From owner-freebsd-net  Wed Dec 29  1:26:40 1999
Delivered-To: freebsd-net@freebsd.org
Received: from verdi.nethelp.no (verdi.nethelp.no [158.36.41.162])
	by hub.freebsd.org (Postfix) with SMTP id 0ECCE15134
	for <freebsd-net@FreeBSD.ORG>; Wed, 29 Dec 1999 01:26:34 -0800 (PST)
	(envelope-from sthaug@nethelp.no)
Received: (qmail 79058 invoked by uid 1001); 29 Dec 1999 09:26:31 +0000 (GMT)
To: justin@apple.com
Cc: fgont@softhome.net, freebsd-net@FreeBSD.ORG
Subject: Re: Two many CRCs?
From: sthaug@nethelp.no
In-Reply-To: Your message of "Tue, 28 Dec 1999 18:13:16 -0800"
References: <199912290213.SAA01337@walker3.apple.com>
X-Mailer: Mew version 1.05+ on Emacs 19.34.2
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Date: Wed, 29 Dec 1999 10:26:31 +0100
Message-ID: <79056.946459591@verdi.nethelp.no>
Sender: owner-freebsd-net@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.org

> So what's really duplicated is the CRC at the media layer and those  
> at the protocol layers.

You can often argue that this isn't duplication - the link layer CRC
only covers one hop, while the TCP checksum is end to end.

Of course, if you have for instance ATM AAL5, you also have an end to
end checksum there. However, there's been research (Craig Partridge
& al) which suggests that the AAL5 CRC and the TCP checksum catch
*different* classes of errors, and that it's really useful to have
both.

Steinar Haug, Nethelp consulting, sthaug@nethelp.no


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