・2021/11/10
Raspberry Pi OS Bullseyeは Raspberry Pi 4 Model Bで 1.8GHzの動作をサポートします
(ラズパイ４Ｂで 1.8GHz動作が可能 config.txtに arm_freq=1800を追加、ただし動作対象の条件有り)
Tags: [Raspberry Pi], [電子工作], [海外通販], [通信販売]
●ラズパイを製品保証を無効にしないでオーバークロックする方法
・2021/05/01
ラズパイを製品保証を無効にしないでオーバークロックする方法
Raspberry Pi Over Clock DO NOT VOID Warranty
● Raspberry Pi OS Bullseyeは Raspberry Pi 4 Model Bで 1.8GHzの動作をサポートします
ラズパイ４Ｂで 1.8GHz動作が可能 config.txtに arm_freq=1800を追加、ただし動作対象の条件有り。
2021/11/28追記 下記のベンチマークの追試確認で、config.txtに arm_freq=1800を追加していない状態でも 1.8GHzで動作している事が確認できました。（＝ 何もしなくても勝手に 1.8GHz動作になる）
Raspberry Pi OS
Debian version 11 bullseye
Bullseye - the new version of Raspberry Pi OS
Bullseye bonus: 1.8GHz Raspberry Pi 4
●ラズパイ４Ｂの 1.8GHzの動作対象の条件
基板の電源部分が下記になっている事。
※ 「基盤」は間違いです。板 Board 「基板」です。
| ラズパイ４Ｂ 8GB | 全部 1.8GHzの動作対象 |
| ラズパイ４Ｂ 4GB | 最近のロットが 1.8GHzの動作対象 |
| ラズパイ４Ｂ 2GB | 最近のロットが 1.8GHzの動作対象 |
| ラズパイ４Ｂ 1GB | 全部動作対象外 |
●ラズパイ４Ｂの 1.8GHzの動作対象の確認方法（目視）
・1.8GHzの動作対象
・2.8GHzの動作対象外
●ラズパイ４Ｂの 1.8GHzの動作対象の確認方法（コマンドライン）
Raspberry Pi Hardware - Raspberry Pi Revision Codes
Raspberry Pi OS Based on Debian version 11 (bullseye)
Release date: October 30th 2021
$ uname -a
Linux raspberrypi 5.10.63-v7l+ #1459 SMP Wed Oct 6 16:41:57 BST 2021 armv7l GNU/Linux
$ cat /proc/cpuinfo
processor : 0
model name : ARMv7 Processor rev 3 (v7l)
BogoMIPS : 108.00
Features : half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm crc32
CPU implementer : 0x41
CPU architecture: 7
CPU variant : 0x0
CPU part : 0xd08
CPU revision : 3
...
Hardware : BCM2711
Revision : d03114
Serial : 10000000xxxxxxxx
Model : Raspberry Pi 4 Model B Rev 1.4
$ cat /proc/cpuinfo | grep Rev
Revision : d03114
Model : Raspberry Pi 4 Model B Rev 1.4
● 1.8GHz動作に対応しているのは「Rev 1.4」以上（2021年 11月現在 Rev 1.4が最新）
| Revision code | Model | Memory | Rev. |
| b03114 | Pi 4B | 2GB | Rev 1.4 |
| c03114 | Pi 4B | 4GB | Rev 1.4 |
| d03114 | Pi 4B | 8GB | Rev 1.4 |
※ 2021年 12月現在 Rev 1.5が最新
ソース
”New” old functionality with Raspberry Pi OS (Legacy)
new Raspberry Pi 4 rev 1.5 (which usually means component changes for supply reasons)
● Raspberry Piの動作周波数を確認する方法 vcgencmd
動作中の「その時」の動作周波数を表示します。
● Raspberry Piに負荷が有る時は 1.5GHz
$ vcgencmd measure_clock arm
frequency(48)=1500345728
※ config.txtに arm_freq=1500を追加
● Raspberry Piに負荷が無い時は 600MHz
$ vcgencmd measure_clock arm
frequency(48)=600117184
● Raspberry Piに負荷が有る時は 1.8GHz
$ vcgencmd measure_clock arm
frequency(48)=1800457088
※ config.txtに arm_freq=1800を追加
※ config.txtに arm_freq=1800を追加していない時も 1.8GHzで動作していました
$ watch vcgencmd measure_clock arm
watchコマンドで連続で表示します（観察時に便利）
● Raspberry Pi 4Bで 1.5GHzと 1.8GHzのベンチマーク比較
/boot/config.txtに下記の行を追加（編集）する。
$ sudo nano /boot/config.txt
# 1.5GHz動作の場合
arm_freq=1500
# 1.8GHz動作の場合
arm_freq=1800
# 2.0GHz動作の場合
arm_freq=2000
# 2.2GHz動作の場合
arm_freq=2200
Linuxのベンチマークで標準的な UnixBenchを使います。
| config.txtは未編集 | config.txtに arm_freq=1800を追加 | |
| 1.5GHz？ | 1.8GHz | UP率（理論値 20% UP） |
| 1 core | 319.0 | 321.0 | UPしていないなぁ？ |
| 4 core | 854.8 | 866.0 | UPしていないなぁ？ |
※ config.txtに arm_freq=1800を追加しなくても 1.8GHzで動いていた説？
ビンゴ！ config.txtに arm_freq=1800を追加していない時も 1.8GHzで動作していました！！
●正しくは下記の表
明示的に arm_freq=1500を追加を追加して 1.5GHzでのベンチマークを測定します。
| config.txtに arm_freq=1500を追加 | config.txtに arm_freq=1800を追加 | config.txtは未編集 | |
| 1.5GHz | 1.8GHz | 1.8GHz | UP率（理論値 20% UP） |
| 1 core | 268.4 | 319.0 | 321.0 | 19.6% UP |
| 4 core | 770.7 | 854.8 | 866.0 | 12.3% UP |
とりあえず、そこそこ納得のいく測定結果が得られました。
File Copy等の I/O周りのボトルネックが有るので 4 coreの成績が CPUの動作速度に比例していないと思われる。
# Benchmark UnixBench byte-unixbench
cd
git clone https://github.com/kdlucas/byte-unixbench --depth 1
cd byte-unixbench
cd UnixBench
./Run
● config.txtに arm_freq=1800の追加無しの状態
※ config.txtに arm_freq=1800を追加しなくても 1.8GHzで動いていた
$ ./Run
make all
make[1]: Entering directory '/home/pi/byte-unixbench/UnixBench'
make distr
make[2]: Entering directory '/home/pi/byte-unixbench/UnixBench'
Checking distribution of files
./pgms exists
./src exists
./testdir exists
./tmp exists
./results exists
make[2]: Leaving directory '/home/pi/byte-unixbench/UnixBench'
make programs
make[2]: Entering directory '/home/pi/byte-unixbench/UnixBench'
make[2]: Nothing to be done for 'programs'.
make[2]: Leaving directory '/home/pi/byte-unixbench/UnixBench'
make[1]: Leaving directory '/home/pi/byte-unixbench/UnixBench'
locale: Cannot set LC_CTYPE to default locale: No such file or directory
locale: Cannot set LC_MESSAGES to default locale: No such file or directory
locale: Cannot set LC_ALL to default locale: No such file or directory
locale: Cannot set LC_CTYPE to default locale: No such file or directory
locale: Cannot set LC_MESSAGES to default locale: No such file or directory
locale: Cannot set LC_ALL to default locale: No such file or directory
sh: 1: 3dinfo: not found
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
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
------------------------------------------------------------------------------
Use directories for:
* File I/O tests (named fs***) = /home/pi/byte-unixbench/UnixBench/tmp
* Results = /home/pi/byte-unixbench/UnixBench/results
------------------------------------------------------------------------------
Use of uninitialized value in printf at ./Run line 1469.
Use of uninitialized value in printf at ./Run line 1470.
Use of uninitialized value in printf at ./Run line 1469.
Use of uninitialized value in printf at ./Run line 1470.
Use of uninitialized value in printf at ./Run line 1469.
Use of uninitialized value in printf at ./Run line 1470.
Use of uninitialized value in printf at ./Run line 1469.
Use of uninitialized value in printf at ./Run line 1470.
Use of uninitialized value in printf at ./Run line 1721.
Use of uninitialized value in printf at ./Run line 1722.
Use of uninitialized value in printf at ./Run line 1721.
Use of uninitialized value in printf at ./Run line 1722.
Use of uninitialized value in printf at ./Run line 1721.
Use of uninitialized value in printf at ./Run line 1722.
Use of uninitialized value in printf at ./Run line 1721.
Use of uninitialized value in printf at ./Run line 1722.
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: raspberrypi: GNU/Linux
OS: GNU/Linux -- 5.10.63-v7l+ -- #1459 SMP Wed Oct 6 16:41:57 BST 2021
Machine: armv7l (unknown)
Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
CPU 0: ARMv7 Processor rev 3 (v7l) (0.0 bogomips)
CPU 1: ARMv7 Processor rev 3 (v7l) (0.0 bogomips)
CPU 2: ARMv7 Processor rev 3 (v7l) (0.0 bogomips)
CPU 3: ARMv7 Processor rev 3 (v7l) (0.0 bogomips)
05:08:30 up 27 min, 3 users, load average: 0.43, 0.14, 0.32; runlevel Oct
------------------------------------------------------------------------
Benchmark Run: Sun Nov 28 2021 05:08:30 - 05:37:00
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 12288814.1 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3042.9 MWIPS (10.0 s, 7 samples)
Execl Throughput 1016.1 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 102300.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 28716.5 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 286287.1 KBps (30.0 s, 2 samples)
Pipe Throughput 124072.9 lps (10.0 s, 7 samples)
Pipe-based Context Switching 35182.9 lps (10.0 s, 7 samples)
Process Creation 1440.6 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2912.6 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 816.6 lpm (60.0 s, 2 samples)
System Call Overhead 582876.8 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 12288814.1 1053.0
Double-Precision Whetstone 55.0 3042.9 553.3
Execl Throughput 43.0 1016.1 236.3
File Copy 1024 bufsize 2000 maxblocks 3960.0 102300.0 258.3
File Copy 256 bufsize 500 maxblocks 1655.0 28716.5 173.5
File Copy 4096 bufsize 8000 maxblocks 5800.0 286287.1 493.6
Pipe Throughput 12440.0 124072.9 99.7
Pipe-based Context Switching 4000.0 35182.9 88.0
Process Creation 126.0 1440.6 114.3
Shell Scripts (1 concurrent) 42.4 2912.6 686.9
Shell Scripts (8 concurrent) 6.0 816.6 1361.0
System Call Overhead 15000.0 582876.8 388.6
========
System Benchmarks Index Score 319.0
------------------------------------------------------------------------
Benchmark Run: Sun Nov 28 2021 05:37:00 - 06:05:35
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables 48512811.4 lps (10.0 s, 7 samples)
Double-Precision Whetstone 11915.2 MWIPS (10.1 s, 7 samples)
Execl Throughput 2970.7 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 187752.1 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 51776.2 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 545529.9 KBps (30.0 s, 2 samples)
Pipe Throughput 472517.0 lps (10.0 s, 7 samples)
Pipe-based Context Switching 135270.7 lps (10.0 s, 7 samples)
Process Creation 6047.7 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 6171.2 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 803.3 lpm (60.1 s, 2 samples)
System Call Overhead 2187432.3 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 48512811.4 4157.1
Double-Precision Whetstone 55.0 11915.2 2166.4
Execl Throughput 43.0 2970.7 690.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 187752.1 474.1
File Copy 256 bufsize 500 maxblocks 1655.0 51776.2 312.8
File Copy 4096 bufsize 8000 maxblocks 5800.0 545529.9 940.6
Pipe Throughput 12440.0 472517.0 379.8
Pipe-based Context Switching 4000.0 135270.7 338.2
Process Creation 126.0 6047.7 480.0
Shell Scripts (1 concurrent) 42.4 6171.2 1455.5
Shell Scripts (8 concurrent) 6.0 803.3 1338.9
System Call Overhead 15000.0 2187432.3 1458.3
========
System Benchmarks Index Score 854.8
● config.txtに arm_freq=1800の追加有りの状態
$ ./Run
make all
make[1]: Entering directory '/home/pi/byte-unixbench/UnixBench'
make distr
make[2]: Entering directory '/home/pi/byte-unixbench/UnixBench'
Checking distribution of files
./pgms exists
./src exists
./testdir exists
./tmp exists
./results exists
make[2]: Leaving directory '/home/pi/byte-unixbench/UnixBench'
make programs
make[2]: Entering directory '/home/pi/byte-unixbench/UnixBench'
make[2]: Nothing to be done for 'programs'.
make[2]: Leaving directory '/home/pi/byte-unixbench/UnixBench'
make[1]: Leaving directory '/home/pi/byte-unixbench/UnixBench'
locale: Cannot set LC_CTYPE to default locale: No such file or directory
locale: Cannot set LC_MESSAGES to default locale: No such file or directory
locale: Cannot set LC_ALL to default locale: No such file or directory
locale: Cannot set LC_CTYPE to default locale: No such file or directory
locale: Cannot set LC_MESSAGES to default locale: No such file or directory
locale: Cannot set LC_ALL to default locale: No such file or directory
sh: 1: 3dinfo: not found
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
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
------------------------------------------------------------------------------
Use directories for:
* File I/O tests (named fs***) = /home/pi/byte-unixbench/UnixBench/tmp
* Results = /home/pi/byte-unixbench/UnixBench/results
------------------------------------------------------------------------------
Use of uninitialized value in printf at ./Run line 1469.
Use of uninitialized value in printf at ./Run line 1470.
Use of uninitialized value in printf at ./Run line 1469.
Use of uninitialized value in printf at ./Run line 1470.
Use of uninitialized value in printf at ./Run line 1469.
Use of uninitialized value in printf at ./Run line 1470.
Use of uninitialized value in printf at ./Run line 1469.
Use of uninitialized value in printf at ./Run line 1470.
Use of uninitialized value in printf at ./Run line 1721.
Use of uninitialized value in printf at ./Run line 1722.
Use of uninitialized value in printf at ./Run line 1721.
Use of uninitialized value in printf at ./Run line 1722.
Use of uninitialized value in printf at ./Run line 1721.
Use of uninitialized value in printf at ./Run line 1722.
Use of uninitialized value in printf at ./Run line 1721.
Use of uninitialized value in printf at ./Run line 1722.
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: raspberrypi: GNU/Linux
OS: GNU/Linux -- 5.10.63-v7l+ -- #1459 SMP Wed Oct 6 16:41:57 BST 2021
Machine: armv7l (unknown)
Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
CPU 0: ARMv7 Processor rev 3 (v7l) (0.0 bogomips)
CPU 1: ARMv7 Processor rev 3 (v7l) (0.0 bogomips)
CPU 2: ARMv7 Processor rev 3 (v7l) (0.0 bogomips)
CPU 3: ARMv7 Processor rev 3 (v7l) (0.0 bogomips)
08:20:58 up 1:06, 4 users, load average: 0.57, 7.52, 6.54; runlevel Nov
------------------------------------------------------------------------
Benchmark Run: Sun Nov 28 2021 08:20:58 - 08:49:21
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 11622439.1 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3043.7 MWIPS (10.0 s, 7 samples)
Execl Throughput 1013.7 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 101646.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 28239.1 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 286597.3 KBps (30.0 s, 2 samples)
Pipe Throughput 124149.2 lps (10.0 s, 7 samples)
Pipe-based Context Switching 35205.6 lps (10.0 s, 7 samples)
Process Creation 1556.6 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2986.7 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 837.3 lpm (60.0 s, 2 samples)
System Call Overhead 585370.8 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 11622439.1 995.9
Double-Precision Whetstone 55.0 3043.7 553.4
Execl Throughput 43.0 1013.7 235.8
File Copy 1024 bufsize 2000 maxblocks 3960.0 101646.5 256.7
File Copy 256 bufsize 500 maxblocks 1655.0 28239.1 170.6
File Copy 4096 bufsize 8000 maxblocks 5800.0 286597.3 494.1
Pipe Throughput 12440.0 124149.2 99.8
Pipe-based Context Switching 4000.0 35205.6 88.0
Process Creation 126.0 1556.6 123.5
Shell Scripts (1 concurrent) 42.4 2986.7 704.4
Shell Scripts (8 concurrent) 6.0 837.3 1395.5
System Call Overhead 15000.0 585370.8 390.2
========
System Benchmarks Index Score 320.4
------------------------------------------------------------------------
Benchmark Run: Sun Nov 28 2021 08:49:21 - 09:17:54
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables 46236201.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 11426.6 MWIPS (10.5 s, 7 samples)
Execl Throughput 2932.0 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 188586.8 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 51679.9 KBps (30.1 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 553475.0 KBps (30.0 s, 2 samples)
Pipe Throughput 458911.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 127062.5 lps (10.0 s, 7 samples)
Process Creation 6128.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 6274.2 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 817.8 lpm (60.2 s, 2 samples)
System Call Overhead 2161890.0 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 46236201.8 3962.0
Double-Precision Whetstone 55.0 11426.6 2077.6
Execl Throughput 43.0 2932.0 681.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 188586.8 476.2
File Copy 256 bufsize 500 maxblocks 1655.0 51679.9 312.3
File Copy 4096 bufsize 8000 maxblocks 5800.0 553475.0 954.3
Pipe Throughput 12440.0 458911.1 368.9
Pipe-based Context Switching 4000.0 127062.5 317.7
Process Creation 126.0 6128.0 486.3
Shell Scripts (1 concurrent) 42.4 6274.2 1479.8
Shell Scripts (8 concurrent) 6.0 817.8 1363.0
System Call Overhead 15000.0 2161890.0 1441.3
========
System Benchmarks Index Score 844.7
● config.txtに arm_freq=1500の追加有りの状態
明示的に 1.5GHzで動かす。
$ ./Run
make all
make[1]: Entering directory '/home/pi/byte-unixbench/UnixBench'
make distr
make[2]: Entering directory '/home/pi/byte-unixbench/UnixBench'
Checking distribution of files
./pgms exists
./src exists
./testdir exists
./tmp exists
./results exists
make[2]: Leaving directory '/home/pi/byte-unixbench/UnixBench'
make programs
make[2]: Entering directory '/home/pi/byte-unixbench/UnixBench'
make[2]: Nothing to be done for 'programs'.
make[2]: Leaving directory '/home/pi/byte-unixbench/UnixBench'
make[1]: Leaving directory '/home/pi/byte-unixbench/UnixBench'
locale: Cannot set LC_CTYPE to default locale: No such file or directory
locale: Cannot set LC_MESSAGES to default locale: No such file or directory
locale: Cannot set LC_ALL to default locale: No such file or directory
locale: Cannot set LC_CTYPE to default locale: No such file or directory
locale: Cannot set LC_MESSAGES to default locale: No such file or directory
locale: Cannot set LC_ALL to default locale: No such file or directory
sh: 1: 3dinfo: not found
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
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
------------------------------------------------------------------------------
Use directories for:
* File I/O tests (named fs***) = /home/pi/byte-unixbench/UnixBench/tmp
* Results = /home/pi/byte-unixbench/UnixBench/results
------------------------------------------------------------------------------
Use of uninitialized value in printf at ./Run line 1469.
Use of uninitialized value in printf at ./Run line 1470.
Use of uninitialized value in printf at ./Run line 1469.
Use of uninitialized value in printf at ./Run line 1470.
Use of uninitialized value in printf at ./Run line 1469.
Use of uninitialized value in printf at ./Run line 1470.
Use of uninitialized value in printf at ./Run line 1469.
Use of uninitialized value in printf at ./Run line 1470.
Use of uninitialized value in printf at ./Run line 1721.
Use of uninitialized value in printf at ./Run line 1722.
Use of uninitialized value in printf at ./Run line 1721.
Use of uninitialized value in printf at ./Run line 1722.
Use of uninitialized value in printf at ./Run line 1721.
Use of uninitialized value in printf at ./Run line 1722.
Use of uninitialized value in printf at ./Run line 1721.
Use of uninitialized value in printf at ./Run line 1722.
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^[[C^[[C
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: raspberrypi: GNU/Linux
OS: GNU/Linux -- 5.10.63-v7l+ -- #1459 SMP Wed Oct 6 16:41:57 BST 2021
Machine: armv7l (unknown)
Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
CPU 0: ARMv7 Processor rev 3 (v7l) (0.0 bogomips)
CPU 1: ARMv7 Processor rev 3 (v7l) (0.0 bogomips)
CPU 2: ARMv7 Processor rev 3 (v7l) (0.0 bogomips)
CPU 3: ARMv7 Processor rev 3 (v7l) (0.0 bogomips)
09:31:55 up 1 min, 4 users, load average: 0.62, 0.24, 0.09; runlevel Nov
------------------------------------------------------------------------
Benchmark Run: Sun Nov 28 2021 09:31:55 - 10:00:49
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 8672198.8 lps (13.3 s, 7 samples)
Double-Precision Whetstone 2532.0 MWIPS (10.0 s, 7 samples)
Execl Throughput 838.1 lps (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 84633.7 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 24643.5 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 253340.5 KBps (30.0 s, 2 samples)
Pipe Throughput 103590.5 lps (10.0 s, 7 samples)
Pipe-based Context Switching 27560.4 lps (10.0 s, 7 samples)
Process Creation 1361.8 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2612.4 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 728.4 lpm (60.0 s, 2 samples)
System Call Overhead 487581.2 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 8672198.8 743.1
Double-Precision Whetstone 55.0 2532.0 460.4
Execl Throughput 43.0 838.1 194.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 84633.7 213.7
File Copy 256 bufsize 500 maxblocks 1655.0 24643.5 148.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 253340.5 436.8
Pipe Throughput 12440.0 103590.5 83.3
Pipe-based Context Switching 4000.0 27560.4 68.9
Process Creation 126.0 1361.8 108.1
Shell Scripts (1 concurrent) 42.4 2612.4 616.1
Shell Scripts (8 concurrent) 6.0 728.4 1214.0
System Call Overhead 15000.0 487581.2 325.1
========
System Benchmarks Index Score 268.4
------------------------------------------------------------------------
Benchmark Run: Sun Nov 28 2021 10:00:49 - 10:29:19
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables 40351324.6 lps (10.0 s, 7 samples)
Double-Precision Whetstone 10136.1 MWIPS (10.0 s, 7 samples)
Execl Throughput 2653.6 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 176064.8 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 47217.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 543648.2 KBps (30.0 s, 2 samples)
Pipe Throughput 404386.5 lps (10.0 s, 7 samples)
Pipe-based Context Switching 114407.9 lps (10.0 s, 7 samples)
Process Creation 5561.8 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 5926.3 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 792.2 lpm (60.2 s, 2 samples)
System Call Overhead 1858779.5 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 40351324.6 3457.7
Double-Precision Whetstone 55.0 10136.1 1842.9
Execl Throughput 43.0 2653.6 617.1
File Copy 1024 bufsize 2000 maxblocks 3960.0 176064.8 444.6
File Copy 256 bufsize 500 maxblocks 1655.0 47217.3 285.3
File Copy 4096 bufsize 8000 maxblocks 5800.0 543648.2 937.3
Pipe Throughput 12440.0 404386.5 325.1
Pipe-based Context Switching 4000.0 114407.9 286.0
Process Creation 126.0 5561.8 441.4
Shell Scripts (1 concurrent) 42.4 5926.3 1397.7
Shell Scripts (8 concurrent) 6.0 792.2 1320.4
System Call Overhead 15000.0 1858779.5 1239.2
========
System Benchmarks Index Score 770.7
● Raspberry Pi Benchmark
Raspberry Pi Benchmark
sudo apt update
curl -L https://raw.githubusercontent.com/aikoncwd/rpi-benchmark/master/rpi-benchmark.sh | sudo bash
明示的に arm_freq=1500を追加を追加して 1.5GHzでのベンチマークを測定します。
| config.txtに arm_freq=1500を追加 | config.txtは未編集（1.8GHzで動作） | |
| 1.5GHz | 1.8GHz | UP率（理論値 20% UP） |
| CPU | 3.76 sec | 3.17 sec | 18.6% UP |
| THREADS | 10.99 sec | 9.24 sec | 19.4% UP |
| MEMORY | 1909.69 MiB/sec | 2266.38 MiB/sec | 19.7% UP |
CPU関係のスコアを比較すると約 19% UPで理論値の 20%とほぼ同じです。
● config.txtに arm_freq=1500の追加有りの状態
明示的に 1.5GHzで動かす。
Raspberry Pi Benchmark Test
Author: AikonCWD
Version: 3.0
temp=58.9'C
arm_freq=1500
core_freq=500
core_freq_min=200
gpu_freq=500
gpu_freq_min=250
sd_clock=50.000 MHz
Running InternetSpeed test...
Ping: 57.835 ms
Download: 309.75 Mbit/s
Upload: 232.58 Mbit/s
Running CPU test...
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
total time: 10.0034s
min: 3.75
avg: 3.76
max: 4.83
temp=66.2'C
Running THREADS test...
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
total time: 10.0098s
min: 10.56
avg: 10.99
max: 22.43
temp=66.2'C
Running MEMORY test...
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
3072.00 MiB transferred (1909.69 MiB/sec)
total time: 1.6013s
min: 0.00
avg: 0.00
max: 0.11
temp=67.2'C
Running HDPARM test...
Timing buffered disk reads: 130 MB in 3.02 seconds = 42.99 MB/sec
temp=61.3'C
Running DD WRITE test...
536870912 bytes (537 MB, 512 MiB) copied, 124.222 s, 4.3 MB/s
temp=56.9'C
Running DD READ test...
536870912 bytes (537 MB, 512 MiB) copied, 12.0106 s, 44.7 MB/s
temp=57.4'C
AikonCWD's rpi-benchmark completed!
● config.txtに arm_freq=1800の追加無しの状態
※ config.txtに arm_freq=1800を追加しなくても 1.8GHzで動いていた
Raspberry Pi Benchmark Test
Author: AikonCWD
Version: 3.0
temp=56.4'C
arm_freq=1800
core_freq=500
core_freq_min=200
gpu_freq=500
gpu_freq_min=250
sd_clock=50.000 MHz
Running InternetSpeed test...
Ping: 59.78 ms
Download: 286.21 Mbit/s
Upload: 187.58 Mbit/s
Running CPU test...
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
total time: 10.0024s
min: 3.12
avg: 3.17
max: 15.64
temp=66.2'C
Running THREADS test...
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
total time: 10.0087s
min: 8.87
avg: 9.24
max: 62.76
temp=66.2'C
Running MEMORY test...
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
3072.00 MiB transferred (2266.38 MiB/sec)
total time: 1.3493s
min: 0.00
avg: 0.00
max: 0.09
temp=67.6'C
Running HDPARM test...
Timing buffered disk reads: 130 MB in 3.02 seconds = 43.02 MB/sec
temp=61.8'C
Running DD WRITE test...
536870912 bytes (537 MB, 512 MiB) copied, 112.44 s, 4.8 MB/s
temp=57.4'C
Running DD READ test...
536870912 bytes (537 MB, 512 MiB) copied, 12.2885 s, 43.7 MB/s
temp=56.9'C
AikonCWD's rpi-benchmark completed!
● Raspberry Piで円周率を 1億桁計算してみる
円周率を1億桁計算しました！ ― その試行錯誤の詳しい経緯と結果 ー
ここの内容をそのまま実行します。
sudo apt-get install libgmp3-dev
# Binary Splitting method によるChudnovsky実装
# 再帰でガーーーーッって行く感じ, 割と直感的に書きました. 以下のコードがこのエントリーの最終的なコードです. ベンチを取るときはこのコードを使用してください.
# のソースコードを入力する
nano pi.c
# 実行する
gcc pi.c -lgmp && ./a.out
計算する桁数は
int main (int argc, char* argv[]) {
の
int digits = 10000;
を書き換える。
| config.txtに arm_freq=1500を追加 | config.txtは未編集（1.8GHzで動作） | |
| 1.5GHz | 1.8GHz | UP率（理論値 20% UP） |
| 1万桁 | 0.006 s | 0.006 s | --- |
| 1000万桁 | 66.017 s | 56.014 s | 17.9% UP |
| 1億桁 | 1071.197 s | 915.646 s | 17.0% UP |
CPU関係のスコアを比較すると約 17% UPで理論値の 20%と近いです。
● config.txtに arm_freq=1800の追加無しの状態
※ config.txtに arm_freq=1800を追加しなくても 1.8GHzで動いていた
● 1万桁
$ grep "digits =" pi.c
int digits = 10000;
$ gcc pi.c -lgmp && ./a.out
0.006 s
0.008 s
● 1000万桁
$ grep "digits =" pi.c
int digits = 10000000;
$ gcc pi.c -lgmp && ./a.out
56.014 s
70.394 s
● 1億桁
$ grep "digits =" pi.c
int digits = 100000000;
$ gcc pi.c -lgmp && ./a.out
915.646 s
1128.108 s
● config.txtに arm_freq=1500の追加有りの状態
明示的に 1.5GHzで動かす。
● 1万桁
$ grep "digits =" pi.c
int digits = 10000;
$ gcc pi.c -lgmp && ./a.out
0.006 s
0.008 s
● 1000万桁
$ grep "digits =" pi.c
int digits = 10000000;
$ gcc pi.c -lgmp && ./a.out
66.017 s
83.004 s
● 1億桁
$ grep "digits =" pi.c
int digits = 100000000;
$ gcc pi.c -lgmp && ./a.out
1071.197 s
1320.592 s
● ラズパイ４Ｂで 2.0GHz動作、config.txtに arm_freq=2000の追加有りの状態
$ sudo apt-get install stress
# Setting up stress (1.0.4-7)
# 4コア同時に負荷テスト
$ stress -c 4
# 私のラズパイ４Ｂは 2.0GHzで安定して動きました
$ vcgencmd measure_clock arm
frequency(48)=2000478464
$ vcgencmd measure_clock arm
frequency(48)=2000531200
● ラズパイ４Ｂで 2.2GHz動作、config.txtに arm_freq=2200の追加有りの状態
$ sudo apt-get install stress
# Setting up stress (1.0.4-7)
# 4コア同時に負荷テスト
$ stress -c 4
# 私のラズパイ４Ｂは 2.2GHzで安定して動きました
$ vcgencmd measure_clock arm
frequency(48)=2200552704
$ vcgencmd measure_clock arm
frequency(48)=2200499968
● Raspberry Pi OS Bullseyeの注意事項
Raspicam apps, the Picamera Python library, or OMXPlayerが Bullseyeでは動きません。
Python Picamera moduleを Bullseyeで使いたい場合は raspi-configで Legacy Camera Enable/disable legacy camera supportを「有効」にします。
sudo raspi-config
3 Interface Options
I1 Legacy Camera Enable/disable legacy camera support
Getting started with the Camera Module
Raspberry Pi Camera Module - Operating System notice
The Python Picamera module is currently not, by default, compatible with the latest version of Raspberry Pi OS (called Bullseye).
To use the Picamera module, you will need to enable legacy support for the camera.
● Raspberry Pi OS Bullseyeは過去のアプリが使えなくなっているので Busterが Legacyとして Raspberry Pi公式が継続サポート
”New” old functionality with Raspberry Pi OS (Legacy)
Raspberry Pi OS - Operating system images
● Raspberry Pi OS (Legacy) Busterの場合
2021年 12月にリリースの Raspberry Pi OS (Legacy) Busterはデフォルトで 1.5GHz動作でした。
Raspberry Pi OS (Legacy) Based on Debian Buster
Release date: December 2nd 2021
Kernel version: 5.10
$ uname -a
Linux raspberrypi 5.10.63-v7l+ #1496 SMP Wed Dec 1 15:58:56 GMT 2021 armv7l GNU/Linux
$ vcgencmd measure_clock arm
frequency(48)=1500398464
$ vcgencmd measure_clock arm
frequency(48)=1500345728
Tags: [Raspberry Pi], [電子工作], [海外通販], [通信販売]
●関連するコンテンツ(この記事を読んだ人は、次の記事も読んでいます)
【メモリ8GB】Raspberry Pi 4 Model B 8GBを KSYで最安値で購入。ベンチマークレビュー
【技適取得】ラズパイ４Ｂ 8GBモデルを入手。従来の Pi3、Pi3B＋と速度比較
【Vulkan】Raspberry Pi 4 Model Bで Vulkanドライバをビルドして 3Dグラフィックのデモを動かす
【v3dv】ラズパイ４Ｂで Vulkan APIを動かす、VK_ICD_FILENAMES broadcom_icd.armv7l.json
Raspberry Pi 4 Model Bに標準搭載の Boot用 EEPROM、ファームウェアのアップデート手順
Raspberry Pi 4 Model Bから新規搭載の Boot用 EEPROMのファームを更新する方法、起動しない場合の復旧方法
NVIDIA Jetson Nano 開発者キットを買ってみた。メモリ容量 4GB LPDDR4 RAM
Jetson Nanoで TensorFlow PyTorch Caffe/Caffe2 Keras MXNet等を GPUパワーで超高速で動かす！
PIP機能付きの 4K対応の 4入力 1出力の HDMIセレクターを買ってみた、HDMI機器が複数有る場合に便利
ピクチャ イン ピクチャ機能付き 4K入力対応の 4入力 1出力 HDMI切り換え機 HDSFX0401P
EDID保持機能付きの 4K対応の 4入力 2出力の マトリックス切り替え HDMIセレクター、液晶画面 2台と使用で最強
TESmart HMA0402A30 マトリックス切り替えで液晶画面 2台に接続できて更に EDID保持の便利機能付き HDMI切り換え機
[HOME]
|
[BACK]
リンクフリー(連絡不要、ただしトップページ以外は Web構成の変更で移動する場合があります)
Copyright (c)
2021 FREE WING,Y.Sakamoto
Powered by 猫屋敷工房 & HTML Generator
http://www.neko.ne.jp/~freewing/raspberry_pi/raspberry_pi_4_bullseye_os_support_1_8ghz/
