From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bernard Pidoux Subject: Re: [AX25] kernel panic Date: Tue, 01 Apr 2008 22:37:02 +0200 Message-ID: <47F29CEE.10307@free.fr> References: <47E68CA1.2040904@free.fr> <20080324205104.GA3721@ami.dom.local> <47E8FCE2.6000308@free.fr> <20080326183538.GA14266@ami.dom.local> <20080328120721.GA11060@ami.dom.local> <47EE3040.3050208@free.fr> <20080329122452.GB3407@ami.dom.local> <47EE38BA.9090500@free.fr> <20080329130451.GC3407@ami.dom.local> <47EF6479.7030702@free.fr> <20080330115958.GA4975@ami.dom.local> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit Cc: Jarek Poplawski , Linux Netdev List , Ralf Baechle DL5RB To: Jarek Poplawski Return-path: Received: from smtp5-g19.free.fr ([212.27.42.35]:43165 "EHLO smtp5-g19.free.fr" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756732AbYDBABY (ORCPT ); Tue, 1 Apr 2008 20:01:24 -0400 In-Reply-To: <20080330115958.GA4975@ami.dom.local> Sender: netdev-owner@vger.kernel.org List-ID: Jarek Poplawski wrote: > On Sun, Mar 30, 2008 at 11:59:21AM +0200, Bernard Pidoux wrote: > ... >> I would not be surprised if ROSE was involved as, remember, I first >> found the bug when I closed an AX25 session established through a level >> 3 ROSE network (FPAC) via AXUDP link (ax25ipd). As I said in my first >> bug report, the bug is not triggered when connecting locally or via AX25 >> call through LAN. Thus ROSE level is involved. >> >> However, I guess that lower layer (AX25) should stand a possible error >> in the upper layer protocol (ROSE). > > I think ROSE breaks some general rules here: bugs trigger in sock.c > code when ROSE packets are destroyed after closing its sockets, but > without proper information on this. > >> ROSE is actually an AX25 packet routing protocol with only three byte >> offset added to AX25 frames (encapsulation). Of course it has its own >> bugs... > > Probably... BTW, my today's patch adds some locking, so there is always > some risk of a lockup or some new lockdep warning. And if accidentally > it works, after testing with these debugging patches there would be > needed later a test of this #5 alone, as well. > > Thanks, > Jarek P. > > I did what you suggested and reversed your patches #2,#3 and #4 to verify that there was no system incompatibilities with the new locks you introduced into rose_release(). Patch #1 has also been removed on another machine running a 64 bit Core2 duo CPU. Three Linux boxes have been are running for more than 24 hours now with only patch #5. They handled a lot of AX25 connections with no problems. Thus, we can reasonably expect that things are going to stay stable. If you want to commit patch#5 I will sustain it very much. I hope that it will be accepted quickly in order to make a coherent system with AX25 patches already applied or not yet applied (if any). Thank you very much Jarek for this very nice job. Best regards, Bernard Pidoux, F6BVP