From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1753876AbbKLK2V (ORCPT <rfc822;w@1wt.eu>);
	Thu, 12 Nov 2015 05:28:21 -0500
Received: from casper.infradead.org ([85.118.1.10]:38609 "EHLO
	casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1752487AbbKLK2U (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Thu, 12 Nov 2015 05:28:20 -0500
Date: Thu, 12 Nov 2015 11:28:12 +0100
From: Peter Zijlstra <peterz@infradead.org>
To: Boqun Feng <boqun.feng@gmail.com>
Cc: Oleg Nesterov <oleg@redhat.com>, mingo@kernel.org,
        linux-kernel@vger.kernel.org, paulmck@linux.vnet.ibm.com,
        corbet@lwn.net, mhocko@kernel.org, dhowells@redhat.com,
        torvalds@linux-foundation.org, will.deacon@arm.com,
        Michael Ellerman <mpe@ellerman.id.au>,
        Benjamin Herrenschmidt <benh@kernel.crashing.org>,
        Paul Mackerras <paulus@samba.org>
Subject: Re: [PATCH 4/4] locking: Introduce smp_cond_acquire()
Message-ID: <20151112102812.GW17308@twins.programming.kicks-ass.net>
References: <20151102132901.157178466@infradead.org>
 <20151102134941.005198372@infradead.org>
 <20151103175958.GA4800@redhat.com>
 <20151111093939.GA6314@fixme-laptop.cn.ibm.com>
 <20151111121232.GN17308@twins.programming.kicks-ass.net>
 <20151111193953.GA23515@redhat.com>
 <20151112070915.GC6314@fixme-laptop.cn.ibm.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20151112070915.GC6314@fixme-laptop.cn.ibm.com>
User-Agent: Mutt/1.5.21 (2012-12-30)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Thu, Nov 12, 2015 at 03:14:51PM +0800, Boqun Feng wrote:
> Hmm.. probably incorrect.. because the ACQUIRE semantics of spin_lock()
> only guarantees that the memory operations following spin_lock() can't
> be reorder before the *LOAD* part of spin_lock() not the *STORE* part,
> i.e. the case below can happen(assuming the spin_lock() is implemented
> as ll/sc loop)
> 
> 	spin_lock(&lock):
> 	  r1 = *lock; // LL, r1 == 0
> 	o = READ_ONCE(object); // could be reordered here.
> 	  *lock = 1; // SC
> 
> This could happen because of the ACQUIRE semantics of spin_lock(), and 
> the current implementation of spin_lock() on PPC allows this happen.

Urgh, you just _had_ to send an email like this, didn't you ;-)

I think AARGH64 does the same thing. They either use LDAXR/STXR, which
also places the ACQUIRE on the load, or use LDADDA (v8.1) which I
suspect does the same, but I'm not entirely sure how to decode these 8.1
atomics yet.

Let me think a little more on this..