Le 2012-12-12 à 17:23, Ran Atkinson <[email protected]> a écrit :
> > On Weds 12th December, Remi Despres wrote, in part: >> Not sure however that the two proposed questions >> are clear enough because today: >> - some IIDs have u = 0, and some have u = 1, >> - some IIDs have g = 0, and some have g = 1. >> The only combination that isn't used is u=g=1. > > While U==G==1 is not *widely deployed* today, My point was only about IIDs of UNICAST addresses, those in scope of RFC 4291. Do you know any deployment with u=g=1 in this context ? RD > it does have a very clear definition, namely: > > (U==1 && G==1) defines a global-scope > multicast identifier. > > This is clear not only from a plain-language > reading of RFC-4291, Section 2.5.1, but also > from the IEEE 802 definition of an EUI-64. > > Further, an IID formed by algorithmically translating > an IEEE 802 MAC Address (e.g. using RFC-4291, > Appendix A), whether using a unicast or multicast > MAC address, is a widely deployed example of an > "interface identifier with universal scope". > > Yours, > > Ran > > > -------------------------------------------------------------------- > IETF IPv6 working group mailing list > [email protected] > Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 > -------------------------------------------------------------------- -------------------------------------------------------------------- IETF IPv6 working group mailing list [email protected] Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
