>>>>> On Tue, 18 Sep 2001 10:47:55 +0200 (CEST),
>>>>> Erik Nordmark <[EMAIL PROTECTED]> said:
> Not very portable.
> Once I wacked away the syntax problems (assuming sin6_len plus a
> bunch of stuff due to Solaris not having all of rfc2292bis)
> I realized that accept.c assumes that for SOCK_RAW ai_protocol is passed
> through from hints to the getaddrinfo result.
> The Solaris getaddrinfo doesn't do this.
Okay, although I'm not sure if you still have an interest on the test
kit, I modified the code as portable as possible according to your
comments. Also, since at least two people tried to use the kit (the
other one said so in a private message), I made a tar+gz ball of the
kit at
ftp://ftp.kame.net/pub/misc/advapitest.tgz
for convenience.
> I don't know if this is a Solaris
> problem since I haven't check what the standards text for getaddrinfo()
> says about SOCK_RAW.
I've checked rfc2553bis-03. The spec seems ambiguous on this point...
> In any case I verified that the socket call doesn't fail on Solaris for
> (PF_INET6, SOCK_RAW, exthdr number).
Hmm, so we have at least two different implementations.
JINMEI, Tatuya
Communication Platform Lab.
Corporate R&D Center, Toshiba Corp.
[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]
--------------------------------------------------------------------
