I'm assuming that there is a space before -comm (" -comm ") because
it's not visible in your message...
If it's crashing in MoteIF its probably trying to open the serial port.
There might be some difference in the environment that you get from exec().
Or there may be different permissions needed.
Check the installation and instructions for javacomm.
I believe it's part of the full T2 install and probably
gets put in tinyos.../tools. It's the interface between
Java and the serial ports. You should be able to google
info about "tinyos-tools" and "tos-install-jni".
MS
sergio mena doce wrote
I'm "fairly sure" that TOSBase doesn't do any received CRC check.
If you look at Packetizer.java::readFramedPacket() you can see
where messages are rejected based on the CRC failure, so this is
probably the place to hack.
I find that jackhammers and banging on pots and pans are good for
creating
e any thing
that can be done with these files:(they seem to be doing something with
crc)
1)tos\platform\mica2\CC1000RadioIntM.nc(I doubt something can be done
here)
2) tos\platform\system\CrcFilter
3) tos\platform\CRCPacket.nc.
As a newbie,i dont know perfectly what is going on?
Many Than
It could be as simple as putting the directories that contain the
"missing" command files into your PATH environment variable.
First, find the files using the "find" command or a FileMangler find.
Second, put the directory into your PATH, either using the Environment
Variables panels hidden benea
The ADCs in the micaz produce 10 linear bits which are stuffed
into a u_int16, where (pretty much) 0x000 is 0 volts at the
sensor input and 0x3FF is the battery voltage (which is used
as the conversion reference). Note that, without external
level shift hardware, there is no concept of a negative
See my just-past-post about the advisability of using FP values...
But you "should" be able to just declare:
float stuff[SIZE];
If you use floats in a TOSMsg payload, MIG should be able to generate
a Java class that interprets them correctly. I poked in there some
time ago and remember seeing
I can't help with the T2 basestation code, but I notice that
you may be confusing two things...
According to the first part of your message, it is
apparently possible to compile BS such that it only
receives packets that are addressed directly TO it.
However at the end you say:
> I just want to
s.
Am I making things too complicated? And is this done automatically?
Someone please correct me if I'm wrong.
Regards,
Kurt
Original Message Follows
From: Michael Schippling <[EMAIL PROTECTED]>
To: Roberto <[EMAIL PROTECTED]>
CC: TinyOS-Help
Subject: Re: [Tinyos-hel
3. RE: Deluge2.0 can't format the flash on telos revision a.
why? (JungDoyeong)
4. Re: Re: Sending float values (Michael Schippling)
5. RE: where is TOSBoot? How can I install and modify it?
(JungDoyeong)
6. He
If you are using Java on the host side you should use Packetizer.java
as your message handling code. This is buried inside
BuildSource.makePacketSource() and MoteIF
The BcastInject.java utility uses MoteIF to send and receive messages.
You can follow through the whole shebang in apps/SimpleCmd
You should probably start by learning the C language. This may help:
http://en.wikibooks.org/wiki/C_Programming
But a quick answer goes something like this:
A 'C' string like "gv" is an array of chars (or bytes to most machines)
so it may be declared and initialized thusly:
char string[] =
I'm struggling with getting tmote sky to work after years of banging
my head on mica/TOS. Using the CD I got from Moteiv a couple months ago,
labeled "boomerang Tmote Tools CD v2.0.5" I loaded the entire system from
scratch and managed to get the Delta demo to compile and run, finally,
on a fai
rce.release();
}
}
}
-
On 10/11/07, *Michael Schippling* < [EMAIL PROTECTED]
<mailto:[EMAIL PROTECTED]>> wrote:
You should probably start by learning the C language. This may help:
http://en.wik
I wonder if you have a timing race condition in your init code?
Just an idea, probably nothing useful...
MS
Nick Soldner wrote:
I posted earlier on this however now I have a bit more intuition as to
the nature of my problem. I have an application which writes data to the
micaz's 512KB flash EE
Tasks are like threads or processes. The task function gets put on
an execution queue and is then fired up sometime later. The post task
command returns, nearly, immediately and only indicates if the function
was put on the queue successfully.
MS
Roberto wrote:
Hi all,
I read the tutorials on t
All those missing header files belong to the "basic system".
If you were doing an actual micaz compile they would be found in:
/usr/local/avr/include
For a simulator compile I would guess that they should come from:
/usr/include
First verify that those directories and files exist. If they d
main.exe is the binary file for the platform to which you are compiling
I suspect that there is another question though?
MS
jose m wrote:
Hello,
A question: Can I compile my code to create a binary file? not intel or
motorola, hexadecimal code. Is the "main.exe" the binary?
Thanks in advance
Capture the output from a complete compile and then just run the
gcc part of it as a script. You may need to put "PFLAGS += -v"
in your makefile to force NCC into verbose mode to get the GCC line.
MS
Zhifeng Lai wrote:
Dear all,
How can I compile “app.c” using avr-gcc directly? Is there any tu
There are quite a number of things wrong with your sample code,
starting with using a TOS_MsgPtr, which should be set to point
to some TOS_Msg sized block of memory, but isn't... But to
answer the actual question, go count up the number of bytes in the
platform/tmote/AM.h file. ( I think it's in t
e
"main.exe" to the microcontroller, is this the right program and works fine?
- Mensaje original
De: Michael Schippling <[EMAIL PROTECTED]>
Para: jose m <[EMAIL PROTECTED]>
CC: tinyos-help@Millennium.Berkeley.EDU
Enviado: sábado 13 de octubre de 2007, 16:16:12
Asunt
his hex dumping?
- Mensaje original ----
De: Michael Schippling <[EMAIL PROTECTED]>
Para: jose m <[EMAIL PROTECTED]>
CC: tinyos-help@Millennium.Berkeley.EDU
Enviado: lunes 15 de octubre de 2007, 19:31:07
Asunto: Re: [Tinyos-help] Compile to binary
I would suppose that it is a loader format,
If you can, run the amp power and ground directly from the battery
or whatever supply you have and put a bunch of filter caps on it.
Also be careful to use a single point ground...multiple ground
wires to one point, not a single wire daisy-chaining from place to
place. Although I suspect some nois
I've heard rumors that some comm components do not do ID filtering.
I know that this works for Mica2's using T1 GenericComm, but it
may be that other flavors' mileage varies. Try searching this list
for some more relevant information.
MS
Yang Bofu wrote:
Hi guys,
I am trying to do a unicast in
I assume you have something like this driving your data acquisition:
call Timer.start(TIMER_REPEAT, 100);
so you (should) get about 10 sample cycles per second.
Are you then sending one sample per message, or aggregating them in
a manner like Oscilloscope? When you say "only 3 packets/sec is
s after around 2mins, but i get
them after exactly 300-301 secs. Which boils down to 3 samples being
sampled instead of 10.(1000 samples collected in 300 secs means around 3
packets per second).
I am using Single hop.
Sincerely,
Ragh
On 10/16/07, *Michael Schippling* <[EMAIL PROTECTED]
&
From: "Michael Schippling" <[EMAIL PROTECTED]>
To: "Yang Bofu" <[EMAIL PROTECTED]>
Cc:
Sent: Tuesday, October 16, 2007 11:16 PM
Subject: Re: [Tinyos-help] AMSend Unicast
I've heard rumors that some comm components do not do ID filtering.
I know that this w
I don't know from TOSSIM, but on real motes the low-level code starts
around UARTM.nc. I don't know if there is an equivalent for the simulator
as it looks like your packet_sim.c file plugs in at a higher level.
But I'm a bit unclear on the purpose. Do you want to write a mote
program that implem
First, none of the "missing" components have anything to do with Java...
Java can live most anywhere on your system independent of TOS.
You can check to see if /bin/tcsh and the like actually exist. If they
do then you might just try using "--nodeps" on the RPM install. But
probably you missed so
h the --nodeps option.
Do I need to set any environment variable for starting the TinyOs?
Regards,
Lalit Shah
-Original Message-
From: Michael Schippling [mailto:[EMAIL PROTECTED]
Sent: Thursday, October 18, 2007 1:58 AM
To: Lalit Shah
Cc: tinyos-help@Millennium.Berkeley.EDU; [EMAIL PROT
The SimpleCmd demo app shows how to send messages from a PC and
receive them on a re-Mote. That may be sort of what you want...
MS
SS C wrote:
Hi all,
I'm trying to send my sensor data through my mote to another receiver.
My sensor data is collected by my MDA320. Right now I programed my
rem
led way.
Thanks in advanced, PLEASE DO help me >.<
On 10/18/07, *Michael Schippling* <[EMAIL PROTECTED]
<mailto:[EMAIL PROTECTED]>> wrote:
I don't know from TOSSIM, but on real motes the low-level code starts
around UARTM.nc. I don't know if there
Kenneth
On 10/19/07, *Michael Schippling* <[EMAIL PROTECTED]
<mailto:[EMAIL PROTECTED]>> wrote:
Well, as I said I don't know anything about TOSSIM so I'm
probably of no use without actually sitting at your computer...
But I'm still not sure why you want to use
the details of the
toscheck output.
I have used the version 1.5 of JDK instead of 1.4 weather it would cause any
problem or not.
Regards,
Lalit Shah
-Original Message-
From: Michael Schippling [mailto:[EMAIL PROTECTED]
Sent: Friday, October 19, 2007 12:46 AM
To: Lalit Shah
from the shell you can redirect both stdout to files or pipes
using something like this:
command > file.txt
command | command2
Similar options are available for stderr and stdin.
See "man bash" for details on "redirection".
MS
Chan kenniel wrote:
Dear all,
May I ask how to capture the
Anybody know if there is an equivalent for T1 with Makerules?
thx
MS
Aurélien Francillon wrote:
mejda chouaieb a écrit :
Hello All,
I want to know if there is a full description of the inner workings of the
TinyOS build system ? if yes please give me a link.
it's well documented here :
ti
Are they making good contact on the USB connectors?
Could be flaky power...
You could also try a "heavier" program that is not USB
related to see if they still flake. That could eliminate
the idea that there is a program size issue.
MS
Ariel Mauricio Nunez Gomez wrote:
Hello everyone,
My envi
Since TOS itself has no pthread lib...hmm...
Are you looking at the simulator code?
MS
Jitender ahuja wrote:
Can someone kindly help to point me why tinyos has thread mutex lock
(in event_queue.c) when as far as I know it has no true concurrency and
no threads too. For e.g., a statement of the
Try "type java" to see where you are getting your Java from.
If it indicates something like "jdk1.5.0_07/bin/java" or
other 1.4, 1.5, 1.6 versions you are probably fineotherwise
you may need to install a modern JDK.
For Graphviz, if "type dot" doesn't find anything, a file search for
the "Gra
OH, one other thing...some (especially) notebooks may not
provide enough current to the USB ports. A powered USB hub
might help that.
MS
Michael Schippling wrote:
Are they making good contact on the USB connectors?
Could be flaky power...
You could also try a "heavier" program that
nup...
There's a vague, environment dependent, relationship between radio signal
strength and distance which you can find a lot about on line, but the
tmotes don't have any internal spacial sensors so they don't know where
they are.
MS
Edmund J wrote:
hi,
Im developing applications for tmote s
I doubt that anything will be easy for people trying to instrument NESC...
But there is a printf() sort of debug interface for 'regular' TOS_Msg
messages, assuming you are not using the message channel for other more
useful work. I forget the details, of course, but searching this list
might turn
.
}
Cheers,
Urs
Yang Bofu schrieb:
Hi,
I still have the question about unicast in T2 (thank Michael
Schippling for your info). I have searched the archive, but can't find
the answer I need. so I post the question again.
I am trying to do a unicast by AMSend in my application. What I di
If "serial_source" has been typedefed to be a pointer,
which is what the subsequent usage tends to indicate,
then this _should_ allocate and init a block of whatever
rather than just a pointer. If that is the case then it
seems to be a poor naming choice. You'll have to chase
the dragon further up
Well, does "com.mathworks.jmi" exist in a class tree or jar file
that you are including in your CLASSPATH? It sounds like something
that MatLab supplies, so I'd go looking in their install first...
MS
Saif A. Al-Hiddabi wrote:
Hi All:
I am trying to set TinyOS to work with Matlab R2007b, b
OK and why I can not get scope to run?
I appreciate your help.
Regards,
SAIF
-----Original Message-
From: Michael Schippling [mailto:[EMAIL PROTECTED]
Sent: Tuesday, October 23, 2007 9:17 PM
To: Saif A. Al-Hiddabi
Cc: tinyos-help@Millennium.Berkeley.EDU
Subject: Re: [Tinyos-help] Error
I will spare the air in respect to MHO of the beauty of this demo app
vis TOSfucation of functionality...
In RfmToIntM.nc::ReceiveIntMsg.receive(TOS_MsgPtr m)
m is the full message being received from another mote.
m->data is the IntMsg 'payload', defined in tos/lib/Counters/IntMsg.h.
The easiest way to prevent TOSBase from sending radio messages is to not
write any messages to it over the UART...it only acts as a passive pass-
though. But if you want to make sure nothing gets transmitted you could
probably disable UARTRcvdTask(). There may be some slick way to wire-it-
out, bu
Does the OscopeMsg struct have a member named "data"?
If not, then don't try to reference it...
Probably if you remove the "->data" from the memcpy() line
you'll get what you want.
MS
Chan kenniel wrote:
Dear all,
I got a compiled problem in the following code, which is quite short and
simpl
I've seen some little plug-in KiloWattHour meters,
but as usual don't remember where...try searching
for "power meter" and stuff like that.
MS
weiping SONG wrote:
Hi all,
Is there a type of sensor for measureing electricity usage? For example,
if we were to use it to monitor teh amount of ele
to it.
I am making use of the XMDA320 code for this.
Is it possible to only sample the relevant ports? and Would this help in
reducing the time taken for program execution?
I also came across the Highfrequencysampling program. Would this code
help in solving this problem?
Regards,
Ragh
On
That is the purpose of the AM type field. You shouldn't have to change
the header, just define different packet types and bodies and use their
[AMMsgtype].receive(TOS_MsgPtr m) methods to implement specific behaviors.
It's a little know secret, but only "length" bytes of the data packet
are transm
The whole kit-n-kabodle is compiled and downloaded in one (or two)
fell swoop(s)...unless you use something like Deluge that allows
you to maintain multiple program images on-board, but even then I
believe they are wholly independent programs. This means that
there is no separate OS 'kernel' or lo
I can't vouch for T2, or even the new-fangled nx_types,
but there used to be support for "float" in MIG, and I
still find float conversion in:
tinyos-1.1.15/tools/java/net/tinyos/message/Message.java
If the support has been 'deprecated' you can always roll
your own by looking at the above exam
This probably isn't all you need, but "msp430" is the chip used on the
telos type Motes. I'd suspect that the string acts to trigger NESCC
into doing necessary controller specific processing.
MS
mejda chouaieb wrote:
Hello All,
Please can someone help me to understand this line in
"tos/platfor
You have stumbled on the -- well, best is not exactly the right word --
way to do embedded program debugging...Step and repeat.
There is a JTAG interface on the programmer board and some googling
may find info about how to use it to run the board under gdb. But that's
not real-time.
The other us
I noticed that someone recently had a matlab issue with starting oscope
and there was a "groupid" argument to be supplied somehow. That looks
to be your problem, and if those messages are cc2420esque then the 7D
is your group. Not much use, but a data-point none-the-less.
MS
Saif A. Al-Hiddabi w
It's probably not referring to files or folders but is used as a flag
to some processing step. Try searching _everything_ for "mmcu" and
"nesc-target" or substrings thereof. Include searching all the
makefiles and nesc files, I'd bet it's in there and will indicate
where you need to add specifics
27;s all I know.
sorry
MS
Saif A. Al-Hiddabi wrote:
How to fix the "groupid" thing?
Thanks
SAIF
-Original Message-
From: Michael Schippling [mailto:[EMAIL PROTECTED]
Sent: Monday, October 29, 2007 1:03 PM
To: Saif A. Al-Hiddabi
Cc: 'Juan Antonio López Riquelme';
I have apparently confused myself with new code...
I'm just starting to do useful work with Boomerang and Tmote sky.
I believe this means TOS1.1.15 plus now-defunct Moteiv addons.
Running the CountUart demo, I get this data stream from Listen:
04 00 00 00 00 00 7E 00 04 7D 00 1C 00 01
in little endian.
Thanks
- jpaek
Michael Schippling wrote:
I have apparently confused myself with new code...
I'm just starting to do useful work with Boomerang and Tmote sky.
I believe this means TOS1.1.15 plus now-defunct Moteiv addons.
Running the CountUart demo, I get this data str
ince Phil wrote his T2 programming manual.
Relevant doc might be Phil's book... and maybe some TEPs? I don't know.
If you think about it, it is just ntoh, hton thing,
and it's not too bad once you get used to it.
it took me a week :-)
Thanks
- jpaek
Michael Schippling wrote:
Yes,
So...basically EVERY platform we are dealing with is little-endian,
-- unless you are still running on a 68K (Mac?) of some kind -- but we
force the network types to be big-end just to increase the processing
requirements?
And the full extent of the doc on this is in the T2WG message that
klueska
Given that you are getting an exec() error, cc1 is probably a pre-processor
pass for the C compiler. There may be some components of your compiler that
are not installed. I'd go look for a manifest of what's supposed to be in
your compiler package. It could also be a 'standard' gcc component that
pon-ended. Good point on the fcfhi-fcflo thing.
ah, yess, standards, we all need standards...I should get some.
thanks
MS
Philip Levis wrote:
On Oct 31, 2007, at 10:27 AM, Michael Schippling wrote:
So...basically EVERY platform we are dealing with is little-endian,
-- unless you are still r
Try [EMAIL PROTECTED]:tmote for your MOTECOM, I believe it will
select some special parsing for tmote messages. At least it let
me receive Delta msgs with Listen.
You'll have to go look at the Delta message structure to see which
bytes are of interest, I don't know offhand.
As to the 3000 valu
0xff 0x0 0x0 ]
Which bytes representing the temperature measurement in the above packet?
Thanks
SAIF
---------
-Original Message-
From: Michael Schippling [mailto:[EMAIL PROTECTED]
Sent: Wednesday, October 31, 2007 5:0
I thought this would be simple, but I'm finding some inconsistent
usages in the demo apps and some of my own code...Anyway the working
examples (say, Oscilloscope) that I have for T1 do this:
In the config file:
OscilloscopeM.CommControl -> GenericComm;
Init the Comm system in the main Module:
oops, and add this to the top of the main Module too:
uses interface StdControl as CommControl;
MS
Michael Schippling wrote:
I thought this would be simple, but I'm finding some inconsistent
usages in the demo apps and some of my own code...Anyway the working
examples (say, Oscillo
I'll take a wild guess...
If that Read interface works like the ADC, what probably
happens is that Read.read() starts a "conversion" but
you don't get a result until Read.readDone() fires.
So you probably want to send your result message from
readDone() to get the latest and greatest.
MS
Aless
; from whatever "local" is into the
"sendbuf" data area and then sending it over the radio.
However I think there are some missing parens and it
should be sizeof(local), but maybe they are optional.
MS
Alessandro Turella wrote:
Michael Schippling ha scritto:
I'll take
and which produce only
08 00 00 00 00 00 7E 00 65 7D 00 00 01 00 00 00 00 00
08 00 00 00 00 00 7E 00 65 7D 00 00 01 00 00 00 00 00
08 00 00 00 00 00 7E 00 65 7D 00 00 01 00 00 00 00 00
...
Any help?
SAIF
-Original Message-
From: Michael Schippling [mailto:[EMAIL PROTECTED]
Sent: We
You should lookup the specs on the standard sensors to see if
they have any useful IR range. If so maybe a filter would do
the trick for you. Otherwise you'll need to find a photo-diode
that is good in the IR that you want and replace one of the
standard sensors -- or add it to the expansion conne
In T1 (I assume T2 is only _slightly_ different in some curious way)...
A header file defines all the message structures and "AM_" types.
Your config file has a line for each message type:
// Host commands to be received
RoboMsgM.RCmdStartMsg -> GenericComm.ReceiveMsg[AM_ROBOCMDSTARTMSG];
You may be able to use the same TOSish mechanism to set IDs.
Here's an example from a recent tmote build:
/opt/tinyos-1.x/tools/make/msp/set-mote-id --objcopy msp430-objcopy
--objdump msp430-objdump --target ihex build/tmote/main.ihex
build/tmote/main.ihex.out-0 0
There is a similar version
You sure it's 1.1.1.3? I've only seen 3 numbers...like 1.1.3 or 1.1.13
If it's 1.1.3, I'd continue to upgrade to something over ..10 just for
your own sanity. But that has nothing to do with your CLASSPATH...
I think the warning is for '.' (period) not dash, unless there is some
unicode foo I don
I just looked at the tmote specs and it shows a bit higher current
when _receiving_ (!!?), but in the range you have measured. Judicious
use of the radio is the issue, also look up "Low Power Listening" and
see if it is of any use.
MS
Leonardo wrote:
Hi Guys,
I have any problem with current con
The tmote Oscilloscope program strikes me as another of the expertly
obfuscated TOS demo apps that is nearly impossible to follow or
modify...IMHO.
I think the business end of the data collection/aggregation happens
in the tinyos-1.x/tos/lib/Oscope/OscopeM.nc file, and you may be
able to reverse-
I googled this:
TOS RSSI localization
and got this:
http://www.cs.berkeley.edu/~jortiz/papers/ee290q_localization.pdf
which may be useful or have relevant refs...
MS
cyberkiwy wrote:
hi!
is there a possibility to get the relative distance between 2 motes
MDA400, and to know if a mote
higher
than transmit power. It's normal with these radios. Decoding requires
much more processing than encoding.
Regards,
Harri
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Michael Schippling
Sent: Wednesday, November 07, 2007 8:40 PM
To: Leo
Actually, it contains everything (or as much as Joe could upload).
For some reason the nascent TOS Doc WG didn't want to announce that
we had it all online again...but you found it none-the-less.
MS
Jeonghoon Kang wrote:
FYI,
Surprising TinyOS Wiki contains something.
http://docs.tinyos.net/i
There used to be a QueuedSend interface in the T1 lib tree...
MS
Murray, Ben wrote:
Hi
I need to send multiple types (am_id_t / len) of mesage but I need to store
them up in a Queue until the mote has permission to send.
How best can I store multiple messages, of varying types/lenths, and
subs
Make sure you are getting a SUCCESS from the send(), otherwise the
message is not sent and no sendDone() is to be expected.
As to reliability, you can use various kinds of ACKs. The choices depend
on your platform, so google around a bit. I also did some brute force
testing and a resend protocol
If you haven't already, start by reading though the doc/tutorial
to get an idea what you are up against.
Then...all mote types have an Analog to Digital Converter (ADC)
which, conveniently enough, has a range from about 0-3.3v so
that should cover how to get the data. Look for modules named ADC.
It's anybody's guess which makefile actually does the deed
(read: I have no clue how to trace it back to the source...),
but the "ncc" command you see when you do a build breaks down
into a number of sub-commands, includng "/usr/bin/nescc" and
"nescc1" which is the most likely culprit. You can see
You need to put those export commands in a startup script file
so they get run everytime you crank up a bash shell.
The most common one is: $HOME/.bashrc
where HOME is usually: /opt/home/[username]
MS
Ittipong Khemapech wrote:
Hi,
Thank you for your reply. I set MAKERULES and it looks good. Ho
a lot.
Best,
Abdullah Kadri, P.Eng., Ph.D. Candidate
The University of Western Ontario
Department of Electrical and Computer Engineering
Phone: (519) 661-2111 ext. 81271
-----Original Message-
From: Michael Schippling [mailto:[EMAIL PROTECTED]
Sent: Friday, November 09, 2007 1:19 PM
To:
You can probably find it under CVS, or look here:
http://www.cems.uvm.edu/~aobrien/tmote/
MS
manu suryavansh wrote:
Hi,
Does any body has the Delta (Moteiv) application packet structure. It
was earlier available online but that link is not working and I couldn't
find it on both the links
Its a known bug in all software everywhere, but TOS doesn't have a
signal() mechanism to catch and release gracefully.
However the point is that many software errors can cause big problems,
running out of memory, running off the end of buffers, overwriting
pointers, etc...Make sure your code does
Pin numbers go from 0 to 7...
Also, assuming you want the same PWM0 as in the mica schematic,
its port B, pin 4. I have a convenient I/O port list online:
http://www.etantdonnes.com/Motes/ATMEGApins.txt
MS
yvonne er wrote:
Dear Sir/ Mdm,
I am using MDA100 for MicaZ to collect data readin
aright aright...BUSERR.
great explanation thought...
thanks
MS
Philip Levis wrote:
On Nov 10, 2007, at 9:07 PM, Jeongyeup Paek wrote:
Maybe there is a misunderstanding...
If you write a code that has segmentation fault,
then the mote will restart.
And I don't think this 'behavior' is a bug
Again, read that tutorial...but the quick answer, in T1 at least, is:
In the config file:
components ADCC;
RobotM.Target -> ADCC.ADC[ROBOADC_TARGET];
In the "RobotM.nc" code:
uses interface ADC as Target;
call Target.getData();
async event result_t Target.dataReady(uint16_t d)
{
My personal belief is that it is like "call" in that no direct return
value is available to the user. My further belief is that the signaled
code is run asynch, so the signaler does not block waiting for it.
I could be wrong (probably) so I hope someone who actually knows will help.
MS
Christop
Now that you mention it, I'm at a bit of a loss to explain where
the micaz ADC code gets it's getData() from...but be assured that
it's there somewhere. Possibly in mica2/ADCREFM.nc. If you do a
"make micaz docs" you may get a diagram of the module interactions.
The fact that there is an ADC inter
Assuming that you are not planing to use ADC3 for anything,
which you probably can't because it isn't connected to
the outside world other than that 10pin anyway, you should
just us it for your GIO bit. I don't know offhand which way
the pins default, but with your short at R14 you may have
both G
You probably need to set your MOTECOM variable to indicate what
platform you are using. For instance mine is:
export MOTECOM="[EMAIL PROTECTED]:tmote"
I'd guess there is a "telsob" as well, but tmote might work as
they are the same thing underneath.
As to your CLASSPATH toscheck issue...since
Search this list for "toscheck CLASSPATH".
Also google up a bit on java CLASSPATH,
it's very similar to the 'regular' PATH.
If you are not familiar with any of it,
doing "man bash" in a shell window will
give you more than you'll ever want.
MS
Abdullah Kadri wrote:
This is what I have in my "t
selectIOFunc() seems to be the thing in T1 that makes
each pin do it's Digital IO function. It may be redundant,
and it certainly isn't documented anywhere that I found...
As to the "shortage on connector pins", I agree. But that
doesn't explain the utility of being able to short two
controller l
in the original context can I do
rval = signal HPLT3capture.fire( count );
to get its result?
thx
MS
Philip Levis wrote:
On Nov 13, 2007, at 10:52 AM, Michael Schippling wrote:
My personal belief is that it is like "call" in that no direct return
value is available to t
t;call" as well...
MS
Philip Levis wrote:
On Nov 13, 2007, at 8:35 PM, Michael Schippling wrote:
So I guess my question is...is signal tantamount to an interrupt,
or is it 'just' a function call? E.g, here's an interrupt routine
that I cobbled together in cargo-cult fash
101 - 200 of 1470 matches
Mail list logo
