From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 15126 invoked by alias); 1 Feb 2002 19:12:25 -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 14971 invoked from network); 1 Feb 2002 19:12:19 -0000 Received: from unknown (HELO fep15-svc.tin.it) (212.216.176.46) by sources.redhat.com with SMTP; 1 Feb 2002 19:12:19 -0000 Received: from unitus.it ([80.116.243.88]) by fep15-svc.tin.it (InterMail vM.4.01.03.13 201-229-121-113) with ESMTP id <20020201191234.QTUT24505.fep15-svc.tin.it@unitus.it>; Fri, 1 Feb 2002 20:12:34 +0100 Message-ID: <3C5AE860.2AA390F1@unitus.it> Date: Fri, 01 Feb 2002 11:12:00 -0000 From: Paolo Carlini X-Mailer: Mozilla 4.79 [en] (X11; U; Linux 2.4.16 i686) X-Accept-Language: en MIME-Version: 1.0 To: Richard Henderson CC: gcc@gcc.gnu.org, aj@suse.de Subject: Re: Loop unrolling-related SPEC regressions? References: <3C5ADF19.6A6865D9@unitus.it> <20020201104619.C10318@redhat.com> <3C5AE35B.9B7B825D@unitus.it> <20020201105740.F10318@redhat.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-SW-Source: 2002-02/txt/msg00025.txt.bz2 Richard Henderson wrote: > On Fri, Feb 01, 2002 at 07:50:03PM +0100, Paolo Carlini wrote: > > Otherwise, we should find one somewhere else. Where?? > > There are other benchmarks. Some of them are on gcc.gnu.org > somewhere (there's a link off the web pages). Try them and > see if we regress -funroll-loops. Ok. I will try to do my best during the weekend. > Note that I have no confidence that -funroll-all-loops is a > useful thing to try. You're overriding the logic in the > unroller that tries to decide if the unrolling would pay off. I see. Perhaps we could ask Andreas to help by running an exceptional SPEC test with -funroll-loops instead (ideally, 2 different runs, pre- and post- the unroller patch). Cheers, Paolo.