[Contiki-developers] msp430 toolchain problem with contiki-2.5/2.x

[Bruno Filipe Marques]

Hi List,

I upgraded my Mac OS X to Lion, and installed XCode 4.1

Also, installed fink and msp430 toolchain (V. 4.5.3).

Using git I upgraded contiki to the last 2.x snapshot

When I compiled examples/ipv6/rpl-udp, I got following compiling errors:

[…]
In file included from ../../../cpu/msp430/./msp430.c:37:0:
/sw/lib/msp430/lib/gcc/msp430/4.5.3/../../../../msp430/include/io.h:1:2: warning: #warning <io.h> is deprecated, please include <msp430.h>
In file included from ../../../cpu/msp430/./msp430.c:38:0:
/sw/lib/msp430/lib/gcc/msp430/4.5.3/../../../../msp430/include/signal.h:43:2: warning: #warning msp430-libc <signal.h> deprecated, using <legacymsp430.h>
../../../cpu/msp430/./msp430.c:39:24: fatal error: sys/unistd.h: No such file or directory
compilation terminated.
make: *** [obj_sky/msp430.o] Error 1

I think that is due to some incongruents in contiki files and the msp430 toolchain.

Can anyone help me on this, perhaps how to apply patchs to correct the files, and where getting those patchs ?

I have my phd research work stoped due to this, and I do not whant to use InstantContiki-2.5.

Another thing is that, I have a ubuntu 11.10 (beta 1) installed on a VM, and using the msp430 toolchain available on it's repos, I get the same error !

Thanks in advance

/bruno
------------------------------------------------------------------------------
All of the data generated in your IT infrastructure is seriously valuable.
Why? It contains a definitive record of application performance, security
threats, fraudulent activity, and more. Splunk takes this data and makes
sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-d2dcopy2
_______________________________________________
Contiki-developers mailing list
Contiki-d...@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/contiki-developers

[Rainer Thome]

Hi!

There is a version with patches for the latest mspgcc toolchain, which
can be found here: https://github.com/pabigot/contiki.

Worked fine for me :-)

Greetz
rainer

2011/9/23 Bruno Filipe Marques <bm...@estv.ipv.pt>:

[Bruno Filipe Marques]

Hi,
Is this applied also to the last contiki-2.x git commitments.

Thanks,
/bruno

[Peter Bigot]

Not yet; see http://article.gmane.org/gmane.os.contiki.devel/9765.

I'm also looking forward to this; putting all that noise in a single
header instead of replicating it everywhere makes a lot of sense.

Peter

[Joakim Eriksson]

Hopefully the wait will be over early next week ;-)

Best regards,
-- Joakim Eriksson, SICS

[Eugen Berlin]

Hi List!

I am experiencing compiling errors too, but not with sky, with zolertia Z1 nodes
as TARGET devices instead.

My current system setup is as follows:

Debian sid with msp430-gcc installed from sid repo:
ii binutils-msp430 2.21~msp20110716-1
ii gcc-msp430 4.5.3~mspgcc-20110716-3
ii msp430-libc 20110612-2
ii msp430mcu 20110613-3

Contiki OS is up-to-date, pulled from git today.

When executing make TARGET=z1 test-adxl345 in the $contiki/examples/z1 folder I
get the following error:

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

[... ...]
msp430-gcc -DCONTIKI=1 -DCONTIKI_TARGET_Z1=1 -Wall -mmcu=msp430x2617 -g -Os
-fno-strict-aliasing -Wall -mmcu=msp430x2617 -g -I. -I../../platform/z1/.
-I../../platform/z1/dev -I../../platform/z1/apps -I../../platform/z1/net
-I../../cpu/msp430/. -I../../cpu/msp430/dev -I../../core/dev -I../../core/lib
-I../../core/net -I../../core/net/mac -I../../core/net/rime -I../../core/net/rpl
-I../../core/sys -I../../core/cfs -I../../core/ctk -I../../core/lib/ctk
-I../../core/loader -I../../core/. -I../../apps/serial-shell -I../../apps/shell
-I../../apps/webserver -I../../apps/webbrowser -I../../apps/irc
-I../../apps/telnet -I../../apps/twitter -I../../apps/powertrace
-I../../apps/collect-view
-DCONTIKI_VERSION_STRING=\"Contiki-2.5-release-79-g50342fa\" -MMD -c
../../platform/z1/./msp430.c -o obj_z1/msp430.o
In file included from ../../platform/z1/./msp430.c:39:0:
/usr/lib/gcc/msp430/4.5.3/../../../../msp430/include/io.h:1:2: warning: #warning

<io.h> is deprecated, please include <msp430.h>

In file included from ../../platform/z1/./msp430.c:40:0:
/usr/lib/gcc/msp430/4.5.3/../../../../msp430/include/signal.h:43:2: warning:

#warning msp430-libc <signal.h> deprecated, using <legacymsp430.h>

../../platform/z1/./msp430.c:41:24: fatal error: sys/unistd.h: No such file or
directory
compilation terminated.
make: *** [obj_z1/msp430.o] Error 1

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

For now, I had to fall back to the msp430-gcc version provided by the Zolertia
team (msp430-gcc-4.4.5 / MSPGCC4_20110312), which is available on their wiki
page:
http://zolertia.sourceforge.net/wiki/index.php/Mainpage:Contiki_installation_Ubuntu

Since my and other groups in the department are actively using Contiki OS on
both Sky and Z1 nodes in research and education, we would be very happy to see
this issue solved. Is there any way we might help with this?

Thanks for reading!
Best regards!

Eugen Berlin

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,

security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.

http://p.sf.net/sfu/splunk-novd2d

[Zinon Zinonos]

Hi,

I am trying to install contiki-2-5 in ubuntu 11.10 (64 bits) and having the same problem with msp430 and contiki-2.5. Is there any solution for this issue?

Regrads,
Zinonas

[David Kopf]

Hi all,

I've been looking at the spacing between contikimac strobes, trying to get the econotag to transmit with the specified 500usec gap.

According to www.jennic.com/files/support_files/JN-AN-1035 Calculating 802-15-4 Data Rates-1v0.pdf the 802.15.4 standard the
minimum interframe separation is either SIFS=192 or LIFS=640 microseconds depending on whether the frame is shorter or longer than
18 bytes, to give radios time to switch between tx and rx and I guess the longer time is also to allow unloading of the rx frame.

I put 10 packets in the tx queue and printed the time taken for tx based on the econotag clock. This showed a 1053 usec gap between
maximum-sized packets, and I thought maybe the mac accelerator was enforcing the LIFS so tried short frames of 5, 10, 15, 20, and 25
bytes.
5 bytes on 10 strobes should add 5*8/250K * 10 = 1600 usec, and the results were
05: 14684 usec
10: 16284 (+1600)
15: 17888 (+1604)
20: 19490 (+1602)
25: 21088 (+1598)

all of which show the 1053 usec gap. So the mac accelerator is not looking at the frame size.

The gap went down to 204 usec when I cleared the ack request flag in the tx frame. The mac accelerator senses this bit and goes
into receive mode with a timeout if no ack is received. The timeout is set to just above what I found was needed to receive raven
ACKS reliably. Maybe something can be done with an added soft timeout that aborts after the 192usec ack turnaround time unless
reception has begun (ack takes 352 usec to transmit) but it still looks like a 700 usec gap is needed to get acks this way.

Adding a hardware cca before tx increases this. While the cca itself should only take 128 usec, the Jennic pdf says the minimum
csma on a clear channel should be 2369 usec based on an initial 140 symbol backoff period. So don't know if just setting the
hardware registers for the short cca plays nice with other radios.

Incidentally I just measured the attached that shows the minimum atmega128rfa1 transmission gap is 352 microseconds if no csma or
autoack is enabled. csma might just squeeze into 500 usec but just waiting for the autoack turnaround would exceed it.

Before making the effort to get acks within the 500 usec window between strobes, should contikimac take this LIFS of 640
microseconds into consideration, and set the rx CCA spacing based on that?