v0.6

Features in This Release

BLAS

• Add generic device, allowing portBLAS backend to run on any SYCL device.

RNG

• Introduced Beta and Gamma distributions support in RNG Device API
• Introduced uint64_t type support for uniform distribution in RNG Device API
• Introduced (u)int8/int16 types support for Bernoulli distribution in RNG Device API

DFT

• Fix correctness issue with ROCm 6.0+.
• Allow portFFT backend to run on any SYCL device.

Sparse BLAS

• Update MKLCPU and MKLGPU backends to the new API following the latest oneAPI specification.
• Add support for cuSPARSE backend.

General updates

• Set libonemkl.so's rpath to $ORIGIN allowing it to load the backend shared objects without setting LD_LIBRARY_PATH
• Use [ext_codeplay_enqueue_native_command].(https://github.com/intel/llvm/blob/bfc4015a3f858ef664e469a9a0dbda97c2d22a1f/sycl/doc/extensions/experimental/sycl_ext_codeplay_enqueue_native_command.asciidoc#differences-with-host_task) extension with all CUDA and ROCm backends, improving the synchronization mechanism.

Known Issues and Limitations

v0.5

Features in This Release

BLAS

• Improved support for ROCm backend

DFT

• Improve specification for compute function overloads, and implement in oneMKL interfaces. #503
• Add FWD/BWD STRIDES API: #514

Documentation update

• Added security.md file which includes guidance on reporting security incidences.
• Updated documentation and README with UXL Foundation details.
• Rewrite build documentation #510

Known Issues and Limitations

v0.4

Features in This Release

BLAS

• Improved support for rocBLAS and portBLAS backends.
• Enabled cuBLAS and rocBLAS backends with AdaptiveCpp compiler.

DFT 

• Enabled portFFT backend for x86 CPU, Intel GPU, Nvidia GPU and AMD GPU for the DFT SYCL APIs
• Enabled the external workspace DFT SYCL API

Sparse BLAS

• Introduced Sparse BLAS interfaces with backend for x86 CPU and Intel GPU

Documentation update:

• Replaced the usage of "Intel project for LLVM* technology" with the approved one "oneAPI DPC++ Compiler" in all documents

• Readme file update:

  • The tables of "Supported Configurations" were reorganized for better readability.
  • Support of Open DPC++ compiler removed for configurations x86 + Intel(R) oneMKL + BLAS|LAPACK
  • Updated the section of "Supported Operating Systems" and compiler version

Known Issues and Limitations

v0.3

Features in This Release

BLAS

• Added NETLIB X86 CPU backend support in BLAS domain for DPC++, LLVM and hipSYCL compilers

• Added Unified Shared Memory (USM) Interfaces support in BLAS domain

• Extended support in BLAS domain with low-precision, mixed precision GEMMs and various batch implementations.

• Extended support in BLAS domain with transposition routines.

• Added cuBLAS Nvidia GPU backend support in BLAS domain for hipSYCL compiler

• Added portBLAS X86 CPU, Intel GPU, Nvidia GPU and AMD GPU backend support in BLAS domain for DPC++ and LLVM compilers.

• Added rocBLAS AMD GPU backend in BLAS domain for LLVM and hipSYCL compilers.

• Extended MKLCPU and MKLGPU backend support in BLAS domain with pointers for scalar parameters.

• Various updates in BLAS domain to align with SYCL 2020 spec and bug fixes

LAPACK

• Introduced LAPACK SYCL APIs support for x86 CPU and Intel GPU with the Intel® oneAPI Math Kernel Library, Nvidia GPU with the cuSOLVER library and AMD GPU with the rocSOLVER library.

DFT 

• Introduced DFT SYCL APIs support for x86 CPU and Intel GPU with the Intel® oneAPI Math Kernel Library, Nvidia GPU with the cuFFT library and AMD GPU with the rocFFT library.

RNG

• Introduced RNG Device API support according to the oneAPI 1.3 specification

Service

• Replaced FindMKL.cmake with MKLConfig.cmake from Intel oneMKL

• Extended support for cmake fetchcontent module

Known Issues and Limitations

• RNG Host API Philox4x32x10 engine may produce incorrect numbers in Skip_Ahead_Ex test in case of mklgpu backend and Intel oneMKL 2024.0 release

• The LAPACK domain for x86 CPU with Intel(R) oneAPI Math Kernel Library backend has build failures with builds of the LLVM* compiler after 2023-09-27. Use builds from 2023-09-26 or earlier.

v0.2

New features:
• Enabled Nvidia support for BLAS
• Added extended BLAS and BATCH support
• Introduced Windows* support for Intel CPU and Intel GPU
• Implemented Linux* dependencies download with Conan* package manager

v0.1

Initial release includes:

• DPC++ BLAS interfaces based on oneMKL specification
• Functional tests
• OS support: Linux*
• Hardware support: Intel CPU and Intel GPU