Washam, Thank you for the detailed reaction. More below.
Le 20 août 2011 à 10:44, Washam Fan a écrit : > Hi, > > I support the seperation of address mapping from tunneled or > translated 4rd mechanisms. Excellent. > I have some comments as below. > 1. Is there any reason you changed 4rd BR to 4rd AFTR? Natural question indeed. This is because the scope is larger than just tunneling mechanisms. Thus, it won't be necessary, in in comparison documents, to play with various acronyms. CPE and AFTR seem to be the most generic ones. > 2. Is 4rd IID for differentiate native ipv6 packets and translated > ipv6 packets? In other words, tunneled 4rd don't need that. Tunnel solutions also need that a CPE receiving an IPv6 packet can differentiate the 4rd packet from any IPv6 packets destined to intra-site hosts. By having the 4rd specific IID, it is achieved without depending on a specific protocol ID for encapsulation. > 3. For the ICMPv4 containing the copy of error packets, i think > BR/AFTR or CPE can do forwarding based on the copy of error packets > unless the copy didn't cover the complete TCP/UDP header. For ICMP > echo request, you should say the ID field should be within the port > set range. All these topics are in section 5.1. It was supposed to be implicit by its being "processed as though it would be a CPE port. It can be made more explicit in a later version. > 4. why omit ipv6 suffix in figure 1? The idea is that the suffix is _after_ the CPE IPv6 suffix, and therefore doesn't need to appear in Fig 1. > 5.don't get why you pad 0 to ipv6 suffix in figure 2? Cases are possible where the previous fields don't reach 64 bits. An example could be added in a later version. > 6. in the paragraphy just below figure 2, shoud you replace 64 with X > in the L=min(14, 64 ...? BTW, it seems to me, the PSID to port sets > mapping is different from draft-murakami-softwire-4rd-00. Yes. This is a consequence of the additional objective of UPnP friendliness. Xiaohong could probably tell more if you need it. > 7. in example a in section 6, how come a mapping rule responding to 2 > cpe ipv6 prefix length? cpe will get confused when it did forwarding. CPE prefix lengths may be different. (It is the Domain prefix length that is given in the rule.) This is key to be able to assign port sets of different sizes (including 64K) to different customers (at least when the CPE-cascade option isn't necessary). Regards, RD _______________________________________________ Softwires mailing list [email protected] https://www.ietf.org/mailman/listinfo/softwires
