[prev in list] [next in list] [prev in thread] [next in thread] 

List:       ipfilter
Subject:    RE: A question about frpr_routing6 function in IPF4.1.22
From:       <Speedo.Chen () zyxel ! com ! tw>
Date:       2007-05-24 8:23:39
Message-ID: 8D5C6A676322304CA638A20DA6DEC6E0F99AA4 () ZyTWBE03 ! ZyXEL ! com
[Download RAW message or body]


Thank you, Darren.

But I am still feeling puzzled here.

IPF in FreeBSD 6.0 (RELENG_6_0)	http://fxr.watson.org/fxr/source//contrib/ipfilter/netinet/fil.c?v=RELENG60
 Line 557-Lin604, IPF does "Nasty extension header length" check for [ routing \
extension header ]. But IPF4.1.22 (line 620) does the same check for [ the extension \
header ] right after routing extension header.  I have no idea why do so since the \
packet may even has no other extension header after routing extension header

Furthermore, if IPF wants to do "Nasty extension header length" check for [ routing \
extension header ].  In RFC2460 page 12, It is ok if there is only one address in \
router extension header,  which means that it is ok if min length of routing \
                extension header (8+16=24)   shorter than 40 bytes(ip6_hdr)
Ex: Take a look of the Figure 2-6 in IPv6 Essentials, Second Edition [O'Reilly]. The \
routing header length has only 24 bytes in trace file


Could someone tell me why needs to check (shift < sizeof(struct ip6_hdr))? 
Or where the related definition of RFC is? Thx a lot.


Thanks again for the quick response to my initial message.


-Y.T.

-----Original Message-----
From: Darren Reed [mailto:darrenr@reed.wattle.id.au] 
Sent: Thursday, May 24, 2007 2:47 PM
To: Yenting Chen -³¯«Û§Ê
Cc: ipfilter@coombs.anu.edu.au
Subject: Re: A question about frpr_routing6 function in IPF4.1.22

Yenting Chen -³¯«Û§Ê wrote:
> Or is there any RFCs specifics that the next header after routing 
> extension header MUST
> 1. longer than 40 bytes?         // ((shift < sizeof(struct ip6_hdr))
> 2. aligned on four bits?          // ((shift - sizeof(struct
> ip6_hdr)) & 15))
Sorry here for which should be 16 bytes alignment.

Yes, these are both defined in the IPv6 RFC.

Darren


[prev in list] [next in list] [prev in thread] [next in thread]