[mc1322x] "atoi" oversize

[Jon Smirl]

On Sat, Jul 3, 2010 at 5:02 PM, Joakim Eriksson <joakime at sics.se> wrote:
> Mariano Alvira skrev 2010-07-03 16:17:
>>
>> On Fri, Jul 02, 2010 at 09:55:54PM -0700, Andrew Pullin wrote:
>>>
>>>  On 7/2/2010 11:04 AM, Mariano Alvira wrote:
>>>
>>> Bleh! No building! Well, I understand the motivation behind it,
>>> especially when working with new chips, and odd chips (like this whole
>>> reverse engineered ROM setup ).
>>
>> Technically I reversed their library only --- the ROM is almost
>> impossible to reverse; it's much harder since the library still had
>> the debugging symbols and wasn't stripped so the functions all had
>> their names still etc...
>>
>> The only ROM calls I use are to interface to the flash and that's
>> becuase I'm lazy.
>>
>>> While we're on the topic, I'm actually super curious what the problem
>>> was with start.S, as a learning exercise. Presumably you are talking
>>> about something that caused the newer versions of the CS toolchain to
>>> segfault when try to assemble?
>>
>> Yes, that one. Here is the patch:
>>
>>
>> http://git.devl.org/?p=malvira/libmc1322x.git;a=commitdiff;h=36361e1eb0eba99c8ab7807fa2f977c80a3be05c;hp=4035dc44f3db1b90cdb0cf51e6ce3d90204294d9
>>
>> I have no clue why it segfaults on the original line (although I
>> understand, sometimes my brain segfaults on the original line). I came
>> up with that first line empirically to make the assembled instruction
>> right. If you squint hard enough you can convince yourself that it
>> makes some kind of sense.
>>
>>> And can you say anything about the new hardware? Cortex M3?
>>
>> Sure! I am planning to do a stm32w board. ST is supporting Contiki and
>> the radio library links directly with GCC --- so no reversing should
>> be necessary (unless there code doesn't work well). The stm32W should
>> be available Q4.
>>
>> I'm planning to do rev2 of a USB stick version of the econotag and
>> will be getting FCC cert. for that. Also a bunch of new modules. I'm
>> planning to do the Xbee and Mbed footprint, a castellated small
>> module, a castellated bigger module that brings out all of the I/O
>> (all 64 pins), and a first rev of a mc13224v with PA.
>>
>> Anyone have any thing on there hardware wishlist?
>
> Any plans for adding serial-ID chip for generating
> MAC address that can be used for IPv6 address autoconfig?

Mar, does Redwire own an 802.15.4 MAC address block? They are
different than an Ethernet block since they are 8 instead 6 digits.

If so you could put them into the reserved area of the flash.


>
> Best regards,
> -- Joakim Eriksson, SICS
>
>
>> -Mar.
>>
>>
>>
>> _______________________________________________
>> mc1322x mailing list
>> mc1322x at devl.org
>> http://devl.org/cgi-bin/mailman/listinfo/mc1322x
>
>
> _______________________________________________
> mc1322x mailing list
> mc1322x at devl.org
> http://devl.org/cgi-bin/mailman/listinfo/mc1322x
>



-- 
Jon Smirl
jonsmirl at gmail.com