Diary/2020-12-14
Ubuntu on Trenz TE0802
こんな感じでビルド https://github.com/miyo/build_linux_te8020
動いたので，UnixBenchとSTREAMで簡単に性能測定．
UnixBenchは，
------------------------------------------------------------------------
Benchmark Run: Sun Jan 28 2018 16:58:27 - 17:26:30
2 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 6368093.6 lps (10.0 s, 7 samples)
Double-Precision Whetstone 1156.5 MWIPS (9.8 s, 7 samples)
Execl Throughput 1550.0 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 130860.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 47744.1 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 236150.0 KBps (30.0 s, 2 samples)
Pipe Throughput 411918.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 76329.0 lps (10.0 s, 7 samples)
Process Creation 3657.7 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2007.3 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 352.7 lpm (60.1 s, 2 samples)
System Call Overhead 606407.4 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 6368093.6 545.7
Double-Precision Whetstone 55.0 1156.5 210.3
Execl Throughput 43.0 1550.0 360.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 130860.5 330.5
File Copy 256 bufsize 500 maxblocks 1655.0 47744.1 288.5
File Copy 4096 bufsize 8000 maxblocks 5800.0 236150.0 407.2
Pipe Throughput 12440.0 411918.1 331.1
Pipe-based Context Switching 4000.0 76329.0 190.8
Process Creation 126.0 3657.7 290.3
Shell Scripts (1 concurrent) 42.4 2007.3 473.4
Shell Scripts (8 concurrent) 6.0 352.7 587.8
System Call Overhead 15000.0 606407.4 404.3
========
System Benchmarks Index Score 349.4
------------------------------------------------------------------------
Benchmark Run: Sun Jan 28 2018 17:26:30 - 17:54:34
2 CPUs in system; running 2 parallel copies of tests
Dhrystone 2 using register variables 12736019.1 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2313.3 MWIPS (9.8 s, 7 samples)
Execl Throughput 2942.4 lps (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 249350.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 88072.8 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 466083.0 KBps (30.0 s, 2 samples)
Pipe Throughput 815794.7 lps (10.0 s, 7 samples)
Pipe-based Context Switching 151154.1 lps (10.0 s, 7 samples)
Process Creation 6492.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2801.4 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 344.1 lpm (60.3 s, 2 samples)
System Call Overhead 1170907.7 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 12736019.1 1091.3
Double-Precision Whetstone 55.0 2313.3 420.6
Execl Throughput 43.0 2942.4 684.3
File Copy 1024 bufsize 2000 maxblocks 3960.0 249350.5 629.7
File Copy 256 bufsize 500 maxblocks 1655.0 88072.8 532.2
File Copy 4096 bufsize 8000 maxblocks 5800.0 466083.0 803.6
Pipe Throughput 12440.0 815794.7 655.8
Pipe-based Context Switching 4000.0 151154.1 377.9
Process Creation 126.0 6492.2 515.3
Shell Scripts (1 concurrent) 42.4 2801.4 660.7
Shell Scripts (8 concurrent) 6.0 344.1 573.4
System Call Overhead 15000.0 1170907.7 780.6
========
System Benchmarks Index Score 619.5
STREAMは，
user@qdev:~/STREAM-master$ ./stream ------------------------------------------------------------- STREAM version $Revision: 5.10 $ ------------------------------------------------------------- This system uses 8 bytes per array element. ------------------------------------------------------------- Array size = 20000000 (elements), Offset = 0 (elements) Memory per array = 152.6 MiB (= 0.1 GiB). Total memory required = 457.8 MiB (= 0.4 GiB). Each kernel will be executed 10 times. The *best* time for each kernel (excluding the first iteration) will be used to compute the reported bandwidth. ------------------------------------------------------------- Number of Threads requested = 2 Number of Threads counted = 2 ------------------------------------------------------------- Your clock granularity/precision appears to be 1 microseconds. Each test below will take on the order of 293097 microseconds. (= 293097 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: 1385.3 0.232937 0.230991 0.234820 Scale: 1369.7 0.235670 0.233622 0.237338 Add: 1234.7 0.390185 0.388745 0.392157 Triad: 1181.9 0.412305 0.406122 0.419152 ------------------------------------------------------------- Solution Validates: avg error less than 1.000000e-13 on all three arrays ------------------------------------------------------------- user@qdev:~/STREAM-master$ OMP_NUM_THREADS=1 ./stream ------------------------------------------------------------- STREAM version $Revision: 5.10 $ ------------------------------------------------------------- This system uses 8 bytes per array element. ------------------------------------------------------------- Array size = 20000000 (elements), Offset = 0 (elements) Memory per array = 152.6 MiB (= 0.1 GiB). Total memory required = 457.8 MiB (= 0.4 GiB). Each kernel will be executed 10 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 439203 microseconds. (= 439203 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: 1149.0 0.280370 0.278513 0.283598 Scale: 1144.7 0.282159 0.279549 0.285550 Add: 958.7 0.501143 0.500660 0.501841 Triad: 938.1 0.512350 0.511683 0.513597 ------------------------------------------------------------- Solution Validates: avg error less than 1.000000e-13 on all three arrays -------------------------------------------------------------
FreeBSD
久しぶりにFreeBSDのセットアップ．実マシンじゃなくてWindows上のVMに．
バージョンは12.2-RELEASE．
生活環境としてpkg installで emacs，Xorg, tightvnc，tgif，samba413，ja-font-migmixなどをインストール．
開発環境としてはgcc，ruby，jdkなどをインストール．
sambaの設定は/usr/local/etc/smb4.conf．
とりあえず，こんな感じに
[global] workgroup = WORKGROUP server string = FreeBSD security = user hosts allow = 192.168. guest ok = no unix charset = UTF-8 dos charset = CP932 max protocol = SMB2 netbios name = microserver create mask = 644 force create mode = 644 directory mask = 775 force directory mode = 775 printing = bsd unix extensions = no nt acl support = yes inherit acls = no map acl inherit = yes map archive = no domain master = no local master = no preferred master = no os level = 0 oplocks = No level2 oplocks = No [homes] comment = Home Directories writable = yes browseable = yes write list = wizard guest ok = no read only = no
実行は，
/usr/local/etc/rc.d/samba_server onestart
とか．次回以降は自動起動したいので，/etc/rc.confに
samba_server_enable="YES" winbindd_enable="YES"
と書いておく．