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 PAA04882; Mon, 4 Feb 2002 15:12:06 +0100 (MET)
X-Authentication-Warning: pauillac.inria.fr: majordomo set sender to owner-caml-list@pauillac.inria.fr using -f
Received: from nez-perce.inria.fr (nez-perce.inria.fr [192.93.2.78]) by pauillac.inria.fr (8.7.6/8.7.3) with ESMTP id PAA03754 for <caml-list@pauillac.inria.fr>; Mon, 4 Feb 2002 15:12:05 +0100 (MET)
Received: from exchange.cs.cornell.edu (exchange.cs.cornell.edu [128.84.97.8])
	by nez-perce.inria.fr (8.11.1/8.11.1) with ESMTP id g14EC4X18116;
	Mon, 4 Feb 2002 15:12:04 +0100 (MET)
content-class: urn:content-classes:message
Subject: RE: [Caml-list] optimizing functors
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Date: Mon, 4 Feb 2002 09:12:03 -0500
Message-ID: <706871B20764CD449DB0E8E3D81C4D4301E90835@opus.cs.cornell.edu>
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
Thread-Topic: [Caml-list] optimizing functors
Thread-Index: AcGte10AFpZ7kzRjTqW5b4eL2ALnrQACiUYw
From: "Michael Hicks" <mhicks@cs.cornell.edu>
To: "David Monniaux" <David.Monniaux@ens.fr>,
        "Xavier Leroy" <xavier.leroy@inria.fr>
Cc: "Liste CAML" <caml-list@inria.fr>
Sender: owner-caml-list@pauillac.inria.fr
Precedence: bulk

I believe that the MLton compiler, which is a whole-program compiler for =
SML, does this sort of optimization.  See =
http://www.sourcelight.com/MLton/.

Mike

> -----Original Message-----
> From: David Monniaux [mailto:David.Monniaux@ens.fr]
> Sent: Monday, February 04, 2002 7:50 AM
> To: Xavier Leroy
> Cc: Liste CAML
> Subject: Re: [Caml-list] optimizing functors
>=20
>=20
> On Sat, 2 Feb 2002, Xavier Leroy wrote:
>=20
> > Yes, but not across functor applications.  More precisely, functions
> > that are passed through a functor parameter cannot be inlined nor
> > called with the optimized "direct" application scheme, they=20
> always go
> > through the generic "indirect-through-closure" application scheme.
>=20
> I realize that the current scheme of implementing modules as=20
> records is
> (relatively) simple and allows easy separate compilation. However, it
> prevents optimizations, as you said.
>=20
> In code containing many modules consisting of a few small=20
> functions called
> by functions in functors, this lack of optimization may be costly.
>=20
> I was thinking of implementing such functors similarly as C++=20
> templates
> (expanding the functor parameters). Has some work been done on this?
>=20
> =20
> David Monniaux            http://www.di.ens.fr/~monniaux
> Laboratoire d'informatique de l'=C9cole Normale Sup=E9rieure,
> Paris, France
>=20
>=20
> -------------------
> Bug reports: http://caml.inria.fr/bin/caml-bugs  FAQ:=20
http://caml.inria.fr/FAQ/
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/
To unsubscribe, mail caml-list-request@inria.fr  Archives: http://caml.inria.fr