HEMM - Hermitian matrix-matrix multiplication

Functions
clblasStatus	clblasChemm (clblasOrder order, clblasSide side, clblasUplo uplo, size_t M, size_t N, cl_float2 alpha, const cl_mem A, size_t offa, size_t lda, const cl_mem B, size_t offb, size_t ldb, cl_float2 beta, cl_mem C, size_t offc, size_t ldc, cl_uint numCommandQueues, cl_command_queue *commandQueues, cl_uint numEventsInWaitList, const cl_event *eventWaitList, cl_event *events)
	Matrix-matrix product of hermitian rectangular matrices with float-complex elements. More...
clblasStatus	clblasZhemm (clblasOrder order, clblasSide side, clblasUplo uplo, size_t M, size_t N, cl_double2 alpha, const cl_mem A, size_t offa, size_t lda, const cl_mem B, size_t offb, size_t ldb, cl_double2 beta, cl_mem C, size_t offc, size_t ldc, cl_uint numCommandQueues, cl_command_queue *commandQueues, cl_uint numEventsInWaitList, const cl_event *eventWaitList, cl_event *events)
	Matrix-matrix product of hermitian rectangular matrices with double-complex elements. More...

Detailed Description

Function Documentation

clblasStatus clblasChemm	(	clblasOrder	order,
		clblasSide	side,
		clblasUplo	uplo,
		size_t	M,
		size_t	N,
		cl_float2	alpha,
		const cl_mem	A,
		size_t	offa,
		size_t	lda,
		const cl_mem	B,
		size_t	offb,
		size_t	ldb,
		cl_float2	beta,
		cl_mem	C,
		size_t	offc,
		size_t	ldc,
		cl_uint	numCommandQueues,
		cl_command_queue *	commandQueues,
		cl_uint	numEventsInWaitList,
		const cl_event *	eventWaitList,
		cl_event *	events
	)

Matrix-matrix product of hermitian rectangular matrices with float-complex elements.

Matrix-matrix products:

• \( C \leftarrow \alpha A B + \beta C \)
• \( C \leftarrow \alpha B A + \beta C \)

Parameters

[in]	order	Row/column order.
[in]	side	The side of triangular matrix.
[in]	uplo	The triangle in matrix being referenced.
[in]	M	Number of rows in matrices B and C.
[in]	N	Number of columns in matrices B and C.
[in]	alpha	The factor of matrix A.
[in]	A	Buffer object storing matrix A.
[in]	offa	Offset of the first element of the matrix A in the buffer object. Counted in elements.
[in]	lda	Leading dimension of matrix A. It cannot be less than M when the side parameter is set to clblasLeft, or less than N when the parameter is set to clblasRight.
[in]	B	Buffer object storing matrix B.
[in]	offb	Offset of the first element of the matrix B in the buffer object. Counted in elements.
[in]	ldb	Leading dimension of matrix B. It cannot be less than N when the order parameter is set to clblasRowMajor, or less than M when it is set to clblasColumnMajor.
[in]	beta	The factor of matrix C.
[out]	C	Buffer object storing matrix C.
[in]	offc	Offset of the first element of the matrix C in the buffer object. Counted in elements.
[in]	ldc	Leading dimension of matrix C. It cannot be less than N when the order parameter is set to clblasRowMajor, or less than M when it is set to clblasColumnMajorOrder.
[in]	numCommandQueues	Number of OpenCL command queues in which the task is to be performed.
[in]	commandQueues	OpenCL command queues.
[in]	numEventsInWaitList	Number of events in the event wait list.
[in]	eventWaitList	Event wait list.
[in]	events	Event objects per each command queue that identify a particular kernel execution instance.

Returns

• clblasSuccess on success;
• clblasNotInitialized if clblasSetup() was not called;
• clblasInvalidValue if invalid parameters are passed:
  • M or N is zero, or
  • any of the leading dimensions is invalid;
  • the matrix sizes lead to accessing outsize of any of the buffers;
• clblasInvalidMemObject if A, B, or C object is invalid, or an image object rather than the buffer one;
• clblasOutOfResources if you use image-based function implementation and no suitable scratch image available;
• clblasOutOfHostMemory if the library can't allocate memory for internal structures;
• clblasInvalidCommandQueue if the passed command queue is invalid;
• clblasInvalidContext if a context a passed command queue belongs to was released;
• clblasInvalidOperation if kernel compilation relating to a previous call has not completed for any of the target devices;
• clblasCompilerNotAvailable if a compiler is not available;
• clblasBuildProgramFailure if there is a failure to build a program executable.

Examples:

example_chemm.cpp.

clblasStatus clblasZhemm	(	clblasOrder	order,
		clblasSide	side,
		clblasUplo	uplo,
		size_t	M,
		size_t	N,
		cl_double2	alpha,
		const cl_mem	A,
		size_t	offa,
		size_t	lda,
		const cl_mem	B,
		size_t	offb,
		size_t	ldb,
		cl_double2	beta,
		cl_mem	C,
		size_t	offc,
		size_t	ldc,
		cl_uint	numCommandQueues,
		cl_command_queue *	commandQueues,
		cl_uint	numEventsInWaitList,
		const cl_event *	eventWaitList,
		cl_event *	events
	)

Matrix-matrix product of hermitian rectangular matrices with double-complex elements.

Matrix-matrix products:

• \( C \leftarrow \alpha A B + \beta C \)
• \( C \leftarrow \alpha B A + \beta C \)

Parameters

[in]	order	Row/column order.
[in]	side	The side of triangular matrix.
[in]	uplo	The triangle in matrix being referenced.
[in]	M	Number of rows in matrices B and C.
[in]	N	Number of columns in matrices B and C.
[in]	alpha	The factor of matrix A.
[in]	A	Buffer object storing matrix A.
[in]	offa	Offset of the first element of the matrix A in the buffer object. Counted in elements.
[in]	lda	Leading dimension of matrix A. It cannot be less than M when the side parameter is set to clblasLeft, or less than N when the parameter is set to clblasRight.
[in]	B	Buffer object storing matrix B.
[in]	offb	Offset of the first element of the matrix B in the buffer object. Counted in elements.
[in]	ldb	Leading dimension of matrix B. It cannot be less than N when the order parameter is set to clblasRowMajor, or less than M when it is set to clblasColumnMajor.
[in]	beta	The factor of matrix C.
[out]	C	Buffer object storing matrix C.
[in]	offc	Offset of the first element of the matrix C in the buffer object. Counted in elements.
[in]	ldc	Leading dimension of matrix C. It cannot be less than N when the order parameter is set to clblasRowMajor, or less than M when it is set to clblasColumnMajorOrder.
[in]	numCommandQueues	Number of OpenCL command queues in which the task is to be performed.
[in]	commandQueues	OpenCL command queues.
[in]	numEventsInWaitList	Number of events in the event wait list.
[in]	eventWaitList	Event wait list.
[in]	events	Event objects per each command queue that identify a particular kernel execution instance.

Returns

• clblasSuccess on success;
• clblasInvalidDevice if a target device does not support floating point arithmetic with double precision;
• the same error codes as the clblasChemm() function otherwise.