From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ben Greear Subject: Re: 3.7.3+: Bad paging request in ip_rcv_finish while running NFS traffic. Date: Wed, 23 Jan 2013 20:26:28 -0800 Message-ID: <5100B7F4.2050505@candelatech.com> References: <50FDADF4.3060601@candelatech.com> <50FDDE35.7070806@candelatech.com> <1358829606.3464.3151.camel@edumazet-glaptop> <50FE2A57.3040804@candelatech.com> <50FEC796.5090404@candelatech.com> <1358875020.3464.4006.camel@edumazet-glaptop> <1358875607.3464.4020.camel@edumazet-glaptop> <50FF102F.2050008@candelatech.com> <50FF4BC9.1060206@candelatech.com> <5100785D.8040101@candelatech.com> <1358985688.12374.1247.camel@edumazet-glaptop> <51007CA8.2050105@candelatech.com> <1358987031.12374.1276.camel@edumazet-glaptop> <51008294.2010201@candelatech.com> <1358988358.12374.1303.camel@edumazet-glaptop> <51008598.4000603@candelatech.com> <1358989242.12374.1323.camel@edumazet-glaptop> <51008900.2010803@candelatech.com> <1358989843.12374.1334.camel@edumazet-glaptop> <1358991906.1 2374.1356.camel@edumazet-glaptop> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: netdev To: Eric Dumazet Return-path: Received: from mail.candelatech.com ([208.74.158.172]:48816 "EHLO ns3.lanforge.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752884Ab3AXE0g (ORCPT ); Wed, 23 Jan 2013 23:26:36 -0500 In-Reply-To: <1358991906.12374.1356.camel@edumazet-glaptop> Sender: netdev-owner@vger.kernel.org List-ID: On 01/23/2013 05:45 PM, Eric Dumazet wrote: > On Wed, 2013-01-23 at 17:10 -0800, Eric Dumazet wrote: > >> Excellent, thats the bug. >> >> I'll send a fix asap, thanks ! > > loopback device doesnt have a qdisc, and unsets IFF_XMIT_DST_RELEASE > > Its hard to believe such an old bug never hit us in the past. > > Probably because most of the time, the packet given to netif_rx() is > immediately processed (and loopback device is hard wired ?) I'll test this in the morning...one thing, maybe it's worth adding a WARN_ON in netif_rx() in case there are other bugs of this nature? The dst objects are also garbage collected, so they have a fairly long valid-ish lifetime after they are logically freed. That might be part of why it was so hard to reproduce the bug... Thanks again for all the help on this bug! Hopefully it will now be well and truly squashed. Ben -- Ben Greear Candela Technologies Inc http://www.candelatech.com