From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751817Ab0IFFht (ORCPT ); Mon, 6 Sep 2010 01:37:49 -0400 Received: from e9.ny.us.ibm.com ([32.97.182.139]:52428 "EHLO e9.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750877Ab0IFFhr (ORCPT ); Mon, 6 Sep 2010 01:37:47 -0400 Date: Sun, 5 Sep 2010 22:37:43 -0700 From: "Paul E. McKenney" To: Neil Brown Cc: Stephen Rothwell , linux-next@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: linux-next: build warning after merge of the rcu tree Message-ID: <20100906053743.GD2463@linux.vnet.ibm.com> Reply-To: paulmck@linux.vnet.ibm.com References: <20100906121408.33946fbe.sfr@canb.auug.org.au> <20100906123255.3eef4311@notabene> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100906123255.3eef4311@notabene> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Sep 06, 2010 at 12:32:55PM +1000, Neil Brown wrote: > On Mon, 6 Sep 2010 12:14:08 +1000 > Stephen Rothwell wrote: > > > Hi Paul, > > > > After merging the rcu tree, today's linux-next build > > (powerpc ppc64_defconfig) produced this warning: > > > > drivers/md/raid1.c: In function 'read_balance': > > drivers/md/raid1.c:445: warning: operation on 'new_disk' may be undefined > > > > I am picking on the rcu tree because the line above has not changed since > > 2005. The line is: > > > > for (rdev = rcu_dereference(conf->mirrors[new_disk].rdev); > > r1_bio->bios[new_disk] == IO_BLOCKED || > > !rdev || !test_bit(In_sync, &rdev->flags) > > || test_bit(WriteMostly, &rdev->flags); > > rdev = rcu_dereference(conf->mirrors[++new_disk].rdev)) { > > > > Where new_disk is being updated in the parameter to the last > > rcu_dereference. > > > > Hi Stephen, > > There is a patch in linux-kernel > "PATCH] md: do not use ++ in rcu_dereference() argument" > > that addresses this - it is a problem in md: I guess rcu as become more > helpful in finding these problems. Indeed, the sparse-based checking is a bit more picky about things, even when you are not using sparse... Thanx, Paul > I'll probably fix it a little differently, but I'll have something in my > for-next for tomorrow. > > Thanks, > NeilBrown