From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 3001 invoked by alias); 28 Apr 2004 00:00:23 -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 2989 invoked by uid 48); 28 Apr 2004 00:00:23 -0000 Date: Wed, 28 Apr 2004 00:42:00 -0000 Message-ID: <20040428000023.2988.qmail@sources.redhat.com> From: "cesarb at nitnet dot com dot br" To: gcc-bugs@gcc.gnu.org In-Reply-To: <20040427233827.15184.cesarb@nitnet.com.br> References: <20040427233827.15184.cesarb@nitnet.com.br> Reply-To: gcc-bugzilla@gcc.gnu.org Subject: [Bug target/15184] [3.3/3.4/3.5 Regression] Direct access to byte inside word not working with -march=pentiumpro X-Bugzilla-Reason: CC X-SW-Source: 2004-04/txt/msg02703.txt.bz2 List-Id: ------- Additional Comments From cesarb at nitnet dot com dot br 2004-04-28 00:00 ------- As does i486 and athlon. It does not seem odd to me; I think it's a collateral effect of something somewhere in the gcc code which tries to avoid using 16-bit words on pentiumpro, as it's known that one of the biggest problems Pentium PRO had was that it was designed with Unix in mind and ran Windows slow exactly because Windows used 16-bit words all the time. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15184