>> I think I have emailed on this more than twice, for me it seems that
>> glibc people misunderstood what is meant by "socklen_t". I *strongly*
>> object to make this unnecessary change (size_t -> socklen_t for
>> hostnamelen/servnamelen/etc).
>Yes I know, I know! But they won't change unless XNS changes... (sigh)
>XNS 5.2: <http://www.opengroup.org/onlinepubs/009619199/>
I'm confused, which people are you trying to convince?
Are you trying to change 2553 to use socklen_t for all the arguments,
or are you trying to correct glibc? If the latter, the mailing list
is a wrong place to ask.
Some description on current situation should help you...
The XNS-RFC incompatibility will be solved when 2553bis document
is finalized. For the current specification, RFC2553 looks
more correct (RFC2553 is more closer to 2553bis). For updated
specification, words in 2553bis will be directly put into XNS so
they will become exactly the same. (correct me if my memory is wrong)
As there is no change between RFC2553 and 2553bis on this particular
"getnameinfo argument" issue, RFC2553 is the correct specification
to look at for this particular issue.
Please try to convince glibc guys, quoting the above if necessary.
itojun
--------------------------------------------------------------------
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]
--------------------------------------------------------------------
