From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jay Vosburgh Subject: Re: [patch for 2.6.24? 1/1] bonding: locking fix Date: Fri, 18 Jan 2008 09:32:44 -0800 Message-ID: <10345.1200677564@death> References: <200801140904.m0E94vJd020425@imap1.linux-foundation.org> <20080114144754.29a448ad.akpm@linux-foundation.org> <16796.1200351673@death> <20080117154243.9ee4265c.akpm@linux-foundation.org> <24607.1200616005@death> <27063.1200620733@death> Cc: Andrew Morton , davem@davemloft.net, jeff@garzik.org, shemminger@linux-foundation.org, netdev@vger.kernel.org, Andy Gospodarek To: Krzysztof Oledzki Return-path: Received: from e3.ny.us.ibm.com ([32.97.182.143]:37206 "EHLO e3.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1761912AbYARRdf (ORCPT ); Fri, 18 Jan 2008 12:33:35 -0500 Received: from d01relay02.pok.ibm.com (d01relay02.pok.ibm.com [9.56.227.234]) by e3.ny.us.ibm.com (8.13.8/8.13.8) with ESMTP id m0IHWlri014015 for ; Fri, 18 Jan 2008 12:32:47 -0500 Received: from d01av02.pok.ibm.com (d01av02.pok.ibm.com [9.56.224.216]) by d01relay02.pok.ibm.com (8.13.8/8.13.8/NCO v8.7) with ESMTP id m0IHWlJ4079762 for ; Fri, 18 Jan 2008 12:32:47 -0500 Received: from d01av02.pok.ibm.com (loopback [127.0.0.1]) by d01av02.pok.ibm.com (8.12.11.20060308/8.13.3) with ESMTP id m0IHWk60017882 for ; Fri, 18 Jan 2008 12:32:47 -0500 In-reply-to: Sender: netdev-owner@vger.kernel.org List-ID: Krzysztof Oledzki wrote: [...] >> which makes the warning go away, but Herbert Xu pointed out that >> there is a potential problem with bond_enslave accessing the mc_lists >> without sufficient locking. It's not the only offender, either, and the >> bond->mc_list references really need to be protected by the bond_lock, >> and the whole thing probably ought to use dev_mc_sync/unsync instead of >> what it does now. >> >> Since the bond_enslave, et al, business isn't a new problem, and >> I've never heard of it being hit, I'm thinking now to just leave the >> bond_enslave part for 2.6.25, and fix the lockdep warning for 2.6.24. > >It is a new problem, as it never happened with <=2.6.23. The lockdep warning is new, but I was referring to the lack of mutexing between bond_enslave, bond_release, etc and the set_multicast_list's use of the mc_lists. There has never been mutexing of the mc_lists in bond_enslave & friends, so that is not a new problem. -J --- -Jay Vosburgh, IBM Linux Technology Center, fubar@us.ibm.com