From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 23414 invoked by alias); 7 Mar 2008 08:31:41 -0000 Received: (qmail 23398 invoked by uid 22791); 7 Mar 2008 08:31:41 -0000 X-Spam-Check-By: sourceware.org Received: from fg-out-1718.google.com (HELO fg-out-1718.google.com) (72.14.220.155) by sourceware.org (qpsmtpd/0.31) with ESMTP; Fri, 07 Mar 2008 08:31:21 +0000 Received: by fg-out-1718.google.com with SMTP id d23so335134fga.28 for ; Fri, 07 Mar 2008 00:31:18 -0800 (PST) Received: by 10.86.27.9 with SMTP id a9mr1113642fga.62.1204878678927; Fri, 07 Mar 2008 00:31:18 -0800 (PST) Received: from scientist-2.lan ( [213.140.22.65]) by mx.google.com with ESMTPS id l12sm3404272fgb.8.2008.03.07.00.31.18 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 07 Mar 2008 00:31:18 -0800 (PST) Message-ID: <47D0FD55.9040603@gnu.org> Date: Fri, 07 Mar 2008 08:31:00 -0000 From: Paolo Bonzini User-Agent: Thunderbird 2.0.0.12 (Macintosh/20080213) MIME-Version: 1.0 To: Ian Lance Taylor CC: GCC Patches Subject: Re: PING: PR/17236, improve long long multiply on x86 (middle-end) References: <47BBDFFE.5030001@gnu.org> <47CFAB57.7060802@gnu.org> <47D02E92.5090404@gnu.org> In-Reply-To: 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: 2008-03/txt/msg00461.txt.bz2 Ian Lance Taylor wrote: > Paolo Bonzini writes: > >>> I'm OK with making this sort of change without a clear argument, but >>> only if you're prepared to do performance testing on three or more >>> primary platforms. I'm not OK with making this sort of change with >>> neither an argument nor performance testing. Right now, as far as I >>> know, you just have a single test case which improves. >> I already had SPEC on x86 and it was neutral, except that half of >> SPECfp was perturbated by the machine load. I should be able to rerun >> it sometime next week. Running i386 and x86_64 would not be very >> different (the RA problems are the same -- %eax/%edx for MUL/DIV and >> %cl for shift counts) and I don't have access to other platforms. > > I think you do need some performance results on other platforms. > There are people who can help you get them. IRA branch already gets the register allocation part right. I'll commit only the parts you approved, the local-alloc.c tweak will remain on hold until we see if IRA gets into 4.4 (as I hope). Thanks! Paolo