---
title: 构建及配置
description: "统一使用matmul进行案例演示，且示例的代码中已经基于插桩指导(https://www.hikunpeng.com/document/detail/zh/kunpengdevps/profiler/profiler/KunpengDevKitCli_0064.html)增加相关插桩代码。1. 下载并安装KML。"
url: https://www.hikunpeng.com/document/detail/zh/kunpengdevps/profiler/demo/KnowHow_0011.html
sourcePath: /source/zh/kunpengdevps/profiler/demo/KnowHow_0011.html
indexId: 3f4c25bf2ffac0ee17673cf9b469fb7a7c787dde3e3c589cd9a440f2af40c9fd55
---
# 构建及配置

统一使用matmul进行案例演示，且示例的代码中已经基于插桩指导(https://www.hikunpeng.com/document/detail/zh/kunpengdevps/profiler/profiler/KunpengDevKitCli_0064.html)增加相关插桩代码。1. 下载并安装KML。
  1 2 3 4 unzip -o BoostKit-kml_2.2.0.zip rpm -ivh boostkit-kml-2.2.0-1.aarch64.rpm export KML_INCLUDE=/usr/local/kml/include/ export KML_LIB=/usr/local/kml/lib/kblas/omp/

2. 下载Demo并编译matmul。
  1 2 3 4 5 git clone xxx cd xxx export RFEVENTS_INCLUDE=/usr/local/devkit/tuner/include/ export RFEVENTS_LIB=/usr/local/devkit/tuner/lib/ make matmul

  “/usr/local/devkit/tuner”为使用鲲鹏DevKit命令行工具的RPM包安装后生成的路径，使用时请根据实际路径自行替换。

  其中matmul的Makefile信息如下：

  1 gcc -g -march=armv8.2-a+fp16fml+simd -O3 -fopenmp $(SOURCES) -o matmul -DROOFLINE_EVENTS -I$(RFEVENTS_INCLUDE) -L$(RFEVENTS_LIB) -lrfevents -DENABLE_KML -I$(KML_INCLUDE) -L$(KML_LIB) -lkblas

3. 配置环境变量LD_LIBRARY_PATH。
  1 export LD_LIBRARY_PATH=$KML_INCLUDE:$KML_LIB:/usr/local/devkit/tuner/lib:$LD_LIBRARY_PATH

4. 根据环境实际物理核数显式指定OMP_NUM_THREADS。
  1 export OMP_NUM_THREADS=128