From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 23578 invoked by alias); 26 Jul 2004 19:30:16 -0000 Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-bugs-owner@gcc.gnu.org Received: (qmail 23571 invoked by uid 48); 26 Jul 2004 19:30:15 -0000 Date: Mon, 26 Jul 2004 19:30:00 -0000 Message-ID: <20040726193015.23570.qmail@sourceware.org> From: "falk at debian dot org" To: gcc-bugs@gcc.gnu.org In-Reply-To: <20040726181156.16721.macro@linux-mips.org> References: <20040726181156.16721.macro@linux-mips.org> Reply-To: gcc-bugzilla@gcc.gnu.org Subject: [Bug c/16721] [3.5 Regression] Accesses to volatile objects optimized away X-Bugzilla-Reason: CC X-SW-Source: 2004-07/txt/msg02955.txt.bz2 List-Id: ------- Additional Comments From falk at debian dot org 2004-07-26 19:30 ------- I don't think that's what the standard says. Even if it did, it is not what people expect. Read accesses can have effects on magic hardware addresses, and you need volatile to keep the compiler from optimizing these away. So I still think this is an important bug. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16721