From owner-freebsd-net@FreeBSD.ORG  Thu Jul 10 21:06:33 2008
Return-Path: <owner-freebsd-net@FreeBSD.ORG>
Delivered-To: net@FreeBSD.org
Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34])
	by hub.freebsd.org (Postfix) with ESMTP id 002171065672;
	Thu, 10 Jul 2008 21:06:32 +0000 (UTC)
	(envelope-from rwatson@FreeBSD.org)
Received: from cyrus.watson.org (cyrus.watson.org [209.31.154.42])
	by mx1.freebsd.org (Postfix) with ESMTP id CF22E8FC1F;
	Thu, 10 Jul 2008 21:06:32 +0000 (UTC)
	(envelope-from rwatson@FreeBSD.org)
Received: from fledge.watson.org (fledge.watson.org [209.31.154.41])
	by cyrus.watson.org (Postfix) with ESMTP id 647D446B49;
	Thu, 10 Jul 2008 17:06:32 -0400 (EDT)
Date: Thu, 10 Jul 2008 22:06:32 +0100 (BST)
From: Robert Watson <rwatson@FreeBSD.org>
X-X-Sender: robert@fledge.watson.org
To: gnn@FreeBSD.org
In-Reply-To: <m2prpljz4k.wl%gnn@neville-neil.com>
Message-ID: <20080710220201.K34050@fledge.watson.org>
References: <m2fxqilpgg.wl%gnn@neville-neil.com>
	<20080710114028.T34050@fledge.watson.org>
	<m2prpljz4k.wl%gnn@neville-neil.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed
Cc: net@FreeBSD.org
Subject: Re: What's the deal with hardware checksum and
	net.inet.udp.checksum?
X-BeenThere: freebsd-net@freebsd.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Networking and TCP/IP with FreeBSD <freebsd-net.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-net>,
	<mailto:freebsd-net-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-net>
List-Post: <mailto:freebsd-net@freebsd.org>
List-Help: <mailto:freebsd-net-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-net>,
	<mailto:freebsd-net-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Thu, 10 Jul 2008 21:06:33 -0000


On Thu, 10 Jul 2008, gnn@FreeBSD.org wrote:

> If the sysctl it turned off on the transmitter then the receiving machine 
> sees UDP checksums of 0.

Right.  If you disable UDP checksumming, we don't generate checksums (hardware 
or software) in udp_output():

         /*
          * Set up checksum and output datagram.
          */
         if (udp_cksum) {
                 if (inp->inp_flags & INP_ONESBCAST)
                         faddr.s_addr = INADDR_BROADCAST;
                 ui->ui_sum = in_pseudo(ui->ui_src.s_addr, faddr.s_addr,
                     htons((u_short)len + sizeof(struct udphdr) + IPPROTO_UDP));
                 m->m_pkthdr.csum_flags = CSUM_UDP;
                 m->m_pkthdr.csum_data = offsetof(struct udphdr, uh_sum);
         } else
                 ui->ui_sum = 0;

You can disable hardware checksums using the -txcsum flag on ifconfig for each 
interface -- once the above-generated mbuf header gets to the IP layer and the 
route out an interface is available, we on-demand generate checksum in 
software if hardware checksums aren't available or are administratively 
disabled.  Vis ip_output():

         m->m_pkthdr.csum_flags |= CSUM_IP;
         sw_csum = m->m_pkthdr.csum_flags & ~ifp->if_hwassist;
         if (sw_csum & CSUM_DELAY_DATA) {
                 in_delayed_cksum(m);
                 sw_csum &= ~CSUM_DELAY_DATA;
         }
         m->m_pkthdr.csum_flags &= ifp->if_hwassist;

It's possible to imagine adding a global sysctl that has slightly different 
policy implications, such as globally disabling hardware checksums, or not 
generating full checksums if the interface doesn't support hardware checksums 
rather than generating them.

Robert N M Watson
Computer Laboratory
University of Cambridge