Subject: Re: proposed kernel change for IP_HDRINCL
To: Bill Fenner <>
From: Matthias Urlichs <>
List: netbsd-bugs
Date: 10/20/1996 23:28:53

Bill Fenner wrote:
>In message <> you write:
>>Currently, when using setsockopt() with the IP_HDRINCL flag, most
>>fields of the IP header are passed into the kernel in network byte
>>order, except for the ip_off and ip_len fields.
>I think that there should be a standard macro defined in <netinet/ip.h>
>that says whether or not this change is in the kernel, so that people
>don't have to send themselves a packet to see whether the kernel byte
>swaps or not.

>Do you have any desire to do anything about raw_input as well?  Various
>fields are messed with on the way in, too.
IMHO, a raw packet should be a _raw_ packet. That means that the kernel
shouldn't fiddle with _anything_.

FWIW, Linux does things the "really raw" way, too; thus, programs which
rely on these things already have the required patches, these days.

Airplanes are interesting toys but of no military value.
                       -- Marechal Ferdinand Foch, Professor of Strategy,
                          Ecole Superieure de Guerre
Matthias Urlichs         \  noris network GmbH  /  Xlink-POP Nürnberg 
Schleiermacherstraße 12   \   Linux+Internet   /   EMail:
90491 Nürnberg (Germany)   \    Consulting+Programming+Networking+etc'ing
   PGP: 1024/4F578875   1B 89 E2 1C 43 EA 80 44  15 D2 29 CF C6 C7 E0 DE
       Click here.    42