From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: [PATCH tip/core/rcu 21/21] drivers/vhost: Remove now-redundant read_barrier_depends() Date: Tue, 5 Dec 2017 21:24:21 +0200 Message-ID: <20171205212053-mutt-send-email-mst@kernel.org> References: <20171201195053.GA23494@linux.vnet.ibm.com> <1512157876-24665-21-git-send-email-paulmck@linux.vnet.ibm.com> <20171205202928-mutt-send-email-mst@kernel.org> <20171205183946.GP3165@worktop.lehotels.local> <20171205204928-mutt-send-email-mst@kernel.org> <20171205191733.GQ3165@worktop.lehotels.local> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <20171205191733.GQ3165@worktop.lehotels.local> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: virtualization-bounces@lists.linux-foundation.org Errors-To: virtualization-bounces@lists.linux-foundation.org To: Peter Zijlstra Cc: tglx@linutronix.de, kvm@vger.kernel.org, fweisbec@gmail.com, jiangshanlai@gmail.com, linux-kernel@vger.kernel.org, rostedt@goodmis.org, josh@joshtriplett.org, dhowells@redhat.com, edumazet@google.com, netdev@vger.kernel.org, mathieu.desnoyers@efficios.com, oleg@redhat.com, dipankar@in.ibm.com, akpm@linux-foundation.org, "Paul E. McKenney" , virtualization@lists.linux-foundation.org, mingo@kernel.org List-Id: virtualization@lists.linuxfoundation.org On Tue, Dec 05, 2017 at 08:17:33PM +0100, Peter Zijlstra wrote: > On Tue, Dec 05, 2017 at 08:57:46PM +0200, Michael S. Tsirkin wrote: > > > I don't see WRITE_ONCE inserting any barriers, release or > > write. > > Correct, never claimed there was. > > Just saying that: > > obj = READ_ONCE(*foo); > val = READ_ONCE(obj->val); > > Never needs a barrier (except on Alpha and we want to make that go > away). Simply because a CPU needs to complete the load of @obj before it > can compute the address &obj->val. Thus the second load _must_ come > after the first load and we get LOAD-LOAD ordering. > > Alpha messing that up is a royal pain, and Alpha not being an > active/living architecture is just not worth the pain of keeping this in > the generic model. > Right. What I am saying is that for writes you need WRITE_ONCE(obj->val, 1); smp_wmb(); WRITE_ONCE(*foo, obj); and this barrier is no longer paired with anything until you realize there's a dependency barrier within READ_ONCE. Barrier pairing was a useful tool to check code validity, maybe there are other, better tools now. -- MST