Sparse utility functions

Contents

Sparse utility functions#

This module contains all sparse utility routines.

The sparse utility routines allow for testing whether matrix data is valid for different matrix formats. These routines do not support execution in a hipGraph context.

rocsparse_check_matrix_csr_buffer_size()#

rocsparse_status rocsparse_scheck_matrix_csr_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const float *csr_val, const rocsparse_int *csr_row_ptr, const rocsparse_int *csr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_dcheck_matrix_csr_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const double *csr_val, const rocsparse_int *csr_row_ptr, const rocsparse_int *csr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_ccheck_matrix_csr_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const rocsparse_float_complex *csr_val, const rocsparse_int *csr_row_ptr, const rocsparse_int *csr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_zcheck_matrix_csr_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const rocsparse_double_complex *csr_val, const rocsparse_int *csr_row_ptr, const rocsparse_int *csr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#

rocsparse_check_matrix_csr_buffer_size computes the required buffer size needed when calling rocsparse_Xcheck_matrix_csr().

Note

This routine does not support execution in a hipGraph context.

Parameters:
Return values:

  • rocsparse_status_success – the operation completed successfully.

  • rocsparse_status_invalid_handle – the library context was not initialized.

  • rocsparse_status_invalid_value – idx_base or matrix_type or uplo or storage is invalid.

  • rocsparse_status_invalid_size – m n or nnz is invalid.

  • rocsparse_status_invalid_pointer – csr_val, csr_row_ptr, csr_col_ind or buffer_size pointer is invalid.

rocsparse_check_matrix_csr()#

rocsparse_status rocsparse_scheck_matrix_csr(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const float *csr_val, const rocsparse_int *csr_row_ptr, const rocsparse_int *csr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_dcheck_matrix_csr(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const double *csr_val, const rocsparse_int *csr_row_ptr, const rocsparse_int *csr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_ccheck_matrix_csr(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const rocsparse_float_complex *csr_val, const rocsparse_int *csr_row_ptr, const rocsparse_int *csr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_zcheck_matrix_csr(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const rocsparse_double_complex *csr_val, const rocsparse_int *csr_row_ptr, const rocsparse_int *csr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#

Check matrix to see if it is valid.

rocsparse_check_matrix_csr checks if the input CSR matrix is valid. It performs basic sanity checks on the input matrix and tries to detect issues in the data. This includes looking for ‘nan’ or ‘inf’ values in the data arrays, invalid column indices or row offsets, whether the matrix is triangular or not, whether there are duplicate indices or whether the column indices are not sorted when they should be. If an issue is found, it is written to the data_status parameter.

Performing the above checks involves two steps. First the user calls rocsparse_Xcheck_matrix_csr_buffer_size in order to determine the required buffer size. The user then allocates this buffer and passes it to rocsparse_Xcheck_matrix_csr. Any issues detected will be written to the data_status parameter which is always a host variable regardless of pointer mode.

// 1 2 0 0
// 0 3 4 0
// 2 0 1 1
// 0 3 0 2
std::vector<int> hcsr_row_ptr = {0, 2, -1, 7, 9}; // <---- invalid ptr array
std::vector<int> hcsr_col_ind = {0, 1, 1, 2, 0, 2, 3, 1, 3};
std::vector<float> hcsr_val = {1, 2, 3, 4, 2, 1, 1, 3, 2};

int m = 4;
int n = 4;
int nnz = 9;

int* dcsr_row_ptr = nullptr;
int* dcsr_col_ind = nullptr;
float* dcsr_val = nullptr;
hipMalloc((void**)&dcsr_row_ptr, sizeof(int) * (m + 1));
hipMalloc((void**)&dcsr_col_ind, sizeof(int) * nnz);
hipMalloc((void**)&dcsr_val, sizeof(float) * nnz);

hipMemcpy(dcsr_row_ptr, hcsr_row_ptr.data(), sizeof(int) * (m + 1), hipMemcpyHostToDevice);
hipMemcpy(dcsr_col_ind, hcsr_col_ind.data(), sizeof(int) * nnz, hipMemcpyHostToDevice);
hipMemcpy(dcsr_val, hcsr_val.data(), sizeof(float) * nnz, hipMemcpyHostToDevice);

rocsparse_handle handle;
rocsparse_create_handle(&handle);

const rocsparse_index_base idx_base = rocsparse_index_base_zero;
const rocsparse_fill_mode fill_mode = rocsparse_fill_mode_upper;
const rocsparse_matrix_type matrix_type = rocsparse_matrix_type_general;
const rocsparse_storage_mode storage_mode = rocsparse_storage_mode_sorted;

rocsparse_data_status data_status;

size_t buffer_size;
rocsparse_scheck_matrix_csr_buffer_size(handle, m, n, nnz, dcsr_val, dcsr_row_ptr, dcsr_col_ind,
    idx_base, matrix_type, fill_mode, storage_mode, &buffer_size);

void* dbuffer = nullptr;
hipMalloc((void**)&dbuffer, buffer_size);

rocsparse_scheck_matrix_csr(handle, m, n, nnz, dcsr_val, dcsr_row_ptr, dcsr_col_ind, idx_base,
    matrix_type, fill_mode, storage_mode, &data_status, dbuffer);

std::cout << "data_status: " << data_status << std::endl;

hipFree(dbuffer);

rocsparse_destroy_handle(handle);

hipFree(dcsr_row_ptr);
hipFree(dcsr_col_ind);
hipFree(dcsr_val);
Example

In this example we want to check whether a CSR matrix has correct row pointer array. The matrix passed is invalid because it contains a -1 entry in the row pointer array.

Note

This routine does not support execution in a hipGraph context.

Parameters:
Return values:

  • rocsparse_status_success – the operation completed successfully.

  • rocsparse_status_invalid_handle – the library context was not initialized.

  • rocsparse_status_invalid_value – idx_base or matrix_type or uplo or storage is invalid.

  • rocsparse_status_invalid_size – m n or nnz is invalid.

  • rocsparse_status_invalid_pointer – csr_val, csr_row_ptr, csr_col_ind, temp_buffer or data_status pointer is invalid.

rocsparse_check_matrix_csc_buffer_size()#

rocsparse_status rocsparse_scheck_matrix_csc_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const float *csc_val, const rocsparse_int *csc_col_ptr, const rocsparse_int *csc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_dcheck_matrix_csc_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const double *csc_val, const rocsparse_int *csc_col_ptr, const rocsparse_int *csc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_ccheck_matrix_csc_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const rocsparse_float_complex *csc_val, const rocsparse_int *csc_col_ptr, const rocsparse_int *csc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_zcheck_matrix_csc_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const rocsparse_double_complex *csc_val, const rocsparse_int *csc_col_ptr, const rocsparse_int *csc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#

rocsparse_check_matrix_csc_buffer_size computes the required buffer size needed when calling rocsparse_Xcheck_matrix_csc().

Note

This routine does not support execution in a hipGraph context.

Parameters:
Return values:

  • rocsparse_status_success – the operation completed successfully.

  • rocsparse_status_invalid_handle – the library context was not initialized.

  • rocsparse_status_invalid_value – idx_base or matrix_type or uplo or storage is invalid.

  • rocsparse_status_invalid_size – m n or nnz is invalid.

  • rocsparse_status_invalid_pointer – csc_val, csc_col_ptr, csc_row_ind or buffer_size pointer is invalid.

rocsparse_check_matrix_csc()#

rocsparse_status rocsparse_scheck_matrix_csc(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const float *csc_val, const rocsparse_int *csc_col_ptr, const rocsparse_int *csc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_dcheck_matrix_csc(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const double *csc_val, const rocsparse_int *csc_col_ptr, const rocsparse_int *csc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_ccheck_matrix_csc(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const rocsparse_float_complex *csc_val, const rocsparse_int *csc_col_ptr, const rocsparse_int *csc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_zcheck_matrix_csc(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const rocsparse_double_complex *csc_val, const rocsparse_int *csc_col_ptr, const rocsparse_int *csc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#

Check matrix to see if it is valid.

rocsparse_check_matrix_csc checks if the input CSC matrix is valid. It performs basic sanity checks on the input matrix and tries to detect issues in the data. This includes looking for ‘nan’ or ‘inf’ values in the data arrays, invalid row indices or invalid column offsets, whether the matrix is triangular or not, whether there are duplicate row indices or whether the row indices are not sorted when they should be. If an issue is found, it is written to the data_status parameter.

Performing the above checks involves two steps. First the user calls rocsparse_Xcheck_matrix_csc_buffer_size in order to determine the required buffer size. The user then allocates this buffer and passes it to rocsparse_Xcheck_matrix_csc. Any issues detected will be written to the data_status parameter which is always a host variable regardless of pointer mode.

// 1 2 0 0
// 0 3 4 0
// 2 0 1 1
// 0 3 0 2
std::vector<int> hcsc_row_ind = {0, 2, 0, 1, 1, 1, 2, 2, 3}; //<---duplicate row index in second column
std::vector<int> hcsc_col_ptr = {0, 2, 5, 7, 9};
std::vector<float> hcsc_val = {1, 2, 2, 3, 3, 4, 1, 1, 2};

int m = 4;
int n = 4;
int nnz = 9;

int* dcsc_row_ind = nullptr;
int* dcsc_col_ptr = nullptr;
float* dcsc_val = nullptr;
hipMalloc((void**)&dcsc_row_ind, sizeof(int) * nnz);
hipMalloc((void**)&dcsc_col_ptr, sizeof(int) * (n + 1));
hipMalloc((void**)&dcsc_val, sizeof(float) * nnz);

hipMemcpy(dcsc_row_ind, hcsc_row_ind.data(), sizeof(int) * nnz, hipMemcpyHostToDevice);
hipMemcpy(dcsc_col_ptr, hcsc_col_ptr.data(), sizeof(int) * (n + 1), hipMemcpyHostToDevice);
hipMemcpy(dcsc_val, hcsc_val.data(), sizeof(float) * nnz, hipMemcpyHostToDevice);

rocsparse_handle handle;
rocsparse_create_handle(&handle);

const rocsparse_index_base idx_base = rocsparse_index_base_zero;
const rocsparse_fill_mode fill_mode = rocsparse_fill_mode_upper;
const rocsparse_matrix_type matrix_type = rocsparse_matrix_type_general;
const rocsparse_storage_mode storage_mode = rocsparse_storage_mode_sorted;

rocsparse_data_status data_status;

size_t buffer_size;
rocsparse_scheck_matrix_csc_buffer_size(handle, m, n, nnz, dcsc_val, dcsc_col_ptr, dcsc_row_ind,
    idx_base, matrix_type, fill_mode, storage_mode, &buffer_size);

void* dbuffer = nullptr;
hipMalloc((void**)&dbuffer, buffer_size);

rocsparse_scheck_matrix_csc(handle, m, n, nnz, dcsc_val, dcsc_col_ptr, dcsc_row_ind, idx_base,
    matrix_type, fill_mode, storage_mode, &data_status, dbuffer);

std::cout << "data_status: " << data_status << std::endl;

hipFree(dbuffer);

rocsparse_destroy_handle(handle);

hipFree(dcsc_row_ind);
hipFree(dcsc_col_ptr);
hipFree(dcsc_val);
Example

In this example we want to check whether a CSC matrix has correct row indices. The matrix passed is invalid because it contains a duplicate entry in the row indices array.

Note

This routine does not support execution in a hipGraph context.

Parameters:
Return values:

  • rocsparse_status_success – the operation completed successfully.

  • rocsparse_status_invalid_handle – the library context was not initialized.

  • rocsparse_status_invalid_value – idx_base or matrix_type or uplo or storage is invalid.

  • rocsparse_status_invalid_size – m n or nnz is invalid.

  • rocsparse_status_invalid_pointer – csc_val, csc_col_ptr, csc_row_ind, temp_buffer or data_status pointer is invalid.

rocsparse_check_matrix_coo_buffer_size()#

rocsparse_status rocsparse_scheck_matrix_coo_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const float *coo_val, const rocsparse_int *coo_row_ind, const rocsparse_int *coo_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_dcheck_matrix_coo_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const double *coo_val, const rocsparse_int *coo_row_ind, const rocsparse_int *coo_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_ccheck_matrix_coo_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const rocsparse_float_complex *coo_val, const rocsparse_int *coo_row_ind, const rocsparse_int *coo_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_zcheck_matrix_coo_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const rocsparse_double_complex *coo_val, const rocsparse_int *coo_row_ind, const rocsparse_int *coo_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#

rocsparse_check_matrix_coo_buffer_size computes the required buffer size needed when calling rocsparse_Xcheck_matrix_coo().

Note

This routine does not support execution in a hipGraph context.

Parameters:
Return values:

  • rocsparse_status_success – the operation completed successfully.

  • rocsparse_status_invalid_handle – the library context was not initialized.

  • rocsparse_status_invalid_value – idx_base or matrix_type or uplo or storage is invalid.

  • rocsparse_status_invalid_size – m n or nnz is invalid.

  • rocsparse_status_invalid_pointer – coo_val, coo_row_ind, coo_col_ind or buffer_size pointer is invalid.

rocsparse_check_matrix_coo()#

rocsparse_status rocsparse_scheck_matrix_coo(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const float *coo_val, const rocsparse_int *coo_row_ind, const rocsparse_int *coo_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_dcheck_matrix_coo(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const double *coo_val, const rocsparse_int *coo_row_ind, const rocsparse_int *coo_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_ccheck_matrix_coo(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const rocsparse_float_complex *coo_val, const rocsparse_int *coo_row_ind, const rocsparse_int *coo_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_zcheck_matrix_coo(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int nnz, const rocsparse_double_complex *coo_val, const rocsparse_int *coo_row_ind, const rocsparse_int *coo_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#

Check matrix to see if it is valid.

rocsparse_check_matrix_coo checks if the input COO matrix is valid. It performs basic sanity checks on the input matrix and tries to detect issues in the data. This includes looking for ‘nan’ or ‘inf’ values in the data arrays, invalid row/column indices, whether the matrix is triangular or not, whether there are duplicate row/column indices or whether the row/column indices are not sorted when they should be. If an issue is found, it is written to the data_status parameter.

Performing the above checks involves two steps. First the user calls rocsparse_Xcheck_matrix_coo_buffer_size in order to determine the required buffer size. The user then allocates this buffer and passes it to rocsparse_Xcheck_matrix_coo. Any issues detected will be written to the data_status parameter which is always a host variable regardless of pointer mode.

// 1 2 0 0
// 0 3 4 0
// 0 0 1 1
// 0 0 0 2
std::vector<int> hcoo_row_ind = {0, 0, -1, 1, 2, 2, 3}; // <---- invalid row index
std::vector<int> hcoo_col_ind = {0, 1, 1, 2, 2, 3, 3};
std::vector<float> hcoo_val = {1, 2, 3, 4, 1, 1, 2};

int m = 4;
int n = 4;
int nnz = 7;

int* dcoo_row_ind = nullptr;
int* dcoo_col_ind = nullptr;
float* dcoo_val = nullptr;
hipMalloc((void**)&dcoo_row_ind, sizeof(int) * nnz);
hipMalloc((void**)&dcoo_col_ind, sizeof(int) * nnz);
hipMalloc((void**)&dcoo_val, sizeof(float) * nnz);

hipMemcpy(dcoo_row_ind, hcoo_row_ind.data(), sizeof(int) * nnz, hipMemcpyHostToDevice);
hipMemcpy(dcoo_col_ind, hcoo_col_ind.data(), sizeof(int) * nnz, hipMemcpyHostToDevice);
hipMemcpy(dcoo_val, hcoo_val.data(), sizeof(float) * nnz, hipMemcpyHostToDevice);

rocsparse_handle handle;
rocsparse_create_handle(&handle);

const rocsparse_index_base idx_base = rocsparse_index_base_zero;
const rocsparse_fill_mode fill_mode = rocsparse_fill_mode_upper;
const rocsparse_matrix_type matrix_type = rocsparse_matrix_type_triangular;
const rocsparse_storage_mode storage_mode = rocsparse_storage_mode_sorted;

rocsparse_data_status data_status;

size_t buffer_size;
rocsparse_scheck_matrix_coo_buffer_size(handle, m, n, nnz, dcoo_val, dcoo_row_ind, dcoo_col_ind,
    idx_base, matrix_type, fill_mode, storage_mode, &buffer_size);

void* dbuffer = nullptr;
hipMalloc((void**)&dbuffer, buffer_size);

rocsparse_scheck_matrix_coo(handle, m, n, nnz, dcoo_val, dcoo_row_ind, dcoo_col_ind, idx_base,
    matrix_type, fill_mode, storage_mode, &data_status, dbuffer);

std::cout << "data_status: " << data_status << std::endl;

hipFree(dbuffer);

rocsparse_destroy_handle(handle);

hipFree(dcoo_row_ind);
hipFree(dcoo_col_ind);
hipFree(dcoo_val);
Example

In this example we want to check whether a COO matrix has correct row indices. The matrix passed is invalid because it contains a -1 entry in the row indices array.

Note

This routine does not support execution in a hipGraph context.

Parameters:
Return values:

  • rocsparse_status_success – the operation completed successfully.

  • rocsparse_status_invalid_handle – the library context was not initialized.

  • rocsparse_status_invalid_value – idx_base or matrix_type or uplo or storage is invalid.

  • rocsparse_status_invalid_size – m n or nnz is invalid.

  • rocsparse_status_invalid_pointer – coo_val, coo_row_ind, coo_col_ind, temp_buffer or data_status pointer is invalid.

rocsparse_check_matrix_gebsr_buffer_size()#

rocsparse_status rocsparse_scheck_matrix_gebsr_buffer_size(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const float *bsr_val, const rocsparse_int *bsr_row_ptr, const rocsparse_int *bsr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_dcheck_matrix_gebsr_buffer_size(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const double *bsr_val, const rocsparse_int *bsr_row_ptr, const rocsparse_int *bsr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_ccheck_matrix_gebsr_buffer_size(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const rocsparse_float_complex *bsr_val, const rocsparse_int *bsr_row_ptr, const rocsparse_int *bsr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_zcheck_matrix_gebsr_buffer_size(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const rocsparse_double_complex *bsr_val, const rocsparse_int *bsr_row_ptr, const rocsparse_int *bsr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#

Check matrix to see if it is valid.

rocsparse_check_matrix_gebsr_buffer_size computes the required buffer size needed when calling rocsparse_Xcheck_matrix_gebsr().

Note

This routine does not support execution in a hipGraph context.

Parameters:
Return values:

  • rocsparse_status_success – the operation completed successfully.

  • rocsparse_status_invalid_handle – the library context was not initialized.

  • rocsparse_status_invalid_value – dir or idx_base or matrix_type or uplo or storage is invalid.

  • rocsparse_status_invalid_size – mb nb nnzb row_block_dim or col_block_dim is invalid.

  • rocsparse_status_invalid_pointer – bsr_val, bsr_row_ptr, bsr_col_ind or buffer_size pointer is invalid.

rocsparse_check_matrix_gebsr()#

rocsparse_status rocsparse_scheck_matrix_gebsr(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const float *bsr_val, const rocsparse_int *bsr_row_ptr, const rocsparse_int *bsr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_dcheck_matrix_gebsr(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const double *bsr_val, const rocsparse_int *bsr_row_ptr, const rocsparse_int *bsr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_ccheck_matrix_gebsr(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const rocsparse_float_complex *bsr_val, const rocsparse_int *bsr_row_ptr, const rocsparse_int *bsr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_zcheck_matrix_gebsr(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const rocsparse_double_complex *bsr_val, const rocsparse_int *bsr_row_ptr, const rocsparse_int *bsr_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#

Check matrix to see if it is valid.

rocsparse_check_matrix_gebsr checks if the input GEBSR matrix is valid. It performs basic sanity checks on the input matrix and tries to detect issues in the data. This includes looking for ‘nan’ or ‘inf’ values in the data arrays, invalid column indices and row offsets, whether the matrix is triangular or not, whether there are duplicate indices or whether the column indices are not sorted when they should be. If an issue is found, it is written to the data_status parameter.

Performing the above checks involves two steps. First the user calls rocsparse_Xcheck_matrix_gebsr_buffer_size in order to determine the required buffer size. The user then allocates this buffer and passes it to rocsparse_Xcheck_matrix_gebsr. Any issues detected will be written to the data_status parameter which is always a host variable regardless of pointer mode.

// 1 2 | 0 0
// 0 3 | 0 0
// ---------
// 4 5 | 7 8
// 0 6 | 0 9
std::vector<int> hbsr_row_ptr = {0, 1, 3};
std::vector<int> hbsr_col_ind = {0, 0, 1};
std::vector<float> hbsr_val = {1, 2, 0, 3, 4, 5, 0, 6, 7, 8, std::numeric_limits<double>::quiet_NaN(), 9}; //<---contains nan

int mb = 2;
int nb = 2;
int nnzb = 3;
int block_dim = 2;

int* dbsr_row_ptr = nullptr;
int* dbsr_col_ind = nullptr;
float* dbsr_val = nullptr;
hipMalloc((void**)&dbsr_row_ptr, sizeof(int) * (mb + 1));
hipMalloc((void**)&dbsr_col_ind, sizeof(int) * nnzb);
hipMalloc((void**)&dbsr_val, sizeof(float) * nnzb * block_dim * block_dim);

hipMemcpy(dbsr_row_ptr, hbsr_row_ptr.data(), sizeof(int) * (mb + 1), hipMemcpyHostToDevice);
hipMemcpy(dbsr_col_ind, hbsr_col_ind.data(), sizeof(int) * nnzb, hipMemcpyHostToDevice);
hipMemcpy(dbsr_val, hbsr_val.data(), sizeof(float) * nnzb * block_dim * block_dim, hipMemcpyHostToDevice);

rocsparse_handle handle;
rocsparse_create_handle(&handle);

const rocsparse_direction direction = rocsparse_direction_row;
const rocsparse_index_base idx_base = rocsparse_index_base_zero;
const rocsparse_fill_mode fill_mode = rocsparse_fill_mode_upper;
const rocsparse_matrix_type matrix_type = rocsparse_matrix_type_triangular;
const rocsparse_storage_mode storage_mode = rocsparse_storage_mode_sorted;

rocsparse_data_status data_status;

size_t buffer_size;
rocsparse_scheck_matrix_gebsr_buffer_size(handle, direction, mb, nb, nnzb, block_dim, block_dim,
    dbsr_val, dbsr_row_ptr, dbsr_col_ind, idx_base, matrix_type, fill_mode, storage_mode, &buffer_size);

void* dbuffer = nullptr;
hipMalloc((void**)&dbuffer, buffer_size);

rocsparse_scheck_matrix_gebsr(handle, direction, mb, nb, nnzb, block_dim, block_dim, dbsr_val, dbsr_row_ptr,
    dbsr_col_ind, idx_base, matrix_type, fill_mode, storage_mode, &data_status, dbuffer);

std::cout << "data_status: " << data_status << std::endl;

hipFree(dbuffer);

rocsparse_destroy_handle(handle);

hipFree(dbsr_row_ptr);
hipFree(dbsr_col_ind);
hipFree(dbsr_val);
Example

In this example we want to check whether a GEBSR matrix has valid values. The matrix passed is invalid because it contains a nan entry in the values array.

Note

This routine does not support execution in a hipGraph context.

Parameters:
Return values:

  • rocsparse_status_success – the operation completed successfully.

  • rocsparse_status_invalid_handle – the library context was not initialized.

  • rocsparse_status_invalid_value – dir or idx_base or matrix_type or uplo or storage is invalid.

  • rocsparse_status_invalid_size – mb nb nnzb row_block_dim or col_block_dim is invalid.

  • rocsparse_status_invalid_pointer – bsr_val, bsr_row_ptr, bsr_col_ind, temp_buffer or data_status pointer is invalid.

rocsparse_check_matrix_gebsc_buffer_size()#

rocsparse_status rocsparse_scheck_matrix_gebsc_buffer_size(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const float *bsc_val, const rocsparse_int *bsc_col_ptr, const rocsparse_int *bsc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_dcheck_matrix_gebsc_buffer_size(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const double *bsc_val, const rocsparse_int *bsc_col_ptr, const rocsparse_int *bsc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_ccheck_matrix_gebsc_buffer_size(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const rocsparse_float_complex *bsc_val, const rocsparse_int *bsc_col_ptr, const rocsparse_int *bsc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_zcheck_matrix_gebsc_buffer_size(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const rocsparse_double_complex *bsc_val, const rocsparse_int *bsc_col_ptr, const rocsparse_int *bsc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#

rocsparse_check_matrix_gebsc_buffer_size computes the required buffer size needed when calling rocsparse_Xcheck_matrix_gebsc().

Note

This routine does not support execution in a hipGraph context.

Parameters:
Return values:

  • rocsparse_status_success – the operation completed successfully.

  • rocsparse_status_invalid_handle – the library context was not initialized.

  • rocsparse_status_invalid_value – dir or idx_base or matrix_type or uplo or storage is invalid.

  • rocsparse_status_invalid_size – mb nb nnzb row_block_dim or col_block_dim is invalid.

  • rocsparse_status_invalid_pointer – bsc_val, bsc_col_ptr, bsc_row_ind or buffer_size pointer is invalid.

rocsparse_check_matrix_gebsc()#

rocsparse_status rocsparse_scheck_matrix_gebsc(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const float *bsc_val, const rocsparse_int *bsc_col_ptr, const rocsparse_int *bsc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_dcheck_matrix_gebsc(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const double *bsc_val, const rocsparse_int *bsc_col_ptr, const rocsparse_int *bsc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_ccheck_matrix_gebsc(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const rocsparse_float_complex *bsc_val, const rocsparse_int *bsc_col_ptr, const rocsparse_int *bsc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_zcheck_matrix_gebsc(rocsparse_handle handle, rocsparse_direction dir, rocsparse_int mb, rocsparse_int nb, rocsparse_int nnzb, rocsparse_int row_block_dim, rocsparse_int col_block_dim, const rocsparse_double_complex *bsc_val, const rocsparse_int *bsc_col_ptr, const rocsparse_int *bsc_row_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#

Check matrix to see if it is valid.

rocsparse_check_matrix_gebsc checks if the input GEBSC matrix is valid. It performs basic sanity checks on the input matrix and tries to detect issues in the data. This includes looking for ‘nan’ or ‘inf’ values in the data arrays, invalid row indices or column offsets, whether the matrix is triangular or not, whether there are duplicate indices or whether the row indices are not sorted when they should be. If an issue is found, it is written to the data_status parameter.

Performing the above checks involves two steps. First the user calls rocsparse_Xcheck_matrix_gebsr_buffer_size in order to determine the required buffer size. The user then allocates this buffer and passes it to rocsparse_Xcheck_matrix_gebsr. Any issues detected will be written to the data_status parameter which is always a host variable regardless of pointer mode.

Note

This routine does not support execution in a hipGraph context.

Parameters:
Return values:

  • rocsparse_status_success – the operation completed successfully.

  • rocsparse_status_invalid_handle – the library context was not initialized.

  • rocsparse_status_invalid_value – dir or idx_base or matrix_type or uplo or storage is invalid.

  • rocsparse_status_invalid_size – mb nb nnzb row_block_dim or col_block_dim is invalid.

  • rocsparse_status_invalid_pointer – bsc_val, bsc_col_ptr, bsc_row_ind, temp_buffer or data_status pointer is invalid.

rocsparse_check_matrix_ell_buffer_size()#

rocsparse_status rocsparse_scheck_matrix_ell_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int ell_width, const float *ell_val, const rocsparse_int *ell_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_dcheck_matrix_ell_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int ell_width, const double *ell_val, const rocsparse_int *ell_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_ccheck_matrix_ell_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int ell_width, const rocsparse_float_complex *ell_val, const rocsparse_int *ell_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#
rocsparse_status rocsparse_zcheck_matrix_ell_buffer_size(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int ell_width, const rocsparse_double_complex *ell_val, const rocsparse_int *ell_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#

rocsparse_check_matrix_ell_buffer_size computes the required buffer size needed when calling rocsparse_Xcheck_matrix_ell().

Note

This routine does not support execution in a hipGraph context.

Parameters:
Return values:

  • rocsparse_status_success – the operation completed successfully.

  • rocsparse_status_invalid_handle – the library context was not initialized.

  • rocsparse_status_invalid_value – idx_base or matrix_type or uplo or storage is invalid.

  • rocsparse_status_invalid_size – m n or ell_width is invalid.

  • rocsparse_status_invalid_pointer – ell_val, ell_col_ind or buffer_size pointer is invalid.

rocsparse_check_matrix_ell()#

rocsparse_status rocsparse_scheck_matrix_ell(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int ell_width, const float *ell_val, const rocsparse_int *ell_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_dcheck_matrix_ell(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int ell_width, const double *ell_val, const rocsparse_int *ell_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_ccheck_matrix_ell(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int ell_width, const rocsparse_float_complex *ell_val, const rocsparse_int *ell_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#
rocsparse_status rocsparse_zcheck_matrix_ell(rocsparse_handle handle, rocsparse_int m, rocsparse_int n, rocsparse_int ell_width, const rocsparse_double_complex *ell_val, const rocsparse_int *ell_col_ind, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#

Check matrix to see if it is valid.

rocsparse_check_matrix_ell checks if the input ELL matrix is valid. It performs basic sanity checks on the input matrix and tries to detect issues in the data. This includes looking for ‘nan’ or ‘inf’ values in the data arrays, invalid column indices, whether there are duplicate indices or whether the column indices are not sorted when they should be. If an issue is found, it is written to the data_status parameter.

Performing the above checks involves two steps. First the user calls rocsparse_Xcheck_matrix_ell_buffer_size in order to determine the required buffer size. The user then allocates this buffer and passes it to rocsparse_Xcheck_matrix_ell. Any issues detected will be written to the data_status parameter which is always a host variable regardless of pointer mode.

Note

This routine does not support execution in a hipGraph context.

Parameters:
Return values:

  • rocsparse_status_success – the operation completed successfully.

  • rocsparse_status_invalid_handle – the library context was not initialized.

  • rocsparse_status_invalid_value – idx_base or matrix_type or uplo or storage is invalid.

  • rocsparse_status_invalid_size – m n or ell_width is invalid.

  • rocsparse_status_invalid_pointer – ell_val, ell_col_ind, temp_buffer or data_status pointer is invalid.

rocsparse_check_matrix_hyb_buffer_size()#

rocsparse_status rocsparse_check_matrix_hyb_buffer_size(rocsparse_handle handle, const rocsparse_hyb_mat hyb, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, size_t *buffer_size)#

Check matrix to see if it is valid.

rocsparse_check_matrix_hyb_buffer_size computes the required buffer size needed when calling rocsparse_check_matrix_hyb.

Note

This routine does not support execution in a hipGraph context.

Parameters:
Return values:

  • rocsparse_status_success – the operation completed successfully.

  • rocsparse_status_invalid_handle – the library context was not initialized.

  • rocsparse_status_invalid_value – idx_base or matrix_type or uplo or storage is invalid.

  • rocsparse_status_invalid_pointer – hyb or buffer_size pointer is invalid.

rocsparse_check_matrix_hyb()#

rocsparse_status rocsparse_check_matrix_hyb(rocsparse_handle handle, const rocsparse_hyb_mat hyb, rocsparse_index_base idx_base, rocsparse_matrix_type matrix_type, rocsparse_fill_mode uplo, rocsparse_storage_mode storage, rocsparse_data_status *data_status, void *temp_buffer)#

Check matrix to see if it is valid.

rocsparse_check_matrix_hyb checks if the input HYB matrix is valid. It performs basic sanity checks on the input matrix and tries to detect issues in the data. This includes looking for ‘nan’ or ‘inf’ values in the data arrays, invalid row/column indices, whether the matrix is triangular or not, whether there are duplicate indices or whether the row/column indices are not sorted when they should be. If an issue is found, it is written to the data_status parameter.

Performing the above checks involves two steps. First the user calls rocsparse_Xcheck_matrix_hyb_buffer_size in order to determine the required buffer size. The user then allocates this buffer and passes it to rocsparse_Xcheck_matrix_hyb. Any issues detected will be written to the data_status parameter which is always a host variable regardless of pointer mode.

Note

This routine does not support execution in a hipGraph context.

Parameters:
Return values:

  • rocsparse_status_success – the operation completed successfully.

  • rocsparse_status_invalid_handle – the library context was not initialized.

  • rocsparse_status_invalid_value – idx_base or matrix_type or uplo or storage is invalid.

  • rocsparse_status_invalid_pointer – hyb or data_status pointer is invalid.

Contents