From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: PROBLEM: pppol2tp over pppoe NULL pointer dereference Date: Tue, 08 Nov 2011 14:00:00 -0500 (EST) Message-ID: <20111108.140000.543570939575214751.davem@davemloft.net> References: <1320191893.4728.13.camel@edumazet-laptop> <20111104.222851.1376278499619626232.davem@davemloft.net> <1320478829.16609.15.camel@edumazet-laptop> Mime-Version: 1.0 Content-Type: Text/Plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: spiked.yar@gmail.com, netdev@vger.kernel.org To: eric.dumazet@gmail.com Return-path: Received: from shards.monkeyblade.net ([198.137.202.13]:46617 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753215Ab1KHTAG convert rfc822-to-8bit (ORCPT ); Tue, 8 Nov 2011 14:00:06 -0500 In-Reply-To: <1320478829.16609.15.camel@edumazet-laptop> Sender: netdev-owner@vger.kernel.org List-ID: =46rom: Eric Dumazet Date: Sat, 05 Nov 2011 08:40:29 +0100 > Le vendredi 04 novembre 2011 =E0 22:28 -0400, David Miller a =E9crit = : >> From: Eric Dumazet >> Date: Wed, 02 Nov 2011 00:58:13 +0100 >>=20 >> > Please try following patch, thanks ! >> >=20 >> > [PATCH] l2tp: handle fragmented skbs in receive path >> >=20 >> > Modern drivers provide skb with fragments, and L2TP doesnt properl= y >> > handles them. >> >=20 >> > Some bad frames can also trigger panics because of insufficent che= cks. >> >=20 >> > Reported-by: Misha Labjuk >> > Signed-off-by: Eric Dumazet >>=20 >> I'm still waiting for testing results of this patch. >=20 > Of course. >=20 > If you prefer, I can submit a smaller patch for the obvious bug first= , > and I can respin the thing when net-next reopens. >=20 > [PATCH] l2tp: fix l2tp_udp_recv_core() >=20 > pskb_may_pull() can change skb->data, so we have to load ptr/optr at = the > right place. >=20 > Signed-off-by: Eric Dumazet Yes, this is easier to digest right now. Applied.