sun0929 / ipp-crypto

Intel(R) Integrated Performance Primitives Cryptography

• Overview
  • Product description
  • Branches description
  • License
  • Online Documentation
• System Requirements to Build Intel(R) IPP Cryptography
  • Common Requirements for All Supported Operating Systems
  • Windows* OS
    • C/C++* Compilers
    • Assembly Compilers
  • Linux* OS
    • C/C++* Compilers
    • Binary Tools
  • macOS*
    • C/C++* Compilers
    • Assembly Compilers
• Building from Source
  • Required Software
  • Build Steps
  • CMake Arguments
    • Common Arguments for All Supported Operating Systems
    • Windows* OS
    • Linux* OS
    • macOS*
• Building an application tied to a specific CPU
• Intel IPP Custom Library Tool
• RSA IFMA Muti-buffer Library
• How to Contribute
• See Also
• Legal Information

Overview

Product description

Intel(R) Integrated Performance Primitives (Intel(R) IPP) Cryptography is a software library that provides a broad range of secure and efficient cryptographic algorithm implementations:

• Symmetric Cryptography Primitive Functions (AES, SM4, TDES, RC4)

• One-Way Hash Primitives (SHA, MD5, SM3)

• Data Authentication Primitive Functions (HMAC, CMAC)

• Public Key Cryptography Functions (RSA, DLP, ECC)

• Finite Field Arithmetic Functions

Intel(R) IPP Cryptography provides multiple function implementations optimized for various Intel(R) CPUs, and can be set up to use dispatching to select the CPU-specific implementation at run time. Dispatching refers to the detection of features supported by the underlying CPU and selecting the corresponding Intel IPP Cryptography binary path.

Branches description

• develop - snapshots of the library under active development. Contains code that may not be fully functional and that Intel may substantially modify in development of a production version.
• ipp_crypto_<release> - source code of the official production release <release>.

License

Intel IPP Cryptography is licensed under Apache License 2.0.

Online Documentation

You can find the latest Intel IPP Cryptography documentation on the Intel(R) Integrated Performance Primitives Documentation web page.

System Requirements to Build Intel(R) IPP Cryptography

The list below contains the system requirements necessary to build Intel IPP Cryptography. We tested the build process of Intel IPP Cryptography only on the operating systems and tools listed below:

Common Requirements for All Supported Operating Systems

• CMake 3.15

• Python 2.7.15

• The Netwide Assembler (NASM) 2.15

Windows* OS

• Windows Server* 2016

C/C++* Compilers

• Intel(R) C++ Compiler 19.0 Update 4 for Windows* OS

• Microsoft Visual C++ Compiler* version 19.16 provided by Microsoft Visual Studio* 2017 version 15.9

Linux* OS

• Red Hat* Enterprise Linux* 7

C/C++* Compilers

• Intel(R) C++ Compiler 19.0 Update 4 for Linux* OS

• GCC 8.3

• GCC 9.1

Binary Tools

• GNU binutils 2.32

macOS*

• macOS* 10.14

C/C++* Compilers

• Intel(R) C++ Compiler 19.0 Update 4 for OS X* OS

Building from Source

Required Software

• C/C++ compiler (see System Requirements)

• Assembly compiler (see System Requirements)

• CMake (see System Requirements)

• Python (see System Requirements)

• Microsoft Visual Studio* (Windows* OS only)

Build Steps

Note that we tested only the build process defined in CMakeLists.txt with the options described in CMake Arguments.

1. Download and install all Required Software.

2. Clone the source code from GitHub* as follows:

   git clone --recursive <repo>
   

3. Set the environment variables for one of the supported C/C++ compilers.
   For Intel(R) Compiler please refer to Intel(R) C++ Compiler Developer Guide and Reference
   For MSVC* Compiler please refer to Use the MSVC toolset from the command line

4. Run CMake* on the command line. Example CMake lines for different operating systems:

   Windows* OS:

   For Intel(R) C++ Compiler:

   cmake CMakeLists.txt -Bbuild -G"Visual Studio 14 2015 Win64" -T"Intel C++ Compiler 19.0"
   

   For MSVC* Compiler:

   cmake CMakeLists.txt -Bbuild -G"Visual Studio 15 2017 Win64"
   

   For Intel(R) C++ Compiler and Visual Studio* 2019:

   cmake CMakeLists.txt -Bbuild -G"Visual Studio 16 2019" -T"Intel C++ Compiler 19.0" -Ax64
   

   For MSVC* Compiler and Visual Studio* 2019:

   cmake CMakeLists.txt -Bbuild -G"Visual Studio 16 2019" -Ax64
   

   Linux* OS:

   For Intel(R) C++ Compiler:

   CC=icc CXX=icpc cmake CMakeLists.txt -Bbuild -DARCH=intel64
   

   For GCC:

   CC=gcc CXX=g++ cmake CMakeLists.txt -Bbuild -DARCH=intel64
   

   macOS*:

   For Intel(R) C++ Compiler:

   CC=icc CXX=icpc cmake CMakeLists.txt -Bbuild -DARCH=intel64
   

   
   The list of supported CMake arguments is available in the CMake Arguments section.

5. Go to the build folder specified when running CMake, then do one of the following:

   • On Windows* OS: open the Microsoft Visual Studio* solution and run a build.

     Important: the process used to build the Microsoft Visual Studio solutions results in debug information being generated for assembly files by default for both Debug and Release configurations. To build Intel IPP Cryptography library binaries without debug information in the Release configuration, follow these steps:

     1. Right-click a project file and select Properties.
     2. In the Configuration drop-down list, select Release.
     3. Select the Microsoft Macro Assembler tab.
     4. Set the value of the Generate Debug Information option to No.
     5. Repeat steps 1-3 for all projects you want to build.

   • On Linux* OS or macOS*: start a build using makefiles.

   At this point, you can shoose to build either static or dynamic libraries. Depending on the value of the -DMERGED_BLD:BOOL option in the cmake call in step 4, you will get a set of separate self-contained libraries optimized for particular platforms, or a dispatched library with all optimizations. See the description of -DMERGED_BLD:BOOL in CMake Arguments for more information.

   Built libraries are located in the <build_dir>/.build/<RELEASE|DEBUG>/lib directory.

CMake Arguments

Common Arguments for All Supported Operating Systems

• -B<build-dir> - defines the build directory. This is the directory where CMake puts the generated Microsoft Visual Studio* solution or makefiles.

• -DARCH=<ia32|intel64> - on Linux* OS and macOS*, defines the target architecture for the build of the Intel IPP Cryptography library. On Windows* OS, use -G/-A instead. See the description of the -G/-A option below for details.

• -DMERGED_BLD:BOOL=<on|off> - optional. Defines the configuration of the Intel IPP Cryptography library to build:

  • -DMERGED_BLD:BOOL=on: default configuration. Build of a dispatched static library with all available optimizations; build of a dispatched dynamic library with all available optimizations; generation of the single-CPU headers (for more details please refer to the section below).

  • -DMERGED_BLD:BOOL=off: build of one static library per optimization; build of one dynamic library per optimization.

• -DPLATFORM_LIST="<platform list>" - optional, works only if -DMERGED_BLD:BOOL=off is set. Sets target platforms for the code to be compiled. See the supported platform list at https://software.intel.com/en-us/ipp-dev-guide-dispatching.

  • example for Linux* OS and the IA-32 architecture: -DPLATFORM_LIST="m7;s8;p8;g9;h9"

  • example for Linux* OS and the Intel(R) 64 architecture: -DPLATFORM_LIST="w7;n8;y8;e9;l9;n0;k0"

Note: you can set C/C++ compilers for Linux* OS and macOS* using the CC and CXX variables before calling cmake. For example:

CC=<path to C compiler> CXX=<path to C++ compiler> cmake <Arguments>

Windows* OS

• -G"<tool-chain-generator>" - defines the native build system CMake will generate from the input files. For example, -G"Visual Studio 12 2013" will generate a solution for the Microsoft Visual Studio* 2013 IDE, -G"Visual Studio 12 2013 Win64" will generate a solution for the Microsoft Visual Studio* 2013 IDE for the Intel(R) 64 architecture.

• -A<x64|Win32> - for Visual Studio* 2019+, defines the target architecture for the build of the Intel IPP Cryptography library.

• -T<Compiler> - defines the compiler for building, for example, -T"Intel C++ Compiler 19.0" defines Intel(R) Compiler 19.0 for building.

Note: Refer to CMake documentation for more information on these options.

Linux* OS

• -DNONPIC_LIB:BOOL=<off|on> - optional. Defines whether the built library will be position-dependent or not:

  • -DNONPIC_LIB:BOOL=off: default. Position-independent code.

  • -DNONPIC_LIB:BOOL=on: position-dependent code.

Building an application tied to a specific CPU

The default build of the Intel(R) IPP Cryptography library (with -DMERGED_BLD:BOOL=on CMake option) provides the merged static library that contains multiple versions of each function optimized for different CPUs. By default, each function is linked to the target application with its all CPU optimization versions. In case when an application or driver needs to be tied to a particular processor to reduce the footprint, it is also possible to link it to only one optimization layer of Intel(R) IPP Cryptography.

For this purpose, there are several CPU-specific headers (each targeted on specific CPU optimization) generated during the merged library build. They are located in the <build_dir>/.build/<RELEASE|DEBUG>/include/autogen directory. To enable linking of processor-specific versions of the library functions, include the appropriate header from the directory above before the primary library header ippcp.h.

Please refer to the article for more details about CPU identification codes used in the header suffixes.

It is important to ensure that both processor and operating system supports full capabilities of the target processor.

Intel IPP Custom Library Tool

With the Intel(R) IPP Custom Library Tool, you can build your own dynamic library containing only the Intel IPP Cryptography functionality that is necessary for your application.

The tool is located in tools/ipp_custom_library_tool_python directory.

Please refer to the tool documentation for more information.

RSA IFMA Muti-buffer Library

It is a library separate from Intel IPP Cryptography. The library provides optimized version of RSA multi-buffer based on Intel(R) Advanced Vector Extensions 512 (Intel(R) AVX-512) Integer Fused Multiply Add (IFMA). Currently, the library is compatible with OpenSSL* only.

The source of RSA multi-buffer is located in a separate folder: sources/ippcp/ifma_rsa_mb.

How to Contribute

We welcome community contributions to Intel IPP Cryptography. If you have an idea how to improve the product, let us know about your proposal via the Intel IPP Forum.

Intel IPP Cryptography is licensed under Apache License, Version 2.0. By contributing to the project, you agree to the license and copyright terms therein and release your contribution under these terms.

See Also

• Intel(R) Integrated Performance Primitives Product Page

• Intel(R) IPP Forum