From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 16185 invoked by alias); 11 Aug 2010 08:51:27 -0000 Received: (qmail 16172 invoked by uid 22791); 11 Aug 2010 08:51:26 -0000 X-SWARE-Spam-Status: No, hits=-1.9 required=5.0 tests=AWL,BAYES_00,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FROM,RCVD_IN_DNSWL_NONE X-Spam-Check-By: sourceware.org Received: from mail-qw0-f47.google.com (HELO mail-qw0-f47.google.com) (209.85.216.47) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Wed, 11 Aug 2010 08:51:20 +0000 Received: by qwg8 with SMTP id 8so9692712qwg.20 for ; Wed, 11 Aug 2010 01:51:18 -0700 (PDT) MIME-Version: 1.0 Received: by 10.224.5.196 with SMTP id 4mr10351682qaw.373.1281516678605; Wed, 11 Aug 2010 01:51:18 -0700 (PDT) Received: by 10.224.28.67 with HTTP; Wed, 11 Aug 2010 01:51:18 -0700 (PDT) In-Reply-To: <4C61864B.7050008@caviumnetworks.com> References: <4C5C20D0.5020105@codesourcery.com> <4C5C6BAA.4030703@codesourcery.com> <4C6162D4.6080605@codesourcery.com> <8739um8mz5.fsf@basil.nowhere.org> <20100810164720.GF6801@basil.fritz.box> <4C61864B.7050008@caviumnetworks.com> Date: Wed, 11 Aug 2010 08:53:00 -0000 Message-ID: Subject: Re: Combine four insns From: Richard Guenther To: David Daney Cc: Andi Kleen , Steven Bosscher , Bernd Schmidt , GCC Patches Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-IsSubscribed: yes Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-patches-owner@gcc.gnu.org X-SW-Source: 2010-08/txt/msg00824.txt.bz2 On Tue, Aug 10, 2010 at 7:03 PM, David Daney wr= ote: > On 08/10/2010 09:47 AM, Andi Kleen wrote: >>>> >>>> To be honest I didn't fully understand why it's ok for -Os >>>> to be 1% slower. A lot of people use -Os. >>> >>> Yes, and they expect the smallest code possible. This patch does seem >>> to help for that. >> >> They also expect reasonable fast compilation. >> > > If they did, their expectation wouldn't have come from having read the > documentation. =A0As far as I can see, the *only* thing one should expect= from > -Os is smaller code. > > We would all like fast compilation, but if you specify -Os, it seems to me > that you are expressing a preference for smaller code. Yes indeed. Richard. > David Daney >