Linux on ARM powered Devices

This post should give an overview on which ARM powered devices GNU/Linux runs and also gives a hint how good it is supported on the respective device. Also Android runs a Linux kernel but does not give a general purpose computing experience to the user and does not host GNU tools.

Here is a list of boards and devices which are affordable for
the home user - devices for industrial use tend to be just too expensive.

Debian GNU/Linux on NAS Devices and the Plug Computer

http://www.cyrius.com/      Get more infos here     

Mostly based on Marvells Orion and Kirkwood platform.
Debian has a long history on NAS devices e.g. from QNAP
and is stable and good supported. Also the popular but discontinued
NSLU2 device runs perfectly Debian and other Linux Distributions.

Debian GNU/Linux on the Raspberry Pi     ARM11

The main Raspberry Pi Distribution is called Raspbian and is based on Debian.

http://www.raspberrypi.org/

Raspbian has good support for the popular Raspberry Pi hardware, but it lacks support for the onboard graphic accelerating capabilities for Xorg. 

Distros on Development Boards 

Ångström/Ubuntu on http://beagleboard.org/               Cortex A8 processor
Ubuntu on http://pandaboard.org/                                Cortex A9 processor dual-core




Debian/Ubuntu on cheap consumer devices and boards        

Mele A1000/A2000(G) media player:                            Cortex A8 processor 
http://rhombus-tech.net/allwinner_a10/a10_image/

PC on a Stick and Mini PC's:
PicUntu on RK3066 devices like MK808                       Cortex A9 processor dual-core          
http://code.google.com/p/rk3066-linux/wiki/PicUntu        early release

Hackberry and Cubieboard
https://www.miniand.com/                                              Cortex A8 processor

OlinuXino
https://www.olimex.com/Products/OLinuXino/                Cortex A8 processor
https://www.olimex.com/Products/OLinuXino/A13/

Wandboard
http://www.wandboard.org/                                           Cortex A9 processor
                                                                                      available as single/dual/quad core

Udoo                                                                                   Cortex A9 processor
http://www.udoo.org/                                                available as dual/quad core

Utilite                                                                                    Cortex A9 processor
http://utilite-computer.com/web/home                       available as single/dual/quad core
Ubuntu 12.04  pre-installed


Ubuntu on Boards from Hardkernel
http://www.hardkernel.com/renewal_2011/main.php    Cortex A9 processor quad-core
 

Ubuntu on the Samsumg Series 3 Chromebook      Cortex A15 processor dual-core

Ubuntu 12.10: (sound not working)
http://chromeos-cr48.blogspot.co.at/2012/10/arm-chrubuntu-1204-alpha-1-now.html 
http://www.phoronix.com/scan.php?page=news_item&px=MTI5Mzg 
Ubuntu 13.04:
http://marcin.juszkiewicz.com.pl/2013/02/14/how-to-install-ubuntu-13-04-on-chromebook/ 
Archlinux:
http://archlinuxarm.org/platforms/armv7/samsung-chromebook

Read More

The Utilite Computer

The Utilite Computer is available
with 1/2/4 Cortex A9 cores clocked at max. 1.2 Ghz.

The Utilite is delivered with Ubuntu 12.04 pre-installed.
It can also run other operating systems like:
Android
http://www.utilite-computer.com/wiki/index.php/Utilite_Android 
Archlinux ARM
http://archlinuxarm.org/platforms/armv7/freescale/utilite

Currently the fastet way to order is: 
http://utilite-computer.com/web/order-utilite-direct

Pre-order in UK:
https://plus.google.com/113210185268665255819/posts/df2sQNxmBj2
https://store.tinygreenpc.com/tiny-green-pcs/utilite-1.html
  
Read More:
http://utilite-computer.com/web/home
https://plus.google.com/u/0/113210185268665255819/posts
https://www.facebook.com/groups/utilite/

Videos:
https://www.youtube.com/channel/UCbVaRkwMGK9FswqQp0XbctQ?feature=watch

First look:
Stephan works on latest XBMC port for Utilite and Wandboard
http://stephan-rafin.net/blog/2013/08/18/first-look-at-utilite/ 
http://stephan-rafin.net/blog/2013/08/23/utilite-xbmc/ 
http://stephan-rafin.net/blog/2013/09/30/xbmc-for-utilite-cec-support-included/ 
http://stephan-rafin.net/blog/2013/11/02/geexbox-for-utilite/
 

Benchmark comparing a Raspberrry Pi and a Utilite Pro with quad core cpu:

Started with:
apt-get install sysbench gtkperf lighttpd      /* on the Raspberry Pi and the Utilite


System 1:
pi@raspberrypi ~ $ uname -a
Linux raspberrypi 3.6.11+ #474 PREEMPT Thu Jun 13 17:14:42 BST 2013 armv6l GNU/Linux
pi@raspberrypi ~ $ sysbench --version
sysbench 0.4.12
512 MB RAM
CPU 700 MHz
Operating System: Raspbian

System 2:
utilite@utilite-ubuntu-desktop:~$ uname -a
Linux utilite-ubuntu-desktop 3.0.35-cm-fx6-3-00055-g8b9354a #92 SMP Mon Sep 9 15:04:27 IST 2013 armv7l armv7l armv7l GNU/Linux

utilite-ubuntu-desktop:~$ sysbench --version
sysbench 0.4.12

2 GB RAM
CPU 4x 1.2 Ghz max (currently at 4x 996 Mhz)
Operating System: Ubuntu 12.04

root@utilite-ubuntu-desktop:cat /sys/devices/system/cpu/*/cpufreq/cpuinfo_max_freq
996000
996000
996000
996000


Results in short:

1. Prime numbers:          /* Test uses 64 Bit Integers for Calculation

Pi:            54 seconds
Utilite Pro:   24 seconds   using 1 CPU core
                6 seconds   using 4 CPU cores


2. Fileio                  /* Test prepares 8GB of data on storage media


Pi:                     100    Kb/sec /* SD Card Class 10
Utilite Pro:              6    MB/sec using one CPU core    /* SSD
                          6.5  MB/sec using 4 CPUs cores    /* SSD
                        /* Please have a look at the hdparm results on next page  



3. gtkperf
Pi:                     100 seconds
Utilite Pro:             85 seconds    /* It looks like circle drawing has a problem on i.MX6



4. Webserver lighttpd 1.4.x with apachebench
Serving a static html page
Ethernet connected to Gigabit Interface

Pi:                      18.9  seconds       
Utilite Pro:              3.4  seconds

  

---------------------------------------------------------------------------------------------------------------------
During a file transfer with SCP or SFTP via Gigabit Ethernet
I got approx. 15 Megabyte per second transfer rate  while the
processor load was about 70% on one processor core.

Compressing an avi file with the xz -zv option gives me a
data compression rate of 470 Kbyte per second while the
processor load was 100% on one processor core.

----------------------------------------------------------------
Tests in detail:


pi@raspberrypi ~ $ cat /proc/cpuinfo
Processor       : ARMv6-compatible processor rev 7 (v6l)
BogoMIPS        : 697.95
Features        : swp half thumb fastmult vfp edsp java tls
CPU implementer : 0x41
CPU architecture: 7
CPU variant     : 0x0
CPU part        : 0xb76
CPU revision    : 7

Hardware        : BCM2708
Revision        : 000f
Serial          : 0000000025cd2f93



utilite@utilite-ubuntu-desktop:~$ cat /proc/cpuinfo
Processor       : ARMv7 Processor rev 10 (v7l)
processor       : 0
BogoMIPS        : 790.52

processor       : 1
BogoMIPS        : 790.52

processor       : 2
BogoMIPS        : 790.52

processor       : 3
BogoMIPS        : 790.52

Features        : swp half thumb fastmult vfp edsp neon vfpv3
CPU implementer : 0x41
CPU architecture: 7
CPU variant     : 0x2
CPU part        : 0xc09
CPU revision    : 10

Hardware        : Compulab CM-FX6
Revision        : 63012
Serial          : 220e89d4d81917c9




pi@raspberrypi ~ $ sudo hdparm -t /dev/root              /* 16 GB Class 10 SD Card
/dev/root:
 Timing buffered disk reads:  62 MB in  3.08 seconds =  20.14 MB/sec


utilite@utilite-ubuntu-desktop:~$ sudo hdparm -t /dev/sda    /* 32 GB internal SSD Drive

/dev/sda:
 Timing buffered disk reads: 386 MB in  3.01 seconds = 128.20 MB/sec



pi@raspberrypi ~ $ sysbench --test=cpu --cpu-max-prime=2000 run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 1

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 2000


Test execution summary:
    total time:                          54.1248s
    total number of events:              10000
    total time taken by event execution: 54.0950
    per-request statistics:
         min:                                  5.34ms
         avg:                                  5.41ms
         max:                                 15.55ms
         approx.  95 percentile:               5.45ms

Threads fairness:
    events (avg/stddev):           10000.0000/0.00
    execution time (avg/stddev):   54.0950/0.00


utilite@utilite-ubuntu-desktop:~$ sysbench --test=cpu --cpu-max-prime=2000 run     /* only using a single CPU
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 1

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 2000


Test execution summary:
    total time:                          24.1038s
    total number of events:              10000
    total time taken by event execution: 24.0932
    per-request statistics:
         min:                                  2.39ms
         avg:                                  2.41ms
         max:                                  7.43ms
         approx.  95 percentile:               2.42ms

Threads fairness:
    events (avg/stddev):           10000.0000/0.00
    execution time (avg/stddev):   24.0932/0.00

pi@raspberrypi ~ $ sysbench --test=cpu --cpu-max-prime=2000 --num-threads=4 run     /* using 4 threads but on only on CPU
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 4

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 2000


Test execution summary:
    total time:                          54.2979s
    total number of events:              10000
    total time taken by event execution: 217.0499
    per-request statistics:
         min:                                  5.34ms
         avg:                                 21.70ms
         max:                                 83.31ms
         approx.  95 percentile:              35.46ms

Threads fairness:
    events (avg/stddev):           2500.0000/0.71
    execution time (avg/stddev):   54.2625/0.02




utilite@utilite-ubuntu-desktop:~$ sysbench --test=cpu --cpu-max-prime=2000 --num-threads=4 run     /* using all 4 CPU's
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 4

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 2000


Test execution summary:
    total time:                          6.0654s
    total number of events:              10000
    total time taken by event execution: 24.2229
    per-request statistics:
         min:                                  2.39ms
         avg:                                  2.42ms
         max:                                 16.13ms
         approx.  95 percentile:               2.40ms

Threads fairness:
    events (avg/stddev):           2500.0000/1.41
    execution time (avg/stddev):   6.0557/0.00




pi@raspberrypi ~ $ sysbench --test=fileio --file-total-size=8G --file-test-mode=rndrw --init -rng=on --max-time=300 --max-requests=0 run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 1
Initializing random number generator from timer.


Extra file open flags: 0
128 files, 64Mb each
8Gb total file size
Block size 16Kb
Number of random requests for random IO: 0
Read/Write ratio for combined random IO test: 1.50
Periodic FSYNC enabled, calling fsync() each 100 requests.
Calling fsync() at the end of test, Enabled.
Using synchronous I/O mode
Doing random r/w test
Threads started!
Time limit exceeded, exiting...
Done.

Operations performed:  1140 Read, 760 Write, 2369 Other = 4269 Total
Read 17.812Mb  Written 11.875Mb  Total transferred 29.688Mb  (101.3Kb/sec)
    6.33 Requests/sec executed

Test execution summary:
    total time:                          300.0960s
    total number of events:              1900
    total time taken by event execution: 3.6853
    per-request statistics:
         min:                                  0.10ms
         avg:                                  1.94ms
         max:                                 67.41ms
         approx.  95 percentile:               3.86ms

Threads fairness:
    events (avg/stddev):           1900.0000/0.00
    execution time (avg/stddev):   3.6853/0.00



utilite@utilite-ubuntu-desktop:~$ sysbench --test=fileio --file-total-size=8G --file-test-mode=rndrw --init-rng=on --max-time=300 --max-requests=0 run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 1
Initializing random number generator from timer.


Extra file open flags: 0
128 files, 64Mb each
8Gb total file size
Block size 16Kb
Number of random requests for random IO: 0
Read/Write ratio for combined random IO test: 1.50
Periodic FSYNC enabled, calling fsync() each 100 requests.
Calling fsync() at the end of test, Enabled.
Using synchronous I/O mode
Doing random r/w test
Threads started!
Time limit exceeded, exiting...
Done.

Operations performed:  69615 Read, 46410 Write, 148480 Other = 264505 Total
Read 1.0622Gb  Written 725.16Mb  Total transferred 1.7704Gb  (6.0428Mb/sec)
  386.74 Requests/sec executed

Test execution summary:
    total time:                          300.0077s
    total number of events:              116025
    total time taken by event execution: 197.1406
    per-request statistics:
         min:                                  0.07ms
         avg:                                  1.70ms
         max:                                 40.14ms
         approx.  95 percentile:               9.33ms

Threads fairness:
    events (avg/stddev):           116025.0000/0.00
    execution time (avg/stddev):   197.1406/0.00





utilite@utilite-ubuntu-desktop:~$ sysbench --test=fileio --file-total-size=8G --file-test-mode=rndrw --init-rng=on --max-time=300 --max-requests=0 --num-threads=4 run
sysbench 0.4.12:  multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 4
Initializing random number generator from timer.


Extra file open flags: 0
128 files, 64Mb each
8Gb total file size
Block size 16Kb
Number of random requests for random IO: 0
Read/Write ratio for combined random IO test: 1.50
Periodic FSYNC enabled, calling fsync() each 100 requests.
Calling fsync() at the end of test, Enabled.
Using synchronous I/O mode
Doing random r/w test
Threads started!
Time limit exceeded, exiting...
(last message repeated 3 times)
Done.

Operations performed:  75844 Read, 50562 Write, 158819 Other = 285225 Total
Read 1.1573Gb  Written 790.03Mb  Total transferred 1.9288Gb  (6.5833Mb/sec)
  421.33 Requests/sec executed

Test execution summary:
    total time:                          300.0175s
    total number of events:              126406
    total time taken by event execution: 271.2085
    per-request statistics:
         min:                                  0.04ms
         avg:                                  2.15ms
         max:                                 75.03ms
         approx.  95 percentile:               9.36ms

Threads fairness:
    events (avg/stddev):           31601.5000/122.09
    execution time (avg/stddev):   67.8021/0.35


Pi:
GtkPerf 0.40 - Starting testing: Sat Sep 21 18:49:13 2013

GtkEntry - time:  0.67
GtkComboBox - time:  9.72
GtkComboBoxEntry - time:  5.69
GtkSpinButton - time:  1.00
GtkProgressBar - time:  0.67
GtkToggleButton - time:  1.12
GtkCheckButton - time:  1.04
GtkRadioButton - time:  1.97
GtkTextView - Add text - time:  8.79
GtkTextView - Scroll - time:  2.40
GtkDrawingArea - Lines - time: 21.40
GtkDrawingArea - Circles - time: 26.93
GtkDrawingArea - Text - time: 17.20
GtkDrawingArea - Pixbufs - time:  1.40
 --- 
Total time: 100.12



Utilite:
GtkPerf 0.40 - Starting testing: Sat Sep 21 18:41:06 2013

GtkEntry - time:  0.10
GtkComboBox - time:  3.30
GtkComboBoxEntry - time:  1.94
GtkSpinButton - time:  0.43
GtkProgressBar - time:  0.39
GtkToggleButton - time:  0.58
GtkCheckButton - time:  0.34
GtkRadioButton - time:  0.67
GtkTextView - Add text - time:  2.60
GtkTextView - Scroll - time:  1.60
GtkDrawingArea - Lines - time:  2.78
GtkDrawingArea - Circles - time: 62.32
GtkDrawingArea - Text - time:  7.43
GtkDrawingArea - Pixbufs - time:  0.95
 --- 
Total time: 85.44



This is ApacheBench, Version 2.3 <$Revision: 655654 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/

Benchmarking 10.42.0.73 (be patient)     /* Raspberry Pi


Server Software:        lighttpd/1.4.31
Server Hostname:        10.42.0.73
Server Port:            80

Document Path:          /test.html
Document Length:        169 bytes

Concurrency Level:      5
Time taken for tests:   18.935 seconds
Complete requests:      10000
Failed requests:        0
Write errors:           0
Total transferred:      4260000 bytes
HTML transferred:       1690000 bytes
Requests per second:    528.13 [#/sec] (mean)
Time per request:       9.467 [ms] (mean)
Time per request:       1.893 [ms] (mean, across all concurrent requests)
Transfer rate:          219.71 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:        0    1   0.1      1       3
Processing:     3    9   1.2      8      28
Waiting:        3    9   1.2      8      28
Total:          4    9   1.2      9      29

Percentage of the requests served within a certain time (ms)
  50%      9
  66%      9
  75%      9
  80%      9
  90%      9
  95%     14
  98%     14
  99%     15
 100%     29 (longest request)



This is ApacheBench, Version 2.3 <$Revision: 655654 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/

Benchmarking 10.42.0.18 (be patient)         /* Utilite Pro


Server Software:        lighttpd/1.4.28
Server Hostname:        10.42.0.18
Server Port:            80

Document Path:          /test.html
Document Length:        169 bytes

Concurrency Level:      5
Time taken for tests:   3.366 seconds
Complete requests:      10000
Failed requests:        0
Write errors:           0
Total transferred:      4240000 bytes
HTML transferred:       1690000 bytes
Requests per second:    2970.98 [#/sec] (mean)
Time per request:       1.683 [ms] (mean)
Time per request:       0.337 [ms] (mean, across all concurrent requests)
Transfer rate:          1230.17 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:        0    0   0.1      0       4
Processing:     1    1   0.7      1      35
Waiting:        0    1   0.7      1      35
Total:          1    2   0.8      2      35

Percentage of the requests served within a certain time (ms)
  50%      2
  66%      2
  75%      2
  80%      2
  90%      2
  95%      3
  98%      3
  99%      4
 100%     35 (longest request)

Read More

Tiny Boards

http://tinyboards.blogspot.com

Read More

The OUYA console, E-Book

The OUYA is a $99 games console powered by a Tegra 3 processor (quadcore ARM).
Do you want to know what can be done with the OUYA and where the limits are ?
There is more than just games for the OUYA, we will like to discover media centers,
web browsers and many other tools and utilities to come.

Debian on OUYA:
http://tuomas.kulve.fi/blog/2013/08/11/debian-on-ouya/

I will try to keep you up to date with news that matters.
Order the E-Book (PDF) for 2 Euro and you will get all facts and news in one place.
The book is a work in progress, You will receive updates per email in regular intervals.
- includes information how to load Apps
  which are not available from the OUYA store (sideloading)
- how to get root access rights
- modify your OUYA (use play store,...)

The delivery of the PDF file may take up to 48 hours.

Read More

The Piface Interface board

The PiFace is actually an I/O extender that uses the SPI bus to extend the I/O capabilities of the Raspberry Pi with 8 extra input and 8 extra output pins.

While all 8 input and 8 ouput capabilities are routed to pins for external connection, some have a double function. 2 of the 8 output pins are used to drive a couple of Relays, and 4 of the 8 input pins are fitted with buttons on the board itself. Further the input pins are directly connected to the I/O extension IC, there is no protection for over voltage in these circuits.

The 8 output pins are driven by a ULN2803A, this means that each output pin can drive up to 500mA (externally supplied) at a maximum of 50V.

In theory you could connect 4 PiFace boards to the RPi, because the address of the SPI I/O extension chip (MCP23S17SP) can be set using 2 jumpers. In reality this will be very hard to achieve because the PiFace is a board that needs to be stacked on top of the RPi and the main connector (the one on the bottom that connects to the GPIO pins on the RPi) is not fitted with a female counter part on the top to enable stacking more boards. I guess they did not do that because the Relays are too high anyway to enable stacking.

http://www.element14.com/community/docs/DOC-52896/l/piface-get-started-guide-feb-2013
http://piface.openlx.org.uk/kernel-driver-delays-and-new-interface-board

Read More

Buch: Der Raspberry Pi Computer

Der Raspberry Pi Computer (4 Euro für das e-Book als PDF Datei):

Der Raspberry Pi ist ein scheckkartengroßer Computer mit USB,Video und
LAN Schnittstelle. Er eignet sich für den Bastler und als Computer
für den Unterricht. Das Buch beschreibt die Installation von Betriebssystemen
auf der Speicherkarte und die Handhabung mit verschiedenen unixartigen
Betriebssystemen wie Debian.

Das Buch zum Raspberry Pi enthält Anleitungen zum Thema

• Installation von Betriebssystemen auf die SD Karte und
  die Benutzung von berryboot als Bootmanager.
  ( Debian bzw. Raspbian, Raspbmc, RISC OS, Arch Linux )
• Anschluß von USB Geräten und deren Diagnose
• Installation  und Administration von Server Anwendungen, wie WEB Server,
  FTP Server, File Server(NFS, SAMBA), SSH Server
• Erweiterung der Ein/Ausgabe Möglichkeiten
  ihres Raspberry Pis (die Piface Interfacekarte).
• Benutzung von VNC zum Remote Zugriff 
• Email von der Kommandozeile oder mit GUI
• Anschluß und Verwendung des Raspberry Pi Kamera Moduls
• Netzwerksicherheit

Bei Interesse können Sie das Buch (50 Seiten, Edition 2.5 August 2013)
gerne hier bestellen.

Die Zusendung des e-Books kann bis zu 48h dauern.

Buchkritik bei macherzin:
http://macherzin.net/blogpost319-Lesenswert-Der-Raspberry-Pi-Computer

In Österreich gibts den Raspberry Pi und Zubehör hier zu kaufen:
http://www.designtools.at


 Viel Spaß mit dem Raspberry Pi,
  Peter Bauer


Peter Bauer
Wildstraße 11
2100 Korneuburg
Österreich

Read More

Raspberry Pi, E-Book

The Raspberry Pi is a credit card sized Computer with USB, Video and Lan
interfaces. Its primary use is for teaching computer science.


E-Book "The Raspberry Pi Computer" (PDF) is available
for 4 Euro (Paypal)  and can be ordered here.

The delivery of the E-book may take up to 48 hours.

This E-Book contains instructions on the subject of:

• Installation of the operating system to the memory card
• Handling of different operating systems like Debian
• Connecting and diagnose USB devices 
• Installation  and Administration of Server applications like WEB Server,
  FTP Server, File Server(NFS, SAMBA), SSH Server
• Usage of the GPIO Interface
• Getting Started with the Piface interface card
• Installation und Administration of the operating systems
  available for the Raspberry Pi ( Debian, Raspbmc, RISC OS, Arch Linux )
• Using berryboot as a boot manager and Operating System installer
• Network security

 Have Fun with the Raspberry Pi,
  Peter Bauer

Peter Bauer
Wildstraße 11
A 2100 Korneuburg
Austria

Read More