search

contiki-os / contiki

The official git repository for Contiki, the open source OS for the Internet of Things
http://www.contiki-os.org/
Other
3.72k stars 2.58k forks source link

hello world example not working on cc2538 #991

Closed TobyEalden closed 9 years ago

TobyEalden commented 9 years ago

I'm a new to contiki so may be missing something obvious, but using the latest master the hello-world (nor any other) example does not work on the cc2538dk board.

I get the following output:

Contiki-2.6-2071-gc169b3e TI SmartRF06 + cc2538EM [hangs...]

Switching to the release-2-7 branch and everything works as expected.

g-oikonomou commented 9 years ago

Hello Toby.

Just one merge ahead of you:

Contiki-2.6-2073-gc76316a
TI SmartRF06 + cc2538EM
 Net: sicslowpan
 MAC: CSMA
 RDC: ContikiMAC
Rime configured with address 00:12:4b:00:ae:26:07:01
Hello, world

What OS are you building on? What toolchain version are you using? I'm on OS X with:

$ arm-none-eabi-gcc -v
Using built-in specs.
COLLECT_GCC=arm-none-eabi-gcc
COLLECT_LTO_WRAPPER=/usr/local/gcc-arm-none-eabi-4_9-2014q4/bin/../lib/gcc/arm-none-eabi/4.9.3/lto-wrapper
Target: arm-none-eabi
[...]
Thread model: single
gcc version 4.9.3 20141119 (release) [ARM/embedded-4_9-branch revision 218278] (GNU Tools for ARM Embedded Processors)
TobyEalden commented 9 years ago

I'm on Linux Mint 17 32-bit, kernel 3.13.0-24-generic, and toolchain:

Using built-in specs. COLLECT_GCC=arm-none-eabi-gcc COLLECT_LTO_WRAPPER=/usr/local/gcc-arm-none-eabi-4_8-2014q3/bin/../lib/gcc/arm-none-eabi/4.8.4/lto-wrapper Target: arm-none-eabi ... Thread model: single gcc version 4.8.4 20140725 (release) [ARM/embedded-4_8-branch revision 213147](GNU Tools for ARM Embedded Processors)

On 12 March 2015 at 21:06, George Oikonomou notifications@github.com wrote:

Hello Toby.

Just one merge ahead of you:

Contiki-2.6-2073-gc76316a TI SmartRF06 + cc2538EM Net: sicslowpan MAC: CSMA RDC: ContikiMAC Rime configured with address 00:12:4b:00:ae:26:07:01 Hello, world

What OS are you building on? What toolchain version are you using? I'm on OS X with:

$ arm-none-eabi-gcc -v Using built-in specs. COLLECT_GCC=arm-none-eabi-gcc COLLECT_LTO_WRAPPER=/usr/local/gcc-arm-none-eabi-4_9-2014q4/bin/../lib/gcc/arm-none-eabi/4.9.3/lto-wrapper Target: arm-none-eabi [...] Thread model: single gcc version 4.9.3 20141119 (release) [ARM/embedded-4_9-branch revision 218278](GNU Tools for ARM Embedded Processors)

— Reply to this email directly or view it on GitHub https://github.com/contiki-os/contiki/issues/991#issuecomment-78617215.

g-oikonomou commented 9 years ago

Just tried with your toolchain version (still on OS X) - again no issues.

Can you try changing the value of SMALL between 0 and 1 either in the hello_example's Makefile or in Makefile.cc2538dk? You will have to make clean after each change.

A little bit of a strange question perhaps, but does your EM have an external antenna?

What version of the chip on your EM? (There should be a sticker at the back saying PG2 or somethng like that)

TobyEalden commented 9 years ago

I changed SMALL from 1 to 0 in Makefile.cc2538dk and rebuilt and it is now working...

What does that indicate?

On 12 March 2015 at 22:51, George Oikonomou notifications@github.com wrote:

Just tried with your toolchain version (still on OS X) - again no issues.

Can you try changing the value of SMALL between 0 and 1 either in the hello_example's Makefile or in Makefile.cc2538dk? You will have to make clean after each change.

A little bit of a strange question perhaps, but does your EM have an external antenna?

What version of the chip on your EM? (There should be a sticker at the back saying PG2 or somethng like that)

— Reply to this email directly or view it on GitHub https://github.com/contiki-os/contiki/issues/991#issuecomment-78672391.

g-oikonomou commented 9 years ago

The optimiser will enable/disable different optimisations (and thus generate different code) between the two cases: SMALL=1 optimises for size -Os, wheras SMALL=0 optimises with -O2

More info #549 and #503

What I fail to get is how both versions boot on my EM, but one of them does not boot on yours...

If you can be bothered, I'd be interested to know if you observe the same situation for the two values of SMALL if you upgrade to a more recent toolchain version (e.g 4.9.x)

Just as I thought that at least that one had been fixed for good :)

TobyEalden commented 9 years ago

Just upgraded the toolchain to 4.9.3 20141119 to no avail - still fails if SMALL = 1, and fine with SMALL = 0

FWIW, the sticker on the back of the EM says:

Texas Instruments Model: CC2538EM (DK) Rev.: 1.2.0.1 SN: 0000 0EBA

There is also a smaller sticker:

NO 1421 PbF

On 12 March 2015 at 23:13, George Oikonomou notifications@github.com wrote:

The optimiser will enable/disable different optimisations (and thus generate different code) between the two cases: SMALL=1 optimises for size -Os, wheras SMALL=0 optimises with -O2

More info #549 https://github.com/contiki-os/contiki/pull/549 and #503 https://github.com/contiki-os/contiki/pull/503

What I fail to get is how both versions boot on my EM, but one of them does not boot on yours...

If you can be bothered, I'd be interested to know if you observe the same situation for the two values of SMALL if you upgrade to a more recent toolchain version (e.g 4.9.x)

Just as I thought that at least that one had been fixed for good :)

— Reply to this email directly or view it on GitHub https://github.com/contiki-os/contiki/issues/991#issuecomment-78682581.

g-oikonomou commented 9 years ago

So your EM doesn't have an external antenna, right? I've got EMs from an older batch: Mine says 1_0_1 and has an external antenna connector. The chip is identical though, as far as I know anyway.

I wonder...

TobyEalden commented 9 years ago

Correct, there is no external antenna...

On 13 March 2015 at 14:29, George Oikonomou notifications@github.com wrote:

So your EM doesn't have an external antenna, right? I've got EMs from an older batch: Mine says 1_0_1 and has an external antenna connector. The chip is identical though, as far as I know anyway.

I wonder...

— Reply to this email directly or view it on GitHub https://github.com/contiki-os/contiki/issues/991#issuecomment-78996140.

g-oikonomou commented 9 years ago

I'll have a chat with TI then and I'll also see if I can grab some of those newer ones. I'm fairly certain a friend of mine nearby has a couple.

If anyone else has the same EMs as those described by Toby, it would be nice to get some success/failure reports.

Cheers

g-oikonomou commented 9 years ago

So I just had a chat with various people who cannot reproduce it either, but one of them said something in the lines of:

"But he should be getting some of the netstack printfs"

Which made a certain degree of sense...

Try getting rid of random_init() please and let me know what happens. Also, it would help if you emailed me your broken .bin and the corresponding .map file. You have my email :)

Cheers

TobyEalden commented 9 years ago

Commenting out random_init() in contiki-main.c made no difference...

On 13 March 2015 at 15:03, George Oikonomou notifications@github.com wrote:

So I just had a chat with various people who cannot reproduce it either, but one of them said something in the lines of:

"But he should be getting some of the netstack printfs"

Which made a certain degree of sense...

Try getting rid of random_init() please and let me know what happens. Also, it would help if you emailed me your broken .bin and the corresponding .map file. You have my email :)

Cheers

— Reply to this email directly or view it on GitHub https://github.com/contiki-os/contiki/issues/991#issuecomment-79020621.

g-oikonomou commented 9 years ago

Turns out this was related with Uniflash / Toby's local setup.