Developer Reference for Intel® oneAPI Math Kernel Library for C
ID
766684
Date
4/28/2026
Public
Getting Help and Support
What’s New
Notational Conventions
oneMKL Library Overview
OpenMP* Offload
BLAS Routines
Sparse BLAS Routines
LAPACK Routines
ScaLAPACK Routines
Sparse Solver Routines
Extended Eigensolver Routines
Vector Mathematical Functions
Statistical Functions
Fourier Transform Functions
PBLAS Routines
Partial Differential Equations Support
Nonlinear Optimization Problem Solvers
Support Functions
BLACS Routines
Data Fitting Functions
Appendix A: Linear Solvers Basics
Appendix B: Routine and Function Arguments
Appendix C: FFTW Interface to Intel® oneMKL
Appendix D: Code Examples
Appendix E: oneMKL Functionality
Bibliography
Glossary
Notices and Disclaimers
?axpby
?axpy_batch
?axpy_batch_strided
?copy_batch
?copy_batch_strided
?dgmm_batch
?dgmm_batch_strided
?gemm_batch
?gemm_batch_strided
gemm_bf16bf16f32
gemm_bf16bf16f32_compute
gemm_f16f16f32
gemm_f16f16f32_compute
gemm_e5m2e5m2f32
gemm_e5m2e5m2f32_compute
gemm_e4m3e4m3f32
gemm_e4m3e4m3f32_compute
gemm_*
?gemm_compute
gemm_*_compute
?gemm_pack
gemm_*_pack
?gemm_pack_get_size , gemm_*_pack_get_size
?gemm3m
?gemm3m_batch
?gemm3m_batch_strided
?gemmt
?gemv_batch
?gemv_batch_strided
?trsm_batch
?trsm_batch_strided
mkl_?imatcopy
mkl_?imatcopy_batch
mkl_?imatcopy_batch_strided
mkl_?omatadd
mkl_?omatadd_batch_strided
mkl_?omatcopy
mkl_?omatcopy_batch
mkl_?omatcopy_batch_strided
mkl_?omatcopy2
mkl_jit_create_?gemm
mkl_jit_destroy
mkl_jit_get_?gemm_ptr
Naming Conventions in Inspector-Executor Sparse BLAS Routines
Sparse Matrix Storage Formats for Inspector-executor Sparse BLAS Routines
Supported Inspector-Executor Sparse BLAS Operations
Two-stage Algorithm in Inspector-Executor Sparse BLAS Routines
Sparse BLAS Reference Types
Matrix Manipulation Routines
Inspector-Executor Sparse BLAS Analysis Routines
Inspector-Executor Sparse BLAS Execution Routines
mkl_sparse_?_create_csr
mkl_sparse_?_create_csc
mkl_sparse_?_create_coo
mkl_sparse_?_create_bsr
mkl_sparse_copy
mkl_sparse_destroy
mkl_sparse_convert_csr
mkl_sparse_convert_csc
mkl_sparse_convert_coo
mkl_sparse_convert_bsr
mkl_sparse_?_dense2csr
mkl_sparse_?_dense2csc
mkl_sparse_?_dense2coo
mkl_sparse_?_dense2bsr
mkl_sparse_?_convert_dense
mkl_sparse_?_export_csr
mkl_sparse_?_export_csc
mkl_sparse_?_export_coo
mkl_sparse_?_export_bsr
mkl_sparse_?_set_value
mkl_sparse_?_update_values
mkl_sparse_order
mkl_sparse_?_lu_smoother
mkl_sparse_?_mv
mkl_sparse_?_trsv
mkl_sparse_?_mm
mkl_sparse_?_trsm
mkl_sparse_?_add
mkl_sparse_spmm
mkl_sparse_?_spmmd
mkl_sparse_sp2m
mkl_sparse_?_sp2md
mkl_sparse_sypr
mkl_sparse_?_syprd
mkl_sparse_?_symgs
mkl_sparse_?_symgs_mv
mkl_sparse_syrk
mkl_sparse_?_syrkd
mkl_sparse_?_dotmv
mkl_sparse_?_sorv
C Interface Conventions for LAPACK Routines
Matrix Layout for LAPACK Routines
Matrix Storage Schemes for LAPACK Routines
Mathematical Notation for LAPACK Routines
Error Analysis
LAPACK Linear Equation Routines
LAPACK Least Squares and Eigenvalue Problem Routines
LAPACK Auxiliary Routines
LAPACK Utility Functions and Routines
LAPACK Test Functions and Routines
Additional LAPACK Routines (Included for Compatibility with Netlib LAPACK)
Matrix Factorization: LAPACK Computational Routines
Solving Systems of Linear Equations: LAPACK Computational Routines
Estimating the Condition Number: LAPACK Computational Routines
Refining the Solution and Estimating Its Error: LAPACK Computational Routines
Matrix Inversion: LAPACK Computational Routines
Matrix Equilibration: LAPACK Computational Routines
Orthogonal Factorizations: LAPACK Computational Routines
Singular Value Decomposition: LAPACK Computational Routines
Symmetric Eigenvalue Problems: LAPACK Computational Routines
Generalized Symmetric-Definite Eigenvalue Problems: LAPACK Computational Routines
Nonsymmetric Eigenvalue Problems: LAPACK Computational Routines
Generalized Nonsymmetric Eigenvalue Problems: LAPACK Computational Routines
Generalized Singular Value Decomposition: LAPACK Computational Routines
Cosine-Sine Decomposition: LAPACK Computational Routines
Linear Least Squares (LLS) Problems: LAPACK Driver Routines
Generalized Linear Least Squares (LLS) Problems: LAPACK Driver Routines
Symmetric Eigenvalue Problems: LAPACK Driver Routines
Nonsymmetric Eigenvalue Problems: LAPACK Driver Routines
Singular Value Decomposition: LAPACK Driver Routines
Cosine-Sine Decomposition: LAPACK Driver Routines
Generalized Symmetric Definite Eigenvalue Problems: LAPACK Driver Routines
Generalized Nonsymmetric Eigenvalue Problems: LAPACK Driver Routines
?lacgv
?lacrm
?syconv
?syr
i?max1
?sum1
?gelq2
?geqr2
?geqrt2
?geqrt3
?getf2
?lacn2
?lacpy
?lakf2
?lange
?lansy
?lanhe
?lantr
LAPACKE_set_nancheck
LAPACKE_get_nancheck
?lapmr
?lapmt
?lapy2
?lapy3
?laran
?larfb
?larfg
?larft
?larfx
?large
?larnd
?larnv
?laror
?larot
?lartgp
?lartgs
?lascl
?lasd0
?lasd1
?lasd2
?lasd3
?lasd4
?lasd5
?lasd6
?lasd7
?lasd8
?lasd9
?lasda
?lasdq
?lasdt
?laset
?lasrt
?laswp
?latm1
?latm2
?latm3
?latm5
?latm6
?latme
?latmr
?lauum
?syswapr
?heswapr
?sfrk
?hfrk
?tfsm
?tfttp
?tfttr
?tpqrt2
?tprfb
?tpttf
?tpttr
?trttf
?trttp
?lacp2
?larcm
mkl_?tppack
mkl_?tpunpack
Systems of Linear Equations: ScaLAPACK Computational Routines
Matrix Factorization: ScaLAPACK Computational Routines
Solving Systems of Linear Equations: ScaLAPACK Computational Routines
Estimating the Condition Number: ScaLAPACK Computational Routines
Refining the Solution and Estimating Its Error: ScaLAPACK Computational Routines
Matrix Inversion: ScaLAPACK Computational Routines
Matrix Equilibration: ScaLAPACK Computational Routines
Orthogonal Factorizations: ScaLAPACK Computational Routines
Symmetric Eigenvalue Problems: ScaLAPACK Computational Routines
Nonsymmetric Eigenvalue Problems: ScaLAPACK Computational Routines
Singular Value Decomposition: ScaLAPACK Driver Routines
Generalized Symmetric-Definite Eigenvalue Problems: ScaLAPACK Computational Routines
p?lacgv
p?max1
pilaver
pmpcol
pmpim2
?combamax1
p?sum1
p?dbtrsv
p?dttrsv
p?gebal
p?gebd2
p?gehd2
p?gelq2
p?geql2
p?geqr2
p?gerq2
p?getf2
p?labrd
p?lacon
p?laconsb
p?lacp2
p?lacp3
p?lacpy
p?laevswp
p?lahrd
p?laiect
p?lamve
p?lange
p?lanhs
p?lansy , p?lanhe
p?lantr
p?lapiv
p?lapv2
p?laqge
p?laqr0
p?laqr1
p?laqr2
p?laqr3
p?laqr5
p?laqsy
p?lared1d
p?lared2d
p?larf
p?larfb
p?larfc
p?larfg
p?larft
p?larz
p?larzb
p?larzc
p?larzt
p?lascl
p?lase2
p?laset
p?lasmsub
p?lasrt
p?lassq
p?laswp
p?latra
p?latrd
p?latrs
p?latrz
p?lauu2
p?lauum
p?lawil
p?org2l/p?ung2l
p?org2r/p?ung2r
p?orgl2/p?ungl2
p?orgr2/p?ungr2
p?orm2l/p?unm2l
p?orm2r/p?unm2r
p?orml2/p?unml2
p?ormr2/p?unmr2
p?pbtrsv
p?pttrsv
p?potf2
p?rot
p?rscl
p?sygs2/p?hegs2
p?sytd2/p?hetd2
p?trord
p?trsen
p?trti2
?lahqr2
?lamsh
?lapst
?laqr6
?lar1va
?laref
?larrb2
?larrd2
?larre2
?larre2a
?larrf2
?larrv2
?lasorte
?lasrt2
?stegr2
?stegr2a
?stegr2b
?stein2
?dbtf2
?dbtrf
?dttrf
?dttrsv
?pttrsv
?steqr2
?trmvt
pilaenv
pilaenvx
pjlaenv
Additional ScaLAPACK Routines
Intel® oneMKL PARDISO - Parallel Direct Sparse Solver Interface
Intel® MKL Parallel Direct Sparse Solver for Clusters
Direct Sparse Solver (DSS) Interface Routines
Typical order for invoking DSS interface routines
Iterative Sparse Solvers based on Reverse Communication Interface (RCI ISS)
Preconditioners based on Incomplete LU Factorization Technique
Sparse Matrix Checker Routines
pardiso
pardisoinit
pardiso_64
mkl_pardiso_pivot
pardiso_getdiag
pardiso_export
pardiso_handle_store
pardiso_handle_restore
pardiso_handle_delete
pardiso_handle_store_64
pardiso_handle_restore_64
pardiso_handle_delete_64
Intel® oneMKL PARDISO Parameters in Tabular Form
pardiso iparm Parameter
PARDISO_DATA_TYPE
vslNewStream
vslNewStreamEx
vsliNewAbstractStream
vsldNewAbstractStream
vslsNewAbstractStream
vslDeleteStream
vslCopyStream
vslCopyStreamState
vslSaveStreamF
vslLoadStreamF
vslSaveStreamM
vslLoadStreamM
vslGetStreamSize
vslLeapfrogStream
vslSkipAheadStream
vslSkipAheadStreamEx
vslGetStreamStateBrng
vslGetNumRegBrngs
Convolution and Correlation Naming Conventions
Convolution and Correlation Data Types
Convolution and Correlation Parameters
Convolution and Correlation Task Status and Error Reporting
Convolution and Correlation Task Constructors
Convolution and Correlation Task Editors
Task Execution Routines
Convolution and Correlation Task Destructors
Convolution and Correlation Task Copiers
Convolution and Correlation Usage Examples
Convolution and Correlation Mathematical Notation and Definitions
Convolution and Correlation Data Allocation
Summary Statistics Naming Conventions
Summary Statistics Data Types
Summary Statistics Parameters
Summary Statistics Task Status and Error Reporting
Summary Statistics Task Constructors
Summary Statistics Task Editors
Summary Statistics Task Computation Routines
Summary Statistics Task Destructor
Summary Statistics Usage Examples
Summary Statistics Mathematical Notation and Definitions
DFTI_PRECISION
DFTI_FORWARD_DOMAIN
DFTI_DIMENSION, DFTI_LENGTHS
DFTI_PLACEMENT
DFTI_FORWARD_SCALE, DFTI_BACKWARD_SCALE
DFTI_NUMBER_OF_USER_THREADS
DFTI_THREAD_LIMIT
DFTI_INPUT_STRIDES, DFTI_OUTPUT_STRIDES
DFTI_NUMBER_OF_TRANSFORMS
DFTI_INPUT_DISTANCE, DFTI_OUTPUT_DISTANCE
DFTI_COMPLEX_STORAGE, DFTI_REAL_STORAGE, DFTI_CONJUGATE_EVEN_STORAGE
DFTI_PACKED_FORMAT
DFTI_WORKSPACE
DFTI_COMMIT_STATUS
DFTI_ORDERING
DFTI_DESTROY_INPUT
Data Fitting Function Naming Conventions
Data Fitting Function Data Types
Mathematical Conventions for Data Fitting Functions
Data Fitting Usage Model
Data Fitting Usage Examples
Data Fitting Function Task Status and Error Reporting
Data Fitting Task Creation and Initialization Routines
Task Configuration Routines
Data Fitting Computational Routines
Data Fitting Task Destructors
df?construct1d df?Construct1D
df?interpolate1d/df?interpolateex1d df?Interpolate1D/df?InterpolateEx1D
df?integrate1d/df?integrateex1d df?Integrate1D/df?IntegrateEx1D
df?searchcells1d/df?searchcellsex1d df?SearchCells1D/df?SearchCellsEx1D
df?interpcallback df?InterpCallBack
df?integrcallback df?IntegrCallBack
df?searchcellscallback df?SearchCellsCallBack
DSS Symmetric Matrix Storage
DSS Nonsymmetric Matrix Storage
DSS Structurally Symmetric Matrix Storage
DSS Distributed Symmetric Matrix Storage
Sparse BLAS CSR Matrix Storage Format
Sparse BLAS CSC Matrix Storage Format
Sparse BLAS Coordinate Matrix Storage Format
Sparse BLAS BSR Matrix Storage Format
Direct Sparse Solver (DSS) Interface Routines
Intel® oneAPI Math Kernel Library (oneMKL) supports the DSS interface, an alternative to the Intel® oneAPI Math Kernel Library (oneMKL) PARDISO interface for the direct sparse solver. The DSS interface implements a group of user-callable routines that are used in the step-by-step solving process and utilizes the general scheme described in Appendix A: Linear Solvers Basics for solving sparse systems of linear equations. This interface also includes one routine for gathering statistics related to the solving process .
The DSS interface also supports the out-of-core (OOC) mode.
Table “DSS Interface Routines” lists the names of the routines and describes their general use.
DSS Interface Routines
Routine |
Description |
|---|---|
Initializes the solver and creates the basic data structures necessary for the solver. This routine must be called before any other DSS routine. |
|
Informs the solver of the locations of the non-zero elements of the matrix. |
|
Based on the non-zero structure of the matrix, computes a permutation vector to reduce fill-in during the factoring process. |
|
Computes the LU , LDL:code:`T` or LL:code:`T` factorization of a real or complex matrix. |
|
Computes the solution vector for a system of equations based on the factorization computed in the previous phase. |
|
Deletes all data structures created during the solving process. |
|
Returns statistics about various phases of the solving process. |
|
mkl_cvt_to_null_terminated_str |
Passes character strings from Fortran routines to C routines. |
To find a single solution vector for a single system of equations with a single right-hand side, invoke the Intel® oneAPI Math Kernel Library (oneMKL) DSS interface routines in this order:
dss_create
dss_define_structure
dss_reorder
dss_factor_real, dss_factor_complex
dss_solve_real, dss_solve_complex
dss_delete
However, in certain applications it is necessary to produce solution vectors for multiple right-hand sides for a given factorization and/or factor several matrices with the same non-zero structure. Consequently, it is sometimes necessary to invoke the Intel® oneAPI Math Kernel Library (oneMKL) sparse routines in an order other than that listed, which is possible using the DSS interface. The solving process is conceptually divided into six phases. Figure “Typical order for invoking DSS interface routines” indicates the typical order in which the DSS interface routines can be invoked.
Typical order for invoking DSS interface routines
See the code examples that use the DSS interface routines to solve systems of linear equations in the Intel® oneAPI Math Kernel Library (oneMKL) installation directory ( dss_*.f dss_*.c ).
examples/solverc/source
examples/solverf/source