From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 7860 invoked by alias); 27 Jul 2002 22:18:13 -0000 Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-owner@gcc.gnu.org Received: (qmail 7842 invoked from network); 27 Jul 2002 22:18:10 -0000 Received: from unknown (HELO sunsite.mff.cuni.cz) (195.113.19.66) by sources.redhat.com with SMTP; 27 Jul 2002 22:18:10 -0000 Received: (from jakub@localhost) by sunsite.mff.cuni.cz (8.11.6/8.11.6) id g6RMI6t10004; Sun, 28 Jul 2002 00:18:06 +0200 Date: Sun, 28 Jul 2002 09:40:00 -0000 From: Jakub Jelinek To: Benjamin Kosnik Cc: gcc@gcc.gnu.org Subject: Re: GCC 3.2 Message-ID: <20020728001806.E20867@sunsite.ms.mff.cuni.cz> Reply-To: Jakub Jelinek References: <200207272156.g6RLuE001756@fillmore.constant.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5.1i In-Reply-To: <200207272156.g6RLuE001756@fillmore.constant.com>; from bkoz@redhat.com on Sat, Jul 27, 2002 at 02:56:14PM -0700 X-SW-Source: 2002-07/txt/msg01334.txt.bz2 On Sat, Jul 27, 2002 at 02:56:14PM -0700, Benjamin Kosnik wrote: > > I've noticed alignment-related bugs in GNATS as well. See libstdc++/6732. Is that really a bug? __attribute__((aligned(16))) is not part of C++ standard, and malloc is not guaranteed to return that much aligned chunks (at least not on all architectures). E.g. glibc malloc by default only guarantees 2*sizeof(size_t) alignment. Jakub