学会使用perf性能分析工具（含移植到arm-linux开发板）

系统：Ubuntu18.04系统
内核版本：5.4.0-150-generic（通过uname -r查看）

一、在ubuntu中使用apt包下载Perf

输入perf命令，提示：

Command 'perf' not found, but can be installed with:

apt install linux-tools-common
Please ask your administrator.

按照提示安装linux-tools-common:

sudo apt-get install linux-tools-common

查看perf -v，提示：

WARNING: perf not found for kernel 5.4.0-150

  You may need to install the following packages for this specific kernel:
    linux-tools-5.4.0-150-generic
    linux-cloud-tools-5.4.0-150-generic

  You may also want to install one of the following packages to keep up to date:
    linux-tools-generic
    linux-cloud-tools-generic

由于linux有不同的发行版本，内核工具也不同，提示的意思是可以下载特定的内核版本下的tools，根据命令行的提示，使用uname -r 查看内核版本，并下载

sudo apt-get install linux-tools-5.4.0-150-generic

再次执行perf -v可以得到当前版本

perf version 5.4.233

参考博文：Perf的安装与简单使用

二、使用源码安装Perf，并移植到arm-linux环境下

perf是一种系统级性能分析工具，它涉及内核调用，所以安装需要知道内核版本。查看arm-linux平台的内核的版本，到官网上去下载特定的内核源码

在开发板终端输入

uname -r
5.4.31

去官网上下载内核源码，可以手动下载，也可以使用wget，以下操作在ubuntu平台

• Linux内核源码下载网站：Index of /sites/ftp.kernel.org/pub/linux/kernel/ (sjtu.edu.cn) 或者 Index of /kernel/linux/kernel/

• wget下载，例如：wget http://ftp.sjtu.edu.cn/sites/ftp.kernel.org/pub/linux/kernel/v5.x/linux-5.4.31.tar.gz

下载完毕之后，解压内核源代码

tar -zxvf linux-5.4.31.tar.gz

进入如下目录

cd linux-5.4.31/tools/perf/

修改Makefile.conf文件，添加LIBC_SUPPORT := 1

源码级安装，如有些依赖包没有安装，得安装一下，Perf源码安装的依赖：ubuntu源码安装性能分析工具perf - 知乎 (zhihu.com)

make -j4 && make install

当前文件夹生成perf应用，

file perf

输出结果：

将编译得到的应用发送到开发板上，运行成功

如果遇到其他问题可以参考博文：
linux perf arm,perf arm交叉编译

perf使用的博文：
perf性能分析工具使用分享
打卡不停更#Linux perf工具使用
perf学习-linux自带性能分析工具

三、使用perf

编写示例程序

#include <stdio.h>
void print(void)
{
  int i = 0;
   	while(1){
   		i++;
   	}
}
int main ()
{
   print();
   return 0;
}

编译目标位可执行文件

gcc -g test.c -o test

使用perf分析性能

perf stat test

出现报错

Error:
You may not have permission to collect stats.

Consider tweaking /proc/sys/kernel/perf_event_paranoid,
which controls use of the performance events system by
unprivileged users (without CAP_SYS_ADMIN).

The current value is 3:

  -1: Allow use of (almost) all events by all users
      Ignore mlock limit after perf_event_mlock_kb without CAP_IPC_LOCK
>= 0: Disallow ftrace function tracepoint by users without CAP_SYS_ADMIN
      Disallow raw tracepoint access by users without CAP_SYS_ADMIN
>= 1: Disallow CPU event access by users without CAP_SYS_ADMIN
>= 2: Disallow kernel profiling by users without CAP_SYS_ADMIN

To make this setting permanent, edit /etc/sysctl.conf too, e.g.:

	kernel.perf_event_paranoid = -1

根据提示到指定文件夹中查找该变量，发现没有，经过百度后替换perf_event_paranoid文件夹内容为-1，注意修改内核文件要加sudo

参考博文：perf have no permission to collect stats

重新执行

perf stat test

输出结果如下

Performance counter stats for 'test':

              0.74 msec task-clock                #    0.484 CPUs utilized          
                 0      context-switches          #    0.000 K/sec                  
                 0      cpu-migrations            #    0.000 K/sec                  
                58      page-faults               #    0.079 M/sec                  
   <not supported>      cycles                                                      
   <not supported>      instructions                                                
   <not supported>      branches                                                    
   <not supported>      branch-misses                                               

       0.001520723 seconds time elapsed

       0.001628000 seconds user
       0.000000000 seconds sys

四、Perf的功能介绍

可以使用许多 perf 命令来获取有关 CPU 使用情况、内存使用情况、I/O 使用情况等信息

>$ perf --help

 usage: perf [--version] [--help] [OPTIONS] COMMAND [ARGS]

 The most commonly used perf commands are:
   annotate        Read perf.data (created by perf record) and display annotated code
   archive         Create archive with object files with build-ids found in perf.data file
   bench           General framework for benchmark suites
   buildid-cache   Manage build-id cache.
   buildid-list    List the buildids in a perf.data file
   c2c             Shared Data C2C/HITM Analyzer.
   config          Get and set variables in a configuration file.
   data            Data file related processing
   diff            Read perf.data files and display the differential profile
   evlist          List the event names in a perf.data file
   ftrace          simple wrapper for kernel's ftrace functionality
   inject          Filter to augment the events stream with additional information
   kallsyms        Searches running kernel for symbols
   kmem            Tool to trace/measure kernel memory properties
   kvm             Tool to trace/measure kvm guest os
   list            List all symbolic event types
   lock            Analyze lock events
   mem             Profile memory accesses
   record          Run a command and record its profile into perf.data #可以记录指定进程的指定事件
   report          Read perf.data (created by perf record) and display the profile #可以生成一份报告，显示前面记录的性能参数
   sched           Tool to trace/measure scheduler properties (latencies)
   script          Read perf.data (created by perf record) and display trace output
   stat            Run a command and gather performance counter statistics  #可以查看指定命令的CPU和内存使用情况
   test            Runs sanity tests.
   timechart       Tool to visualize total system behavior during a workload
   top             System profiling tool.
   version         display the version of perf binary
   probe           Define new dynamic tracepoints
   trace           strace inspired tool

 See 'perf help COMMAND' for more information on a specific command.