[EMAIL PROTECTED] wrote:
> 
> >=====================================================
> >Description:
> >remove comment which may not be correct; i've experienced unaligned header
> >on bpf taps with netbsd 1.5Y.
> >Modified files:
> >       File: tcpdump/print-ip.c; Revision: 1.102;
> >       Date: 2001/11/16 02:17:36; Author: itojun; Lines: (+1 -3)
> >       File: tcpdump/print-ip6.c; Revision: 1.21;
> >       Date: 2001/11/16 02:17:36; Author: itojun; Lines: (+1 -3)
> >-------------------------------
> >
> >Why is this a bug in tcpdump and not a bug in NetBSD 1.5Y?
> 
>         i'm not saying it is a bug in tcpdump.  is it mandatory for BPF to
>         return an aligned packet?  if so, it is a bug in netbsd.

What exactly do you mean by an "aligned" packet? If you mean the sort of
obvious superficial business that the link header starts on a 2- or 4-byte
boundary, that still won't help you in the general case if you're hoping to
access unaligned data structures in the packet payload. Some link headers, as
I'm sure you're aware, have pretty strange lengths -- look at the FDDI header
structure for example. FDDI llc encapsulation results in a 13-octet header, so
the alignment of the link header won't help you at the IP layer or above.

Maybe I don't understand the issue well enough...?

-- 
Jefferson Ogata <[EMAIL PROTECTED]>
NOAA Computer Incident Response Team (N-CIRT) <[EMAIL PROTECTED]>
-
This is the TCPDUMP workers list. It is archived at
http://www.tcpdump.org/lists/workers/index.html
To unsubscribe use mailto:[EMAIL PROTECTED]?body=unsubscribe

Reply via email to