Integrated Performance Primitives(IPP)的基本概述与安装

名扬569

于 2023-07-14 17:02:15 发布

阅读量1.6k

点赞数

文章标签： IPP intel 介绍

本文链接：https://blog.csdn.net/qq_29809823/article/details/118021715

版权

IntelIPP是一个多线程软件库，提供用于多媒体、数据处理和加密等功能，支持多种操作系统和处理器。它利用现代处理器的硬件特性，如MMX到AVX-512指令集。OneAPI是Intel的一个项目，IPP是其一部分，旨在提供跨平台的统一编程体验。文章还介绍了如何在Ubuntu上安装和测试IntelIPP库。

摘要由CSDN通过智能技术生成

intel ipp库（Integrated Performance Primitives）

概述

来自维基百科
https://zh.wikipedia.org/wiki/英特尔集成性能原语

一个多线程软件函式库，包含用于多媒体和数据处理应用的函数，支持英特尔和兼容处理器，可用于Linux、macOS、Windows和Android操作系统
Intel IPP被划分为四大主要处理组：信号（带有线性阵列或矢量数据）、图像（具有典型颜色空间的2D阵列）、矩阵（用于矩阵运算的nxm数组）和密码学。
其入口点中一半为矩阵类型，三分之一为信号类型，其余是图像和密码类型。Intel IPP函数分为4种数据类型：数据类型包括8u (8-bit unsigned)、8s (8-bit signed)、16s、32f (32-bit floating-point)、64f等。
该库利用处理器的包括MMX、SSE、SSE2、SSE3、SSSE3、SSE4、AVX、AVX2、AVX-512、AES-NI和多核心处理器特性。[3]Intel IPP包含下列函数：
- 视频解码/编码
- 音频解码/编码
- JPEG/JPEG2000/JPEG XR
- 计算机视觉
- 密码学
- 数据压缩
- 图像颜色转换
- 图像处理
- 光线追踪和渲染
- 信号处理
- 语音编码
- 语音识别
- 字符串处理
- 矢量和矩阵数学

IPPI

ippi即是IPP库的Image Processing部分

IPPS

ipps即是IPP库的Signal and Data Processing部分

OneAPI

OneAPI是intel搞的一个项目，听名字就知道了，intel想开发一个api，可以在各个平台使用。
而IPP库是Oneapi的一个子集，Oneapi的所有子集在intel的官网可以找到，比如这个链接

这里搬运并整理一下网页上“What’s Included”部分。

Intel® oneAPI Base Toolkit (Base Kit)，The Intel® oneAPI Base Toolkit (Base Kit) is a core set of tools and libraries for developing high-performance, data-centric applications across diverse architectures. It features an industry-leading C++ compiler and the Data Parallel C++ (DPC++) language, an evolution of C++ for heterogeneous computing.
Add-on Toolkits for Specialized Workloads，Domain-specific libraries and the Intel® Distribution for Python* provide drop-in acceleration across relevant architectures. Enhanced profiling, design assistance, and debug tools complete the kit.

组件列表

Intel® oneAPI Collective Communications Library
Intel® oneAPI Data Analytics Library
Intel® oneAPI Deep Neural Network Library
Intel® oneAPI DPC++/C++ Compiler
Intel® oneAPI DPC++ Library
Intel® oneAPI Math Kernel Library
Intel® oneAPI Threading Building Blocks
Intel® oneAPI Video Processing Library
Intel® Advisor
Intel® Distribution for GDB*
Intel® Distribution for Python*
Intel® DPC++ Compatibility Tool
Intel® FPGA Add-on for oneAPI Base Toolkit
Intel® Integrated Performance Primitives （即IPP库）
Intel® VTune™ Profiler

文档

windows

get-started-with-ipp-for-oneapi-windows
Developer Guide(dev-guide-ipp-for-oneapi)

linux

install-using-package-managers

安装

注意检查系统要求 https://software.intel.com/content/www/us/en/develop/articles/intel-oneapi-base-toolkit-system-requirements.html，比如ubuntu18.04以上才能用

ubuntu 18.04

使用离线安装

参考教程
https://software.intel.com/content/www/us/en/develop/documentation/installation-guide-for-intel-oneapi-toolkits-linux/top/installation/install-with-command-line.html#install-with-command-line_INTERACTIVE

wget https://registrationcenter-download.intel.com/akdlm/irc_nas/17769/l_BaseKit_p_2021.2.0.2883_offline.sh
sudo sh ./l_BaseKit_p_2021.1.0.2659_offline.sh -s -a --silent --eula accept # 我觉得静默安装方便一些

参考安装过程如下

lmy@ubuntu-lmy:/media/lmy/Data/ubuntu/ipp$ sudo sh ./l_BaseKit_p_2021.1.0.2659_offline.sh -s -a --silent --eula accept
QStandardPaths: XDG_RUNTIME_DIR not set, defaulting to '/tmp/runtime-root'
Checking system requirements...
Done.
Wait while the installer is preparing...
Done.
Launching the installer...
QStandardPaths: XDG_RUNTIME_DIR not set, defaulting to '/tmp/runtime-root'
Intel® Graphics Compute Runtime for OpenCL™ not found.
For developing and running applications for GPU on this system, the latest version of the Intel® Graphics Driver is required. Please reference the GPU driver section at <a href="https://dgpu-docs.intel.com/installation-guides/index.html">https://dgpu-docs.intel.com/installation-guides/index.html</a> about GPU driver version(s) and installation information.
Start installation flow...
Installed Location: /opt/intel/oneapi
Installation has successfully completed
Log file: /opt/intel/oneapi/logs/installer.install.intel.oneapi.lin.basekit.product,v=2021.1.0-2659.2021.06.19.22.41.16.813721.log

Set Environment Variables

link

$ source <install-dir>/setvars.sh

测试

link
link2
https://software.intel.com/content/www/us/en/develop/articles/porting-guide-for-icc-users-to-dpcpp-or-icx.html

icc = intel c++ compiler

icx is one of icc

#include <stdio.h>
#include "ipp.h"

#define PRINT_INFO(feature, text) printf("  %-30s= ", #feature); \
    printf("%c\t%c\t", (cpuFeatures & feature) ? 'Y' : 'N', (enabledFeatures & feature) ? 'Y' : 'N'); \
    printf( #text "\n")

int main(int argc, char* argv[])
{
    const       IppLibraryVersion *libVersion;
    IppStatus   status;
    Ipp64u      cpuFeatures, enabledFeatures;

    ippInit();                      /* Initialize Intel(R) IPP library */
    libVersion = ippGetLibVersion();/* Get Intel(R) IPP library version info */
    printf("%s %s\n", libVersion->Name, libVersion->Version);

    status = ippGetCpuFeatures(&cpuFeatures, 0);/* Get CPU features and features enabled with selected library level */
    if (ippStsNoErr != status) return status;
    enabledFeatures = ippGetEnabledCpuFeatures();
    printf("Features supported: by CPU\tby Intel(R) IPP\n");
    printf("------------------------------------------------\n");
    PRINT_INFO(ippCPUID_MMX,        Intel(R) Architecture MMX technology supported);
    PRINT_INFO(ippCPUID_SSE,        Intel(R) Streaming SIMD Extensions);
    PRINT_INFO(ippCPUID_SSE2,       Intel(R) Streaming SIMD Extensions 2);
    PRINT_INFO(ippCPUID_SSE3,       Intel(R) Streaming SIMD Extensions 3);
    PRINT_INFO(ippCPUID_SSSE3,      Supplemental Streaming SIMD Extensions 3);
    PRINT_INFO(ippCPUID_MOVBE,      Intel(R) MOVBE instruction);
    PRINT_INFO(ippCPUID_SSE41,      Intel(R) Streaming SIMD Extensions 4.1);
    PRINT_INFO(ippCPUID_SSE42,      Intel(R) Streaming SIMD Extensions 4.2);
    PRINT_INFO(ippCPUID_AVX,        Intel(R) Advanced Vector Extensions instruction set);
    PRINT_INFO(ippAVX_ENABLEDBYOS,  Intel(R) Advanced Vector Extensions instruction set is supported by OS);
    PRINT_INFO(ippCPUID_AES,        Intel(R) AES New Instructions);
    PRINT_INFO(ippCPUID_CLMUL,      Intel(R) CLMUL instruction);
    PRINT_INFO(ippCPUID_RDRAND,     Intel(R) RDRAND instruction);
    PRINT_INFO(ippCPUID_F16C,       Intel(R) F16C new instructions);
    PRINT_INFO(ippCPUID_AVX2,       Intel(R) Advanced Vector Extensions 2 instruction set);
    PRINT_INFO(ippCPUID_ADCOX,      Intel(R) ADOX/ADCX new instructions);
    PRINT_INFO(ippCPUID_RDSEED,     Intel(R) RDSEED instruction);
    PRINT_INFO(ippCPUID_PREFETCHW,  Intel(R) PREFETCHW instruction);
    PRINT_INFO(ippCPUID_SHA,        Intel(R) SHA new instructions);
    PRINT_INFO(ippCPUID_AVX512F,    Intel(R) Advanced Vector Extensions 512 Foundation instruction set);
    PRINT_INFO(ippCPUID_AVX512CD,   Intel(R) Advanced Vector Extensions 512 CD instruction set);
    PRINT_INFO(ippCPUID_AVX512ER,   Intel(R) Advanced Vector Extensions 512 ER instruction set);
    PRINT_INFO(ippCPUID_AVX512PF,   Intel(R) Advanced Vector Extensions 512 PF instruction set);
    PRINT_INFO(ippCPUID_AVX512BW,   Intel(R) Advanced Vector Extensions 512 BW instruction set);
    PRINT_INFO(ippCPUID_AVX512VL,   Intel(R) Advanced Vector Extensions 512 VL instruction set);
    PRINT_INFO(ippCPUID_AVX512VBMI, Intel(R) Advanced Vector Extensions 512 Bit Manipulation instructions);
    PRINT_INFO(ippCPUID_MPX,        Intel(R) Memory Protection Extensions);
    PRINT_INFO(ippCPUID_AVX512_4FMADDPS,    Intel(R) Advanced Vector Extensions 512 DL floating-point single precision);
    PRINT_INFO(ippCPUID_AVX512_4VNNIW,      Intel(R) Advanced Vector Extensions 512 DL enhanced word variable precision);
    PRINT_INFO(ippCPUID_KNC,        Intel(R) Xeon Phi(TM) Coprocessor);
    PRINT_INFO(ippCPUID_AVX512IFMA, Intel(R) Advanced Vector Extensions 512 IFMA (PMADD52) instruction set);
    PRINT_INFO(ippAVX512_ENABLEDBYOS,       Intel(R) Advanced Vector Extensions 512 is supported by OS);
    return 0;
}

compile code

$ icx ipptest.cpp -o ipptest -I$IPPROOT/include -L$IPPROOT/lib/intel64 -lippcore
$ ./ipptest

my@ubuntu-lmy:~/myfile/ipp$ ./ipptest 
ippCore 2021.1 (r0xe13a8835)
Features supported: by CPU	by Intel(R) IPP
------------------------------------------------
  ippCPUID_MMX                  = Y	Y	Intel(R) Architecture MMX technology supported
  ippCPUID_SSE                  = Y	Y	Intel(R) Streaming SIMD Extensions
  ippCPUID_SSE2                 = Y	Y	Intel(R) Streaming SIMD Extensions 2
  ippCPUID_SSE3                 = Y	Y	Intel(R) Streaming SIMD Extensions 3
  ippCPUID_SSSE3                = Y	Y	Supplemental Streaming SIMD Extensions 3
  ippCPUID_MOVBE                = Y	Y	Intel(R) MOVBE instruction
  ippCPUID_SSE41                = Y	Y	Intel(R) Streaming SIMD Extensions 4.1
  ippCPUID_SSE42                = Y	Y	Intel(R) Streaming SIMD Extensions 4.2
  ippCPUID_AVX                  = Y	Y	Intel(R) Advanced Vector Extensions instruction set
  ippAVX_ENABLEDBYOS            = Y	Y	Intel(R) Advanced Vector Extensions instruction set is supported by OS
  ippCPUID_AES                  = Y	Y	Intel(R) AES New Instructions
  ippCPUID_CLMUL                = Y	Y	Intel(R) CLMUL instruction
  ippCPUID_RDRAND               = Y	Y	Intel(R) RDRAND instruction
  ippCPUID_F16C                 = Y	Y	Intel(R) F16C new instructions
  ippCPUID_AVX2                 = Y	Y	Intel(R) Advanced Vector Extensions 2 instruction set
  ippCPUID_ADCOX                = Y	Y	Intel(R) ADOX/ADCX new instructions
  ippCPUID_RDSEED               = Y	Y	Intel(R) RDSEED instruction
  ippCPUID_PREFETCHW            = Y	Y	Intel(R) PREFETCHW instruction
  ippCPUID_SHA                  = N	N	Intel(R) SHA new instructions
  ippCPUID_AVX512F              = N	N	Intel(R) Advanced Vector Extensions 512 Foundation instruction set
  ippCPUID_AVX512CD             = N	N	Intel(R) Advanced Vector Extensions 512 CD instruction set
  ippCPUID_AVX512ER             = N	N	Intel(R) Advanced Vector Extensions 512 ER instruction set
  ippCPUID_AVX512PF             = N	N	Intel(R) Advanced Vector Extensions 512 PF instruction set
  ippCPUID_AVX512BW             = N	N	Intel(R) Advanced Vector Extensions 512 BW instruction set
  ippCPUID_AVX512VL             = N	N	Intel(R) Advanced Vector Extensions 512 VL instruction set
  ippCPUID_AVX512VBMI           = N	N	Intel(R) Advanced Vector Extensions 512 Bit Manipulation instructions
  ippCPUID_MPX                  = Y	Y	Intel(R) Memory Protection Extensions
  ippCPUID_AVX512_4FMADDPS      = N	N	Intel(R) Advanced Vector Extensions 512 DL floating-point single precision
  ippCPUID_AVX512_4VNNIW        = N	N	Intel(R) Advanced Vector Extensions 512 DL enhanced word variable precision
  ippCPUID_KNC                  = N	N	Intel(R) Xeon Phi(TM) Coprocessor
  ippCPUID_AVX512IFMA           = N	N	Intel(R) Advanced Vector Extensions 512 IFMA (PMADD52) instruction set
  ippAVX512_ENABLEDBYOS         = N	N	Intel(R) Advanced Vector Extensions 512 is supported by OS

compile GetBilinearTransform.c

icx GetBilinearTransform.c -o GetBilinearTransform -I$IPPROOT/include -L$IPPROOT/lib/intel64 -lippi -lipps -lippcore

run

$ ./GetBilinearTransform 
Exit status 0 (ippStsNoErr: No errors)

名扬569

关注

0
点赞
踩
3

收藏

觉得还不错? 一键收藏
0
评论
复制链接

分享到 QQ

分享到新浪微博

扫一扫