Intel® Integrated Performance Primitives Release Notes for Intel® oneAPI Base Toolkit

Where to Find the Release

To get the product, download the Intel® oneAPI Base Toolkit.
For questions or technical support, visit Intel® oneAPI Products Support.

New in This Release

The Intel® IPP zlib 1.2.11 patch available with Intel® IPP 2021.6 enables building the well-known zlib library (https://zlib.net/) with Intel® IPP to speed up several data compression and decompression operations. This Intel® IPP zlib 1.2.11 patch file includes a mitigation which alleviates the memory corruption issue reported as CVE-2018-25032 against zlib 1.2.11. Intel® IPP provides a code check with zlib 1.2.11 patch for the bytes that override the overlay, which is separated and does resolve the memory corruption issue.

NOTE for NuGet Package Manager Users: There will be a delay in providing IPP NuGet package for the version 2021.2 release. We are working to get package size that work within NuGet size limits.  Because of this, Intel® IPP packages for 2021.2 will not be available at the oneAPI version 2021.2 release.  We hope to have these uploaded soon. Please check back for information on these packages.  If you do not use NuGet package manager you are not affected.

KNOWN ISSUE
After installing the oneAPI Base Toolkit 2022.1, compiling applications with Win32 platform settings that require Intel® Integrated Performance Primitives (Intel® IPP) will fail. 32-bit Intel® IPP on Windows* OS are provided separately as part of Intel® oneAPI Base Toolkit 32-bit package. It can be downloaded here as an add-on. In order to build the Intel® IPP examples projects for "x86" configurations on Windows*, Intel® oneAPI Base Toolkit 32-bit package has to be installed.

2021.6.1

System Requirements  Bug Fix Log

No changes in this release.

2021.6

System Requirements  Bug Fix Log

• Extended optimization for Intel® IPP Cryptography NIST Elliptic Curve functions P-256/P-384/P-521 and SM2 algorithms for 3rd Generation of Intel® Xeon® Processor Scalable and next generation Intel® Xeon® Scalable Processor, code named 'Sapphire Rapids'.
• Added build support for Intel® IPP Cryptography library and Multi-buffer(MB) library with Microsoft Visual C++ Compiler* version 19.30 provided by Microsoft Visual Studio* 2022 version 17.0.
• Improved Intel® IPP Cryptography NIST Elliptic Curve functions input parameters by adding keys boundary check.
• Improved Intel® IPP Cryptography AES-GCM functions input parameters by adding input text boundary check.
• Extended optimization Intel® IPP Image Processing ippiFilterBorder function of for 3rd Generation of Intel® Xeon® Processor Scalable and next generation Intel® Xeon® Scalable Processor, code named 'Sapphire Rapids'.
• Extended and improved support for Intel® IPP Image Processing ippiFilterMedian function with 8-bit unsigned.

2021.5

System Requirements  Bug Fix Log

• Extended optimization for Intel® IPP Cryptography AES-GCM for smaller packets size for 3rd Generation Intel® Xeon® Processor Scalable and next generation Intel® Xeon® Scalable Processor, code named 'Sapphire Rapids'.
• Enhanced Intel® IPP Image Processing ipprResize function with 16s support for 3D resizing functionality.

2021.4

System Requirements  Bug Fix Log

• Enabled Intel® IPP Cryptography Multi-buffer(MB) functions for 3rd Generation Intel® Xeon® Processor Scalable and 10th Gen Intel® Core™ Processors for the following:
  • SM4 block cipher standard modes (OFB, CFB, ECB, CBC, CTR)
  • ECDSA Ed25519 Verify API
  • ECDH SM2 and ECDSA SM2 Sign universal and SSL APIs
  • 1Kb, 2Kb, 3Kb and 4Kb modular exponentiations
• Enabled 16-bit float Discrete Fourier transform (DFT) and Fast Fourier transform (FFT) functions within Intel® IPP optimized for next generation Intel® Xeon® Scalable Processor, code named 'Sapphire Rapids' focused on various lengths supported in 5G environment.
• Extended optimization for Intel® IPP Image and Signal Processing functions for 3rd Generation Intel® Xeon® Processor Scalable and 10th Gen Intel® Core™ Processors for the following:
  • Multi-rate FIR filtering
  • Resize Supersampling (one-channel, floating-point)
  • Resize Antialiasing (one-channel, double-precision floating-point)
• Added support and optimization for LZ4HC 1.9.2 and LZ4HC 1.9.3 version in Intel® IPP Data compression.
• Added support for Intel® IPP Cryptography crypto_mb library build within BoringSSL* for 3rd Generation Intel® Xeon® Processor Scalable and 10th Gen Intel® Core™ Processors..
• Added and optimized modular exponentiation in Intel® IPP Cryptography Multi-buffer(MB) library.
• Added support for Intel® IPP Image Processing 3D WarpAffine functions to the Threading Layer(TL) API.

2021.3

System Requirements  Bug Fix Log

• Enabled ECDSA Ed25519sign crypto multi buffer API within Intel® IPP Cryptography for 3rd Generation Intel® Xeon® Processor Scalable and 10th Gen Intel® Core™ Processors.
• Enabled RSA single buffer 3k, 4k within Intel® IPP Cryptography for 3rd Generation Intel® Xeon® Processor Scalable and 10th Gen Intel® Core™ Processors.
• Extended optimization for Intel® IPP Data Compression LZ4 decompression.
• Enhanced Intel® IPP Image Processing ippiResizeAntialiasing and ippiResize functions with 64f support for one-channel images.
• Extended optimization for Intel® IPP Image Processing ippiResizeAntialiasing_64f function for Intel® Advanced Vector Extensions 2 (Intel® AVX2).
• Extended optimization for Intel® IPP Image Processing ippiResizeLanczos* and ippiResizeCubic* functions for Intel® Advanced Vector Extensions 512 (Intel® AVX-512).
• Extended Platform-Aware Functions list in Intel® IPP Image Processing domains with ippiPyramid* functions.
• Fixed patching issue encountered with Intel® IPP Data Compression LZ4 1.9.3 patch.
• Fixed Intel® IPP Cryptography AES-GCM decryption incorrect tag issue while dispatching on processors supported with Intel® Advanced Vector Extensions 512 (Intel® AVX-512).

2021.2

System Requirements  Bug Fix Log

• Added crypto_mb as a separate library within Intel® IPP Cryptography in Intel® oneAPI Base Toolkit 2021.2 release.
• Added new Intel® IPP Cryptography Multi-buffer(MB) library functions for the following:
  • ECDHE and ECDSA with new Instruction Set Architecture(ISA) for the NIST curve p521r1.
  • ECDSA verify with new Instruction Set Architecture(ISA) for the NIST curve p384r1, p256r1 and p521r1.
  • SM3 with new Instruction Set Architecture(ISA).
• Added new Intel® IPP Cryptography pre-defined hash algorithm APIs ippsHashMethodGetSize and ippsHashMethodInit.
• Implemented precised bilateral filter in Intel® IPP Image Processing domain for image smoothing based on iterative least square method.
• Enhanced Intel® IPP Image Processing ippiResizeAntialiasing and ippiResize functions with MirrorR border type support.
• Added floating-point shift routine to existing Intel® IPP Image Processing ippiResizeSuper function.
• Extended optimizations for Intel® IPP Image Processing and Color Coversions functions for selected algorithms for Intel® Advanced Vector Extensions 512 (Intel® AVX-512).
• Added support and optimization for LZ4 1.9.3 version in Intel® IPP Data compression.
• Added new K1 architecture code in Intel® IPP for 3rd Gen Intel® Xeon® Scalable Processor.

2021.1 Initial Release

• Renamed Intel® IPP Threading Layer Library for all operating systems. OpenMP* and Intel® oneAPI Threading Building Blocks(Intel® oneTBB) had identical suffix: "_tl" (e.g. ippcore_tl.lib).
  • New names have additional suffix for each configuration:
    • “_tl_omp” for OpenMP* configuration (e.g. ippcore_tl_omp.lib).
    • “_tl_tbb” for Intel® oneTBB configuration (e.g. ippcore_tl_tbb.lib).
• Extended optimization for Intel® IPP Cryptography SM4 (ECB, CBC, CFB, OFB, CRT) on 10th Generation Intel® Core™ processor family.
• Added new universal 16-bit CRC functions to compute CRC8, CRC16, CRC24 and CRC32 checksums.
• Enabled Intel® IPP Threading Layer for Intel® IPP Image Processing ippiCrossCorrNorm* functions.
• Extended optimzation for Intel® IPP Signal Processing ippsHilbert_32f32fc function on double precision counterparts.
• Reinstated and extended optimization for Intel® IPP Image Processing ippiComplement* function for Intel® Advanced Vector Extensions 512 (Intel® AVX-512), Intel® Advanced Vector Extensions 2 (Intel® AVX2) and Intel® Streaming SIMD Extensions 4.2 (Intel® SSE4.2).
• Added support for custom dispatcher in the Intel® IPP Custom Library Tool.
• Extended optimization for Intel® IPP Image Processing ippiFilterSeparable* function for Intel® IPP data type Ipp32f and 8u.

Unsupported or Discontinued Features

• Deprecated support for Microsoft Visual C++ Compiler* version 19.16 provided by Microsoft Visual Studio* 2017 version 15.9 from Intel® IPP Cryptography. If you use this version of Compiler for building Intel® IPP Cryptography library, please plan on migrating to a newer supported version of Microsoft Visual C++ Compiler*.
• Intel® IPP deprecated support for multi-threaded libraries.
• Dropped SSSE3 optimized variant in Intel® IPP Core due to unsupported hardware in macOS* systems.
• Dropped performance tests for Intel® IPP.

Known Issues and Limitations

• Please avoid using Intel® IPP Cryptography ippsGFpECSharedSecretDH function with NIST P-384 Elliptic Curve. It computes incorrect shared key when using with NIST P-384 Elliptic Curve for some edge inputs.
• This RSA Multi-Buffer API is deprecated from Intel® IPP Cryptography and is removed since 2021.2 release. It is recommended to switch to crypto_mb library. If you have any concerns, open a ticket and provide feedback at Intel® online support center.
• Intel® IPP Custom Library Tool cannot see the path to Intel® Compilers after setting environment variables due to different versions of Intel® IPP and Intel® Compilers. The workaround is to auto build the dynamic library in GUI mode. Path to the Intel® Compilers should be provided manually.
• DEPRECATION NOTICE
  • Intel® C++ Compiler Classic (icc) is deprecated and will be removed in a oneAPI release in the second half of 2023. Intel recommends that customers transition now to using the LLVM-based Intel® oneAPI DPC++/C++ Compiler (icx) for continued Windows* and Linux* support, new language support, new language features, and optimizations. Note that starting with release 2021.7, macOS* support is limited to Mac* computers with Intel® Processors. For more information on icx, see the Intel® oneAPI DPC++/C++ Compiler Developer Guide and Reference.

Known Installation and Configuration Issues

See the Intel® oneAPI Base Toolkit Release Notes for additional details and issues common to multiple toolkits.

Notices and Disclaimers

Intel technologies may require enabled hardware, software or service activation.

No product or component can be absolutely secure.

Your costs and results may vary.

© Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries. Other names and brands may be claimed as the property of others.

No license (express or implied, by estoppel or otherwise) to any intellectual property rights is granted by this document.

The products described may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request.

Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade.