[Tinyos-devel] potential micaz flash issues

[Janos Sallai]

Philip Levis:
>On Mar 5, 2008, at 10:38 AM, Janos Sallai wrote:
>
>> I have recently learned that newer micaz motes and OEM modules use  
>> a newer version of the at45db, which is not fully backwards  
>> compatible with the old version.
>>
>>
>>
>> From my experiences with the iris mote, which has the new version  
>> (at45db041d), the micaz code would not work with the new chip.  
>> Could anybody verify this by running the flash test apps on a  
>> recently purchased micaz?
>>
>
>Is there any way for software to detect which chip a node has?
>This is going to be pretty hairy, if you have to include both chip  
>implementations and switch between then at runtime. Or are the  
>differences trivial? It would be nice to avoid the 433/900MHz mica2  
>debacle.

Yes, it is possible to programmatically detect which chip the mote has.
I've just skimmed the manuals, and it appears that it is possible to
come up with a single piece of code that works on both revisions, so we
wouldn't need to branch based on the chip ID.

The problem is that with the older version it was possible to check if
the chip is ready by toggling the clock and checking a pin. This feature
was removed from the at45db041d from revision G (after February, 2007).
I think that checking for ready can be done by reading the status
register on all at45db versions.

Before touching the code, could somebody check if this problem exists at
all? 

Phil