From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753438Ab0C2TWK (ORCPT ); Mon, 29 Mar 2010 15:22:10 -0400 Received: from e7.ny.us.ibm.com ([32.97.182.137]:60966 "EHLO e7.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753328Ab0C2TWC (ORCPT ); Mon, 29 Mar 2010 15:22:02 -0400 Date: Mon, 29 Mar 2010 12:21:59 -0700 From: "Paul E. McKenney" To: David Howells Cc: Trond.Myklebust@netapp.com, linux-nfs@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] NFS: Fix RCU warnings in nfs_inode_return_delegation_noreclaim() [ver #2] Message-ID: <20100329192159.GM2569@linux.vnet.ibm.com> Reply-To: paulmck@linux.vnet.ibm.com References: <20100319022527.GC2894@linux.vnet.ibm.com> <20100318133302.29754.1584.stgit@warthog.procyon.org.uk> <19192.1269889348@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <19192.1269889348@redhat.com> 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, Mar 29, 2010 at 08:02:28PM +0100, David Howells wrote: > Paul E. McKenney wrote: > > > > spin_lock(&clp->cl_lock); > > > - if (rcu_dereference(nfsi->delegation) != NULL) { > > > + if (nfsi->delegation != NULL) { > > > > And this one. I thought that Trond said that clp->cl_lock protects > > this one, in which case this should work: > > > > if (rcu_dereference_check(nfsi->delegation, > > lockdep_is_held(&clp->cl_lock)) != NULL) { > > If clp->cl_lock protects this pointer, why the need for rcu_dereference_check() > at all? The check is redundant since the line above gets the very lock we're > checking for. Because Arnd Bergmann is working on a set of patches that makes sparse complain if you access an RCU-protected pointer directly, without using some flavor of rcu_dereference(). So your approach would work for the moment, but would need another change, probably in the 2.6.35 timeframe. > > > - if (rcu_dereference(nfsi->delegation) != NULL) { > > > + if (nfsi->delegation != NULL) { > > > > And this one, although the check for cp->cl_lock obviously won't work here. > > > > > spin_lock(&clp->cl_lock); > > > delegation = nfs_detach_delegation_locked(nfsi, NULL); > > > spin_unlock(&clp->cl_lock); > > On this one, why does nfsi->delegation need a memory barrier interpolating > afterwards? It has an implicit one in the form of the spin_lock() immediately > after, if the value of the pointer wasn't NULL. What two memory accesses is > the memory barrier ordering? > > Ditto on the next one. I must defer to Trond on this one. Thanx, Paul