From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 15069 invoked by alias); 21 Nov 2003 10:42:27 -0000 Mailing-List: contact gcc-help-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-help-owner@gcc.gnu.org Received: (qmail 15060 invoked from network); 21 Nov 2003 10:42:25 -0000 Received: from unknown (HELO mx5.informatik.uni-tuebingen.de) (134.2.12.32) by sources.redhat.com with SMTP; 21 Nov 2003 10:42:25 -0000 Received: from localhost (loopback [127.0.0.1]) by mx5.informatik.uni-tuebingen.de (Postfix) with ESMTP id 90D98125; Fri, 21 Nov 2003 11:42:24 +0100 (NFT) Received: from mx3.informatik.uni-tuebingen.de ([134.2.12.26]) by localhost (mx5 [134.2.12.32]) (amavisd-new, port 10024) with ESMTP id 16134-01; Fri, 21 Nov 2003 11:42:23 +0100 (NFT) Received: from juist (semeai [134.2.15.66]) by mx3.informatik.uni-tuebingen.de (Postfix) with ESMTP id 9746313B; Fri, 21 Nov 2003 11:42:22 +0100 (NFT) Received: from falk by juist with local (Exim 3.36 #1 (Debian)) id 1AN8k1-0000KN-00; Fri, 21 Nov 2003 11:42:21 +0100 To: "Kaspar Thommen" Cc: gcc-help@gcc.gnu.org Subject: Re: g++ V3.3.1 not faster with __restrict References: <24123.1069410397@www48.gmx.net> From: Falk Hueffner Date: Fri, 21 Nov 2003 10:42:00 -0000 In-Reply-To: <24123.1069410397@www48.gmx.net> Message-ID: <87he0y0zqq.fsf@student.uni-tuebingen.de> User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.5 (cabbage) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Virus-Scanned: by amavisd-new (McAfee AntiVirus) at informatik.uni-tuebingen.de X-SW-Source: 2003-11/txt/msg00280.txt.bz2 "Kaspar Thommen" writes: > To my surprise, the __restrict version is not faster than the conventional > one. In fact, they show the same speed. Do I have to specify any compilation > option to tell g++ to really take advantage from the __restrict information? No, although probably -O2 is required to show any difference. > Didn't find anything in the docs however. I tried using -o, -o2, > -o3, giving always equal speed for both functions. Try -funroll-loops or -funroll-all-loops, this produces more possibilities to take advantage of the non-aliasing. -- Falk