From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 20450 invoked by alias); 16 Mar 2004 06:33:20 -0000 Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-owner@gcc.gnu.org Received: (qmail 20435 invoked from network); 16 Mar 2004 06:33:13 -0000 Received: from unknown (HELO mail-out3.apple.com) (17.254.13.22) by sources.redhat.com with SMTP; 16 Mar 2004 06:33:13 -0000 Received: from mailgate1.apple.com (a17-128-100-225.apple.com [17.128.100.225]) by mail-out3.apple.com (8.12.11/8.12.11) with ESMTP id i2G6Nu56027946 for ; Mon, 15 Mar 2004 22:23:56 -0800 (PST) Received: from relay2.apple.com (relay2.apple.com) by mailgate1.apple.com (Content Technologies SMTPRS 4.3.6) with ESMTP id ; Mon, 15 Mar 2004 22:23:56 -0800 Received: from [17.112.72.132] ([17.112.72.132]) by relay2.apple.com (8.12.11/8.12.11) with ESMTP id i2G6NrVw018220; Tue, 16 Mar 2004 06:23:54 GMT In-Reply-To: <20040316033313.GA5426@redhat.com> References: <7F249017-7628-11D8-A9FC-003065FC08E4@spies.com> <20040315223807.GA4850@redhat.com> <74CCC879-76D5-11D8-B445-000393A91CAA@apple.com> <20040315232029.GB5053@redhat.com> <1F5C2707-76DD-11D8-B445-000393A91CAA@apple.com> <1079395578.3173.423.camel@localhost.localdomain> <20040316004507.GC5053@redhat.com> <1079406088.3173.426.camel@localhost.localdomain> <20040316033313.GA5426@redhat.com> Mime-Version: 1.0 (Apple Message framework v612) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: <62E6550B-7712-11D8-8789-000393A91CAA@apple.com> Content-Transfer-Encoding: 7bit Cc: "gcc@gcc.gnu.org list" , Diego Novillo From: Devang Patel Subject: Re: [lno] [RFC] if-conversion and auto vectorizer Date: Tue, 16 Mar 2004 06:33:00 -0000 To: Richard Henderson X-SW-Source: 2004-03/txt/msg00780.txt.bz2 On Mar 15, 2004, at 7:33 PM, Richard Henderson wrote: > One thing that might should be checked before marking various > variables for ssa-renaming, however, is that you get > > if (x_1 < y_1) > x_2 = a, y_2 = b; > x_3 = PHI(x_1, x_2) > y_3 = PHI(y_1, y_2) > ==> > x_3 = (x_1 < y_1 ? a : x_1) > y_3 = (x_1 < y_1 ? b : y_1) > > Note that x_1 outlives the creation of x_3, which is new to > this transformed code. This mere fact might argue for the > creation of a boolean temporary for the comparison. Yes. I also need to simplify boolean exp. -- Devang