[net2-wg] Fwd: lis and ctp

[Omprakash Gnawali]

Could be due to documentation...

- om_p

On Wed, Aug 5, 2009 at 7:45 PM, Philip Levis<pal at cs.stanford.edu> wrote:
> CollectionSenderC is too complex?
>
> Phil
>
> On Aug 5, 2009, at 19:31, Omprakash Gnawali <gnawali at usc.edu> wrote:
>
>> ---------- Forwarded message ----------
>> From: Roy Shea <royshea at gmail.com>
>> Date: Wed, Aug 5, 2009 at 9:50 AM
>> Subject: Re: lis and ctp
>> To: Omprakash Gnawali <gnawali at usc.edu>
>>
>>
>> Howdy,
>>
>>> I noticed that you use ctp to haul lis logs. In general, what is your
>>> experience using ctp? How can we improve it?
>>
>> Overall, I like it.  CTP works and is the correct fit for the job.  My
>> notes are:
>>
>> Does not play nice with Avrora.  Perhaps only the CC24200 radio model?
>> I can't remember precisely.  This is not CTP's fault, but it is a bummer
>> that users of CTP should be aware of.
>>
>> I think there should be a higher level interface.  And I may have missed
>> that :-)  But in the common case, the user probably wants to send data
>> (AMSend.send) and know when they can send data again (AMSend.sendDone).
>> That's it.  I would guess that many users of CTP recreate this, so it
>> would be nice to provide a default configuration of CTP that provides
>> this behavior (perhaps configure queue size and end gateway ID during
>> wiring).  Users needing more advanced features (snooping, not simply
>> dumping to UART at the collection point) could always drop back down and
>> do their own thing.
>>
>> Peace,
>> -Roy
>>
>> _______________________________________________
>> net2-wg mailing list
>> net2-wg at millennium.berkeley.edu
>> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/net2-wg
>