From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753171AbcA2Wbc (ORCPT ); Fri, 29 Jan 2016 17:31:32 -0500 Received: from mailgw.movielink.com.au ([175.103.28.14]:52344 "EHLO mailgw.movielink.com.au" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752097AbcA2Wbb (ORCPT ); Fri, 29 Jan 2016 17:31:31 -0500 Date: Fri, 29 Jan 2016 02:22:53 -0800 From: "Paul E. McKenney" To: Will Deacon Cc: Peter Zijlstra , "Maciej W. Rozycki" , David Daney , =?iso-8859-1?Q?M=E5ns_Rullg=E5rd?= , Ralf Baechle , linux-kernel@vger.kernel.org, torvalds@linux-foundation.org, boqun.feng@gmail.com Subject: Re: [RFC][PATCH] mips: Fix arch_spin_unlock() Message-ID: <20160129102253.GG4503@linux.vnet.ibm.com> Reply-To: paulmck@linux.vnet.ibm.com References: <5644D7B5.6020009@caviumnetworks.com> <20160127114348.GF2390@arm.com> <20160127145421.GT6357@twins.programming.kicks-ass.net> <20160127152158.GJ2390@arm.com> <20160127233836.GQ4503@linux.vnet.ibm.com> <20160128095718.GC30928@arm.com> <20160128223131.GV4503@linux.vnet.ibm.com> <20160129095958.GA4541@arm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20160129095958.GA4541@arm.com> User-Agent: Mutt/1.5.21 (2010-09-15) X-Movielink-MailScanner-Information: Please contact the ISP for more information X-Movielink-MailScanner-ID: 84CC39912C6.A3B36 X-Movielink-MailScanner: Found to be clean X-Movielink-MailScanner-From: paulmck@linux.vnet.ibm.com Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Jan 29, 2016 at 09:59:59AM +0000, Will Deacon wrote: > On Thu, Jan 28, 2016 at 02:31:31PM -0800, Paul E. McKenney wrote: [ . . . ] > > For Linux in general, this is a question: How strict do we want to be > > about matching the type of write with the corresponding read? My > > default approach is to initially be quite strict and loosen as needed. > > Here "quite strict" might mean requiring an rcu_assign_pointer() for > > the write and rcu_dereference() for the read, as opposed to (say) > > ACCESS_ONCE() for the read. (I am guessing that this would be too > > tight, but it makes a good example.) > > > > Thoughts? > > That sounds broadly sensible to me and allows rcu_assign_pointer and > rcu_dereference to be used as drop-in replacements for release/acquire > where local transitivity isn't required. However, I don't think we can > rule out READ_ONCE/WRITE_ONCE interactions as they seem to be used > already in things like the osq_lock (albeit without the address > dependency). Agreed. So in the most strict case that I can imagine anyone putting up with, we have the following pairings: o smp_store_release() -> smp_load_acquire() (locally transitive) o smp_store_release() -> lockless_dereference() (???) o smp_store_release() -> READ_ONCE(); if o rcu_assign_pointer() -> rcu_dereference() o smp_mb(); WRITE_ONCE() -> READ_ONCE(); (globally transitive) o synchronize_rcu(); WRITE_ONCE() -> READ_ONCE(); (globally transitive) o synchronize_rcu(); WRITE_ONCE() -> rcu_read_lock(); READ_ONCE() (strange and wonderful properties) Seem reasonable, or am I missing some? Thanx, Paul -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.