From mboxrd@z Thu Jan 1 00:00:00 1970 From: Patrick McHardy Subject: Re: [OOPS PATCH 1/1] netfilter: fix OOPS in flush_expectations() Date: Fri, 11 Oct 2013 15:35:41 +0100 Message-ID: <20131011143539.GA5276@macbook.localnet> References: <20131011140204.916097373@eitzenberger.org> <20131011140440.339579297@eitzenberger.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: netfilter-devel@vger.kernel.org To: Holger Eitzenberger Return-path: Received: from stinky.trash.net ([213.144.137.162]:45748 "EHLO stinky.trash.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751968Ab3JKOft (ORCPT ); Fri, 11 Oct 2013 10:35:49 -0400 Content-Disposition: inline In-Reply-To: <20131011140440.339579297@eitzenberger.org> Sender: netfilter-devel-owner@vger.kernel.org List-ID: On Fri, Oct 11, 2013 at 04:02:05PM +0200, Holger Eitzenberger wrote: > This is the initial report I got: > > [ 2886.953175] BUG: unable to handle kernel paging request at 00100100 > [ 2886.956435] IP: [] flush_expectations+0x68/0x85 [nf_conntrack_sip] > [ 2886.956435] *pde = 00000000 > [ 2886.956435] Oops: 0000 [001] SMP > ... > [ 2886.956435] Pid: 5606, comm: red_server.plc Tainted: G O > 3.3.8-79.g20f5c30-smp 001 Astaro AG ASG/i845GV-W83627HF > [ 2886.956435] EIP: 0060:[] EFLAGS: 00210246 CPU: 0 > [ 2886.956435] EIP is at flush_expectations+0x68/0x85 [nf_conntrack_sip] > [ 2886.956435] EAX: 00000000 EBX: 00100100 ECX: 00000000 EDX: effdc0a0 > [ 2886.956435] ESI: 00100100 EDI: 00000001 EBP: 00000001 ESP: f5c0bd54 > [ 2886.956435] DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068 > [ 2886.956435] Process red_server.plc (pid: 5606, ti=f5c0a000 task=f5da2a20 task.ti=efc62000) > [ 2886.956435] Stack: > [ 2886.956435] f490b948 00000001 00000197 f45f4f00 f88a5918 f5c0bde0 f5c0bddc 0000001c > [ 2886.956435] 00000014 f88a72a8 0000015d f5c0bddc 00000001 f88a472e f5c0bddc f5c0bde0 > [ 2886.956435] 00000001 00000197 00000014 f490b948 f45f4f00 f88a72a8 00000197 00000001 > > Which is due to nf_conntrack_expect.lnode hlist entry not being reset > to NULL after being removed from the list in hlist_del(), but instead to > LIST_POISON1. And because of this hlist_for_each_entry_safe() does > not terminate correctly. > > Therefore change nf_ct_unlink_expect_report() to use __hlist_del() > instead. We should be holding the conntrack lock here and in flush_expectations(), Not sure what I'm missing here, but if locking were used correctly, this shouldn't be happening. > > Signed-off-by: Holger Eitzenberger > > Index: linux-stable-3.8.y/net/netfilter/nf_conntrack_expect.c > =================================================================== > --- linux-stable-3.8.y.orig/net/netfilter/nf_conntrack_expect.c > +++ linux-stable-3.8.y/net/netfilter/nf_conntrack_expect.c > @@ -51,7 +51,7 @@ void nf_ct_unlink_expect_report(struct n > hlist_del_rcu(&exp->hnode); > net->ct.expect_count--; > > - hlist_del(&exp->lnode); > + __hlist_del(&exp->lnode); > master_help->expecting[exp->class]--; > > nf_ct_expect_event_report(IPEXP_DESTROY, exp, pid, report); > > -- > To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html