M, int N, int K, unsigned batch_A=0, unsigned batch_B=0)
Dot product: Device pointer interface
See also Linear Algebra Parameters: [out] h_res The output. (Host pointer); [in] len The length of the vectors [in] d_A Device pointer to the first vector [in] d_B Device pointer to the second vector
afError	af_dot_S (float *h_res, unsigned len, const float *d_A, const float *d_B)
afError	af_dot_D (double *h_res, unsigned len, const double *d_A, const double *d_B)
afError	af_dot_C (cuComplex *h_res, unsigned len, const cuComplex *d_A, const cuComplex *d_B)
afError	af_dot_Z (cuDoubleComplex *h_res, unsigned len, const cuDoubleComplex *d_A, const cuDoubleComplex *d_B)

---

Function Documentation

afError af_gemm_S	(	char	TRANSA,
		char	TRANSB,
		int	M,
		int	N,
		int	K,
		float	ALPHA,
		const float *	d_A,
		int	LDA,
		const float *	d_B,
		int	LDB,
		float	BETA,
		float *	d_C,
		int	LDC,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

Single precision, Real matrix multiply.

afError af_gemm_D	(	char	TRANSA,
		char	TRANSB,
		int	M,
		int	N,
		int	K,
		double	ALPHA,
		const double *	d_A,
		int	LDA,
		const double *	d_B,
		int	LDB,
		double	BETA,
		double *	d_C,
		int	LDC,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

Double precision, Real matrix multiply.

afError af_gemm_C	(	char	TRANSA,
		char	TRANSB,
		int	M,
		int	N,
		int	K,
		cuComplex	ALPHA,
		const cuComplex *	d_A,
		int	LDA,
		const cuComplex *	d_B,
		int	LDB,
		cuComplex	BETA,
		cuComplex *	d_C,
		int	LDC,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

Single precision, Complex matrix multiply.

afError af_gemm_Z	(	char	TRANSA,
		char	TRANSB,
		int	M,
		int	N,
		int	K,
		cuDoubleComplex	ALPHA,
		const cuDoubleComplex *	d_A,
		int	LDA,
		const cuDoubleComplex *	d_B,
		int	LDB,
		cuDoubleComplex	BETA,
		cuDoubleComplex *	d_C,
		int	LDC,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

Double precision, Complex matrix multiply.

afError af_matmul_SS	(	const float *	d_A,
		const float *	d_B,
		float *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

Mixed precision, mixed complexity matrix multiply.

afError af_matmul_DS	(	const double *	d_A,
		const float *	d_B,
		float *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_matmul_SD	(	const float *	d_A,
		const double *	d_B,
		float *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_matmul_DD	(	const double *	d_A,
		const double *	d_B,
		double *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_matmul_CS	(	const cuComplex *	d_A,
		const float *	d_B,
		cuComplex *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_matmul_ZS	(	const cuDoubleComplex *	d_A,
		const float *	d_B,
		cuComplex *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_matmul_CD	(	const cuComplex *	d_A,
		const double *	d_B,
		cuComplex *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_matmul_ZD	(	const cuDoubleComplex *	d_A,
		const double *	d_B,
		cuDoubleComplex *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_matmul_SC	(	const float *	d_A,
		const cuComplex *	d_B,
		cuComplex *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_matmul_DC	(	const double *	d_A,
		const cuComplex *	d_B,
		cuComplex *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_matmul_SZ	(	const float *	d_A,
		const cuDoubleComplex *	d_B,
		cuComplex *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_matmul_DZ	(	const double *	d_A,
		const cuDoubleComplex *	d_B,
		cuDoubleComplex *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_matmul_CC	(	const cuComplex *	d_A,
		const cuComplex *	d_B,
		cuComplex *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_matmul_ZC	(	const cuDoubleComplex *	d_A,
		const cuComplex *	d_B,
		cuComplex *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_matmul_CZ	(	const cuComplex *	d_A,
		const cuDoubleComplex *	d_B,
		cuComplex *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_matmul_ZZ	(	const cuDoubleComplex *	d_A,
		const cuDoubleComplex *	d_B,
		cuDoubleComplex *	d_C,
		int	M,
		int	N,
		int	K,
		unsigned	batch_A = 0,
		unsigned	batch_B = 0
	)

afError af_dot_S	(	float *	h_res,
		unsigned	len,
		const float *	d_A,
		const float *	d_B
	)

afError af_dot_D	(	double *	h_res,
		unsigned	len,
		const double *	d_A,
		const double *	d_B
	)

afError af_dot_C	(	cuComplex *	h_res,
		unsigned	len,
		const cuComplex *	d_A,
		const cuComplex *	d_B
	)

afError af_dot_Z	(	cuDoubleComplex *	h_res,
		unsigned	len,
		const cuDoubleComplex *	d_A,
		const cuDoubleComplex *	d_B
	)