From owner-sc22wg14+sc22wg14-domo2=www.open-std.org@open-std.org Thu Oct 29 06:21:54 2015 Return-Path: X-Original-To: sc22wg14-domo2 Delivered-To: sc22wg14-domo2@www.open-std.org Received: by www.open-std.org (Postfix, from userid 521) id 31E713587E0; Thu, 29 Oct 2015 06:21:54 +0100 (CET) Delivered-To: sc22wg14@open-std.org Received: from mga11.intel.com (mga11.intel.com [192.55.52.93]) by www.open-std.org (Postfix) with ESMTP id 470853568B2 for ; Thu, 29 Oct 2015 06:21:48 +0100 (CET) Received: from orsmga002.jf.intel.com ([10.7.209.21]) by fmsmga102.fm.intel.com with ESMTP; 28 Oct 2015 22:21:46 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.20,213,1444719600"; d="scan'208";a="837689360" Received: from orsmsx102.amr.corp.intel.com ([10.22.225.129]) by orsmga002.jf.intel.com with ESMTP; 28 Oct 2015 22:21:45 -0700 Received: from orsmsx160.amr.corp.intel.com (10.22.226.43) by ORSMSX102.amr.corp.intel.com (10.22.225.129) with Microsoft SMTP Server (TLS) id 14.3.248.2; Wed, 28 Oct 2015 22:21:45 -0700 Received: from orsmsx103.amr.corp.intel.com ([169.254.2.240]) by ORSMSX160.amr.corp.intel.com ([169.254.13.248]) with mapi id 14.03.0248.002; Wed, 28 Oct 2015 22:21:45 -0700 From: "Nelson, Clark" To: "sc22wg14@open-std.org" Subject: N1956: volatile semantics Thread-Topic: N1956: volatile semantics Thread-Index: AdESCbL/dOKA7ANdRdSS4LDK+4Owug== Date: Thu, 29 Oct 2015 05:21:45 +0000 Message-ID: <38C37E44FD352B44ABFC58410B0790D090036AAC@ORSMSX103.amr.corp.intel.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.22.254.138] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Sender: owner-sc22wg14@open-std.org Precedence: bulk I don't know whether we're going to get back to this tomorrow, and even if = we do, I don't know if Martin will be there. So I'll post my comments. In the first place, I compared the C standard to the C++ standard where Mar= tin proposes to modify the former. In only one of those places does the C++= standard use "lvalue" instead of "object". So I definitely think the C++ s= tandard should be modified along the same lines. Also, I think it might be better, from an editorial perspective, to define = a new term, perhaps "volatile access", roughly as "use of a volatile lvalue= ", and then use that where needed. So, for example, we would wind up with s= omething like "What constitutes a volatile access is implementation-defined= ." In the first edit, to 5.1.2.3p2, the existing clause "modifying an object" = was omitted -- inadvertently, I assume. I noticed that 5.1.2.3p4, Douglas' particularly problematic statement, is e= xpressed differently in the C++ standard. In the first place, it's a footno= te. The complete text of the footnote is as follows: "This provision is sometimes called the "as-if" rule, because an implementa= tion is free to disregard any requirement of this International Standard as= long as the result is /as if/ the requirement had been obeyed, as far as c= an be determined from the observable behavior of the program. For instance,= an actual implementation need not evaluate part of an expression if it can= deduce that its value is not used and that no side effects affecting the o= bservable behavior of the program are produced." I'm not necessarily suggesting that the C standard should explicitly mentio= n the "as-if" rule, even in a footnote. But note that the second sentence, = which corresponds directly to the sentence Martin proposes to modify, doesn= 't even contain the word "volatile", and (again) is definitely not normativ= e. We should consider emulating this. Clark