hipsparsesdense2csc Interface Reference

hipsparsesdense2csc Interface Reference#

HIPFORT API Reference: hipfort_hipsparse::hipsparsesdense2csc Interface Reference

hipsparseXdense2csc functions convert the matrix A in dense format into a sparse matrix in CSC format. More...

Public Member Functions
integer(kind(hipsparse_status_success)) function	hipsparsesdense2csc_ (handle, m, n, descr, a, ld, nnzpercolumn, cscval, cscrowind, csccolptr)

integer(kind(hipsparse_status_success)) function	hipsparsesdense2csc_rank_0 (handle, m, n, descr, a, ld, nnzpercolumn, cscval, cscrowind, csccolptr)

integer(kind(hipsparse_status_success)) function	hipsparsesdense2csc_rank_1 (handle, m, n, descr, a, ld, nnzpercolumn, cscval, cscrowind, csccolptr)

integer(kind(hipsparse_status_success)) function	hipsparsesdense2csc_full_rank (handle, m, n, descr, a, ld, nnzpercolumn, cscval, cscrowind, csccolptr)

Detailed Description

hipsparseXdense2csc functions convert the matrix A in dense format into a sparse matrix in CSC format.

Given a dense, column-ordered, matrix A with leading dimension ld, where ld>=m, hipsparseXdense2csc converts the matrix to a sparse CSC format matrix. All the parameters are assumed to have been preallocated by the user and the arrays are filled in based on number of non-zeros per row, which can be pre-computed with hipsparseSnnz "hipsparseXnnz()". Users can set the desired index base in the output CSC matrix by setting it in the hipsparseMatDescr_t. See hipsparseSetMatIndexBase().

As an example, if using index base zero (that is, the default) and the dense matrix:

\[ \begin{bmatrix} 1 & 0 & 0 & 2 \\% 3 & 4 & 0 & 0 \\% 5 & 0 & 6 & 7 \end{bmatrix} \]

where the A values have column ordering with leading dimension ld=m:

\[ \text{A} &= \begin{bmatrix} 1 & 3 & 5 & 0 & 4 & 0 & 0 & 0 & 6 & 2 & 0 & 7 \end{bmatrix} \\% \]

the conversion results in the CSC arrays:

\[ \begin{align} \text{cscRowInd} &= \begin{bmatrix} 0 & 1 & 2 & 1 & 2 & 0 & 2 \end{bmatrix} \\% \text{cscColPtr} &= \begin{bmatrix} 0 & 3 & 4 & 5 & 7 \end{bmatrix} \\% \text{cscVal} &= \begin{bmatrix} 1 & 3 & 5 & 4 & 6 & 2 & 7 \end{bmatrix} \\% \end{align} \]

This function works very similar to hipsparseSdense2csr "hipsparseXdense2csr()". See hipsparseSdense2csr() for a code example.

Note: This function is executed asynchronously with respect to the host and can return control to the application on the host before the entire result is ready.

Deprecated:: This function is deprecated when using the CUDA backend (CUDA 11.0+) and will be removed in CUDA 12.0. This deprecation does not apply to the ROCm backend.

Parameters

[in]	handle	- handle to the hipSPARSE library context queue.
[in]	m	- number of rows of the dense matrix `A`. Must be non-negative.
[in]	n	- number of columns of the dense matrix `A`. Must be non-negative.
[in]	descr	- the descriptor of the dense matrix `A`. The supported matrix type is `HIPSPARSE_MATRIX_TYPE_GENERAL` and any valid value of the `hipsparseIndexBase_t`.
[in]	A	- array of dimensions (`ld`, `n`).
[in]	ld	- leading dimension of dense array `A`. Must be at least `m`.
[in]	nnzPerColumn	- array of size `n` containing the number of non-zero elements per column.
[out]	cscVal	- array of nnz ( = `cscColPtr`[n] - `cscColPtr`[0] ) nonzero elements of matrix `A`.
[out]	cscRowInd	- integer array of nnz ( = `cscColPtr`[n] - `cscColPtr`[0] ) column indices of the non-zero elements of matrix `A`.
[out]	cscColPtr	- integer array of `n+1` elements that contains the start of every column and the end of the last column plus one.

Return values

HIPSPARSE_STATUS_SUCCESS	the operation completed successfully.
HIPSPARSE_STATUS_NOT_INITIALIZED	`handle` is not initialized.
HIPSPARSE_STATUS_INVALID_VALUE	`handle`, `descr`, `A`, `nnzPerColumn`, `cscVal`, `cscColPtr`, or `cscRowInd` is nullptr, `m` or `n` is negative, or `ld` is invalid.

Member Function/Subroutine Documentation