From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 22155 invoked by alias); 19 Aug 2009 15:09:26 -0000 Received: (qmail 22145 invoked by uid 22791); 19 Aug 2009 15:09:25 -0000 X-SWARE-Spam-Status: No, hits=-2.7 required=5.0 tests=AWL,BAYES_00,SPF_PASS X-Spam-Check-By: sourceware.org Received: from smtp-out.google.com (HELO smtp-out.google.com) (216.239.33.17) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Wed, 19 Aug 2009 15:09:15 +0000 Received: from spaceape23.eur.corp.google.com (spaceape23.eur.corp.google.com [172.28.16.75]) by smtp-out.google.com with ESMTP id n7JF9CHT017748 for ; Wed, 19 Aug 2009 16:09:12 +0100 Received: from rv-out-0506.google.com (rvbf6.prod.google.com [10.140.82.6]) by spaceape23.eur.corp.google.com with ESMTP id n7JF98lO010002 for ; Wed, 19 Aug 2009 08:09:09 -0700 Received: by rv-out-0506.google.com with SMTP id f6so1300209rvb.1 for ; Wed, 19 Aug 2009 08:09:08 -0700 (PDT) Received: by 10.140.125.18 with SMTP id x18mr3629484rvc.250.1250694548525; Wed, 19 Aug 2009 08:09:08 -0700 (PDT) Received: from localhost.localdomain.google.com (dhcp-172-22-126-192.mtv.corp.google.com [172.22.126.192]) by mx.google.com with ESMTPS id f42sm958183rvb.22.2009.08.19.08.09.07 (version=TLSv1/SSLv3 cipher=RC4-MD5); Wed, 19 Aug 2009 08:09:07 -0700 (PDT) To: "Amker.Cheng" Cc: gcc@gcc.gnu.org Subject: Re: Question about the difference between two instruction scheduling passes References: From: Ian Lance Taylor Date: Wed, 19 Aug 2009 23:26:00 -0000 In-Reply-To: (Amker Cheng's message of "Wed\, 19 Aug 2009 22\:44\:22 +0800") Message-ID: User-Agent: Gnus/5.11 (Gnus v5.11) Emacs/22.3 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-System-Of-Record: true 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/msg00345.txt.bz2 "Amker.Cheng" writes: > I'm currently studying implementation of instruction sched in gcc. > > it is possible to schedule insns directly from queue in case > there is nothing better to do and there are still vacant dispatch slots > in the current cycle. > > Gcc only does this work in the second pass, but what's the point? > Is it wrong or just not necessary in the first sched pass? The first scheduling pass runs before register allocation and reload so it's going to get generally trashed anyhow. There no point to being very precise in the first pass, it's just a rough approximation. Ian