From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from sc8-sf-mx2-b.sourceforge.net ([10.3.1.12] helo=sc8-sf-mx2.sourceforge.net) by sc8-sf-list1.sourceforge.net with esmtp (Cipher TLSv1:DES-CBC3-SHA:168) (Exim 3.31-VA-mm2 #1 (Debian)) id 1AMuZ9-0007yq-00 for ; Thu, 20 Nov 2003 11:34:11 -0800 Received: from smtp004.mail.ukl.yahoo.com ([217.12.11.35]) by sc8-sf-mx2.sourceforge.net with smtp (Exim 4.24) id 1AMuYz-0002f0-11 for user-mode-linux-devel@lists.sourceforge.net; Thu, 20 Nov 2003 11:34:01 -0800 From: BlaisorBlade Subject: Re: [uml-devel] Host panic when UML reads host /dev/shm References: <200311201915.29188.blaisorblade_spam@yahoo.it> <200311201956.39623.Frank.dekervel@student.kuleuven.ac.Be> In-Reply-To: <200311201956.39623.Frank.dekervel@student.kuleuven.ac.Be> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Disposition: inline Message-Id: <200311202037.14480.blaisorblade_spam@yahoo.it> Sender: user-mode-linux-devel-admin@lists.sourceforge.net Errors-To: user-mode-linux-devel-admin@lists.sourceforge.net List-Help: List-Post: List-Subscribe: , List-Id: The user-mode Linux development list List-Unsubscribe: , List-Archive: Date: Thu, 20 Nov 2003 20:37:14 +0100 Content-Transfer-Encoding: quoted-printable To: user-mode-linux-devel@lists.sourceforge.net Alle 19:56, gioved=EC 20 novembre 2003, Frank Dekervel ha scritto: > Op Thursday 20 November 2003 19:15, schreef BlaisorBlade: > > I reported that if I leave an UML sleeping, it segfaults(once I saw the > > killed by SIGSEGV from the shell, the other time nothing, maybe I had > > detached the job from the shell). > > > > Now I've seen that what happens is that the host kernel oopses. I'm > > posting it here to collect a bit more infos before posting it to the > > LKML, and to know why only UML is able to trigger it. Since the bug see= ms > > related only to /dev/shm, maybe you could have some enlightenment about > > it and how UML uses shared memory to make this happen. > > i saw the same crash. Only an oops on linux or the same oops? If the second, it's useful news > i worked around it by not using hostfs ... So, probably there was a process runned from inside UML triggering the bug. In fact, I'm seeing now 4:40, the time the oops happens, is when the slackw= are=20 every day runs updatedb. A little transcript from a UML session: root [~: slack90: 1 (0)] # mount /dev/ubda on / type ext3 (rw) devpts on /dev/pts type devpts (rw,gid=3D5,mode=3D620) proc on /proc type proc (rw) host on /mnt/host type hostfs (rw,/) root [~: slack90: 1 (0)] # ls /mnt/host/dev/sh sheep_net shm shmiq root [~: slack90: 1 (0)] # ls /mnt/host/dev/shm The host shell reported Segmentation fault(wrongly), so all the crashes I h= ad=20 were due to this. I can even get the processes in D state(even a ls /dev/sh= m=20 suffices). Thanks, Frank. Your message was very useful and enlightening. I think hostfs reads files just as any normal process on the host, right? O= r=20 it does anything special or at least unusual? Could the unusual thing be the context switching of SKAS? I'll test if it happens within TT mode. Jeff, this seems easy reproduceable= ,=20 so try it, please. Bye --=20 cat <