From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 31896 invoked by alias); 17 Aug 2009 20:58:44 -0000 Received: (qmail 31888 invoked by uid 22791); 17 Aug 2009 20:58:44 -0000 X-SWARE-Spam-Status: No, hits=-2.6 required=5.0 tests=BAYES_00,SPF_PASS X-Spam-Check-By: sourceware.org Received: from mail-yx0-f200.google.com (HELO mail-yx0-f200.google.com) (209.85.210.200) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Mon, 17 Aug 2009 20:58:37 +0000 Received: by yxe38 with SMTP id 38so6349026yxe.0 for ; Mon, 17 Aug 2009 13:58:34 -0700 (PDT) Received: by 10.91.76.5 with SMTP id d5mr3104323agl.68.1250542714628; Mon, 17 Aug 2009 13:58:34 -0700 (PDT) Received: from lawn-143-215-96-168.lawn.gatech.edu (lawn-143-215-96-168.lawn.gatech.edu [143.215.96.168]) by mx.google.com with ESMTPS id 6sm3670234agb.68.2009.08.17.13.58.33 (version=TLSv1/SSLv3 cipher=RC4-MD5); Mon, 17 Aug 2009 13:58:33 -0700 (PDT) Message-Id: <63D3DBD8-D267-449A-82CF-A6F7A28DCF44@ithinksw.com> From: Alexander Strange To: gcc@gcc.gnu.org In-Reply-To: Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v936) Subject: Re: "-fno-unswitch-loops" option have no effect? Date: Mon, 17 Aug 2009 22:16:00 -0000 References: X-IsSubscribed: yes Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-owner@gcc.gnu.org X-SW-Source: 2009-08/txt/msg00304.txt.bz2 On Aug 17, 2009, at 10:50 AM, Bernd Roesch wrote: >> [..] >> There is no need for much detail, just compile any bigger >> sourcecode with >> -O3 -fno-.. and after that with -O2 .... When the object files are >> not >> identical, something is wrong and I can start a bug report. > > how much is size diffrent ? > > You use > -finline-functions -fpredictive-commoning -fgcse-after-reload > -ftree-vectorize > > right ? > > I see in Gcc source that there is more set on -O3. > flag_ipa_cp_clone = opt3; > if (flag_ipa_cp_clone) > flag_ipa_cp = 1; > > .but i > think you can easy test, compile the ffmpeg, if it give no internal > compiler error with -fno........ then it work. > > here is from source opt.c what is set on opt3 > > opt3 = (optimize >= 3); > flag_predictive_commoning = opt3; > flag_inline_functions = opt3; > flag_unswitch_loops = opt3; > flag_gcse_after_reload = opt3; > flag_tree_vectorize = opt3; > flag_ipa_cp_clone = opt3; > if (flag_ipa_cp_clone) > flag_ipa_cp = 1; That doesn't cover all of it; for instance, complete loop unrolling increases code size more on -O3. That means you run into http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40992 with ffmpeg's asm enabled, but turning it down (--param max- completely-peel-times=2) actually seems slightly faster even with plain C on i386.