From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 799 invoked by alias); 22 Apr 2002 14:21:52 -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 30042 invoked from network); 22 Apr 2002 14:20:16 -0000 Received: from unknown (HELO cygnus.com) (205.180.83.203) by sources.redhat.com with SMTP; 22 Apr 2002 14:20:16 -0000 Received: from porcupine.cygnus.com (romulus.sfbay.redhat.com [172.16.27.251]) by runyon.cygnus.com (8.8.7-cygnus/8.8.7) with ESMTP id HAA20258 for ; Mon, 22 Apr 2002 07:20:14 -0700 (PDT) Received: from porcupine.cygnus.com (localhost.localdomain [127.0.0.1]) by porcupine.cygnus.com (8.12.2/8.12.2) with ESMTP id g3MEOSnp014172; Mon, 22 Apr 2002 08:24:30 -0600 Received: from porcupine.cygnus.com (law@localhost) by porcupine.cygnus.com (8.12.2/8.12.2/Submit) with ESMTP id g3MEOQvS014167; Mon, 22 Apr 2002 08:24:26 -0600 To: Carlo Wood cc: Michel LESPINASSE , Jan Hubicka , Andreas Jaeger , gcc list Subject: Re: GCC performance regression - up to 20% ? Reply-To: law@redhat.com From: law@redhat.com In-reply-to: Your message of Mon, 22 Apr 2002 15:58:01 +0200. <20020422155801.A21747@alinoe.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 22 Apr 2002 07:34:00 -0000 Message-ID: <14166.1019485465@porcupine.cygnus.com> X-SW-Source: 2002-04/txt/msg01085.txt.bz2 In message <20020422155801.A21747@alinoe.com>, Carlo Wood writes: > > > > Then I tried to figure out where the slowdown is, using gprof. And > > this is where things get really interesting: gprof tells me that the > > code compiled with 3.1 is faster, but 'time' tells me that the user > > time spent executing that code is higher with 3.1 than with 2.95. I'm > > not sure what to make of this, but I think this might give you some > > clues, so I'll describe it in more detail. I'm not sure what the > > overhead is, but it seems to be right in gprof's blind spot. > > gprof "measures" the time that a function takes by probing > which function the program is executing about every 20 ms. > >From that it builds up a statistical histogram. > > I wish there would be a more precise profiler that uses the > hardware counters. Does anyone know of one? Hmm, I remember > a Subject: line on the PAPI mailinglist that mentioned gprof, > but I deleted it. I think it asked the same question: whether > or not there existed a 'gprof' that used PAPI. You might look at oprofile. jeff