From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 445 invoked by alias); 28 Oct 2010 21:14:23 -0000 Received: (qmail 437 invoked by uid 22791); 28 Oct 2010 21:14:22 -0000 X-SWARE-Spam-Status: No, hits=-2.0 required=5.0 tests=AWL,BAYES_00,DKIM_SIGNED,DKIM_VALID,FREEMAIL_FROM,RCVD_IN_DNSWL_NONE X-Spam-Check-By: sourceware.org Received: from mail-fx0-f47.google.com (HELO mail-fx0-f47.google.com) (209.85.161.47) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Thu, 28 Oct 2010 21:14:18 +0000 Received: by fxm3 with SMTP id 3so2461697fxm.20 for ; Thu, 28 Oct 2010 14:14:16 -0700 (PDT) Received: by 10.223.69.140 with SMTP id z12mr147988fai.142.1288300456221; Thu, 28 Oct 2010 14:14:16 -0700 (PDT) Received: from yakj.usersys.redhat.com (s209p8.home.99maxprogres.cz [85.93.118.17]) by mx.google.com with ESMTPS id b15sm723350fah.4.2010.10.28.14.14.14 (version=TLSv1/SSLv3 cipher=RC4-MD5); Thu, 28 Oct 2010 14:14:15 -0700 (PDT) Message-ID: <4CC9E7A5.8040404@gnu.org> Date: Fri, 29 Oct 2010 01:04:00 -0000 From: Paolo Bonzini User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.9) Gecko/20100907 Fedora/3.1.3-1.fc13 Lightning/1.0b3pre Mnenhy/0.8.3 Thunderbird/3.1.3 MIME-Version: 1.0 Newsgroups: gmane.comp.gcc.patches To: Eric Botcazou CC: Tom de Vries , gcc-patches@gcc.gnu.org, Bernd Schmidt Subject: Re: new sign/zero extension elimination pass References: <4CBC698B.3080204@codesourcery.com> <201010221030.30363.ebotcazou@adacore.com> In-Reply-To: <201010221030.30363.ebotcazou@adacore.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit 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: 2010-10/txt/msg02450.txt.bz2 On 10/22/2010 10:30 AM, Eric Botcazou wrote: > If not, why does it not use > the DF framework instead of recomputing DU chains manually? Rather than on DU chains (which are a very expensive problem just because of its asymptotic complexity, so it's better to use it only on small regions such as loops), I'd be picky about usage of note_uses, which can be very simply replaced by a loop over the DF_INSN_USES vector. Paolo From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 1020 invoked by alias); 28 Oct 2010 21:14:34 -0000 Received: (qmail 1011 invoked by uid 22791); 28 Oct 2010 21:14:33 -0000 X-SWARE-Spam-Status: No, hits=-2.0 required=5.0 tests=AWL,BAYES_00,SPF_HELO_PASS,T_RP_MATCHES_RCVD X-Spam-Check-By: sourceware.org Received: from lo.gmane.org (HELO lo.gmane.org) (80.91.229.12) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Thu, 28 Oct 2010 21:14:30 +0000 Received: from list by lo.gmane.org with local (Exim 4.69) (envelope-from ) id 1PBZnk-0002eu-OK for gcc-patches@gcc.gnu.org; Thu, 28 Oct 2010 23:14:24 +0200 Received: from s209p8.home.99maxprogres.cz ([85.93.118.17]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Thu, 28 Oct 2010 23:14:24 +0200 Received: from bonzini by s209p8.home.99maxprogres.cz with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Thu, 28 Oct 2010 23:14:24 +0200 To: gcc-patches@gcc.gnu.org From: Paolo Bonzini Subject: Re: new sign/zero extension elimination pass Date: Fri, 29 Oct 2010 01:33:00 -0000 Message-ID: <4CC9E7A5.8040404@gnu.org> References: <4CBC698B.3080204@codesourcery.com> <201010221030.30363.ebotcazou@adacore.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Tom de Vries , gcc-patches@gcc.gnu.org, Bernd Schmidt User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.9) Gecko/20100907 Fedora/3.1.3-1.fc13 Lightning/1.0b3pre Mnenhy/0.8.3 Thunderbird/3.1.3 In-Reply-To: <201010221030.30363.ebotcazou@adacore.com> 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: 2010-10/txt/msg02451.txt.bz2 Message-ID: <20101029013300.V7WK0ShInCvjI2tUC5Ctomb1FI7n5eEU9P441bHNaNI@z> On 10/22/2010 10:30 AM, Eric Botcazou wrote: > If not, why does it not use > the DF framework instead of recomputing DU chains manually? Rather than on DU chains (which are a very expensive problem just because of its asymptotic complexity, so it's better to use it only on small regions such as loops), I'd be picky about usage of note_uses, which can be very simply replaced by a loop over the DF_INSN_USES vector. Paolo