目录
- 一、概括
- 二、stream
一、概括
嵌入式开发中对要设计的产品、立项的项目进行设计时，往往需要对关键芯片进行性能评估，本文主要总结基于linux系统的产品在性能评估时的工具使用总结，在aarch64(arm64平台下测试)，板卡根文件系统为debian系统。
工具列表如下：
| 名称 | 作用 | git源码链接 |
|---|---|---|
| lmbench | 带宽测评，反应时间测评 | https://github.com/redrose2100/lmbench.git |
| stream | 内存带宽（每秒通过的字节数）测试 | https://github.com/jeffhammond/STREAM.git |
| unixbench | 测试 unix 系统基本性能，测试的结果不仅仅只是CPU,内存,或者磁盘为基准,还取决于硬件,操作系统版本,编译器. | https://github.com/kdlucas/byte-unixbench.git |
| cyclictest 和 stress-ng | 实时性测试 | 压力工具 git clone https://github.com/ColinIanKing/stress-ng.git ，测试工具git://git.kernel.org/pub/scm/utils/rt-tests/rt-tests.git |
二、stream
1、编译
修改Makefile为以下
CC ?= gcc
CFLAGS = -O3 -fno-PIC -mcmodel=large -fopenmp -DSTREAM_ARRAY_SIZE=200000000 -DNTIMES=30
all: stream
clean:
rm -f stream *.o
stream: stream.c
$(CC) $(CFLAGS) stream.c -o stream
export CC=aarch64-linux-gnu-gcc
make
2、将编译后的stream拷贝到嵌入式板卡中
3、运行测试
单线程
export OMP_NUM_THREADS=1
./stream > stream-result-1thread.txt
多线程,以8线程为例，这里cpu核数为8，如果是单核单线程的话最大支持到8
export OMP_NUM_THREADS=8
export GOMP_CPU_AFFINITY=0-7
./stream > stream-result-8thread.txt
4、运行出现的问题
提示如下错误
./stream: error while loading shared libraries: libgomp.so.1: cannot open shared object file: No such file or directory
debian环境下 dpkg -i libgomp1_8.3.0-6_arm64.deb
再次
buildroot环境下编译时打开对此包编译的选项；
5、结果和解释
单线程
-------------------------------------------------------------
STREAM version $Revision: 5.10 $
-------------------------------------------------------------
This system uses 8 bytes per array element.
-------------------------------------------------------------
Array size = 200000000 (elements), Offset = 0 (elements)
Memory per array = 1525.9 MiB (= 1.5 GiB).
Total memory required = 4577.6 MiB (= 4.5 GiB).
Each kernel will be executed 30 times.
The *best* time for each kernel (excluding the first iteration)
will be used to compute the reported bandwidth.
-------------------------------------------------------------
Number of Threads requested = 1
Number of Threads counted = 1
-------------------------------------------------------------
Your clock granularity/precision appears to be 1 microseconds.
Each test below will take on the order of 347944 microseconds.
(= 347944 clock ticks)
Increase the size of the arrays if this shows that
you are not getting at least 20 clock ticks per test.
-------------------------------------------------------------
WARNING -- The above is only a rough guideline.
For best results, please be sure you know the
precision of your system timer.
-------------------------------------------------------------
Function Best Rate MB/s Avg time Min time Max time
Copy: 11397.9 0.280894 0.280754 0.281033
Scale: 10245.3 0.312539 0.312339 0.313669
Add: 8855.1 0.542250 0.542060 0.542685
Triad: 8857.6 0.542100 0.541906 0.542925
-------------------------------------------------------------
Solution Validates: avg error less than 1.000000e-13 on all three arrays
-------------------------------------------------------------
多线程
-------------------------------------------------------------
STREAM version $Revision: 5.10 $
-------------------------------------------------------------
This system uses 8 bytes per array element.
-------------------------------------------------------------
Array size = 200000000 (elements), Offset = 0 (elements)
Memory per array = 1525.9 MiB (= 1.5 GiB).
Total memory required = 4577.6 MiB (= 4.5 GiB).
Each kernel will be executed 30 times.
The *best* time for each kernel (excluding the first iteration)
will be used to compute the reported bandwidth.
-------------------------------------------------------------
Number of Threads requested = 8
Number of Threads counted = 8
-------------------------------------------------------------
Your clock granularity/precision appears to be 1 microseconds.
Each test below will take on the order of 339367 microseconds.
(= 339367 clock ticks)
Increase the size of the arrays if this shows that
you are not getting at least 20 clock ticks per test.
-------------------------------------------------------------
WARNING -- The above is only a rough guideline.
For best results, please be sure you know the
precision of your system timer.
-------------------------------------------------------------
Function Best Rate MB/s Avg time Min time Max time
Copy: 14378.2 0.223113 0.222559 0.223810
Scale: 12578.0 0.257082 0.254413 0.260384
Add: 10312.8 0.468002 0.465440 0.470596
Triad: 8938.6 0.542479 0.536994 0.548937
-------------------------------------------------------------
Solution Validates: avg error less than 1.000000e-13 on all three arrays
-------------------------------------------------------------
说明
关注以下四行的每秒字节数和时间
Function Best Rate MB/s Avg time Min time Max time
Copy: 14378.2 0.223113 0.222559 0.223810
Scale: 12578.0 0.257082 0.254413 0.260384
Add: 10312.8 0.468002 0.465440 0.470596
Triad: 8938.6 0.542479 0.536994 0.548937
