From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755005Ab3CSM6O (ORCPT ); Tue, 19 Mar 2013 08:58:14 -0400 Received: from mx1.redhat.com ([209.132.183.28]:11515 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754675Ab3CSM6N (ORCPT ); Tue, 19 Mar 2013 08:58:13 -0400 Message-ID: <514860DC.9020404@redhat.com> Date: Tue, 19 Mar 2013 13:58:04 +0100 From: Daniel Borkmann User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/17.0 Thunderbird/17.0 MIME-Version: 1.0 To: Eric Dumazet CC: Jason Wang , David Miller , netdev@vger.kernel.org, linux-kernel@vger.kernel.org, mst@redhat.com, edumazet@google.com Subject: Re: [PATCH net-next 2/2] net: reset transport header if it was not set before transmission References: <1363333305-54398-1-git-send-email-jasowang@redhat.com> <1363333305-54398-2-git-send-email-jasowang@redhat.com> <1363399851.29475.52.camel@edumazet-glaptop> <20130317.121318.1869489921021513474.davem@davemloft.net> <51482F55.8020900@redhat.com> <1363695234.21184.36.camel@edumazet-glaptop> In-Reply-To: <1363695234.21184.36.camel@edumazet-glaptop> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 03/19/2013 01:13 PM, Eric Dumazet wrote: > On Tue, 2013-03-19 at 17:26 +0800, Jason Wang wrote: >> On 03/18/2013 12:13 AM, David Miller wrote: >>> From: Eric Dumazet >>> Date: Fri, 15 Mar 2013 19:10:51 -0700 >>> >>>> Any way we can avoid adding this to fast path, for people not using >>>> macvtap and ixgbe ? >>> Likewise I'd rather see macvtap be responsible for fixing this up by >>> setting the transport header properly, and therfore sending well >>> formed packets to the rest of the stack. >> >> Ok, haven't checked all other possibility but looks like packet needs to >> be fixed also. > > Daniel, could you post your patches if ready ? Yes, will post them in a couple of minutes. > Jason, I believe you could reuse existing flow dissector once Daniel > patches are in.