gmane.os.contiki.devel

Re: Contiki in the real world: from badgerstolight bulbs

月饼 — 2013-06-19T13:12:52

So cool! Proud of you!

Tim
------------------ Original ------------------
From:  "Adam Dunkels"<adam< at >thingsquare.com>;
Date:  Wed, Jun 19, 2013 09:04 PM
To:  "Contiki developer mailing list"<contiki-developers< at >lists.sourceforge.net>; 

Subject:  [Contiki-developers] Contiki in the real world: from badgers tolight bulbs

Hi all,

one of the cool things with Contiki has always been that it has such a wide range of uses. Until now, however, we haven't really had a place to show this off.

Today, this changes! We have compiled a list of real, deployed Contiki-based products and systems on the Contiki website. The list covers a lot of ground: from badgers to light bulbs, from data centers to city streets, and more.

Check it out: http://contiki-os.blogspot.se/2013/06/contikis-internet-of-things-from.html

/adam

Contiki in the real world: from badgers tolight bulbs

Adam Dunkels — 2013-06-19T13:04:30

Hi all,

one of the cool things with Contiki has always been that it has such a wide
range of uses. Until now, however, we haven't really had a place to show
this off.

Today, this changes! We have compiled a list of real, deployed
Contiki-based products and systems on the Contiki website. The list covers
a lot of ground: from badgers to light bulbs, from data centers to city
streets, and more.

Check it out:
http://contiki-os.blogspot.se/2013/06/contikis-internet-of-things-from.html

/adam

Re: Uncrustify code style config

Peter A. Bigot — 2013-06-19T12:47:14

Are you the one that decides?  (I ask that as an outsider who truly 
doesn't know, not as a challenge.)
I've read many issue and pull comments over the past week, and have seen 
the comments you mention.  They do nothing to refute my "suspicion" of 
inconsistent application.

As an outsider, I expect my contributions to be reviewed so they can be 
merged.  The closeness ("pickiness") of that review is affected by any 
antipathy the reviewer holds towards me.  Absent a standard tool and 
configuration by which style conformance is defined there is no 
objectivity, and I could be made to jump through hoops that do not apply 
to reviewers who are entitled to merge their own pulls without even 
waiting for a review.

It's a question of Contiki's governance and intent to lower the barriers 
that face potential contributors by eliminating variability of aspects 
of acceptance that can be made objective.  I'm just probing at that 
question.

Peter


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

Re: Uncrustify code style config

Peter A. Bigot — 2013-06-19T12:47:17

I'm not sure whether astyle would work.  The problem, as you note, is 
the degree to which Contiki's definition of "correct style" differs from 
pre-existing standards, for which the various tools generally support 
only minor variations.  In my own work I've chosen to mostly conform to 
1TBS with exceptions that are supported by my chosen conformance tool 
(see http://pabigot.github.io/bsp430/coding.html).  I define conformance 
by what that tool produces (astyle supersedes emacs ccmode), not an 
example document or textual description.


My experience from both sides of the fence is that this doesn't hold up 
in practice.  Simply stripping trailing whitespace from files would make 
them much more compatible with git patch exchange, and would be visibly 
unnoticed, but would have a huge impact on anybody who has made local 
changes (I ran into this in recent TimeLine.java enhancements; I haven't 
gone looking specifically but don't recall any conventions for Java, 
Makefile, and shell script style). Style-

Re: Uncrustify code style config

Karl Palsson — 2013-06-19T12:16:26

And you're updating the requirements document for that right now right?  So that it
actually does become a requirement, and not something someone said on a list some day
that no-body can find anymore?

I'm more than happy to work on improving an massaging the code into documented,
compilable, functioning compliance, but it does require someone to state what the
requirements would actually _be_

Sincerely,
Karl P

------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev

Re: Contikimac timing clarification for CC2530

David Kopf — 2013-06-19T12:18:10

Your packets may be too short for reliable detection by the two CCAs which 
might straddle the RF packet. If all the timings are correct they should 
not, (34+6
)*32 = 1.2 msec of RF per packet. But timings are not always correct :)

There is a pad option in contimac, look for SHORTEST_PACKET_SIZE which 
defaults to 43. That adds a byte header to store the actual length so has to 
be done on both sides. With 6lowpan address compression can be bypassed so 
no extra header needed (wireshark does not like it). Simplest test is to 
increase the length of the "Hello World" string.

Or the ACK timing may be off, either too fast or too slow. I don't know much 
about that hardware; with the Econotag I used the Raven hardware autoack 
timing as the gold standard and varied the econotag hardware delay for 
reliability.  Software delays can depend on interrupt latency so might be 
expected to fail more often.

-----Original Message----- 
From: SUTARIYA MAHESH
Sent: Wednesday, June 19, 2013 7:38 AM
To: Contiki developer

Re: Uncrustify code style config

George Oikonomou — 2013-06-19T11:51:27

Hi Peter


Having seen Contiki's code format guidelines, do you think astyle could be configured accordingly? I should probably point out here that there are some more code style requirements not documented in said page. The spaces (or lack thereof) around stars in declarations, casts and dereferences spring to mind here.


The problem with both tools at the moment is that, in addition to fixing incorrect code style, they also change from correct to different correct (for instance, annoyingly, indent can't be told to leave comments alone). Even worse, both tools also change correct to incorrect. This is what I am trying to fix with uncrustify but I'm not quite there yet, one of the reasons being a bug in uncrustify itself.


I would hazard a 'probably not' guess here


I don't think this would be required but my opinion is that any pull request that fixes code style would be welcomed.
George



------------------------------------------------------------------------------
This SF.net email is sponsored by Wi

Re: Contikimac timing clarification for CC2530

SUTARIYA MAHESH — 2013-06-19T11:38:40

Hi

< at >David,
Yes phase optimization is turned off using WITH_PHASE_OPTIMIZATION define
set to 0. I also cross checked it by placing #error inside phase
optimization code.

Two reason due to which packet doesn't reach destination is
1)I am not receiving ack(small packet received on transmit- I presume its
ack) on sendor side from receiver during loss of packet case which I
usually receive. - occurs more than 2nd reason. Please check attached log
for this case. As can be seen, I don't receive ack during last packet
transmit.

2)I am receiving Bad sync error on sendor size, which may be during read of
ack.

Regards,
Mahesh


On Wed, Jun 19, 2013 at 4:13 PM, David Kopf <dak664-2p+qKb8Fl0QN+BqQ9rBEUg< at >public.gmane.org> wrote:

------
Sender
------
Send packet
RF: Prepare 0x25 bytes
RF: data = 61ccab49548a050e01004b1200b1100e01004b1200000c48656c6c6f20576f726c6421
RF: Read
RF: read (0x05 bytes) = 0200ab
Send packet
RF: Prepare 0x25 bytes
RF: data = 61ccac49548a050e01004b1200b1100e01004b1200000c48656c6c6f20576f726c642

Re: Uncrustify code style config

Oliver Schmidt — 2013-06-19T11:18:54

Hi,


I'd consider this desirable. However I don't intend to do it.


This suspicion is wrong. If you follow the discussions on recent pull
requests you'll find quite some comments (and fixing commits)
regarding code style.

Regards,
Oliver

------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev

Re: Uncrustify code style config

Peter A. Bigot — 2013-06-19T10:50:06

I haven't tried the uncrustify version (I personally like astyle for 
this sort of thing), but was curious about the prevalence of compliance 
with 
https://github.com/contiki-os/contiki/wiki/Code-Contributions#code-formatting 
within the baseline.  Running the two tools over the current master branch:

1150/1250 .c files (546/969 .h files) modified by indent
1132/1250 .c files (705/969 .h files) modified by uncrustify

Once an acceptable tool and configuration is identified, is there any 
intent of reformatting the entire Contiki baseline in one massive change 
to introduce consistency?

If not, is there an intent to require that pull requests that 
significantly change files that are non-compliant begin with a 
indentation-only commit which brings the file into conformance, allowing 
the changes-with-effect to be clearly identified while slowly moving the 
baseline to the standard layout?

I'm not proposing either approach.  I just suspect that in the absence 
of a plan, coding style compliance will simply

Re: Contikimac timing clarification for CC2530

David Kopf — 2013-06-19T10:43:16

Are you sure you turned off phase optimization? That is the only thing I can 
think of that would cause  packet loss to increase with delay between 
packets (due to clock drift).

When single 802.15.4 packets are sent, each channel check cycle is 
independent and should result in one unicast packet being reliably 
transferred with an average time of half the channel cycle. By RF gap I mean 
the blank interval between tx strobes, if the rx does two CCAs with a 
slightly larger time separation it is guaranteed to detect a signal.

CCA_CHECK_TIME contributes to the CCA interval, but the hardware may take a 
millisecond to wake and do a CCA , or a millisecond delay for autoack 
detect. You need to actually time the hardware using rtimers or an 
oscilloscope. See e.g. 
https://github.com/contiki-os/contiki/wiki/RDC-Phase-optimization and 
https://github.com/contiki-os/contiki/wiki/Mobile-Econotag

-----Original Message----- 
From: SUTARIYA MAHESH
Sent: Wednesday, June 19, 2013 3:31 AM
To: Contiki developer mailin

Re: Contikimac timing clarification for CC2530

SUTARIYA MAHESH — 2013-06-19T07:31:29

Hi David,

I was trying to get 100% output with your suggestion.
I could get better result after disabling phase optimization at a cost of
power. But still loss occurs sometimes.
I set Inter packet interval to 0.4ms, When I send packet at lesser rate(say
packet at every 5 second) then error rate is high. When I send packet at an
interval of approximately 1 second, then it gives me good result and packet
reaches most of the time. So if actual packet sending interval is high()
then will it affect transmission error rate?

I guess rx CCA interval you mention is CCA_CHECK_TIME in contikimac.c which
might needs adjustment.
What you mean by tx RF gap?

Regards,
Mahesh

On Wed, Jun 19, 2013 at 12:19 PM, SUTARIYA MAHESH
<maheshsutariya-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>wrote:

------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev_______________________________________________
Conti

Re: Contikimac timing clarification for CC2530

Antonio Rosa — 2013-06-19T07:10:48

Thank you very much, and sorry for my english.

Regards, Antonio Rosa.

2013/6/19 SUTARIYA MAHESH <maheshsutariya-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>

------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev_______________________________________________
Contiki-developers mailing list
Contiki-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/contiki-developers

Re: Contikimac timing clarification for CC2530

SUTARIYA MAHESH — 2013-06-19T06:49:32

Hi Antonio,

I am having this configuration. Even I am getting overflow after few
operation. But just concentrating on main issue right now.

channel check rate: 8(default)
rdc: contikimac
mac:csma(default)

Regards,
Mahesh

On Wed, Jun 19, 2013 at 12:11 PM, Antonio Rosa <
antoniorosarodriguez-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org> wrote:

------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev_______________________________________________
Contiki-developers mailing list
Contiki-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/contiki-developers

Re: Contikimac timing clarification for CC2530

Antonio Rosa — 2013-06-19T06:41:24

Hi,

I have this configuration in project-config.h:

#ifndef PROJECT_CONF_H_
#define PROJECT_CONF_H_

#define BUTTON_SENSOR_CONF_ON 1
#define UIP_CONF_ICMP6        1


#define NETSTACK_CONF_RDC_CHANNEL_CHECK_RATE 16

#define NETSTACK_CONF_RDC contikimac_driver

#define NETSTACK_CONF_MAC csma_driver


this configuration generates a stack over flow in my cc2530 platform. I'm
using RPL, contikimac_driver and a RDC setting with 16 an csma driver.

Regards, Antonio Rosa.

2013/6/18 Antonio Rosa <antoniorosarodriguez-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>

------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev_______________________________________________
Contiki-developers mailing list
Contiki-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/contiki-developers

netcat tunslip6 with udp_echo_server

karim fathallah — 2013-06-19T01:08:24

hello
i am using rpl_udp example in my application. but i want to send message
from base station to the udp_server by using netcat. i connect on the root
by tunslip6 but netcat can't connect on the udp server.

i tried to modify the code of rpl_border_router but it is not clear how it
work.
 thanks a lot for your help

Re: change the target in contiki examples‏

mostafa rashad — 2013-06-18T20:10:45

Thanks a lot Mr. David 

       ------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev_______________________________________________
Contiki-developers mailing list
Contiki-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/contiki-developers

Re: Button cc2530 does not generatepushbuttonevent

Antonio Rosa — 2013-06-18T17:15:53

Ok,

Tell me if you get new results.

Regards, Antonio Rosa.

2013/6/18 Vladimir Alemasov <alemv.c550-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>

------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev_______________________________________________
Contiki-developers mailing list
Contiki-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/contiki-developers

Re: Contikimac timing clarification for CC2530

Antonio Rosa — 2013-06-18T17:13:30

At last, about I refer to this RIME stack :
http://dunkels.com/adam/dunkels07rime.pdf

Regards, Antonio Rosa.

2013/6/18 SUTARIYA MAHESH <maheshsutariya-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>

------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev_______________________________________________
Contiki-developers mailing list
Contiki-developers-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f< at >public.gmane.org
https://lists.sourceforge.net/lists/listinfo/contiki-developers

Re: Contikimac timing clarification for CC2530

Antonio Rosa — 2013-06-18T17:12:13

Hi,

I have been reading your proyect-conf.h file in
contiki<https://github.com/cepro/contiki>/
examples <https://github.com/cepro/contiki/tree/master/examples> /
serial-lprf-mac<https://github.com/cepro/contiki/tree/master/examples/serial-lprf-mac>.
In order to set a RDCMAC driver or a MAC driver, there is this guide with
all the steps necessaries:
https://github.com/contiki-os/contiki/wiki/Change-mac-or-radio-duty-cycling-protocols

And in your makefile file, you have added this code line:


CFLAGS+=-DPROJECT_CONF_H



I don't know if this is the same that this:


 CFLAGS += -DPROJECT_CONF_H=\"project-conf.h\"


Regards, Antonio Rosa.


2013/6/18 SUTARIYA MAHESH <maheshsutariya-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org>

------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev_______________________________________________
Contiki-developers mailing list
Contiki-developers-5NWGOfrQm

Re: change the target in contiki examples‏

David Kopf — 2013-06-18T15:02:48

The various defines are in httpd.h, WEBSERVER_CONF_NANO==2 will enable all 
the cgi code for the corresponding target. You may need to modify ajax.js 
and ajax.shtml to correctly process the data, it seems to be configured for 
the econotag at present.

I have not tested it in a while and would not be surprised if the javascript 
needs fundamental changes. Debugging that is most easily done using the 
minimal-net webserver and the Firefox Web Developer add-on.

-----Original Message----- 
From: mostafa rashad
Sent: Tuesday, June 18, 2013 9:16 AM
To: Contiki developer mailing list
Subject: Re: [Contiki-developers] change the target in contiki examples‏



Thanks a lot Mr.David for replying & for help

I really don't know from where the problem of addressing

I only do the command of
root< at >instant-contiki:/home/user# ip -6 address add bbbb::0/64 dev usb0
root< at >instant-contiki:/home/user#ifconfig usb0 -arp

but for now l would like to skip that problem & start to ask another 
question

In the page of ajax only