From mboxrd@z Thu Jan 1 00:00:00 1970 Received: (from majordomo@localhost) by pauillac.inria.fr (8.7.6/8.7.3) id MAA10489; Wed, 2 Jul 2003 12:26:37 +0200 (MET DST) Received: from concorde.inria.fr (concorde.inria.fr [192.93.2.39]) by pauillac.inria.fr (8.7.6/8.7.3) with ESMTP id MAA09832 for ; Wed, 2 Jul 2003 12:26:35 +0200 (MET DST) Received: from nef.ens.fr (nef.ens.fr [129.199.96.32]) by concorde.inria.fr (8.11.1/8.11.1) with ESMTP id h62AQZf02628 for ; Wed, 2 Jul 2003 12:26:35 +0200 (MET DST) Received: from di.ens.fr (di.ens.fr [129.199.99.1]) by nef.ens.fr (8.12.9/1.01.28121999) with ESMTP id h62AQVko001388 for ; Wed, 2 Jul 2003 12:26:31 +0200 (CEST) Received: from basilic.ens.fr (monniaux@basilic [129.199.99.48]) by di.ens.fr (8.12.9/jb-1.1) id h62AQTjv013391 for ; Wed, 2 Jul 2003 12:26:29 +0200 (MEST) Received: from localhost (monniaux@localhost) by basilic.ens.fr (8.11.0/jb-1.1) id h62AQSX17737 for ; Wed, 2 Jul 2003 12:26:28 +0200 (MEST) X-Authentication-Warning: basilic.ens.fr: monniaux owned process doing -bs Date: Wed, 2 Jul 2003 12:26:27 +0200 (MEST) From: David Monniaux X-Sender: monniaux@basilic.ens.fr To: Liste CAML Subject: Re: [Caml-list] FP's and HyperThreading Processors In-Reply-To: <20030613210751.A2264@pauillac.inria.fr> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT X-Virus-Scanned: by amavisd-milter (http://amavis.org/) X-Spam: no; 0.00; monniaux:01 caml-list:01 allocating:01 12%:99 unboxed:01 floats:01 profile:98 comparison:02 computations:03 symbolic:03 slightly:03 l'ecole:04 profiler:04 seems:05 probably:05 Sender: owner-caml-list@pauillac.inria.fr Precedence: bulk For all that's worth: we run a program doing lots of symbolic computations and allocating lots of small memory blocks (< a few k) on a variety of machines. The P4 performs roughly proportionally to its frequency compared to the P3; the Athlon is slightly faster than the P3. We ran the program under Linux + OProfile. Roughly 12% of the time is spent within the GC, representing roughly 30% of the cache faults. Surprisingly, it seems that accesses to unboxed floats in some comparison procedures take up 10% of the time and 4% of the page faults. "Profile - don't speculate.". You should probably get OProfile or any other kind of profiler that enables you to count events such as cache faults. David Monniaux http://www.di.ens.fr/~monniaux Laboratoire d'informatique de l'École Normale Supérieure, Paris, France ------------------- To unsubscribe, mail caml-list-request@inria.fr Archives: http://caml.inria.fr Bug reports: http://caml.inria.fr/bin/caml-bugs FAQ: http://caml.inria.fr/FAQ/ Beginner's list: http://groups.yahoo.com/group/ocaml_beginners