hipsparsesgemvi Interface Reference

• hipfort_hipsparse
• hipsparsesgemvi

Public Member Functions | List of all members

hipfort_hipsparse::hipsparsesgemvi Interface Reference

Dense matrix sparse vector multiplication. More...

Public Member Functions
integer(kind(hipsparse_status_success)) function	hipsparsesgemvi_ (handle, transa, m, n, alpha, a, lda, nnz, x, xind, beta, y, idxbase, pbuffer)
integer(kind(hipsparse_status_success)) function	hipsparsesgemvi_rank_0 (handle, transa, m, n, alpha, a, lda, nnz, x, xind, beta, y, idxbase, pbuffer)
integer(kind(hipsparse_status_success)) function	hipsparsesgemvi_rank_1 (handle, transa, m, n, alpha, a, lda, nnz, x, xind, beta, y, idxbase, pbuffer)
integer(kind(hipsparse_status_success)) function	hipsparsesgemvi_full_rank (handle, transa, m, n, alpha, a, lda, nnz, x, xind, beta, y, idxbase, pbuffer)

Detailed Description

Dense matrix sparse vector multiplication.

hipsparseXgemvi multiplies the scalar \(\alpha\) with a dense \(m \times n\) matrix \(A\) and the sparse vector \(x\) and adds the result to the dense vector \(y\) that is multiplied by the scalar \(\beta\), such that

\[ y := \alpha \cdot op(A) \cdot x + \beta \cdot y, \]

with

\[ op(A) = \left\{ \begin{array}{ll} A, & \text{if transA == HIPSPARSE_OPERATION_NON_TRANSPOSE} \end{array} \right. \]

Performing the above operation involves two steps. First, the user calls hipsparseSgemvi_bufferSize "hipsparseXgemvi_bufferSize()" to determine the size of the temporary storage buffer. Next, the user allocates this temporary buffer and passes it to hipsparseXgemvi to complete the computation. After all calls to hipsparseXgemvi are complete, the temporary storage buffer can be freed.

Note

This function is non-blocking and executed asynchronously with respect to the host. It can return before the actual computation has finished.

Currently, only transA == HIPSPARSE_OPERATION_NON_TRANSPOSE is supported.

Parameters

[in]	handle	- handle to the hipSPARSE library context queue.
[in]	transA	- matrix operation type.
[in]	m	- number of rows of the dense matrix.
[in]	n	- number of columns of the dense matrix.
[in]	alpha	- scalar \(\alpha\).
[in]	A	- pointer to the dense matrix.
[in]	lda	- leading dimension of the dense matrix.
[in]	nnz	- number of non-zero entries in the sparse vector.
[in]	x	- array of nnz elements containing the values of the sparse vector.
[in]	xInd	- array of nnz elements containing the indices of the sparse vector.
[in]	beta	- scalar \(\beta\).
[in,out]	y	- array of m elements ( \(op(A) == A\)) or n elements ( \(op(A) == A^T\) or \(op(A) == A^H\)).
[in]	idxBase	- HIPSPARSE_INDEX_BASE_ZERO or HIPSPARSE_INDEX_BASE_ONE.
[in]	pBuffer	- temporary storage buffer.

Return values

HIPSPARSE_STATUS_SUCCESS	the operation completed successfully.
HIPSPARSE_STATUS_INVALID_VALUE	handle, m, n, lda, nnz, alpha, A, x, xInd, beta, y, or pBuffer is invalid.
HIPSPARSE_STATUS_NOT_SUPPORTED	transA != HIPSPARSE_OPERATION_NON_TRANSPOSE or hipsparseMatrixType_t != HIPSPARSE_MATRIX_TYPE_GENERAL.

Member Function/Subroutine Documentation

◆ hipsparsesgemvi_()

integer(kind(hipsparse_status_success)) function hipfort_hipsparse::hipsparsesgemvi::hipsparsesgemvi_	(	type(c_ptr), value	handle,
		integer(kind(hipsparse_operation_non_transpose)), value	transa,
		integer(c_int), value	m,
		integer(c_int), value	n,
		real(c_float)	alpha,
		type(c_ptr), value	a,
		integer(c_int), value	lda,
		integer(c_int), value	nnz,
		type(c_ptr), value	x,
		type(c_ptr), value	xind,
		real(c_float)	beta,
		type(c_ptr), value	y,
		integer(kind(hipsparse_index_base_zero)), value	idxbase,
		type(c_ptr), value	pbuffer
	)

◆ hipsparsesgemvi_full_rank()

integer(kind(hipsparse_status_success)) function hipfort_hipsparse::hipsparsesgemvi::hipsparsesgemvi_full_rank	(	type(c_ptr)	handle,
		integer(kind(hipsparse_operation_non_transpose))	transa,
		integer(c_int)	m,
		integer(c_int)	n,
		real(c_float)	alpha,
		real(c_float), dimension(:,:), target	a,
		integer(c_int)	lda,
		integer(c_int)	nnz,
		real(c_float), dimension(:), target	x,
		integer(c_int), dimension(:), target	xind,
		real(c_float)	beta,
		real(c_float), dimension(:), target	y,
		integer(kind(hipsparse_index_base_zero))	idxbase,
		type(c_ptr)	pbuffer
	)

◆ hipsparsesgemvi_rank_0()

integer(kind(hipsparse_status_success)) function hipfort_hipsparse::hipsparsesgemvi::hipsparsesgemvi_rank_0	(	type(c_ptr)	handle,
		integer(kind(hipsparse_operation_non_transpose))	transa,
		integer(c_int)	m,
		integer(c_int)	n,
		real(c_float)	alpha,
		real(c_float), target	a,
		integer(c_int)	lda,
		integer(c_int)	nnz,
		real(c_float), target	x,
		integer(c_int), target	xind,
		real(c_float)	beta,
		real(c_float), target	y,
		integer(kind(hipsparse_index_base_zero))	idxbase,
		type(c_ptr)	pbuffer
	)

◆ hipsparsesgemvi_rank_1()

integer(kind(hipsparse_status_success)) function hipfort_hipsparse::hipsparsesgemvi::hipsparsesgemvi_rank_1	(	type(c_ptr)	handle,
		integer(kind(hipsparse_operation_non_transpose))	transa,
		integer(c_int)	m,
		integer(c_int)	n,
		real(c_float)	alpha,
		real(c_float), dimension(:), target	a,
		integer(c_int)	lda,
		integer(c_int)	nnz,
		real(c_float), dimension(:), target	x,
		integer(c_int), dimension(:), target	xind,
		real(c_float)	beta,
		real(c_float), dimension(:), target	y,
		integer(kind(hipsparse_index_base_zero))	idxbase,
		type(c_ptr)	pbuffer
	)

---

The documentation for this interface was generated from the following file:

• hipfort_hipsparse.F90