1. mt08

    Posted

    mt08
Changes in title
+[メモ] らずぱい3: aarch64 openSUSEで、UnixBench
Changes in tags
Changes in body
Source | HTML | Preview
@@ -0,0 +1,515 @@
+## 概要
+- らずぱい3で、aarch64のUnixBenchも、とりたい
+- openSUSEのRPi3向け配布イメージは、aarch64対応している
+
+ ```shell-session:aarch64
+# uname -a
+Linux linux.suse 4.4.49-16-default #1 SMP Sun Feb 19 17:40:35 UTC 2017 (70e9954) aarch64 aarch64 aarch64 GNU/Linux
+```
+
+![RPi3_openSUSE.png](https://qiita-image-store.s3.amazonaws.com/0/75594/0609b910-6335-d970-eb89-b534b34de8ed.png)
+
+## 環境
+- Raspberry Pi 3
+- openSUSE-Leap42.2-ARM-XFCE-raspberrypi3.aarch64.raw.xz<br>
+[HCL:Raspberry_Pi3](https://en.opensuse.org/HCL:Raspberry_Pi3) ページから <br>> Installing the 64-bit openSUSE Leap image > XFCEイメージ選択
+ - デフォルトログイン > ユーザ: **root** / パスワード: **linux**
+- UART:115200bpsで、接続するのが、簡単かも。
+
+## 手順
+1. いつものツールでmicroSDにやく。
+2. 起動
+3. sshでログイン<br>`/etc/sshd/sshd_config`をrootでログイン可能+パスワード使用可能、と編集して、`systemctl restart sshd`
+4. パッケージ導入は、`zypper install ほにゃらら` で。
+
+
+## BYTE UNIX Benchmarks (Version 5.1.3)
+1. ここら辺をインストール<br>`zypper install make git-core gcc`
+2. Makefileを編集 (`-march=native -mtune=native`があるとエラー出る)
+
+ ```diff
+ diff --git a/UnixBench/Makefile b/UnixBench/Makefile
+ index d6868ba..1dd7e42 100644
+ --- a/UnixBench/Makefile
+ +++ b/UnixBench/Makefile
+ @@ -87,7 +87,7 @@ OPTON = -O3 -ffast-math
+ OSNAME:=$(shell uname -s)
+
+ ifeq ($(OSNAME),Linux)
+ -OPTON += -march=native -mtune=native
+ +#OPTON += -march=native -mtune=native
+ endif
+ ifeq ($(OSNAME),Darwin)
+ # (adjust flags or comment out this section for older versions of XCode or OS X)
+ ```
+
+
+### 結果: Benchmark of linux.suse / openSUSE 42.2 (aarch64) on Wed Mar 22 2017</h2>
+<p><b>BYTE UNIX Benchmarks (Version 5.1.3)</b></p>
+
+<h3>Test System Information</h3>
+<p><table>
+<tr>
+ <td><b>System:</b></td>
+ <td colspan=2>linux.suse: openSUSE 42.2 (aarch64)</td>
+</tr><tr>
+ <td><b>OS:</b></td>
+ <td colspan=2>GNU/Linux -- 4.4.49-16-default -- #1 SMP Sun Feb 19 17:40:35 UTC 2017 (70e9954)</td>
+</tr><tr>
+ <td><b>Machine:</b></td>
+ <td colspan=2>aarch64: aarch64</td>
+</tr><tr>
+ <td><b>Language:</b></td>
+ <td colspan=2>en_US.utf8 (charmap="UTF-8", collate="UTF-8")</td>
+</tr>
+<tr>
+ <td><b>Uptime:</b></td>
+ <td colspan=2>20:08pm up 0:07, 2 users, load average: 1.12, 0.80, 0.42; runlevel 1970-01-01</td>
+</tr>
+</table></p>
+
+<p><hr/></p>
+<h3>Benchmark Run: 4 CPUs; 1 parallel process</h3>
+<p>Time: 20:08:40 - 20:37:56; 29m 16s</p>
+
+<h4>System Benchmarks</h4>
+<p><table width="100%">
+<tr>
+ <th align=left>Test</th>
+ <th align=right>Score</th>
+ <th align=left>Unit</th>
+ <th align=right>Time</th>
+ <th align=right>Iters.</th>
+ <th align=right>Baseline</th>
+ <th align=right>Index</th>
+</tr>
+<tr>
+ <td><b>Dhrystone 2 using register variables</b></td>
+ <td align=right><tt>3379589.7</tt></td>
+ <td align=left><tt>lps</tt></td>
+ <td align=right><tt>10.0 s</tt></td>
+ <td align=right><tt>7</tt></td>
+ <td align=right><tt>116700.0</tt></td>
+ <td align=right><tt>289.6</tt></td>
+</tr>
+<tr>
+ <td><b>Double-Precision Whetstone</b></td>
+ <td align=right><tt>610.0</tt></td>
+ <td align=left><tt>MWIPS</tt></td>
+ <td align=right><tt>12.0 s</tt></td>
+ <td align=right><tt>7</tt></td>
+ <td align=right><tt>55.0</tt></td>
+ <td align=right><tt>110.9</tt></td>
+</tr>
+<tr>
+ <td><b>Execl Throughput</b></td>
+ <td align=right><tt>1049.4</tt></td>
+ <td align=left><tt>lps</tt></td>
+ <td align=right><tt>29.6 s</tt></td>
+ <td align=right><tt>2</tt></td>
+ <td align=right><tt>43.0</tt></td>
+ <td align=right><tt>244.0</tt></td>
+</tr>
+<tr>
+ <td><b>File Copy 1024 bufsize 2000 maxblocks</b></td>
+ <td align=right><tt>109187.0</tt></td>
+ <td align=left><tt>KBps</tt></td>
+ <td align=right><tt>30.0 s</tt></td>
+ <td align=right><tt>2</tt></td>
+ <td align=right><tt>3960.0</tt></td>
+ <td align=right><tt>275.7</tt></td>
+</tr>
+<tr>
+ <td><b>File Copy 256 bufsize 500 maxblocks</b></td>
+ <td align=right><tt>39170.3</tt></td>
+ <td align=left><tt>KBps</tt></td>
+ <td align=right><tt>30.0 s</tt></td>
+ <td align=right><tt>2</tt></td>
+ <td align=right><tt>1655.0</tt></td>
+ <td align=right><tt>236.7</tt></td>
+</tr>
+<tr>
+ <td><b>File Copy 4096 bufsize 8000 maxblocks</b></td>
+ <td align=right><tt>270879.3</tt></td>
+ <td align=left><tt>KBps</tt></td>
+ <td align=right><tt>30.0 s</tt></td>
+ <td align=right><tt>2</tt></td>
+ <td align=right><tt>5800.0</tt></td>
+ <td align=right><tt>467.0</tt></td>
+</tr>
+<tr>
+ <td><b>Pipe Throughput</b></td>
+ <td align=right><tt>212769.9</tt></td>
+ <td align=left><tt>lps</tt></td>
+ <td align=right><tt>10.0 s</tt></td>
+ <td align=right><tt>7</tt></td>
+ <td align=right><tt>12440.0</tt></td>
+ <td align=right><tt>171.0</tt></td>
+</tr>
+<tr>
+ <td><b>Pipe-based Context Switching</b></td>
+ <td align=right><tt>57328.4</tt></td>
+ <td align=left><tt>lps</tt></td>
+ <td align=right><tt>10.0 s</tt></td>
+ <td align=right><tt>7</tt></td>
+ <td align=right><tt>4000.0</tt></td>
+ <td align=right><tt>143.3</tt></td>
+</tr>
+<tr>
+ <td><b>Process Creation</b></td>
+ <td align=right><tt>2708.7</tt></td>
+ <td align=left><tt>lps</tt></td>
+ <td align=right><tt>30.0 s</tt></td>
+ <td align=right><tt>2</tt></td>
+ <td align=right><tt>126.0</tt></td>
+ <td align=right><tt>215.0</tt></td>
+</tr>
+<tr>
+ <td><b>Shell Scripts (1 concurrent)</b></td>
+ <td align=right><tt>1438.3</tt></td>
+ <td align=left><tt>lpm</tt></td>
+ <td align=right><tt>60.0 s</tt></td>
+ <td align=right><tt>2</tt></td>
+ <td align=right><tt>42.4</tt></td>
+ <td align=right><tt>339.2</tt></td>
+</tr>
+<tr>
+ <td><b>Shell Scripts (8 concurrent)</b></td>
+ <td align=right><tt>419.0</tt></td>
+ <td align=left><tt>lpm</tt></td>
+ <td align=right><tt>60.1 s</tt></td>
+ <td align=right><tt>2</tt></td>
+ <td align=right><tt>6.0</tt></td>
+ <td align=right><tt>698.3</tt></td>
+</tr>
+<tr>
+ <td><b>System Call Overhead</b></td>
+ <td align=right><tt>247097.0</tt></td>
+ <td align=left><tt>lps</tt></td>
+ <td align=right><tt>10.0 s</tt></td>
+ <td align=right><tt>7</tt></td>
+ <td align=right><tt>15000.0</tt></td>
+ <td align=right><tt>164.7</tt></td>
+</tr>
+<tr>
+ <td colspan=6><b>System Benchmarks Index Score:</b></td>
+ <td align=right><b><tt>246.0</tt></b></td>
+</tr>
+</table></p>
+
+<p><hr/></p>
+<h3>Benchmark Run: 4 CPUs; 4 parallel processes</h3>
+<p>Time: 20:37:56 - 21:06:55; 28m 59s</p>
+
+<h4>System Benchmarks</h4>
+<p><table width="100%">
+<tr>
+ <th align=left>Test</th>
+ <th align=right>Score</th>
+ <th align=left>Unit</th>
+ <th align=right>Time</th>
+ <th align=right>Iters.</th>
+ <th align=right>Baseline</th>
+ <th align=right>Index</th>
+</tr>
+<tr>
+ <td><b>Dhrystone 2 using register variables</b></td>
+ <td align=right><tt>23353355.0</tt></td>
+ <td align=left><tt>lps</tt></td>
+ <td align=right><tt>10.0 s</tt></td>
+ <td align=right><tt>7</tt></td>
+ <td align=right><tt>116700.0</tt></td>
+ <td align=right><tt>2001.1</tt></td>
+</tr>
+<tr>
+ <td><b>Double-Precision Whetstone</b></td>
+ <td align=right><tt>4196.7</tt></td>
+ <td align=left><tt>MWIPS</tt></td>
+ <td align=right><tt>9.9 s</tt></td>
+ <td align=right><tt>7</tt></td>
+ <td align=right><tt>55.0</tt></td>
+ <td align=right><tt>763.0</tt></td>
+</tr>
+<tr>
+ <td><b>Execl Throughput</b></td>
+ <td align=right><tt>2651.0</tt></td>
+ <td align=left><tt>lps</tt></td>
+ <td align=right><tt>29.9 s</tt></td>
+ <td align=right><tt>2</tt></td>
+ <td align=right><tt>43.0</tt></td>
+ <td align=right><tt>616.5</tt></td>
+</tr>
+<tr>
+ <td><b>File Copy 1024 bufsize 2000 maxblocks</b></td>
+ <td align=right><tt>278507.9</tt></td>
+ <td align=left><tt>KBps</tt></td>
+ <td align=right><tt>30.0 s</tt></td>
+ <td align=right><tt>2</tt></td>
+ <td align=right><tt>3960.0</tt></td>
+ <td align=right><tt>703.3</tt></td>
+</tr>
+<tr>
+ <td><b>File Copy 256 bufsize 500 maxblocks</b></td>
+ <td align=right><tt>76708.0</tt></td>
+ <td align=left><tt>KBps</tt></td>
+ <td align=right><tt>30.0 s</tt></td>
+ <td align=right><tt>2</tt></td>
+ <td align=right><tt>1655.0</tt></td>
+ <td align=right><tt>463.5</tt></td>
+</tr>
+<tr>
+ <td><b>File Copy 4096 bufsize 8000 maxblocks</b></td>
+ <td align=right><tt>671709.4</tt></td>
+ <td align=left><tt>KBps</tt></td>
+ <td align=right><tt>30.0 s</tt></td>
+ <td align=right><tt>2</tt></td>
+ <td align=right><tt>5800.0</tt></td>
+ <td align=right><tt>1158.1</tt></td>
+</tr>
+<tr>
+ <td><b>Pipe Throughput</b></td>
+ <td align=right><tt>1345451.5</tt></td>
+ <td align=left><tt>lps</tt></td>
+ <td align=right><tt>10.0 s</tt></td>
+ <td align=right><tt>7</tt></td>
+ <td align=right><tt>12440.0</tt></td>
+ <td align=right><tt>1081.6</tt></td>
+</tr>
+<tr>
+ <td><b>Pipe-based Context Switching</b></td>
+ <td align=right><tt>246604.4</tt></td>
+ <td align=left><tt>lps</tt></td>
+ <td align=right><tt>10.0 s</tt></td>
+ <td align=right><tt>7</tt></td>
+ <td align=right><tt>4000.0</tt></td>
+ <td align=right><tt>616.5</tt></td>
+</tr>
+<tr>
+ <td><b>Process Creation</b></td>
+ <td align=right><tt>7088.1</tt></td>
+ <td align=left><tt>lps</tt></td>
+ <td align=right><tt>30.0 s</tt></td>
+ <td align=right><tt>2</tt></td>
+ <td align=right><tt>126.0</tt></td>
+ <td align=right><tt>562.5</tt></td>
+</tr>
+<tr>
+ <td><b>Shell Scripts (1 concurrent)</b></td>
+ <td align=right><tt>3253.0</tt></td>
+ <td align=left><tt>lpm</tt></td>
+ <td align=right><tt>60.1 s</tt></td>
+ <td align=right><tt>2</tt></td>
+ <td align=right><tt>42.4</tt></td>
+ <td align=right><tt>767.2</tt></td>
+</tr>
+<tr>
+ <td><b>Shell Scripts (8 concurrent)</b></td>
+ <td align=right><tt>440.3</tt></td>
+ <td align=left><tt>lpm</tt></td>
+ <td align=right><tt>60.3 s</tt></td>
+ <td align=right><tt>2</tt></td>
+ <td align=right><tt>6.0</tt></td>
+ <td align=right><tt>733.8</tt></td>
+</tr>
+<tr>
+ <td><b>System Call Overhead</b></td>
+ <td align=right><tt>1562405.9</tt></td>
+ <td align=left><tt>lps</tt></td>
+ <td align=right><tt>10.0 s</tt></td>
+ <td align=right><tt>7</tt></td>
+ <td align=right><tt>15000.0</tt></td>
+ <td align=right><tt>1041.6</tt></td>
+</tr>
+<tr>
+ <td colspan=6><b>System Benchmarks Index Score:</b></td>
+ <td align=right><b><tt>808.6</tt></b></td>
+</tr>
+</table></p>
+
+<p><hr/></p>
+
+
+```txt:コンソールログ
+
+ # # # # # # # ##### ###### # # #### # #
+ # # ## # # # # # # # ## # # # # #
+ # # # # # # ## ##### ##### # # # # ######
+ # # # # # # ## # # # # # # # # #
+ # # # ## # # # # # # # ## # # # #
+ #### # # # # # ##### ###### # # #### # #
+
+ Version 5.1.3 Based on the Byte Magazine Unix Benchmark
+
+ Multi-CPU version Version 5 revisions by Ian Smith,
+ Sunnyvale, CA, USA
+ January 13, 2011 johantheghost at yahoo period com
+
+
+1 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
+
+1 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
+
+1 x Execl Throughput 1 2 3
+
+1 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
+
+1 x File Copy 256 bufsize 500 maxblocks 1 2 3
+
+1 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
+
+1 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
+
+1 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
+
+1 x Process Creation 1 2 3
+
+1 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
+
+1 x Shell Scripts (1 concurrent) 1 2 3
+
+1 x Shell Scripts (8 concurrent) 1 2 3
+
+4 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
+
+4 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
+
+4 x Execl Throughput 1 2 3
+
+4 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
+
+4 x File Copy 256 bufsize 500 maxblocks 1 2 3
+
+4 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
+
+4 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
+
+4 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
+
+4 x Process Creation 1 2 3
+
+4 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
+
+4 x Shell Scripts (1 concurrent) 1 2 3
+
+4 x Shell Scripts (8 concurrent) 1 2 3
+
+========================================================================
+ BYTE UNIX Benchmarks (Version 5.1.3)
+
+ System: linux.suse: openSUSE 42.2 (aarch64)
+ OS: GNU/Linux -- 4.4.49-16-default -- #1 SMP Sun Feb 19 17:40:35 UTC 2017 (70e9954)
+ Machine: aarch64 (aarch64)
+ Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
+ 20:08pm up 0:07, 2 users, load average: 1.12, 0.80, 0.42; runlevel 1970-01-01
+
+------------------------------------------------------------------------
+Benchmark Run: Wed Mar 22 2017 20:08:40 - 20:37:56
+4 CPUs in system; running 1 parallel copy of tests
+
+Dhrystone 2 using register variables 3379589.7 lps (10.0 s, 7 samples)
+Double-Precision Whetstone 610.0 MWIPS (12.0 s, 7 samples)
+Execl Throughput 1049.4 lps (29.6 s, 2 samples)
+File Copy 1024 bufsize 2000 maxblocks 109187.0 KBps (30.0 s, 2 samples)
+File Copy 256 bufsize 500 maxblocks 39170.3 KBps (30.0 s, 2 samples)
+File Copy 4096 bufsize 8000 maxblocks 270879.3 KBps (30.0 s, 2 samples)
+Pipe Throughput 212769.9 lps (10.0 s, 7 samples)
+Pipe-based Context Switching 57328.4 lps (10.0 s, 7 samples)
+Process Creation 2708.7 lps (30.0 s, 2 samples)
+Shell Scripts (1 concurrent) 1438.3 lpm (60.0 s, 2 samples)
+Shell Scripts (8 concurrent) 419.0 lpm (60.1 s, 2 samples)
+System Call Overhead 247097.0 lps (10.0 s, 7 samples)
+
+System Benchmarks Index Values BASELINE RESULT INDEX
+Dhrystone 2 using register variables 116700.0 3379589.7 289.6
+Double-Precision Whetstone 55.0 610.0 110.9
+Execl Throughput 43.0 1049.4 244.0
+File Copy 1024 bufsize 2000 maxblocks 3960.0 109187.0 275.7
+File Copy 256 bufsize 500 maxblocks 1655.0 39170.3 236.7
+File Copy 4096 bufsize 8000 maxblocks 5800.0 270879.3 467.0
+Pipe Throughput 12440.0 212769.9 171.0
+Pipe-based Context Switching 4000.0 57328.4 143.3
+Process Creation 126.0 2708.7 215.0
+Shell Scripts (1 concurrent) 42.4 1438.3 339.2
+Shell Scripts (8 concurrent) 6.0 419.0 698.3
+System Call Overhead 15000.0 247097.0 164.7
+ ========
+System Benchmarks Index Score 246.0
+
+------------------------------------------------------------------------
+Benchmark Run: Wed Mar 22 2017 20:37:56 - 21:06:55
+4 CPUs in system; running 4 parallel copies of tests
+
+Dhrystone 2 using register variables 23353355.0 lps (10.0 s, 7 samples)
+Double-Precision Whetstone 4196.7 MWIPS (9.9 s, 7 samples)
+Execl Throughput 2651.0 lps (29.9 s, 2 samples)
+File Copy 1024 bufsize 2000 maxblocks 278507.9 KBps (30.0 s, 2 samples)
+File Copy 256 bufsize 500 maxblocks 76708.0 KBps (30.0 s, 2 samples)
+File Copy 4096 bufsize 8000 maxblocks 671709.4 KBps (30.0 s, 2 samples)
+Pipe Throughput 1345451.5 lps (10.0 s, 7 samples)
+Pipe-based Context Switching 246604.4 lps (10.0 s, 7 samples)
+Process Creation 7088.1 lps (30.0 s, 2 samples)
+Shell Scripts (1 concurrent) 3253.0 lpm (60.1 s, 2 samples)
+Shell Scripts (8 concurrent) 440.3 lpm (60.3 s, 2 samples)
+System Call Overhead 1562405.9 lps (10.0 s, 7 samples)
+
+System Benchmarks Index Values BASELINE RESULT INDEX
+Dhrystone 2 using register variables 116700.0 23353355.0 2001.1
+Double-Precision Whetstone 55.0 4196.7 763.0
+Execl Throughput 43.0 2651.0 616.5
+File Copy 1024 bufsize 2000 maxblocks 3960.0 278507.9 703.3
+File Copy 256 bufsize 500 maxblocks 1655.0 76708.0 463.5
+File Copy 4096 bufsize 8000 maxblocks 5800.0 671709.4 1158.1
+Pipe Throughput 12440.0 1345451.5 1081.6
+Pipe-based Context Switching 4000.0 246604.4 616.5
+Process Creation 126.0 7088.1 562.5
+Shell Scripts (1 concurrent) 42.4 3253.0 767.2
+Shell Scripts (8 concurrent) 6.0 440.3 733.8
+System Call Overhead 15000.0 1562405.9 1041.6
+ ========
+System Benchmarks Index Score 808.6
+
+linux:~/byte-unixbench/UnixBench # uname -a
+Linux linux.suse 4.4.49-16-default #1 SMP Sun Feb 19 17:40:35 UTC 2017 (70e9954) aarch64 aarch64 aarch64 GNU/Linux
+linux:~/byte-unixbench/UnixBench # cat /etc/issue
+Welcome to openSUSE Leap 42.2 - Kernel \r (\l).
+
+
+linux:~/byte-unixbench/UnixBench # cat /proc/cpuinfo
+processor : 0
+BogoMIPS : 38.40
+Features : fp asimd evtstrm crc32
+CPU implementer : 0x41
+CPU architecture: 8
+CPU variant : 0x0
+CPU part : 0xd03
+CPU revision : 4
+
+processor : 1
+BogoMIPS : 38.40
+Features : fp asimd evtstrm crc32
+CPU implementer : 0x41
+CPU architecture: 8
+CPU variant : 0x0
+CPU part : 0xd03
+CPU revision : 4
+
+processor : 2
+BogoMIPS : 38.40
+Features : fp asimd evtstrm crc32
+CPU implementer : 0x41
+CPU architecture: 8
+CPU variant : 0x0
+CPU part : 0xd03
+CPU revision : 4
+
+processor : 3
+BogoMIPS : 38.40
+Features : fp asimd evtstrm crc32
+CPU implementer : 0x41
+CPU architecture: 8
+CPU variant : 0x0
+CPU part : 0xd03
+CPU revision : 4
+
+linux:~/byte-unixbench/UnixBench #
+```