From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-patches-return-270227-listarch-gcc-patches=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 19017 invoked by alias); 10 Aug 2010 16:13:45 -0000
Received: (qmail 19000 invoked by uid 22791); 10 Aug 2010 16:13:42 -0000
X-SWARE-Spam-Status: No, hits=-1.9 required=5.0	tests=AWL,BAYES_00,T_RP_MATCHES_RCVD
X-Spam-Check-By: sourceware.org
Received: from one.firstfloor.org (HELO one.firstfloor.org) (213.235.205.2)    by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Tue, 10 Aug 2010 16:13:37 +0000
Received: from basil.firstfloor.org (p5B3C956D.dip0.t-ipconnect.de [91.60.149.109])	by one.firstfloor.org (Postfix) with ESMTP id 65B9E1A9804F;	Tue, 10 Aug 2010 18:13:35 +0200 (CEST)
Received: by basil.firstfloor.org (Postfix, from userid 1000)	id D9D86B2504; Tue, 10 Aug 2010 18:13:34 +0200 (CEST)
From: Andi Kleen <andi@firstfloor.org>
To: Steven Bosscher <stevenb.gcc@gmail.com>
Cc: Richard Guenther <richard.guenther@gmail.com>,  Bernd Schmidt <bernds@codesourcery.com>,  GCC Patches <gcc-patches@gcc.gnu.org>
Subject: Re: Combine four insns
References: <4C5C20D0.5020105@codesourcery.com>	<AANLkTimmPSgpS=1RcHqZD94=yN+kT-MqNd9-i2_Ywc=t@mail.gmail.com>	<4C5C6BAA.4030703@codesourcery.com>	<AANLkTikyui9waDT8hv_30xRtBHOvb85phZhbVaO+uEs3@mail.gmail.com>	<4C6162D4.6080605@codesourcery.com>	<AANLkTim=8XK9ZVjzvJBKerQtuCB3qfsk_zR_30PcQqdn@mail.gmail.com>	<AANLkTik0r-56QNrQ-QLf4SF0r1K2WWwR=0TX4OjhSvhG@mail.gmail.com>
Date: Tue, 10 Aug 2010 16:27:00 -0000
In-Reply-To: <AANLkTik0r-56QNrQ-QLf4SF0r1K2WWwR=0TX4OjhSvhG@mail.gmail.com>	(Steven Bosscher's message of "Tue, 10 Aug 2010 17:05:28 +0200")
Message-ID: <8739um8mz5.fsf@basil.nowhere.org>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-patches.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-patches/>
List-Post: <mailto:gcc-patches@gcc.gnu.org>
List-Help: <mailto:gcc-patches-help@gcc.gnu.org>
Sender: gcc-patches-owner@gcc.gnu.org
X-SW-Source: 2010-08/txt/msg00773.txt.bz2

Steven Bosscher <stevenb.gcc@gmail.com> writes:

> On Tue, Aug 10, 2010 at 4:39 PM, Richard Guenther
> <richard.guenther@gmail.com> wrote:
>> You could enable the heuristic with optimize < 3 && !optimize_size
>> (thus keep combining everything at -O3 and -Os).
>
> Better s/optimize_size/optimize_bb_for_size(BLOCK_FOR_INSN (i0))/

That would mean that a -O2 build where all BBs are cold for some
reason would be 1% slower, right?

To be honest I didn't fully understand why it's ok for -Os 
to be 1% slower. A lot of people use -Os.

It seems more like a -O3 only thing for me.

-Andi

-- 
ak@linux.intel.com -- Speaking for myself only.