[riot-users] STM32F3Discovery + CC1101 radio

Viktor Váczi viktor.vaczi.90 at gmail.com
Thu Jan 28 17:49:06 CET 2016

I've solved the problem. The problem was in the cc110x gpio initialization.
You should use:
const cc110x_params_t cc110x_params[] = {
        .spi  = 0,
        .cs   = GPIO_PIN(0,3),
        .gdo0 = GPIO_PIN(0,0),
        .gdo1 = GPIO_PIN(0,6),
        .gdo2 = GPIO_PIN(0,2)
instead of pure numbers.

2016-01-28 16:57 GMT+01:00 Viktor Váczi <viktor.vaczi.90 at gmail.com>:

> I forgot to mention that the debugger says that the CPU catched a hard
> fault exception.
> 2016-01-28 14:06 GMT+01:00 Viktor Váczi <viktor.vaczi.90 at gmail.com>:
>> Thanks for the reply!
>> I was using PIN3 originally, but I've tried a ton of other pins. They
>> worked in the GPIO example.
>> I've added an interrutpt disable before the gpio inits, it has the same
>> error: http://pastebin.com/cBuMFyWH
>> 2016-01-28 12:22 GMT+01:00 Viktor Váczi <viktor.vaczi.90 at gmail.com>:
>>> Hi!
>>> I'm trying to get STM32F3discovery board to work with a cc1101 radio,
>>> and I'm getting an error while initializing it.
>>> I've only included the necessary modules in the boards makefile +
>>> created the cc110x_params.h file with the appropriate setting.
>>> The whole thing freezes up in the netdev2_cc110x_setup function. It goes
>>> for a while, then it calls cc110x_setup, and the first step is to
>>> initialize the CS gpio pin, so it calls gpio_init(dev->params.cs,
>>> The error occures at this line: port->PUPDR |= (pushpull << (2 *
>>> pin_num));
>>> I've ran a debugger an its seems that its trying to do a contexts switch.
>>> The GPIO test example works just fine I get this output on the serial
>>> console: http://pastebin.com/eGQLqisc
>>>  What to do/where to go? I've just started using ARM MCUs.
>>> Here is the github repo: https://github.com/blaisehorvath/RIOT I've
>>> changed the stm32f3discovery in boards.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.riot-os.org/pipermail/users/attachments/20160128/be7d3eed/attachment-0001.html>

More information about the users mailing list