From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754633Ab3EVRqN (ORCPT ); Wed, 22 May 2013 13:46:13 -0400 Received: from e39.co.us.ibm.com ([32.97.110.160]:34403 "EHLO e39.co.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754021Ab3EVRqL (ORCPT ); Wed, 22 May 2013 13:46:11 -0400 Date: Wed, 22 May 2013 10:45:32 -0700 From: "Paul E. McKenney" To: Roman Gushchin Cc: Eric Dumazet , Dipankar Sarma , zhmurov@yandex-team.ru, linux-kernel@vger.kernel.org, netdev@vger.kernel.org, "David S. Miller" , Alexey Kuznetsov , James Morris , Hideaki YOSHIFUJI , Patrick McHardy Subject: Re: [PATCH v2] rcu: fix a race in hlist_nulls_for_each_entry_rcu macro Message-ID: <20130522174532.GC3431@linux.vnet.ibm.com> Reply-To: paulmck@linux.vnet.ibm.com References: <20130521120906.GD3578@linux.vnet.ibm.com> <1369143885.3301.221.camel@edumazet-glaptop> <519B8908.9080007@yandex-team.ru> <1369150693.3301.233.camel@edumazet-glaptop> <519BB90B.6080706@yandex-team.ru> <1369188080.3301.268.camel@edumazet-glaptop> <1369201765.3301.299.camel@edumazet-glaptop> <519CB2D8.103@yandex-team.ru> <1369225837.3301.324.camel@edumazet-glaptop> <519CC2FB.2010006@yandex-team.ru> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <519CC2FB.2010006@yandex-team.ru> User-Agent: Mutt/1.5.21 (2010-09-15) X-TM-AS-MML: No X-Content-Scanned: Fidelis XPS MAILER x-cbid: 13052217-3620-0000-0000-000002A141C6 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, May 22, 2013 at 05:07:07PM +0400, Roman Gushchin wrote: > On 22.05.2013 16:30, Eric Dumazet wrote: > >On Wed, 2013-05-22 at 15:58 +0400, Roman Gushchin wrote: > > > >>+/* > >>+ * Same as ACCESS_ONCE(), but used for accessing field of a structure. > >>+ * The main goal is preventing compiler to store &ptr->field in a register. > > > >But &ptr->field is a constant during the whole duration of > >udp4_lib_lookup2() and could be in a register, in my case field is at > >offset 0, and ptr is a parameter (so could be in a 'register') > > > >The bug you found is that compiler caches the indirection (ptr->field) > >into a register, not that compiler stores &ptr->field into a register. > > > >>+ */ > >>+#define ACCESS_FIELD_ONCE(PTR, FIELD) (((volatile typeof(*PTR) *)PTR)->FIELD) > >>+ > > > >Here we force the compiler to consider ptr as volatile, but semantically > >it is not required in rcu_dereference(ptr->field) > > Actually, we need to mark an "address of a place" where the field value is > located as volatile before dereferencing. I have no idea how to do it in another way, > except using multiple casts and offsetof's, but, IMHO, it will be even more complex: > ACCESS_ONCE(typeof(&ptr->field)((char*)ptr + offsetof(typeof(*ptr), field))) Why not just ACCESS_ONCE(ptr->field)? Or if it is the thing that ptr->field points to that is subject to change, ACCESS_ONCE(*ptr->field)? Or rcu_dereference(ptr->field), as appropriate? Thanx, Paul > >We want field to be reloaded, not ptr. > > > >So yes, the patch appears to fix the bug, but it sounds not logical to > >me. > > > > May be we can enhance it by providing better/more detailed comments here? > Have you any suggestions? > > Thanks, > Roman > -- > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > Please read the FAQ at http://www.tux.org/lkml/ >