Miscellaneous - 5.2 English - 68552

AOCL API Guide (68552)

Document ID
68552
Release Date
2025-12-29
Version
5.2 English
enum aoclsparse_operation#

Indicate the operation type performed on a matrix.

Values:

enumerator aoclsparse_operation_none#

No operation is performed on the matrix.

enumerator aoclsparse_operation_transpose#

Operate with transpose.

enumerator aoclsparse_operation_conjugate_transpose#

Operate with conjugate transpose.

typedef struct _aoclsparse_itsol_handle *aoclsparse_itsol_handle#

Optimization handle.

This type of handle is a container box for storing problem data and optional parameter values. it must be initialized using aoclsparse_itsol_s_init, and should be destroyed after using it with aoclsparse_itsol_destroy. For double precision data types use aoclsparse_itsol_d_init.

For more details, refer to Solver chapter introduction Iterative Solver Suite (itsol).

enum aoclsparse_ilu_type#

Specify the type of Incomplete LU (ILU) factorization.

Indicates the type of factorization to perform.

Values:

enumerator aoclsparse_ilu0#

Incomplete LU with zero fill-in, ILU(0).

enumerator aoclsparse_ilup#

Incomplete LU with thresholding, ILU(p). Not implemented in this release.

enum aoclsparse_request#

Request stages for API that perform sparse matrix products.

This list describes the possible request types used by matrix product kernels such as aoclsparse_csr2m.

Values:

enumerator aoclsparse_stage_nnz_count#

Perform only first stage of analysis and computation. No result is returned but it is useful when optimizing for multiple calls.

enumerator aoclsparse_stage_finalize#

Perform computation. After this stage the product result is returned. Needs to follow after a call with aoclsparse_stage_nnz_count request.

enumerator aoclsparse_stage_full_computation#

Indicates to perform the entire computation in a single call.

enum aoclsparse_sor_type#

List of successive over-relaxation types.

This is a list of supported SOR types that are supported by aoclsparse_dsorv (or other variants function).

Values:

enumerator aoclsparse_sor_forward#

Forward sweep.

enumerator aoclsparse_sor_backward#

Backward sweep.

enumerator aoclsparse_sor_symmetric#

Symmetric preconditioner.

enum aoclsparse_memory_usage#

List of memory utilization policy.

This is a list of supported aoclsparse_memory_usage() types that are used by optimization routine.

Values:

enumerator aoclsparse_memory_usage_minimal#

Allocate memory only for auxiliary structures.

enumerator aoclsparse_memory_usage_unrestricted#

Allocate memory upto matrix size for appropriate sparse format conversion. Default value.