Using hipSPARSE

Using hipSPARSE#

This topic discusses how to use hipSPARSE, including a discussion of device and stream management, storage formats, and pointer mode.

HIP device management#

Before starting a HIP kernel, you can call hipSetDevice() to set a device. The system uses the default device if you don’t call the function. Unless you explicitly call hipSetDevice() to specify another device, HIP kernels are always launched on device 0. This HIP (and CUDA) device management approach is not specific to the hipSPARSE library. hipSPARSE honors this approach and assumes you have already set the preferred device before a hipSPARSE routine call.

After you set the device, you can create a handle with hipsparseCreate(). Subsequent hipSPARSE routines take this handle as an input parameter. hipSPARSE only queries the specified device (using hipGetDevice()). You are responsible for providing a valid device to hipSPARSE and ensuring device safety. If it’s not a valid device, hipSPARSE returns an error message.

To change to another device, you must destroy the current handle using hipsparseDestroy(), then create another handle using hipsparseCreate(), specifying another device.

Note

hipSetDevice() and hipGetDevice() are not part of the hipSPARSE API. They are part of the HIP runtime API for device management.

HIP stream management#

HIP kernels are always launched in a queue (also known as a stream). If you don’t explicitly specify a stream, the system provides and maintains a default stream, which you cannot create or destroy. However, you can freely create new streams (using hipStreamCreate()) and bind them to the hipSPARSE handle using hipsparseSetStream(). The hipSPARSE routines invoke HIP kernels. A hipSPARSE handle is always associated with a stream, which hipSPARSE passes to the kernels inside the routine. One hipSPARSE routine only takes one stream in a single invocation. If you create a stream, you are responsible for destroying it. See the HIP stream management API for more information.

Asynchronous execution#

Except for functions that allocate memory themselves, preventing asynchronicity, all hipSPARSE library functions are non-blocking and execute asynchronously with respect to the host, unless otherwise stated. These functions might return before the actual computation has finished. To force synchronization, use either hipDeviceSynchronize() or hipStreamSynchronize(). This ensures that all previously executed hipSPARSE functions on the device or stream have been completed.

Multiple streams and multiple devices#

If a system has multiple HIP devices, you can run multiple hipSPARSE handles concurrently. However, you cannot run a single hipSPARSE handle on different discrete devices because each handle is associated with a particular device. A new handle must be created for each additional device.

Interface examples#

The hipSPARSE interface is compatible with the rocSPARSE and NVIDIA CUDA cuSPARSE-v2 APIs. Porting a CUDA application that calls the CUDA cuSPARSE API to an application that calls the hipSPARSE API is relatively straightforward. For example, the hipSPARSE SCSRMV API interface is as follows:

hipsparseStatus_t
hipsparseScsrmv(hipsparseHandle_t handle,
               hipsparseOperation_t transA,
               int m, int n, int nnz, const float *alpha,
               const hipsparseMatDescr_t descrA,
               const float *csrValA,
               const int *csrRowPtrA, const int *csrColIndA,
               const float *x, const float *beta,
               float *y);

hipSPARSE assumes matrix A and vectors x and y are allocated in the GPU memory space and filled with data. You are responsible for copying data to and from the host and device memory.

Storage formats#

This section describes the supported matrix storage formats.

Note

The different storage formats support indexing with a base of 0 or 1, as described in Storage schemes and indexing base.

COO storage format#

The Coordinate (COO) storage format represents an \(m \times n\) matrix by:

m

Number of rows (integer).

n

Number of columns (integer).

nnz

Number of non-zero elements (integer).

coo_val

Array of nnz elements containing the data (floating point).

coo_row_ind

Array of nnz elements containing the row indices (integer).

coo_col_ind

Array of nnz elements containing the column indices (integer).

The COO matrix is expected to be sorted by row indices and column indices per row. Furthermore, each pair of indices should appear only once. Consider the following \(3 \times 5\) matrix and the corresponding COO structures, with \(m = 3, n = 5\) and \(\text{nnz} = 8\) using zero-based indexing:

\[\begin{split}A = \begin{pmatrix} 1.0 & 2.0 & 0.0 & 3.0 & 0.0 \\ 0.0 & 4.0 & 5.0 & 0.0 & 0.0 \\ 6.0 & 0.0 & 0.0 & 7.0 & 8.0 \\ \end{pmatrix}\end{split}\]

where

\[\begin{split}\begin{array}{ll} \text{coo_val}[8] & = \{1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0\} \\ \text{coo_row_ind}[8] & = \{0, 0, 0, 1, 1, 2, 2, 2\} \\ \text{coo_col_ind}[8] & = \{0, 1, 3, 1, 2, 0, 3, 4\} \end{array}\end{split}\]

COO (AoS) storage format#

The Coordinate (COO) Array of Structure (AoS) storage format represents an \(m \times n\) matrix by:

m

Number of rows (integer).

n

Number of columns (integer).

nnz

Number of non-zero elements (integer).

coo_val

Array of nnz elements containing the data (floating point).

coo_ind

Array of 2 * nnz elements containing alternating row and column indices (integer).

The COO (AoS) matrix is expected to be sorted by row indices and column indices per row. Furthermore, each pair of indices should appear only once. Consider the following \(3 \times 5\) matrix and the corresponding COO (AoS) structures, with \(m = 3, n = 5\) and \(\text{nnz} = 8\) using zero-based indexing:

\[\begin{split}A = \begin{pmatrix} 1.0 & 2.0 & 0.0 & 3.0 & 0.0 \\ 0.0 & 4.0 & 5.0 & 0.0 & 0.0 \\ 6.0 & 0.0 & 0.0 & 7.0 & 8.0 \\ \end{pmatrix}\end{split}\]

where

\[\begin{split}\begin{array}{ll} \text{coo_val}[8] & = \{1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0\} \\ \text{coo_ind}[16] & = \{0, 0, 0, 1, 0, 3, 1, 1, 1, 2, 2, 0, 2, 3, 2, 4\} \\ \end{array}\end{split}\]

CSR storage format#

The Compressed Sparse Row (CSR) storage format represents an \(m \times n\) matrix by:

m

Number of rows (integer).

n

Number of columns (integer).

nnz

Number of non-zero elements (integer).

csr_val

Array of nnz elements containing the data (floating point).

csr_row_ptr

Array of m+1 elements that point to the start of every row (integer).

csr_col_ind

Array of nnz elements containing the column indices (integer).

The CSR matrix is expected to be sorted by column indices within each row. Furthermore, each pair of indices should appear only once. Consider the following \(3 \times 5\) matrix and the corresponding CSR structures, with \(m = 3, n = 5\) and \(\text{nnz} = 8\) using one-based indexing:

\[\begin{split}A = \begin{pmatrix} 1.0 & 2.0 & 0.0 & 3.0 & 0.0 \\ 0.0 & 4.0 & 5.0 & 0.0 & 0.0 \\ 6.0 & 0.0 & 0.0 & 7.0 & 8.0 \\ \end{pmatrix}\end{split}\]

where

\[\begin{split}\begin{array}{ll} \text{csr_val}[8] & = \{1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0\} \\ \text{csr_row_ptr}[4] & = \{1, 4, 6, 9\} \\ \text{csr_col_ind}[8] & = \{1, 2, 4, 2, 3, 1, 4, 5\} \end{array}\end{split}\]

CSC storage format#

The Compressed Sparse Column (CSC) storage format represents an \(m \times n\) matrix by:

m

Number of rows (integer).

n

Number of columns (integer).

nnz

Number of non-zero elements (integer).

csc_val

Array of nnz elements containing the data (floating point).

csc_col_ptr

Array of n+1 elements that point to the start of every column (integer).

csc_row_ind

Array of nnz elements containing the row indices (integer).

The CSC matrix is expected to be sorted by row indices within each column. Furthermore, each pair of indices should appear only once. Consider the following \(3 \times 5\) matrix and the corresponding CSC structures, with \(m = 3, n = 5\) and \(\text{nnz} = 8\) using one-based indexing:

\[\begin{split}A = \begin{pmatrix} 1.0 & 2.0 & 0.0 & 3.0 & 0.0 \\ 0.0 & 4.0 & 5.0 & 0.0 & 0.0 \\ 6.0 & 0.0 & 0.0 & 7.0 & 8.0 \\ \end{pmatrix}\end{split}\]

where

\[\begin{split}\begin{array}{ll} \text{csc_val}[8] & = \{1.0, 6.0, 2.0, 4.0, 5.0, 3.0, 7.0, 8.0\} \\ \text{csc_col_ptr}[6] & = \{1, 3, 5, 6, 8, 9\} \\ \text{csc_row_ind}[8] & = \{1, 3, 1, 2, 2, 1, 3, 3\} \end{array}\end{split}\]

BSR storage format#

The Block Compressed Sparse Row (BSR) storage format represents an \((mb \cdot \text{bsr_dim}) \times (nb \cdot \text{bsr_dim})\) matrix by:

mb

Number of block rows (integer).

nb

Number of block columns (integer).

nnzb

Number of non-zero blocks (integer).

bsr_val

Array of nnzb * bsr_dim * bsr_dim elements containing the data (floating point). Blocks can be stored in column-major or row-major format.

bsr_row_ptr

Array of mb+1 elements that point to the start of every block row (integer).

bsr_col_ind

Array of nnzb elements containing the block column indices (integer).

bsr_dim

Dimension of each block (integer).

The BSR matrix is expected to be sorted by column indices within each row. If \(m\) or \(n\) are not evenly divisible by the block dimension, then zeros are padded to the matrix, such that \(mb = (m + \text{bsr_dim} - 1) / \text{bsr_dim}\) and \(nb = (n + \text{bsr_dim} - 1) / \text{bsr_dim}\). Consider the following \(4 \times 3\) matrix and the corresponding BSR structures, with \(\text{bsr_dim} = 2, mb = 2, nb = 2\) and \(\text{nnzb} = 4\) using zero-based indexing and column-major storage:

\[\begin{split}A = \begin{pmatrix} 1.0 & 0.0 & 2.0 \\ 3.0 & 0.0 & 4.0 \\ 5.0 & 6.0 & 0.0 \\ 7.0 & 0.0 & 8.0 \\ \end{pmatrix}\end{split}\]

with the blocks \(A_{ij}\)

\[\begin{split}A_{00} = \begin{pmatrix} 1.0 & 0.0 \\ 3.0 & 0.0 \\ \end{pmatrix}, A_{01} = \begin{pmatrix} 2.0 & 0.0 \\ 4.0 & 0.0 \\ \end{pmatrix}, A_{10} = \begin{pmatrix} 5.0 & 6.0 \\ 7.0 & 0.0 \\ \end{pmatrix}, A_{11} = \begin{pmatrix} 0.0 & 0.0 \\ 8.0 & 0.0 \\ \end{pmatrix}\end{split}\]

such that

\[\begin{split}A = \begin{pmatrix} A_{00} & A_{01} \\ A_{10} & A_{11} \\ \end{pmatrix}\end{split}\]

with arrays represented as

\[\begin{split}\begin{array}{ll} \text{bsr_val}[16] & = \{1.0, 3.0, 0.0, 0.0, 2.0, 4.0, 0.0, 0.0, 5.0, 7.0, 6.0, 0.0, 0.0, 8.0, 0.0, 0.0\} \\ \text{bsr_row_ptr}[3] & = \{0, 2, 4\} \\ \text{bsr_col_ind}[4] & = \{0, 1, 0, 1\} \end{array}\end{split}\]

GEBSR storage format#

The General Block Compressed Sparse Row (GEBSR) storage format represents an \((mb \cdot \text{bsr_row_dim}) \times (nb \cdot \text{bsr_col_dim})\) matrix by:

mb

Number of block rows (integer).

nb

Number of block columns (integer).

nnzb

Number of non-zero blocks (integer).

bsr_val

Array of nnzb * bsr_row_dim * bsr_col_dim elements containing the data (floating point). Blocks can be stored in column-major or row-major format.

bsr_row_ptr

Array of mb+1 elements that point to the start of every block row (integer).

bsr_col_ind

Array of nnzb elements containing the block column indices (integer).

bsr_row_dim

Row dimension of each block (integer).

bsr_col_dim

Column dimension of each block (integer).

The GEBSR matrix is expected to be sorted by column indices within each row. If \(m\) is not evenly divisible by the row block dimension or \(n\) is not evenly divisible by the column block dimension, then zeros are padded to the matrix, such that \(mb = (m + \text{bsr_row_dim} - 1) / \text{bsr_row_dim}\) and \(nb = (n + \text{bsr_col_dim} - 1) / \text{bsr_col_dim}\). Consider the following \(4 \times 5\) matrix and the corresponding GEBSR structures, with \(\text{bsr_row_dim} = 2\), \(\text{bsr_col_dim} = 3\), \(mb = 2\), \(nb = 2\), and \(\text{nnzb} = 4\) using zero-based indexing and column-major storage:

\[\begin{split}A = \begin{pmatrix} 1.0 & 0.0 & 0.0 & 2.0 & 0.0 \\ 3.0 & 0.0 & 4.0 & 0.0 & 0.0 \\ 5.0 & 6.0 & 0.0 & 7.0 & 0.0 \\ 0.0 & 0.0 & 8.0 & 0.0 & 9.0 \\ \end{pmatrix}\end{split}\]

with the blocks \(A_{ij}\)

\[\begin{split}A_{00} = \begin{pmatrix} 1.0 & 0.0 & 0.0 \\ 3.0 & 0.0 & 4.0 \\ \end{pmatrix}, A_{01} = \begin{pmatrix} 2.0 & 0.0 & 0.0 \\ 0.0 & 0.0 & 0.0 \\ \end{pmatrix}, A_{10} = \begin{pmatrix} 5.0 & 6.0 & 0.0 \\ 0.0 & 0.0 & 8.0 \\ \end{pmatrix}, A_{11} = \begin{pmatrix} 7.0 & 0.0 & 0.0 \\ 0.0 & 9.0 & 0.0 \\ \end{pmatrix}\end{split}\]

such that

\[\begin{split}A = \begin{pmatrix} A_{00} & A_{01} \\ A_{10} & A_{11} \\ \end{pmatrix}\end{split}\]

with arrays represented as

\[\begin{split}\begin{array}{ll} \text{bsr_val}[24] & = \{1.0, 3.0, 0.0, 0.0, 0.0, 4.0, 2.0, 0.0, 0.0, 0.0, 0.0, 0.0, 5.0, 0.0, 6.0, 0.0, 0.0, 8.0, 7.0, 0.0, 0.0, 9.0, 0.0, 0.0\} \\ \text{bsr_row_ptr}[3] & = \{0, 2, 4\} \\ \text{bsr_col_ind}[4] & = \{0, 1, 0, 1\} \end{array}\end{split}\]

ELL storage format#

The Ellpack-Itpack (ELL) storage format represents an \(m \times n\) matrix by:

m

Number of rows (integer).

n

Number of columns (integer).

ell_width

Maximum number of non-zero elements per row (integer)

ell_val

Array of m * ell_width elements containing the data (floating point).

ell_col_ind

Array of m * ell_width elements containing the column indices (integer).

The ELL matrix is assumed to be stored in column-major format. Rows with less than ell_width non-zero elements are padded with zeros (ell_val) and \(-1\) (ell_col_ind). Consider the following \(3 \times 5\) matrix and the corresponding ELL structures, with \(m = 3, n = 5\), and \(\text{ell_width} = 3\) using zero-based indexing:

\[\begin{split}A = \begin{pmatrix} 1.0 & 2.0 & 0.0 & 3.0 & 0.0 \\ 0.0 & 4.0 & 5.0 & 0.0 & 0.0 \\ 6.0 & 0.0 & 0.0 & 7.0 & 8.0 \\ \end{pmatrix}\end{split}\]

where

\[\begin{split}\begin{array}{ll} \text{ell_val}[9] & = \{1.0, 4.0, 6.0, 2.0, 5.0, 7.0, 3.0, 0.0, 8.0\} \\ \text{ell_col_ind}[9] & = \{0, 1, 0, 1, 2, 3, 3, -1, 4\} \end{array}\end{split}\]

HYB storage format#

The Hybrid (HYB) storage format represents an \(m \times n\) matrix by:

m

Number of rows (integer).

n

Number of columns (integer).

nnz

Number of non-zero elements of the COO part (integer).

ell_width

Maximum number of non-zero elements per row of the ELL part (integer).

ell_val

Array of m * ell_width elements containing the ELL-part data (floating point).

ell_col_ind

Array of m * ell_width elements containing the ELL-part column indices (integer).

coo_val

Array of nnz elements containing the COO-part data (floating point).

coo_row_ind

Array of nnz elements containing the COO-part row indices (integer).

coo_col_ind

Array of nnz elements containing the COO-part column indices (integer).

The HYB format is a combination of the ELL and COO sparse matrix formats. Typically, the regular part of the matrix is stored in ELL storage format and the irregular part of the matrix is stored in COO storage format. Three different partitioning schemes can be applied when converting a CSR matrix to a matrix in HYB storage format. For further details on the partitioning schemes, see hipsparseHybPartition_t.

Storage schemes and indexing base#

hipSPARSE supports 0-based and 1-based indexing. The index base is selected by the hipsparseIndexBase_t type, which is either passed as a standalone parameter or as part of the hipsparseMatDescr_t type.

Dense vectors are represented with a 1D array, stored linearly in memory. Sparse vectors are represented by a 1D data array stored linearly in memory that holds all non-zero elements and a 1D indexing array stored linearly in memory that holds the positions of the corresponding non-zero elements.

Pointer mode#

The auxiliary functions hipsparseSetPointerMode() and hipsparseGetPointerMode() are used to set and get the value of the state variable hipsparsePointerMode_t. If hipsparsePointerMode_t is equal to HIPSPARSE_POINTER_MODE_HOST, then scalar parameters must be allocated on the host. If hipsparsePointerMode_t is equal to HIPSPARSE_POINTER_MODE_DEVICE, then scalar parameters must be allocated on the device.

There are two types of scalar parameter:

  1. Scaling parameters, such as alpha and beta, that are used, for example, in hipsparseScsrmv() and hipsparseSbsrmv()

  2. Scalar results from functions such as hipsparseSdoti() or hipsparseCdotci()

For scalar parameters such as alpha and beta, memory can be allocated on the host heap or stack when hipsparsePointerMode_t is equal to HIPSPARSE_POINTER_MODE_HOST. The kernel launch is asynchronous, and if the scalar parameter is on the heap, it can be freed after the return from the kernel launch. When hipsparsePointerMode_t is equal to HIPSPARSE_POINTER_MODE_DEVICE, the scalar parameter must not be changed until the kernel completes.

For scalar results, when hipsparsePointerMode_t is equal to HIPSPARSE_POINTER_MODE_HOST, the function blocks the CPU until the GPU has copied the result back to the host. When hipsparsePointerMode_t is equal to HIPSPARSE_POINTER_MODE_DEVICE, the function returns after the asynchronous launch. Similar to the vector and matrix results, the scalar result is only available when the kernel has completed execution.