Carsten, Might I suggest an additional design objective for the document: if a code subset or function can be identified that takes up significant code space but might be implemented in hardware, then that might provide motivation for a controller vendor to implement such a function in hardware and free up code space since that controller might then have a competitive advantage. For such a thing to happen, it would be important to define bounds, like single or a limited number of sockets and/or use of only a subset of TCP/IP protocols.
Regards, Robin Uyeshiro Disclosure: I have my name (along with many others) on a few hardware implementation patents of TCP/IP from work in a former life. I have no connection with any hardware vendors currently. I receive no income from the patents except reimbursements for minor administrative activities. -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Carsten Bormann Sent: Tuesday, January 24, 2012 2:30 AM To: [email protected] Subject: [Lwip] Version -01 of LWIG guidance draft Happy new year (both Gregorian and Chinese)! Slightly late for the occasion, I have submitted a new version of the LWIG guidance draft. Please enjoy at: http://tools.ietf.org/html/draft-bormann-lwig-guidance-01 If you just want to see what's new: http://tools.ietf.org/rfcdiff?url2=draft-bormann-lwig-guidance-01 I want to have at least one more iteration before March, so we have a good document to discuss in Paris. If you have additional material, or comments on the current draft, now would be a good time to send it to the list (or to me). Grüße, Carsten _______________________________________________ Lwip mailing list [email protected] https://www.ietf.org/mailman/listinfo/lwip _______________________________________________ Lwip mailing list [email protected] https://www.ietf.org/mailman/listinfo/lwip
