From owner-freebsd-net@FreeBSD.ORG Wed Jun 6 19:13:09 2012 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 67F7D1065673 for ; Wed, 6 Jun 2012 19:13:09 +0000 (UTC) (envelope-from philipp_subx@redfish-solutions.com) Received: from mail.redfish-solutions.com (mail.redfish-solutions.com [66.232.79.143]) by mx1.freebsd.org (Postfix) with ESMTP id 351FC8FC1A for ; Wed, 6 Jun 2012 19:13:09 +0000 (UTC) Received: from macbook.redfish-solutions.com ([192.168.1.17]) (authenticated bits=0) by mail.redfish-solutions.com (8.14.5/8.14.5) with ESMTP id q56JD21M001849 (version=TLSv1/SSLv3 cipher=DHE-RSA-CAMELLIA256-SHA bits=256 verify=NO) for ; Wed, 6 Jun 2012 13:13:03 -0600 Message-ID: <4FCFABBE.7030008@redfish-solutions.com> Date: Wed, 06 Jun 2012 13:13:02 -0600 From: Philip Prindeville User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:12.0) Gecko/20120428 Thunderbird/12.0.1 MIME-Version: 1.0 To: freebsd-net@freebsd.org References: <1D03F00E-2777-4B0B-8E1C-860EA115B6AF@lurchi.franken.de> <7731AEE8-D1BF-4297-8F75-454149D3E303@lurchi.franken.de> <5A4A73CF-95FB-4B33-9652-2E4FD517CF5F@lurchi.franken.de> <5399FCC6-BDC4-47DD-9948-ADD16264EE46@lurchi.franken.de> <8D7E9322-0973-4657-B3FE-8A480C0E0304@lurchi.franken.de> In-Reply-To: <8D7E9322-0973-4657-B3FE-8A480C0E0304@lurchi.franken.de> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Scanned-By: MIMEDefang 2.73 on 192.168.1.3 Subject: Re: IP_RECVTOS X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 06 Jun 2012 19:13:09 -0000 On 6/6/12 11:54 AM, Michael Tuexen wrote: > On Jun 6, 2012, at 7:17 PM, Adrian Chadd wrote: > >> Hi, >> >> For TCP, I've seen the network layer change things (eg setting bits on >> incoming traffic to mark which interface it came in on), so you can't >> guarantee the outbound ToS == inbound ToS. > I've not said that inbound == outbound... I just wanted to make clear > that there is a socket option of setting the TOS byte for outgoing > traffic. The TCP stack might overwrite the ECN bits... > > I need the IP_RECVTOS functionality for implementing a transport stack supporting > ECN in userland, actually running the SCTP kernel sources (with some glue code) > in userland on top of UDP... > > Best regards > Michael What about UDT? Can you patch that as well? For certain services (like backups or AFP), it might be useful for a server to be able to detect a QoS-friendly client and reciprocate in kind. -Philip