From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 8065 invoked by alias); 27 Sep 2006 02:12:59 -0000 Received: (qmail 8024 invoked by alias); 27 Sep 2006 02:12:50 -0000 Date: Wed, 27 Sep 2006 02:12:00 -0000 Message-ID: <20060927021250.8023.qmail@sourceware.org> X-Bugzilla-Reason: CC References: Subject: [Bug tree-optimization/28778] [4.0/4.1/4.2 Regression] alias bug with cast and call clobbered In-Reply-To: Reply-To: gcc-bugzilla@gcc.gnu.org To: gcc-bugs@gcc.gnu.org From: "dberlin at dberlin dot org" 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 X-SW-Source: 2006-09/txt/msg02516.txt.bz2 List-Id: ------- Comment #41 from dberlin at gcc dot gnu dot org 2006-09-27 02:12 ------- Subject: Re: [4.0/4.1/4.2 Regression] alias bug with cast and call clobbered On 26 Sep 2006 15:57:28 -0000, pcarlini at suse dot de wrote: > > > ------- Comment #40 from pcarlini at suse dot de 2006-09-26 15:57 ------- > (In reply to comment #38) > > We have our reasons to enable strict aliasing by default. > > In my opinion, this is a central point. I think we should try to explain what > strict aliasing buys from the point of view of the optimizers. In my > experience, quite a few people do not quite understand *why* those aliasing > rules are there in the first place and why lately GCC (or any modern compiler, > for that matter, of course, that's the point) tries to be as strict as > possible: otherwise, those people would be more willing to learn the rules and > appreciate our (I should say, yours, Danny and compiler people) hard efforts in > this area. > And I would be more than happy to discuss this (there is plenty of empirical evidence as to how much it buys us, etc), just not in a bug report :) Bring it to gcc@ -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28778