hipsparsesbsrxmv Interface Reference

hipsparsesbsrxmv Interface Reference#

HIPFORT API Reference: hipfort_hipsparse::hipsparsesbsrxmv Interface Reference

Sparse matrix vector multiplication with mask operation using the BSR storage format. More...

Public Member Functions
integer(kind(hipsparse_status_success)) function	hipsparsesbsrxmv_ (handle, dir, trans, sizeofmask, mb, nb, nnzb, alpha, descr, bsrval, bsrmaskptr, bsrrowptr, bsrendptr, bsrcolind, blockdim, x, beta, y)

integer(kind(hipsparse_status_success)) function	hipsparsesbsrxmv_rank_0 (handle, dir, trans, sizeofmask, mb, nb, nnzb, alpha, descr, bsrval, bsrmaskptr, bsrrowptr, bsrendptr, bsrcolind, blockdim, x, beta, y)

integer(kind(hipsparse_status_success)) function	hipsparsesbsrxmv_rank_1 (handle, dir, trans, sizeofmask, mb, nb, nnzb, alpha, descr, bsrval, bsrmaskptr, bsrrowptr, bsrendptr, bsrcolind, blockdim, x, beta, y)

Detailed Description

Sparse matrix vector multiplication with mask operation using the BSR storage format.

hipsparseXbsrxmv multiplies the scalar \(\alpha\) with a sparse \((mb \times \text{blockDim}) \times (nb \times \text{blockDim})\) modified matrix, defined in BSR storage format, and the dense vector \(x\) and adds the result to the dense vector \(y\) that is multiplied by the scalar \(\beta\), such that

\[ y := \left( \alpha \cdot op(A) \cdot x + \beta \cdot y \right)\left( \text{mask} \right), \]

with

\[ op(A) = \left\{ \begin{array}{ll} A, & \text{if trans == HIPSPARSE_OPERATION_NON_TRANSPOSE} \\% A^T, & \text{if trans == HIPSPARSE_OPERATION_TRANSPOSE} \\% A^H, & \text{if trans == HIPSPARSE_OPERATION_CONJUGATE_TRANSPOSE} \end{array} \right. \]

The \(\text{mask}\) is defined as an array of block row indices. The input sparse matrix is defined with a modified BSR storage format where the beginning and the end of each row is defined with two arrays, bsrRowPtr and bsr_end_ptr (both of size mb), rather the usual bsrRowPtr of size mb+1.

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 trans == HIPSPARSE_OPERATION_NON_TRANSPOSE is supported, and blockDim == 1 is not supported.

Deprecated:: This function is deprecated when using the CUDA backend (CUDA 12.0+) and will be removed in CUDA 13.0. This deprecation does not apply to the ROCm backend.

Parameters

[in]	handle	- handle to the hipSPARSE library context queue.
[in]	dir	- matrix storage of BSR blocks.
[in]	trans	- matrix operation type.
[in]	sizeOfMask	- number of updated block rows of the array `y`. Must be non-negative and not greater than `mb`.
[in]	mb	- number of block rows of the sparse BSR matrix. Must be non-negative.
[in]	nb	- number of block columns of the sparse BSR matrix. Must be non-negative.
[in]	nnzb	- number of non-zero blocks of the sparse BSR matrix. Must be non-negative.
[in]	alpha	- scalar \(\alpha\).
[in]	descr	- descriptor of the sparse BSR matrix. Currently, only `HIPSPARSE_MATRIX_TYPE_GENERAL` is supported.
[in]	bsrVal	- array of `nnzb` blocks of the sparse BSR matrix.
[in]	bsrMaskPtr	- array of `sizeOfMask` elements that give the indices of the updated block rows.
[in]	bsrRowPtr	- array of `mb` elements that point to the start of every block row of the sparse BSR matrix.
[in]	bsrEndPtr	- array of `mb` elements that point to the end of every block row of the sparse BSR matrix.
[in]	bsrColInd	- array of `nnzb` elements containing the block column indices of the sparse BSR matrix.
[in]	blockDim	- block dimension of the sparse BSR matrix. Must be greater than 1.
[in]	x	- array of `nbblockDim` elements ( \(op(A) = A\)) or `mbblockDim` elements ( \(op(A) = A^T\) or \(op(A) = A^H\)).
[in]	beta	- scalar \(\beta\).
[in,out]	y	- array of `mbblockDim` elements ( \(op(A) = A\)) or `nbblockDim` elements ( \(op(A) = A^T\) or \(op(A) = A^H\)).

Return values

HIPSPARSE_STATUS_SUCCESS	the operation completed successfully.
HIPSPARSE_STATUS_NOT_INITIALIZED	`handle` is not initialized.
HIPSPARSE_STATUS_INVALID_VALUE	`handle`, `descr`, `alpha`, or `beta` is nullptr, `mb`, `nb`, `nnzb`, or `sizeOfMask` is negative, `sizeOfMask` is greater than `mb`, `blockDim` is less than or equal to 1, or `bsrVal`, `bsrMaskPtr`, `bsrRowPtr`, `bsrEndPtr`, `bsrColInd`, `x`, or `y` is nullptr.
HIPSPARSE_STATUS_ARCH_MISMATCH	the device is not supported.
HIPSPARSE_STATUS_NOT_SUPPORTED	`trans` is not `HIPSPARSE_OPERATION_NON_TRANSPOSE`, or `hipsparseMatrixType_t` is not `HIPSPARSE_MATRIX_TYPE_GENERAL`.

Member Function/Subroutine Documentation