From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 27213 invoked by alias); 8 Mar 2008 14:19:09 -0000 Received: (qmail 27203 invoked by uid 22791); 8 Mar 2008 14:19:09 -0000 X-Spam-Check-By: sourceware.org Received: from atrey.karlin.mff.cuni.cz (HELO atrey.karlin.mff.cuni.cz) (195.113.31.123) by sourceware.org (qpsmtpd/0.31) with ESMTP; Sat, 08 Mar 2008 14:18:45 +0000 Received: by atrey.karlin.mff.cuni.cz (Postfix, from userid 4018) id BCB98F007A; Sat, 8 Mar 2008 15:18:42 +0100 (CET) Date: Sat, 08 Mar 2008 14:19:00 -0000 From: Jan Hubicka To: Jan Hubicka Cc: Mark Mitchell , GCC Patches Subject: Re: [PATCH, middle-end] Switch initializations conversion (take four) Message-ID: <20080308141842.GK24550@atrey.karlin.mff.cuni.cz> References: <20080229160939.GA29616@dhcp64.suse.cz> <47C865E9.8050903@codesourcery.com> <20080304202129.GA15391@dhcp64.suse.cz> <20080308141654.GL13246@atrey.karlin.mff.cuni.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20080308141654.GL13246@atrey.karlin.mff.cuni.cz> User-Agent: Mutt/1.5.13 (2006-08-11) 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: 2008-03/txt/msg00534.txt.bz2 > > On Fri, Feb 29, 2008 at 12:07:05PM -0800, Mark Mitchell wrote: > > > Optimization is a quantitative question: the whole point is to make the > > > program smaller, faster, etc. So, all non-obvious optimization patches > > > should come with quantitative information, in addition to qualitative > > > information. > > > > > > In particular, which benchmark(s), on which platform(s), with which flags > > > did you use to measure improvement? And how much improvement did you see? > > > And, what benchmark(s) on which platforms which which flags did you use to > > > measure costs? And how high were the costs? > > > > So far I have only managed to find out how many and what switches get > > converted during GCC bootstrap (this time all languages including ADA) > > on i386 Linux. 177 conversions take place (59 twice and 1 once during > > the whole process). The exact switches as reported by EXPR_LOCATION > > are listed in http://misc.jamborm.net/cswtch.log > > Hi, > I've dropped the patch to periodic tester (vangelis) for last 2 days. > The compile time seems to be off noise (first CPU2000 build was 4 ^^ in ;) > seconds slower, other 40 seconds faster than the build just before that > might be result of mainline change) as are performance results. > > There are some code size improvements, though interestingly GCC > benchmark size grew from 2242->2246Kb (in peak so might be more > inlining). Perl changes 938->935Kb, Mesa 692Kb->688Kb. > > Honza