Christos Antonopoulos | 1 Aug 2007 14:38
Picon

problem creating a contiki type mote in cooja

Hi all,

 This is a cooja related question. The problem appears when I try to create a Contiki mote type.

Even if I set only the cooja_test_process I get these messages back.

 
I must note that these messages appear after some comments out System.out's in tools/cooja/java/se/sics/cooja/contikimote/ContikiMoteTypeDialog.java:1471 (Fredrik's hind)

 
[java]  INFO [compilation thread] (?:?) - -- Compiling --
[java]  INFO [compilation thread] (?:?) - Project dirs:  ../apps/mrm
[java]  INFO [compilation thread] (?:?) - Project sources:  testcooja.c cooja-radio.c radio-uip.c init-net-uip.c
[java]  INFO [compilation thread] (?:?) - Compiler flags:  -I../apps/mrm
[java] Full command:
[java] /root/contiki-2.x/tools/cooja/build/./> make obj_cooja/mtype1.library -f ../../../Makefile.include
[java] Environment:

[java] CONTIKI=../../..
[java] TARGET=cooja
[java] TYPEID=mtype1
[java] LD_ARGS_1=

[java] LD_ARGS_2=
[java] EXTRA_CC_ARGS= -I../apps/mrm
[java] SYMBOLS=
[java] CC=gcc
[java] LD=ld
[java] PROJECTDIRS= ../apps/mrm
[java] PROJECT_SOURCEFILES= testcooja.c cooja-radio.c radio-uip.c init-net-uip.c
[java] PATH=/usr/java/jre1.6.0_02/bin:/usr/lib/qt-3.3/bin:/usr/kerberos/sbi                                                                            n:/usr/kerberos/bin:/usr/lib/ccache:/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin:/root/bin:/sbin
[java] FATAL [compilation thread] (?:?) - Make file returned error: 2

 
If some has any suggestion or idea of what could be wrong or what I could try please let me know, it would be very very helpful.

Thank you in advance,
Christos

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Contiki-developers mailing list
Contiki-developers@...
https://lists.sourceforge.net/lists/listinfo/contiki-developers
shane wood | 2 Aug 2007 03:13
Picon

contiki v1x

 run a web server using contiki v1.0 it still has the mouse support.
Its fantastic runs flawlessly, now it would be great to have more apps
but changing it to contiki v2 is unnessary.
 
Time would be better spent on improving the version we have and creating
more apps to run on it for the apple and c64.
 
I am not a programmer and dont have a clue on how to even start improving
it the best i can do to help would be to test it on a collection of c64s and accessories.
 
The contiki developers have done us a great service it would be nice now if we can improve
on what is a great operating system.
 
If i can help in anyway you would find me very active in testing for the c64 i have just finished
setting up phpbb2 on my debian box and would gladly start a permanent contiki forum on it.
 
 
 
Thanks   Shane Wood  www.c64web.com
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Contiki-developers mailing list
Contiki-developers@...
https://lists.sourceforge.net/lists/listinfo/contiki-developers
Andy Jarrell | 2 Aug 2007 05:31
Picon
Gravatar

Thanks for the summary

Thanks, Oliver, for the summary.  That was very helpful as I was
wondering where it would be worth plugging in.  No fun reinventing the
wheel and much more fun improving on the already existing base.
Thanks!!

Andy

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
Oliver Schmidt | 2 Aug 2007 08:07
Picon

Re: Thanks for the summary

Hi Andy,

> Thanks, Oliver, for the summary.

You're welcome :-)

Best, Oliver

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
Scheers Bart | 6 Aug 2007 15:56
Picon
Favicon

SPI on the Sky platform

 

 

Hi ,

 

My name is Bart Scheers. I'am working at the Royal Military Academy of Belgium. We use contiki on the Sky platform for some months now. For the moment we have 3 projects running.

One of the projects is about measuring  throughput over the IEEE802.15.4 air-interface on the nodes. In this project we  have some problems. It seems that the bottleneck in the throughput between 2 Sky nodes is not the air interface but the SPI interface between the MSP430 and the CC2420. (the transfer of 127 bytes on the SPI takes longer than the inter frame space (LIFS)).

 

Question: How can we best speed up the baudrate of the SPI interface without influencing the rest of the code (contiki and uIP)?

 

 

 

Thanks for your time,

 

Bart

 

-----------------------------------------------

Maj Bart SCHEERS

E-mail: bart.scheers-vJY35iZFB3CzQB+pC5nmwQ@public.gmane.org

Bemilcom: 9-2428-6626

Tel: 02/742.66.26

Fax: 02/742.66.22

-----------------------------------------------

 

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Contiki-developers mailing list
Contiki-developers@...
https://lists.sourceforge.net/lists/listinfo/contiki-developers
duane ellis | 7 Aug 2007 02:06

Re: SPI on the Sky platform

Scheers Bart wrote:

 >> [[helpme - my msp430 is so slow]]

here are some fairly generic things to look at, they are not "contiki" 
specific.

Q1: There are many MSP430 chips.
    Are you using the one with a hardware SPI - or a software GPI bitbang?
   You did not say.

Q2:   The CC2420 - (data sheet I have) says the SPI clock can run at 10mhz.
    How fast have you measured it with your scope?
    If it is bit banged - perhaps that can be fixed - little assembler.
    Or perhaps the hardware SPI baud rate is well - poorly chosen.

I have had similar situations on other things (MMC cards)
     The CPU clock ran at 48mhz - and the thing has to be an integer 
division of that.
     Divide by 2 = is too fast (24mhz, MMC card max=20mhz)
     Divide by 4 = 12mhz - huge through put loss.

     Might seem dumb - but - slowing the system clock from 48 to 40mhz...
     gives div-by-2  an exact 10mhz clock - fixing the bottle neck.

Q3: Is the problem the measured SPI baud rate being very slow (ie: bit bang)
    If so - I'd look at the SPI implementation - look for delay code
    that is used to throttle the SPI clock - simplify it.

Q4: Or - the SPI clock is fast... It is contiki that takes for ever 
getting the packet ready.

here's how to figure out where the problems are:

Here is a dirty rotten trick:
    Can you find a couple GPIO spare pins?
    At key places in your code - wiggle the GPIO pins.
         Start with a LONG wiggle, then a series of short long - 
binary-ish code.
       At key points in your code insert a debug_wiggle( 5 ), or 
debug_wiggle(3);
   Or - if you have a DAC of some sort...
       Output different voltages.

   Watch these signals on a digital scope or timing logic analyzer.

    Sometimes - if you know the sequence of events - just wiggling a pin 
0/1 - is enough
    each event toggles the pin, you just need to measure each segment on 
the scope screen.

In effect, the scope draws a nice 'time line of events' You are just 
putting markers along the time line by wiggling pins. You can use that 
time line to figure out what is taking so long.

The DAC output is kind of cool and simple - *AND* it looks like one of 
those "WHICH TASK IS RUNNING" graphs that fancy operating systems have.

Sometimes - depending on your system configuration the operating system 
may not have a high enough resolution timer, or maybe - you need to use 
a higher resolution timer.

-Duane.

    
>  
>  
> Hi ,
>  
> My name is Bart Scheers. I'am working at the Royal Military Academy of
> Belgium. We use contiki on the Sky platform for some months now. For the
> moment we have 3 projects running. 
> One of the projects is about measuring  throughput over the IEEE802.15.4
> air-interface on the nodes. In this project we  have some problems. It seems
> that the bottleneck in the throughput between 2 Sky nodes is not the air
> interface but the SPI interface between the MSP430 and the CC2420. (the
> transfer of 127 bytes on the SPI takes longer than the inter frame space
> (LIFS)).
>  
> Question: How can we best speed up the baudrate of the SPI interface without
> influencing the rest of the code (contiki and uIP)?
>  
>  
>  
> Thanks for your time,
>  
> Bart
>  
> ----------------------------------------------- 
> Maj Bart SCHEERS 
> E-mail:  <mailto:bart.scheers@...>
bart.scheers@... 
> Bemilcom: 9-2428-6626 
> Tel: 02/742.66.26 
> Fax: 02/742.66.22 
> ----------------------------------------------- 
>  
>
>   
> ------------------------------------------------------------------------
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> ------------------------------------------------------------------------
>
> _______________________________________________
> Contiki-developers mailing list
> Contiki-developers@...
> https://lists.sourceforge.net/lists/listinfo/contiki-developers
>   

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
Jack Peacock | 7 Aug 2007 02:18

Re: SPI and MSP430

-----Original Message-----
From: contiki-developers-bounces@...
[mailto:contiki-developers-bounces@...] 
Q2:   The CC2420 - (data sheet I have) says the SPI clock can run at 10mhz.

The MSP430 with hardware SPI is limited to 1/2 of MCLK (8mhz), or 4Mhz for
all but the newest MSP430 versions, so you can't run at 10Mhz SPI clock.  I
haven't looked at the CC2420 code, but is the SPI data transfer interrupt or
polled?  At max 4Mhz speed the interrupt overhead is substantial, might be
better to poll it.
  Jack Peacock

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
Simon Berg | 7 Aug 2007 12:29
Picon
Favicon

Porting contiki to ZX Spectrum

Inspired by the C64 port, I started porting Contiki to my ZX Spectrum.
So far I've been using the 2.x branch, but some recent mesages on this
list makes me wonder wether this is a good idea or if I should go with
1.x . 
My ultimate goal is to get something like a webbrowser or a VNC client
to work. I noticed that there are such things in the apps directory.
Will they work under 2.x ?

So far I'm close to being able to send IP packets to my Spectrum (4kbps,
half duplex).

simon

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
Oliver Schmidt | 7 Aug 2007 11:30
Picon

Re: Porting contiki to ZX Spectrum

Hi Simon,

> Inspired by the C64 port, I started porting Contiki to my ZX Spectrum.

Great :-)

> So far I've been using the 2.x branch, but some recent mesages on this
> list makes me wonder wether this is a good idea or if I should go with
> 1.x .

By now not surprisingly I'd go for 1.x for the reasons pointed out before.

BTW: One additional reason not mentioned before is that the 2.x build
system much more complicated / harder to understand than the 1.x one.
And just like with the source code "1.x alike" builds don't benefit
from those changes. Have a look at

http://contiki.cvs.sourceforge.net/contiki/contiki-2.x/platform/win32/Makefile?revision=1.5&view=markup

and

http://contiki.cvs.sourceforge.net/contiki/contiki-2.x/platform/win32/Makefile.win32?revision=1.10&view=markup

to see what gnumake tricks were necessary to _CLEANLY_ the 2.x build
system create a "1.x alike" output.

> My ultimate goal is to get something like a webbrowser or a VNC client
> to work. I noticed that there are such things in the apps directory.
> Will they work under 2.x ?

I created a "1.x alike" port of 2.x to Win32. It's primary goal is to
make it very easy to proove that "1.x alike" usage of the 2.x codebase
is (and stays) in fact possible.

So the answer to your question is yes.

Best, Oliver

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
Groepaz | 7 Aug 2007 11:42
Picon

Re: Porting contiki to ZX Spectrum

On Tuesday 07 August 2007 11:30, Oliver Schmidt wrote:
> BTW: One additional reason not mentioned before is that the 2.x build
> system much more complicated / harder to understand than the 1.x one.
> And just like with the source code "1.x alike" builds don't benefit
...
> to see what gnumake tricks were necessary to _CLEANLY_ the 2.x build
> system create a "1.x alike" output.
...
> I created a "1.x alike" port of 2.x to Win32. It's primary goal is to
> make it very easy to proove that "1.x alike" usage of the 2.x codebase
> is (and stays) in fact possible.

infact, the new buildsystem is what stopped me from proceeding with the c64 
port of 2.x ... it heavily relies on gnumake, and also certain properties of 
the gcc preprocessor. with just cc65 its not possible to build it cleanly 
(you'd have to use the preprocessor from gcc atleast, which i didnt want to). 
and i strongly doubt you'd have much fun with sdcc either (which i guess you 
would be using for a speccy port?).

so uhm yes, go for 1.x ... check the gameboy port on cvs, it should atleast 
contain some necessary tweaks for sdcc :) in theory all you need to do is 
write a conio style lib (which is just a handful functions) and link the 
stuff together =P

--

-- 

http://www.hitmen-console.org
http://www.gc-linux.org/docs/yagcd.html
http://www.pokefinder.org
http://ftp.pokefinder.org

Java is high performance. By high performance we mean adequate. By adequate we 
mean slow. 
<Mr. Bunny>

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/

Gmane