From owner-freebsd-current@FreeBSD.ORG Wed Jan 21 12:38:51 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 9251F16A4CE; Wed, 21 Jan 2004 12:38:51 -0800 (PST) Received: from transport.cksoft.de (transport.cksoft.de [62.111.66.27]) by mx1.FreeBSD.org (Postfix) with ESMTP id EA9F743D60; Wed, 21 Jan 2004 12:38:30 -0800 (PST) (envelope-from bzeeb-lists@lists.zabbadoz.net) Received: from transport.cksoft.de (localhost [127.0.0.1]) by transport.cksoft.de (Postfix) with ESMTP id 926961FF9A6; Wed, 21 Jan 2004 21:38:28 +0100 (CET) Received: by transport.cksoft.de (Postfix, from userid 66) id 025081FF931; Wed, 21 Jan 2004 21:38:26 +0100 (CET) Received: by mail.int.zabbadoz.net (Postfix, from userid 1060) id 64579155DD; Wed, 21 Jan 2004 20:38:08 +0000 (UTC) Received: from localhost (localhost [127.0.0.1]) by mail.int.zabbadoz.net (Postfix) with ESMTP id 593F9155AB; Wed, 21 Jan 2004 20:38:08 +0000 (UTC) Date: Wed, 21 Jan 2004 20:38:08 +0000 (UTC) From: "Bjoern A. Zeeb" X-X-Sender: bz@e0-0.zab2.int.zabbadoz.net To: Jun-ichiro itojun Hagino In-Reply-To: Message-ID: References: <20040115041435.7B739A6@coconut.itojun.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Virus-Scanned: by AMaViS cksoft-s20020300-20031204bz on transport.cksoft.de cc: core@kame.net cc: Hajimu UMEMOTO cc: "Bjoern A. Zeeb" cc: FreeBSD current mailing list Subject: Re: [PATCH] IPSec fixes X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 21 Jan 2004 20:38:51 -0000 On Sat, 17 Jan 2004, Bjoern A. Zeeb wrote: Hi, > why do we mark all the other SPs not in sptree[] dead ? I think this > is not correct. We shouldn't walk sptailq but sptree in > key_spdflush(). So at this point forget the above and have a look > at the following patch (you may ingnore the inlining) > (can also be found at > http://sources.zabbadoz.net/freebsd/patchset/118-ipsec-flush-fix.diff > ) sorry for replying to myself. I have gone through KAME CVS and checked where this walk through sptailq has been committed: http://www.kame.net/dev/cvsweb2.cgi/kame/kame/sys/netkey/key.c.diff?r1=1.303&r2=1.304&f=h cleanup, avoid double-loop I think it just happened somewhat erroneously with the cleanup with the good thought of simplifying and improving code. sptree[] does not hold all the entries as sptailq does as stated in my other mail. For me sptree[] holds 18 entries out of 54 in sptailq. Fully walking both will thus not give the same result. Reverting first half of this change will do the same as my patch does (in key_spdflush()) modulo persist checking. Perhaps this information might help to get this thing sorted out and closed soon. TIA. -- Greetings Bjoern A. Zeeb bzeeb at Zabbadoz dot NeT 56 69 73 69 74 http://www.zabbadoz.net/