I'd expect duplicate the codes just like we did for RFC 6282. This way, if a 6LoRH moved to page 1, we do ,not need to switch back to page 0. But is there a use case for it?
Pascal > -----Original Message----- > From: Michael Richardson <[email protected]> > Sent: mardi 21 mai 2019 15:50 > To: Pascal Thubert (pthubert) <[email protected]> > Cc: [email protected]; Laurent Toutain <[email protected]> > Subject: Re: [6lo] recoverable fragments: use page 1 > > > Pascal Thubert (pthubert) <[email protected]> wrote: > > During the review by Laurent, a question came up on whether we could > use page > > 1 for fragments. > > This implies to me that we'd allocate code points for fragments from page 1. > > > At the moment the fragment is always before the 6lo routing > > header so there is no point. But say that we source route the fragments > end > > to end with a 6loRH, e.g., for projected DAO. Then supporting page 1 as > well > > would save a page switch. > > What exactly does it mean to support page 1? > Duplicate code points? > Starting in page 1 mode for fragments? > Allocating all fragment code points from page 1? > > Maybe we should have a document about source routing fragments end to > end for projected DAO, and then we can see the full picture. > > -- > Michael Richardson <[email protected]>, Sandelman Software Works - > = IPv6 IoT consulting =- > > _______________________________________________ 6lo mailing list [email protected] https://www.ietf.org/mailman/listinfo/6lo
