hipsparsescsrmm Interface Reference

hipsparsescsrmm Interface Reference#

HIPFORT API Reference: hipfort_hipsparse::hipsparsescsrmm Interface Reference

Sparse matrix dense matrix multiplication using the CSR storage format. More...

Public Member Functions
integer(kind(hipsparse_status_success)) function	hipsparsescsrmm_ (handle, transa, m, n, k, nnz, alpha, descra, csrsortedvala, csrsortedrowptra, csrsortedcolinda, b, ldb, beta, c, ldc)

integer(kind(hipsparse_status_success)) function	hipsparsescsrmm_rank_0 (handle, transa, m, n, k, nnz, alpha, descra, csrsortedvala, csrsortedrowptra, csrsortedcolinda, b, ldb, beta, c, ldc)

integer(kind(hipsparse_status_success)) function	hipsparsescsrmm_rank_1 (handle, transa, m, n, k, nnz, alpha, descra, csrsortedvala, csrsortedrowptra, csrsortedcolinda, b, ldb, beta, c, ldc)

integer(kind(hipsparse_status_success)) function	hipsparsescsrmm_full_rank (handle, transa, m, n, k, nnz, alpha, descra, csrsortedvala, csrsortedrowptra, csrsortedcolinda, b, ldb, beta, c, ldc)

Detailed Description

Sparse matrix dense matrix multiplication using the CSR storage format.

hipsparseXcsrmm multiplies the scalar \(\alpha\) with a sparse \(m \times k\) matrix \(A\), defined in CSR storage format, and the column-oriented dense \(k \times n\) matrix \(B\) and adds the result to the column-oriented dense \(m \times n\) matrix \(C\) that is multiplied by the scalar \(\beta\), such that

\[ C := \alpha \cdot op(A) \cdot B + \beta \cdot C, \]

with

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

for(i = 0; i < ldc; ++i)
{
    for(j = 0; j < n; ++j)
    {
        C[i][j] = beta * C[i][j];
 
        for(k = csrRowPtr[i]; k < csrRowPtr[i + 1]; ++k)
        {
            C[i][j] += alpha * csrVal[k] * B[csrColInd[k]][j];
        }
    }
}

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

Deprecated:: This function is deprecated when using the CUDA backend (CUDA 10.0+) and will be removed in CUDA 11.0. This deprecation does not apply to the ROCm backend.

Parameters

[in]	handle	- handle to the hipSPARSE library context queue.
[in]	transA	- matrix \(A\) operation type.
[in]	m	- number of rows of the sparse CSR matrix \(A\). Must be non-negative.
[in]	n	- number of columns of the dense matrix \(op(B)\) and \(C\). Must be non-negative.
[in]	k	- number of columns of the sparse CSR matrix \(A\). Must be non-negative.
[in]	nnz	- number of non-zero entries of the sparse CSR matrix \(A\). Must be non-negative.
[in]	alpha	- scalar \(\alpha\).
[in]	descrA	- descriptor of the sparse CSR matrix \(A\). Currently, only `HIPSPARSE_MATRIX_TYPE_GENERAL` is supported.
[in]	csrSortedValA	- array of `nnz` elements of the sparse CSR matrix \(A\).
[in]	csrSortedRowPtrA	- array of `m+1` elements that point to the start of every row of the sparse CSR matrix \(A\).
[in]	csrSortedColIndA	- array of `nnz` elements containing the column indices of the sparse CSR matrix \(A\).
[in]	B	- array of dimension `ldbn` ( \(op(B) == B\)), `ldbk` otherwise.
[in]	ldb	- leading dimension of \(B\), must be at least \(\max{(1, k)}\) ( \(op(B) == B\)), \(\max{(1, n)}\) otherwise.
[in]	beta	- scalar \(\beta\).
[in,out]	C	- array of dimension `ldc*n`.
[in]	ldc	- leading dimension of \(C\), must be at least \(\max{(1, m)}\) ( \(op(A) == A\)), \(\max{(1, k)}\) otherwise.

Return values

HIPSPARSE_STATUS_SUCCESS	the operation completed successfully.
HIPSPARSE_STATUS_NOT_INITIALIZED	`handle` is not initialized.
HIPSPARSE_STATUS_INVALID_VALUE	`handle`, `descrA`, `alpha`, or `beta` is nullptr, `m`, `n`, `k`, or `nnz` is negative, `ldb` or `ldc` is invalid, or `csrSortedValA`, `csrSortedRowPtrA`, `csrSortedColIndA`, `B`, or `C` is nullptr.
HIPSPARSE_STATUS_ARCH_MISMATCH	the device is not supported.
HIPSPARSE_STATUS_NOT_SUPPORTED	`hipsparseMatrixType_t` is not `HIPSPARSE_MATRIX_TYPE_GENERAL`.

Member Function/Subroutine Documentation