龙芯3A6000性能测试——UnixBench跑分

1 下载UnixBench源码

下载UnixBench-5.1.2
地址：

https://soft.lnmp.com/test/unixbench/

2 解压与编译

2.1 解压

将tar.gz压缩包移动到自己的文件夹下。
进入当前目录的终端，解压：

tar -xzvf unixbench-5.1.2.tar.gz

解压完成后，进入文件夹。

cd unixbench-5.1.2

2.2 修改Makefile

使用gedit编辑Makefile，终端输入：

gedit Makefile

在gedit编辑器内，注释掉GRAPHIC_TESTS = defined，不进行显卡测试，保存。

# GRAPHICS TESTS: Uncomment the definition of "GRAPHIC_TESTS" to enable
# the building of the graphics benchmarks.  This will require the
# X11 libraries on your system.
#
# Comment the line out to disable these tests.
# GRAPHIC_TESTS = defined

2.3 编译

回到终端，编译程序：

make

2.4 运行

运行性能测试：

./Run

等待约30多分钟，输出结果：

   BYTE UNIX Benchmarks (Version 5.1.2)

   System: ren: GNU/Linux
   OS: GNU/Linux -- 4.19.0-19-loongson-3 -- #1 SMP 4.19.190.8.14 Thu Aug 24 08:54:20 UTC 2023
   Machine: loongarch64 (loongarch64)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: Loongson-3A6000-HV (0.0 bogomips)
          
   CPU 1: Loongson-3A6000-HV (0.0 bogomips)
          
   CPU 2: Loongson-3A6000-HV (0.0 bogomips)
          
   CPU 3: Loongson-3A6000-HV (0.0 bogomips)
          
   CPU 4: Loongson-3A6000-HV (0.0 bogomips)
          
   CPU 5: Loongson-3A6000-HV (0.0 bogomips)
          
   CPU 6: Loongson-3A6000-HV (0.0 bogomips)
          
   CPU 7: Loongson-3A6000-HV (0.0 bogomips)
          
   15:12:58 up 35 min,  1 user,  load average: 1.64, 2.07, 1.47; runlevel 

------------------------------------------------------------------------
Benchmark Run: 二 1月 30 2024 15:12:58 - 15:40:57
8 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       44026633.8 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     4856.6 MWIPS (9.3 s, 7 samples)
Execl Throughput                               6954.8 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1142505.6 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          318176.3 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       2684418.0 KBps  (30.0 s, 2 samples)
Pipe Throughput                             2126858.1 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 353275.7 lps   (10.0 s, 7 samples)
Process Creation                              15124.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  14375.6 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   5117.3 lpm   (60.0 s, 2 samples)
System Call Overhead                        3159763.7 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   44026633.8   3772.6
Double-Precision Whetstone                       55.0       4856.6    883.0
Execl Throughput                                 43.0       6954.8   1617.4
File Copy 1024 bufsize 2000 maxblocks          3960.0    1142505.6   2885.1
File Copy 256 bufsize 500 maxblocks            1655.0     318176.3   1922.5
File Copy 4096 bufsize 8000 maxblocks          5800.0    2684418.0   4628.3
Pipe Throughput                               12440.0    2126858.1   1709.7
Pipe-based Context Switching                   4000.0     353275.7    883.2
Process Creation                                126.0      15124.8   1200.4
Shell Scripts (1 concurrent)                     42.4      14375.6   3390.5
Shell Scripts (8 concurrent)                      6.0       5117.3   8528.9
System Call Overhead                          15000.0    3159763.7   2106.5
                                                                   ========
System Benchmarks Index Score                                        2231.8

------------------------------------------------------------------------
Benchmark Run: 二 1月 30 2024 15:40:57 - 16:09:08
8 CPUs in system; running 8 parallel copies of tests

Dhrystone 2 using register variables      196841849.2 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                    36083.7 MWIPS (9.4 s, 7 samples)
Execl Throughput                              35339.7 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1210752.9 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          323595.9 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       3620712.3 KBps  (30.0 s, 2 samples)
Pipe Throughput                            11239607.5 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                1994861.1 lps   (10.0 s, 7 samples)
Process Creation                              73916.9 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  46910.1 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   6970.6 lpm   (60.0 s, 2 samples)
System Call Overhead                       22652905.7 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0  196841849.2  16867.3
Double-Precision Whetstone                       55.0      36083.7   6560.7
Execl Throughput                                 43.0      35339.7   8218.5
File Copy 1024 bufsize 2000 maxblocks          3960.0    1210752.9   3057.5
File Copy 256 bufsize 500 maxblocks            1655.0     323595.9   1955.3
File Copy 4096 bufsize 8000 maxblocks          5800.0    3620712.3   6242.6
Pipe Throughput                               12440.0   11239607.5   9035.1
Pipe-based Context Switching                   4000.0    1994861.1   4987.2
Process Creation                                126.0      73916.9   5866.4
Shell Scripts (1 concurrent)                     42.4      46910.1  11063.7
Shell Scripts (8 concurrent)                      6.0       6970.6  11617.7
System Call Overhead                          15000.0   22652905.7  15101.9
                                                                   ========
System Benchmarks Index Score                                        7135.3

由于运行测试时，我的电脑还在允许其他程序，因此可能会对分数有一定影响。