In your previous mail you wrote:
At 3:41 PM +0100 3/16/01, Francis Dupont wrote:
>=> don't worry, there is no future use of those bits (in the past
>there were some proposals which were rejected for many strong reasons).
>But please don't apply this advice to site-local addresses.
My advice is the same for link-locals and site-locals: use a full 128-bit
compare to recognize your own addresses. That will continue to work,
whether or not sites end up using non-zero values in the high-order
part of their site-locals.
=> I believe the initial issue was a bit different: check or not check
that the must-be-zero bits are zero or not. Usually you should not
check them (ie. only do what you advice) but if you suspect some bugs
you may add a check in order to track them, in such a case if MBZ bits
become no more MBZ you'll be in trouble. My advice was this should not
happen for link-local addresses (so a check may safely be added and
forgotten) but not for site-local addresses (so if a check is added
then it must not be forgotten, usually it will be forgotten so it
should not be added)...
Regards
[EMAIL PROTECTED]
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
