From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 28399 invoked by alias); 17 May 2002 23:38:42 -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 28391 invoked from network); 17 May 2002 23:38:40 -0000 Received: from unknown (HELO mallaury.noc.nerim.net) (62.4.17.82) by sources.redhat.com with SMTP; 17 May 2002 23:38:40 -0000 Received: from tetto.gentiane.org (espie.gentiane.org [62.212.102.210]) by mallaury.noc.nerim.net (Postfix) with ESMTP id 78DE162D01; Sat, 18 May 2002 01:38:36 +0200 (CEST) Received: from tetto.gentiane.org (espie@localhost [127.0.0.1]) by tetto.gentiane.org (8.12.1/8.12.1) with ESMTP id g4HNcbqj000812; Sat, 18 May 2002 01:38:38 +0200 (CEST) Received: (from espie@localhost) by tetto.gentiane.org (8.12.1/8.12.1/Submit) id g4HNcaQ5022548; Sat, 18 May 2002 01:38:36 +0200 (CEST) Date: Fri, 17 May 2002 17:08:00 -0000 From: Marc Espie To: "David O'Brien" Cc: Loren James Rittle , gcc@gcc.gnu.org Subject: Re: GCC 3.1 Release Message-ID: <20020518013835.A26088@tetto.liafa.jussieu.fr> Reply-To: espie@nerim.net References: <200204162351.g3GNpha21092@quatramaran.ens.fr> <20020517140210.A16176@tetto.liafa.jussieu.fr> <200205172204.g4HM4IvB044134@latour.rsch.comm.mot.com> <20020517163500.A4119@dragon.nuxi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20020517163500.A4119@dragon.nuxi.com>; from obrien@FreeBSD.org on Fri, May 17, 2002 at 04:35:00PM -0700 X-SW-Source: 2002-05/txt/msg01564.txt.bz2 On Fri, May 17, 2002 at 04:35:00PM -0700, David O'Brien wrote: > On Fri, May 17, 2002 at 05:04:18PM -0500, Loren James Rittle wrote: > > >> FWIW, FreeBSD is adopting GCC 3.1 as system compiler for -CURRENT which > > >> will become FreeBSD 5.0 later this year. > ... > > I have not personally tested it > > but perhaps David could post the exact time difference he saw when he > > made this change. > > I really don't have any machines to compare with. My own went from 2.95 > to 3.1 in a peice meal fashion as I tested various things. Sorry. Well, unless you run tests, I will tend to assume you have the same catastrophic behavior that I see. Of course, since FreeBSD only targets fast architectures, this is less of a problem... For that matter, this is a mistake we've made once, between gcc 2.8 and gcc 2.95 (which was already a slowdown), which is the reason why we are much more careful when doing compiler upgrades now.