From: =?ISO-8859-1?Q?Holger_Hoffst=E4tte?= Subject: Re: Repeatable ext4 oops with 3.6.0 (regression) Date: Thu, 04 Oct 2012 23:11:21 +0200 Message-ID: <506DFB79.1040307@googlemail.com> References: <20121002133642.GD22777@quack.suse.cz> <20121004130119.GH4641@quack.suse.cz> <506DABDD.7090105@googlemail.com> <20121004173425.GA15405@thunk.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Jan Kara , linux-ext4@vger.kernel.org, linux-mm@kvack.org To: Theodore Ts'o Return-path: Received: from mho-04-ewr.mailhop.org ([204.13.248.74]:58699 "EHLO mho-02-ewr.mailhop.org" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1754988Ab2JDVLZ (ORCPT ); Thu, 4 Oct 2012 17:11:25 -0400 In-Reply-To: <20121004173425.GA15405@thunk.org> Sender: linux-ext4-owner@vger.kernel.org List-ID: (dear -mm: please see http://thread.gmane.org/gmane.comp.file-systems.ext4/34665 for the origins of this oops) On 10/04/12 19:34, Theodore Ts'o wrote: > On Thu, Oct 04, 2012 at 05:31:41PM +0200, Holger Hoffst=E4tte wrote: >=20 >> So armed with multiple running shells I finally managed to save the = dmesg >> to NFS. It doesn't get any more complete than this and again shows t= he >> ext4 stacktrace from before. So maybe it really is generic kmem corr= uption >> and ext4 looking at symlinks/inodes is just the victim. >=20 > That certainly seems to be the case. As near as I can tell from the Good to know. Unfortunately I'm still at a loss why apparently only gthumb can trigger this; I had not noticed any other problems with 3.6.= 0 before that (ran it for half a day, desktop use). > So it's very likely that the crash in __kmalloc() is probably caused > by the internal slab/slub data structures getting scrambled. =46or giggles I rebuilt with SLAB instead of SLUB, but no luck; same segfault and delayed oopsie. I also collected an strace, but I cannot really see anything out of the ordinary - it starts, loads things, traverses directories and then segfaults. I've put the earlier full dmesg and the strace into http://hoho.dyndns.org/~holger/ext4-oops-3.6.0/ - maybe it helps someon= e else. Any suggestions for memory debugging? I saw several options in the kernel config (under "Kernel Hacking") but was not sure what to enable. Holger -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html