From owner-freebsd-net  Tue Apr  9 17:18:54 2002
Delivered-To: freebsd-net@freebsd.org
Received: from sheffield.cnchost.com (sheffield.concentric.net [207.155.252.12])
	by hub.freebsd.org (Postfix) with ESMTP id 080DA37B41B
	for <net@freebsd.org>; Tue,  9 Apr 2002 17:18:50 -0700 (PDT)
Received: from bitblocks.com (adsl-209-204-185-216.sonic.net [209.204.185.216])
	by sheffield.cnchost.com
	id UAA13907; Tue, 9 Apr 2002 20:18:47 -0400 (EDT)
	[ConcentricHost SMTP Relay 1.14]
Message-ID: <200204100018.UAA13907@sheffield.cnchost.com>
To: Julian Elischer <julian@elischer.org>
Cc: net@freebsd.org
Subject: Re: IP checksup update 
In-reply-to: Your message of "Tue, 09 Apr 2002 16:02:01 PDT."
             <Pine.BSF.4.21.0204091557570.57350-100000@InterJet.elischer.org> 
Date: Tue, 09 Apr 2002 17:18:47 -0700
From: Bakul Shah <bakul@bitblocks.com>
Sender: owner-freebsd-net@FreeBSD.ORG
Precedence: bulk
List-ID: <freebsd-net.FreeBSD.ORG>
List-Archive: <http://docs.freebsd.org/mail/> (Web Archive)
List-Help: <mailto:majordomo@FreeBSD.ORG?subject=help> (List Instructions)
List-Subscribe: <mailto:majordomo@FreeBSD.ORG?subject=subscribe%20freebsd-net>
List-Unsubscribe: <mailto:majordomo@FreeBSD.ORG?subject=unsubscribe%20freebsd-net>
X-Loop: FreeBSD.org

> to update a checksum when changing a word
> 
>   #define FIXSUM16(c, op, np)                                 \
>     do {                                                      \
>       (c) -= (u_int16_t) ~*((u_int16_t *) (op));              \
>       if ((c) < 0) {                                          \
>         (c) += 0xffff;                                        \
>       }                                                       \
>       (c) -= (u_int16_t)  *((u_int16_t *) (np));              \
>       if ((c) < 0) {                                          \
>         (c) += 0xffff;                                        \
>       }                                                       \
>     } while (0)
>
> The question is:
> apparently there are several "tricks" with this checksum
> and + and - 0. Does anyone know how to test if this macro gets it
> right?

Looks like a straight forward implementation of rfc1624 eqn 4.
"if (c < 0) c += 0xffff;" is needed to turn two's complement
subtraction to one's complement.

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-net" in the body of the message