From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 26891 invoked by alias); 25 Feb 2008 22:30:21 -0000 Received: (qmail 26880 invoked by uid 22791); 25 Feb 2008 22:30:21 -0000 X-Spam-Check-By: sourceware.org Received: from e36.co.us.ibm.com (HELO e36.co.us.ibm.com) (32.97.110.154) by sourceware.org (qpsmtpd/0.31) with ESMTP; Mon, 25 Feb 2008 22:29:54 +0000 Received: from d03relay04.boulder.ibm.com (d03relay04.boulder.ibm.com [9.17.195.106]) by e36.co.us.ibm.com (8.13.8/8.13.8) with ESMTP id m1PMTos0004046 for ; Mon, 25 Feb 2008 17:29:50 -0500 Received: from d03av02.boulder.ibm.com (d03av02.boulder.ibm.com [9.17.195.168]) by d03relay04.boulder.ibm.com (8.13.8/8.13.8/NCO v8.7) with ESMTP id m1PMToG4197468 for ; Mon, 25 Feb 2008 15:29:50 -0700 Received: from d03av02.boulder.ibm.com (loopback [127.0.0.1]) by d03av02.boulder.ibm.com (8.12.11.20060308/8.13.3) with ESMTP id m1PMToTM029743 for ; Mon, 25 Feb 2008 15:29:50 -0700 Received: from [192.168.1.102] (vorma.rchland.ibm.com [9.10.86.174]) by d03av02.boulder.ibm.com (8.12.11.20060308/8.12.11) with ESMTP id m1PMTkeM029533; Mon, 25 Feb 2008 15:29:49 -0700 Subject: Re: [PATCH] PR35371 GCSE loses track of REG_POINTER attribute From: Peter Bergner To: gcc-patches@gcc.gnu.org In-Reply-To: <20080225222624.GA26857@vervain.rchland.ibm.com> References: <20080225222624.GA26857@vervain.rchland.ibm.com> Content-Type: text/plain Date: Mon, 25 Feb 2008 23:13:00 -0000 Message-Id: <1203978585.21677.52.camel@otta> Mime-Version: 1.0 X-Mailer: Evolution 2.12.1 Content-Transfer-Encoding: 7bit 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: 2008-02/txt/msg01236.txt.bz2 On Mon, 2008-02-25 at 16:26 -0600, Peter Bergner wrote: > While tracking down an unrelated performance problem, I noticed that > for the following simple loop, we generate indexed form stores and that > we get the ordering of the operands wrong which causes a slowdown on > POWER6 (see PR28690 for more details). The culprit here is GCSE. Sorry, I forgot to show the loop. It's posted in the PR though: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35371 Peter