From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-return-50430-listarch-gcc=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 24807 invoked by alias); 22 Apr 2002 14:11:44 -0000
Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Archive: <http://gcc.gnu.org/ml/gcc/>
List-Post: <mailto:gcc@gcc.gnu.org>
List-Help: <http://gcc.gnu.org/ml/>
Sender: gcc-owner@gcc.gnu.org
Received: (qmail 24798 invoked from network); 22 Apr 2002 14:11:40 -0000
Received: from unknown (HELO mx1.informatik.uni-tuebingen.de) (134.2.12.5)
  by sources.redhat.com with SMTP; 22 Apr 2002 14:11:40 -0000
Received: from semeai.informatik.uni-tuebingen.de (semeai [134.2.15.66])
	by mx1.informatik.uni-tuebingen.de (Postfix) with ESMTP id 87D7B439
	for <gcc@gcc.gnu.org>; Mon, 22 Apr 2002 16:11:39 +0200 (MST)
Received: (from hueffner@localhost)
	by  semeai.informatik.uni-tuebingen.de (8.11.6+Sun/8.10.2) id g3MEBcb08149;
	Mon, 22 Apr 2002 16:11:38 +0200 (MEST)
X-Authentication-Warning: semeai. informatik.uni-tuebingen.de: hueffner set sender to falk.hueffner@student.uni-tuebingen.de using -f
X-Face: >Q)4Pn.JVfRz{G(G_eIkykbZGG\)2mk8:5a"{^Mk07iC#F.t2L7h<Sa|7Zr1_L8[nbiq:8F
 %o\(7>|]{*cFg$GEPDdun~+UTjG(^4z<_Ksw%L-\w0xDmUR~<zsnGH]&sK=M\Y=;U4XZ"z)[CX6I6d
 _p
To: gcc list <gcc@gcc.gnu.org>
Subject: Re: GCC performance regression - up to 20% ?
References: <20020421005718.GA16378@zoy.org>
	<20020421113238.GC16602@atrey.karlin.mff.cuni.cz>
	<20020422061937.GA27171@zoy.org> <20020422155801.A21747@alinoe.com>
From: Falk Hueffner <falk.hueffner@student.uni-tuebingen.de>
Date: Mon, 22 Apr 2002 07:11:00 -0000
In-Reply-To: <20020422155801.A21747@alinoe.com>
Message-ID: <st2pu0rn7qd.fsf@student.uni-tuebingen.de>
User-Agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Acadia)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-SW-Source: 2002-04/txt/msg01082.txt.bz2

Carlo Wood <carlo@alinoe.com> writes:

> 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.

http://oprofile.sourceforge.net/

Only runs on i386 Linux and needs a kernel module, though.

        Falk