Visible to Intel only — GUID: GUID-F4CA1212-D975-452B-8CC1-5B40993B2A73
What's New
Introduction to the Intel® oneAPI Math Kernel Library (oneMKL) BLAS and LAPACK with DPC++
Overview of Intel® oneMKL BLAS Routines for Data Parallel C++
Overview of Intel® oneAPI Math Kernel Library (oneMKL) Sparse BLAS for DPC++
Overview of Intel® oneAPI Math Kernel Library (oneMKL) LAPACK for DPC++
Data Types
Matrix Storage
Error Handling
BLAS Routines
Sparse BLAS Routines
LAPACK Routines
Vector Mathematical Functions
Random Number Generators
Summary Statistics
Fourier Transform Functions
Data Fitting
Bibliography
Appendix A: oneMKL Functionality
Notices and Disclaimers
Sparse BLAS Matrix Handle Contract between User and Library
Sparse BLAS Supported Data and Integer Types
Sparse Storage Formats
oneapi::mkl::sparse::init_matrix_handle
oneapi::mkl::sparse::release_matrix_handle
oneapi::mkl::sparse::set_csr_data
oneapi::mkl::sparse::set_matrix_property
oneapi::mkl::sparse::optimize_gemv
oneapi::mkl::sparse::optimize_trmv
oneapi::mkl::sparse::optimize_trsv
oneapi::mkl::sparse::gemv
oneapi::mkl::sparse::gemvdot
oneapi::mkl::sparse::symv
oneapi::mkl::sparse::trmv
oneapi::mkl::sparse::trsv
oneapi::mkl::sparse::gemm
oneapi::mkl::sparse::init_matmat_descr
oneapi::mkl::sparse::set_matmat_data
oneapi::mkl::sparse::get_matmat_data
oneapi::mkl::sparse::release_matmat_descr
oneapi::mkl::sparse::matmat
oneapi::mkl::sparse::omatcopy
oneapi::mkl::sparse::sort_matrix
gebrd
gebrd (USM Version)
gebrd_scratchpad_size
gels_batch (Buffer Strided Version)
gels_batch (USM Strided Version)
gels_batch_scratchpad_size (Strided Version)
geqrf
geqrf (USM Version)
geqrf_batch (Buffer Strided Version)
geqrf_batch (Group Version)
geqrf_batch (USM Strided Version)
geqrf_batch_scratchpad_size (Group Version)
geqrf_batch_scratchpad_size (Strided Version)
geqrf_scratchpad_size
gerqf
gerqf (USM Version)
gerqf_scratchpad_size
gesvd
gesvd (USM Version)
gesvd_scratchpad_size
getrf
getrf (USM Version)
getrf_batch (Buffer Strided Version)
getrf_batch (Group Version)
getrf_batch (USM Strided Version)
getrf_batch_scratchpad_size (Group Version)
getrf_batch_scratchpad_size (Strided Version)
getrf_scratchpad_size
getrfnp_batch (Buffer Strided Version)
getrfnp_batch (Group Version)
getrfnp_batch (USM Strided Version)
getrfnp_batch_scratchpad_size (Group Version)
getrfnp_batch_scratchpad_size (Strided Version)
getri
getri (USM Version)
getri_batch (Buffer Strided Version)
getri_batch (Group Version)
getri_batch (USM Strided Version)
getri_batch_scratchpad_size (Group Version)
getri_batch_scratchpad_size (Strided Version)
getri_batch (Out-of-place, Buffer Strided Version)
getri_batch (Out-of-place, USM Strided Version)
getri_batch_scratchpad_size (Strided Version)
getri_scratchpad_size
getrs
getrs (USM Version)
getrs_batch (Buffer Strided Version)
getrs_batch (Group Version)
getrs_batch (USM Strided Version)
getrs_batch_scratchpad_size (Group Version)
getrs_batch_scratchpad_size (Strided Version)
getrs_scratchpad_size
getrsnp_batch (Buffer Strided Version)
getrsnp_batch (USM Strided Version)
getrsnp_batch_scratchpad_size (Strided Version)
heevd
heevd (USM Version)
heevd_scratchpad_size
hegvd
hegvd (USM Version)
hegvd_scratchpad_size
hetrd
hetrd (USM Version)
hetrd_scratchpad_size
hetrf
hetrf (USM Version)
hetrf_scratchpad_size
orgbr
orgbr (USM Version)
orgbr_scratchpad_size
orgqr
orgqr (USM Version)
orgqr_batch (Buffer Strided Version)
orgqr_batch (Group Version)
orgqr_batch (USM Strided Version)
orgqr_batch_scratchpad_size (Group Version)
orgqr_batch_scratchpad_size (Strided Version)
orgqr_scratchpad_size
orgtr
orgtr (USM Version)
orgtr_scratchpad_size
ormqr
ormqr (USM Version)
ormqr_scratchpad_size
ormrq
ormrq (USM Version)
ormrq_scratchpad_size
ormtr
ormtr (USM Version)
ormtr_scratchpad_size
potrf
potrf (USM Version)
potrf_batch (Buffer Strided Version)
potrf_batch (Group Version)
potrf_batch (USM Strided Version)
potrf_batch_scratchpad_size (Group Version)
potrf_batch_scratchpad_size (Strided Version)
potrf_scratchpad_size
potri
potri (USM Version)
potri_scratchpad_size
potrs
potrs (USM Version)
potrs_batch (Buffer Strided Version)
potrs_batch (Group Version)
potrs_batch (USM Strided Version)
potrs_batch_scratchpad_size (Group Version)
potrs_batch_scratchpad_size (Strided Version)
potrs_scratchpad_size
syevd
syevd (USM Version)
syevd_scratchpad_size
sygvd
sygvd (USM Version)
sygvd_scratchpad_size
sytrd
sytrd (USM Version)
sytrd_scratchpad_size
sytrf
sytrf (USM Version)
sytrf_scratchpad_size
trtrs
trtrs (USM Version)
trtrs_scratchpad_size
ungbr
ungbr (USM Version)
ungbr_scratchpad_size
ungqr
ungqr (USM Version)
ungqr_batch (Buffer Strided Version)
ungqr_batch (Group Version)
ungqr_batch (USM Strided Version)
ungqr_batch_scratchpad_size (Group Version)
ungqr_batch_scratchpad_size (Strided Version)
ungqr_scratchpad_size
ungtr
ungtr (USM Version)
ungtr_scratchpad_size
unmqr
unmqr (USM Version)
unmqr_scratchpad_size
unmrq
unmrq (USM Version)
unmrq_scratchpad_size
unmtr
unmtr (USM Version)
unmtr_scratchpad_size
oneapi::mkl::rng::mrg32k3a
oneapi::mkl::rng::philox4x32x10
oneapi::mkl::rng::mcg31m1
oneapi::mkl::rng::mcg59
oneapi::mkl::rng::r250
oneapi::mkl::rng::wichmann_hill
oneapi::mkl::rng::mt19937
oneapi::mkl::rng::sfmt19937
oneapi::mkl::rng::mt2203
oneapi::mkl::rng::ars5
oneapi::mkl::rng::sobol
oneapi::mkl::rng::niederreiter
oneapi::mkl::rng::nondeterministic
Distributions Template Parameter Method
oneapi::mkl::rng::uniform (Continuous)
oneapi::mkl::rng::gaussian
oneapi::mkl::rng::exponential
oneapi::mkl::rng::laplace
oneapi::mkl::rng::weibull
oneapi::mkl::rng::cauchy
oneapi::mkl::rng::rayleigh
oneapi::mkl::rng::lognormal
oneapi::mkl::rng::gumbel
oneapi::mkl::rng::gamma
oneapi::mkl::rng::beta
oneapi::mkl::rng::chi_square
oneapi::mkl::rng::gaussian_mv
oneapi::mkl::rng::uniform (Discrete)
oneapi::mkl::rng::uniform_bits
oneapi::mkl::rng::bits
oneapi::mkl::rng::bernoulli
oneapi::mkl::rng::geometric
oneapi::mkl::rng::binomial
oneapi::mkl::rng::hypergeometric
oneapi::mkl::rng::poisson
oneapi::mkl::rng::poisson_v
oneapi::mkl::rng::negative_binomial
oneapi::mkl::rng::multinomial
oneapi::mkl::rng::device::uniform (Continuous)
oneapi::mkl::rng::device::gaussian
oneapi::mkl::rng::device::lognormal
oneapi::mkl::rng::device::exponential
oneapi::mkl::rng::device::uniform (Discrete)
oneapi::mkl::rng::device::bits
oneapi::mkl::rng::device::poisson
oneapi::mkl::rng::device::bernoulli
oneapi::mkl::stats::raw_sum
oneapi::mkl::stats::central_sum
oneapi::mkl::stats::central_sum with User-provided Mean
oneapi::mkl::stats::raw_moment
oneapi::mkl::stats::central_moment
oneapi::mkl::stats::central_moment with User-provided Mean
oneapi::mkl::stats::mean
oneapi::mkl::stats::variation
oneapi::mkl::stats::variation with User-provided Mean
oneapi::mkl::stats::skewness
oneapi::mkl::stats::skewness with User-provided Mean
oneapi::mkl::stats::kurtosis
oneapi::mkl::stats::kurtosis with User-provided Mean
oneapi::mkl::stats::min
oneapi::mkl::stats::max
oneapi::mkl::stats::min_max
descriptor<precision, domain>
descriptor<precision, domain>::set_value
descriptor<precision, domain>::get_value
descriptor<precision, domain>::commit
compute_forward<typename descriptor_type, typename data_type>
compute_backward<typename descriptor_type, typename data_type>
User-Allocated Workspaces
Visible to Intel only — GUID: GUID-F4CA1212-D975-452B-8CC1-5B40993B2A73
Random Number Generators Device Routines
The main purpose of Device routines is to make them callable from your DPC++ kernels; however, there are no limitations to be called from the Host. For example:
sycl::queue queue; queue.submit([&](sycl::handler& cgh) { cgh.parallel_for(range,[=](...) { mkl::rng::device::routine(...); // calling routine from user's kernel code }); }); mkl::rng::device::routine(...); // calling routine from host
RNG Device APIs Structure
RNG Device API Structure