From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Delivered-To: caml-list@yquem.inria.fr Received: from concorde.inria.fr (concorde.inria.fr [192.93.2.39]) by yquem.inria.fr (Postfix) with ESMTP id F27BDBC84 for ; Wed, 30 Mar 2005 11:11:09 +0200 (CEST) Received: from KVIW08.KVI.nl (KVIW08.KVI.nl [129.125.15.28]) by concorde.inria.fr (8.13.0/8.13.0) with ESMTP id j2U8BcCn006775 for ; Wed, 30 Mar 2005 10:11:38 +0200 Received: from KVIR52.KVI.nl ("port 37512"@KVIR52.KVI.nl [129.125.37.116]) by KVI.nl (PMDF V6.2-X17 #30869) with ESMTP id <01LMHYBNB2MSAEL0QU@KVI.nl> for caml-list@yquem.inria.fr; Wed, 30 Mar 2005 10:11:33 +0200 (MET DST) Received: from KVIW13.KVI.nl by KVIR52.KVI.nl (AvMailGate-2.0.2-10) id 18471-0BF008BB; Wed, 30 Mar 2005 10:11:32 +0200 Received: from kvip88 ("port 33040"@KVIP88.KVI.nl [129.125.15.152]) by KVI.nl (PMDF V6.2-X17 #30869) with ESMTP id <01LMHYBJR35QAEL0QU@KVI.nl> for caml-list@yquem.inria.fr; Wed, 30 Mar 2005 10:11:28 +0200 (MET DST) Date: Wed, 30 Mar 2005 10:11:23 +0200 From: "Alexander S. Usov" Subject: Re: [Caml-list] 32- and 64-bit performance In-reply-to: <424A593A.5050608@barettadeit.com> To: caml-list@yquem.inria.fr Message-id: <200503301011.23639.A.S.Usov@kvi.nl> Organization: KVI MIME-version: 1.0 Content-type: text/plain; charset=iso-8859-1 Content-transfer-encoding: 7BIT Content-disposition: inline User-Agent: KMail/1.8 X-AntiVirus: checked by AntiVir MailGate (version: 2.0.2-10; AVE: 6.30.0.7; VDF: 6.30.0.54; host: kvi.nl) References: <200503300340.15874.jon@ffconsultancy.com> <424A593A.5050608@barettadeit.com> X-Miltered: at concorde with ID 424A5F3A.000 by Joe's j-chkmail (http://j-chkmail.ensmp.fr)! X-Spam: no; 0.00; caml-list:01 baretta:01 ocamlopt:01 46,:98 wrote:01 wrote:01 slower:01 alexander:02 alexander:02 simplest:02 benchmark:02 alex:03 generally:03 compiled:04 suppose:05 X-Spam-Checker-Version: SpamAssassin 3.0.2 (2004-11-16) on yquem.inria.fr X-Spam-Status: No, score=0.0 required=5.0 tests=none autolearn=disabled version=3.0.2 X-Spam-Level: On Wednesday 30 March 2005 09:46, Alex Baretta wrote: > Jon Harrop wrote: > > So ocamlopt does seem to generate significantly better code in these > > examples, particularly when they are floating point intensive. Also, only > > one test is slower in 64-bit, due to its heavy use of trees. > > Why do you suppose is there *any* benchmark faster in 32 bit mode than > in 64 bit mode on the Athlon64? Since the AMD64 architecture is > generally better than IA32--were it only for the additional registers--I > would expect all benchmarks to run as fast or faster when compiled to > the AMD64 instruction set. A one of the simplest reasons -- somewhat increased memory use. And memory access is incredibly expensive novadays. -- Best regards, Alexander.