From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alex Subject: Re: Bug 75571 Date: Thu, 05 Jun 2014 12:22:12 +0700 Message-ID: <1401945732.29664.6.camel@anpws> References: <1401085231.1775.9.camel@anpws> <1401870692.29664.1.camel@anpws> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org To: Cong Wang Return-path: Received: from mxnat3.880.ru ([194.8.85.227]:6439 "EHLO stk-gw0.880.ru" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1750876AbaFEFV7 (ORCPT ); Thu, 5 Jun 2014 01:21:59 -0400 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: =D0=92 =D0=A1=D1=80, 04/06/2014 =D0=B2 17:27 -0700, Cong Wang =D0=BF=D0= =B8=D1=88=D0=B5=D1=82: > Thanks for testing! >=20 > Thinking a bit more, the question is why > skb_push(skb->dev->hard_header_len) is not correct here? It looks > like we don't calculate tunnel->dev->hard_header_len correctly. But > ipip_tunnel_bind_dev() should do the right thing less your setup is n= ot > correct, in that case your ipip tunnel should not receive any packet. > We should be safe. >=20 > Which is the interface routing to 192.168.18.17 on your system? Which > driver does it use? Its a stock e1000e from kernel with no special options. No jumbo frames= , etc. lspci says: Ethernet controller: Intel Corporation 82572EI Gigabit Ethernet Controller (Copper) (rev 06) IPIP tunnel runs on top of this interface.