From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 25298 invoked by alias); 25 Jan 2002 13:14:29 -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 25246 invoked from network); 25 Jan 2002 13:14:23 -0000 Received: from unknown (HELO vlsi1.ultra.nyu.edu) (128.122.140.213) by sources.redhat.com with SMTP; 25 Jan 2002 13:14:23 -0000 Received: by vlsi1.ultra.nyu.edu (4.1/1.34) id AA11170; Fri, 25 Jan 02 08:09:53 EST Date: Fri, 25 Jan 2002 07:05:00 -0000 From: kenner@vlsi1.ultra.nyu.edu (Richard Kenner) Message-Id: <10201251309.AA11170@vlsi1.ultra.nyu.edu> To: gdr@codesourcery.com Subject: Re: g++ and aliasing bools Cc: gcc@gcc.gnu.org X-SW-Source: 2002-01/txt/msg01616.txt.bz2 Optimizations have high priorities (cf. regular patches and discussions about such and such Intermdiate Language Tranformations). It may be that an optimization like alias analysis doesn't have the highest priority but that doesn't mean notbody isn't consider it -- Indeed, I strongly suspect that if it were possible to do better alias set handling in C++, it would produce a far larger performance improvement than any of the Intermediate Language Transformations being discussed and would also likely be a lot less work.