https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=4014
--- Comment #27 from Stephen Fisher <steve@xxxxxxxxxxxxxxxxxx> 2010-09-16 15:47:40 MDT ---
(In reply to comment #25)
> So if ICMP is actually an integral part of IP as they say, and Appendix B of
> RFC 791 dictates Big-Endian for IP, then it stands to reason that ICMP MUST use
> Big-Endian as well, would it not? Well that's my interpretation.
This is worth digging further into to see if there was a discussion before this
was changed in NetBSD and if it has ever come up in other BSD flavors.
> > NetBSD's rationale of changing this from getpid() in CVS revision 1.76 of
> > ping.c was "do not disclose endian" back in 2004.
>
> But if they had instead used, "ident = htons(getpid() & 0xFFFF);", then
> endian-ness would not have been disclosed.
True. According to their cvs log, the initial change made it
"htons(arc4random())" and then another person took the htons() out since they
didn't feel it was necessary to convert the byte order of a random value :).
--
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
