From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <daniel.buenzli@erratique.ch>
X-Original-To: caml-list@sympa.inria.fr
Delivered-To: caml-list@sympa.inria.fr
Received: from mail4-relais-sop.national.inria.fr (mail4-relais-sop.national.inria.fr [192.134.164.105])
	by sympa.inria.fr (Postfix) with ESMTPS id 095B67ED26
	for <caml-list@sympa.inria.fr>; Thu,  7 Jun 2012 19:34:58 +0200 (CEST)
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: AtABAGDl0E9KN1ZKnGdsb2JhbABFhU6udQEBAQEBCAsJCRQngkKBCwImAkkWG4gDBJdujj6SaIEjjnQyYAOaXRONCQ
X-IronPort-AV: E=Sophos;i="4.75,732,1330902000"; 
   d="scan'208";a="146947420"
Received: from mail6.webfaction.com (HELO smtp.webfaction.com) ([74.55.86.74])
  by mail4-smtp-sop.national.inria.fr with ESMTP; 07 Jun 2012 19:34:57 +0200
Received: from [192.168.0.101] (121-233.197-178.cust.bluewin.ch [178.197.233.121])
	by smtp.webfaction.com (Postfix) with ESMTP id EABF621078E6
	for <caml-list@inria.fr>; Thu,  7 Jun 2012 12:34:54 -0500 (CDT)
Date: Thu, 7 Jun 2012 19:34:49 +0200
From: =?utf-8?Q?Daniel_B=C3=BCnzli?= <daniel.buenzli@erratique.ch>
To: caml-list <caml-list@inria.fr>
Message-ID: <CE0EBC4C252E4B178BD00227099BA4D4@erratique.ch>
X-Mailer: sparrow 1.6.1 (build 1081.52)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
Subject: [Caml-list] Local functions with arguments or closures ?

Hello, 

In the past I remember having indirectly benchmarked two different styles for writing local functions :

    let f x = 
      let v = ... in
      let rec loop x v = ... in 
      loop x v 

or 

    let f x = 
        let v = ... in
        let rec loop () = ... (* uses v and x *) in 
        loop ()

without being able to reach a conclusion. Is there any particular style that is definitively faster ? Does it maybe depend on the number of arguments and/or on whether loop is recursive or not ? 

This question keeps coming back in my mind when I write local functions... I'm sure someone with some knowledge of the compiler's internals can provide a more reasonable answer than benchmarks.  

Best,

Daniel