From mboxrd@z Thu Jan 1 00:00:00 1970 From: Patrick McHardy Subject: Re: [PATCH 1/2] IPv6: conntrack: Use protocol-related initialization routine to initial queues of IPv6 connection track Date: Tue, 26 Jan 2010 13:46:46 +0100 Message-ID: <4B5EE436.3040204@trash.net> References: <4B5E53EE.9010703@cn.fujitsu.com> <20100125.233254.35824860.davem@davemloft.net> <4B5EE2C6.6010802@cn.fujitsu.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: David Miller , yasuyuki.kozakai@toshiba.co.jp, netfilter-devel@vger.kernel.org, netdev@vger.kernel.org To: Shan Wei Return-path: Received: from stinky.trash.net ([213.144.137.162]:33663 "EHLO stinky.trash.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751914Ab0AZMqw (ORCPT ); Tue, 26 Jan 2010 07:46:52 -0500 In-Reply-To: <4B5EE2C6.6010802@cn.fujitsu.com> Sender: netfilter-devel-owner@vger.kernel.org List-ID: Shan Wei wrote: > David Miller wrote, at 01/26/2010 03:32 PM: >> From: Shan Wei >> Date: Tue, 26 Jan 2010 10:31:10 +0800 >> >>> IPv6 connection track and IPv6 stack separately use a different que= ue to=20 >>> manage received fragments. The former uses nf_ct_frag6_queue struct= ure,=20 >>> the latter uses frag_queue structure. >>> >>> When creating new queue for IPv6 connection track, ip6_frag_init()=20 >>> that belongs to IPv6 stack is called to initial nf_ct_frag6_queue s= tructure.=20 >>> This broken the saddr&daddr member in nf_ct_frag6_queue, and then h= ash value=20 >>> generated by nf_hashfn() is not equal with that generated by fq_fin= d().=20 >>> So, a new received fragment can't be inserted to right queue. >>> =20 >>> The patch fixes the bug with protocol-related initialization routin= e. >>> The patch-set have been tested. >>> >>> Signed-off-by: Shan Wei >> This breakage was recently introduced by: >> >> commit 0b5ccb2ee250136dd7385b1c7da28417d0d4d32d >> Author: Patrick McHardy >> Date: Tue Dec 15 16:59:18 2009 +0100 >> >> ipv6: reassembly: use seperate reassembly queues for conntrack a= nd local delivery >=20 > Yes, this patch adds user member to frag_queue structure=EF=BC=8Cbut = not to nf_ct_frag6_queue structure. Oops, sorry. Not sure why I missed this, I've successfully tested that change multiple times. > Please ignore the patch-set.=20 > Can you apply the following patch(bug-fix) to your net-tree? >=20 > -- > [PATCH]IPv6: conntrack: Add member of user to nf_ct_frag6_queue stru= cture >=20 > The commit 0b5ccb2(title:ipv6: reassembly: use seperate reassembly qu= eues for=20 > conntrack and local delivery) has broken the saddr&&daddr member of=20 > nf_ct_frag6_queue when creating new queue. And then hash value > generated by nf_hashfn() was not equal with that generated by fq_find= ().=20 > So, a new received fragment can't be inserted to right queue. >=20 > The patch fixes the bug with adding member of user to nf_ct_frag6_que= ue structure. >=20 > Signed-off-by: Shan Wei Acked-by: Patrick McHardy -- To unsubscribe from this list: send the line "unsubscribe netfilter-dev= el" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html