caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: Basile STARYNKEVITCH <basile@starynkevitch.net>
To: Carlos Pita <cpitadev@yahoo.com.ar>
Cc: caml-list@yquem.inria.fr
Subject: Re: [Caml-list] Why + vs +. but "fake" parametric polymorphism for <
Date: Thu, 12 Oct 2006 07:49:39 +0200	[thread overview]
Message-ID: <20061012054939.GA31702@ours.starynkevitch.net> (raw)
In-Reply-To: <1160631680.7649.27.camel@monad>

Le Thu, Oct 12, 2006 at 02:41:20AM -0300, Carlos Pita écrivait/wrote:
> Umh, soon after writing my previous post I found out an interesting
> chapter in the tutorial giving an exact example of asm code generated by
> the compiler for comparison operator <:
> 
> # let max a b =
>   if a > b then a else b;;
> val max : 'a -> 'a -> 'a = <fun>
> 
> ===>
>         [...]
>         ; Call the C "greaterthan" function (in the OCaml library).
>         pushl   %ebx
>         pushl   %eax
>         movl    $greaterthan, %eax
>         call    caml_c_call
> .L102:
>         addl    $8, %esp
>         ; If the C "greaterthan" function returned 1, jump to .L100
>         [...]
> 
> Pretty expensive for a simple int comparison I would say.


If you explicitly type it as an integer compare 
   let max (a : int) b = if a > b then a else b;;
you get much more efficient code (AMD64 linux ocaml 3.09.2 Debian/Sid)


        .globl  camlMaxi__max_57
camlMaxi__max_57:
.L101:
        cmpq    %rbx, %rax
        jle     .L100
        ret
        .align  4
.L100:
        movq    %rbx, %rax
        ret
        .text

-- 
Basile STARYNKEVITCH         http://starynkevitch.net/Basile/ 
email: basile<at>starynkevitch<dot>net 
aliases: basile<at>tunes<dot>org = bstarynk<at>nerim<dot>net
8, rue de la Faïencerie, 92340 Bourg La Reine, France


  reply	other threads:[~2006-10-12  5:49 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-10-12  5:19 Carlos Pita
2006-10-12  5:41 ` [Caml-list] " Carlos Pita
2006-10-12  5:49   ` Basile STARYNKEVITCH [this message]
2006-10-12  5:53   ` Jonathan Roewen
2006-10-12  6:10     ` Carlos Pita
  -- strict thread matches above, loose matches on Subject: below --
2006-10-12  5:18 Carlos Pita
2006-10-12  5:45 ` [Caml-list] " Jacques Garrigue
2006-10-12  5:58   ` Carlos Pita
2006-10-12  6:08     ` Jonathan Roewen
     [not found]       ` <452DF46C.802@fmf.uni-lj.si>
2006-10-12 14:26         ` Carlos Pita
2006-10-13 11:56       ` Diego Olivier FERNANDEZ PONS
2006-10-13 12:14         ` Gerd Stolpmann
2006-10-13 12:46           ` Diego Olivier FERNANDEZ PONS
2006-10-13 13:01             ` Luc Maranget
2006-10-13 13:15               ` Diego Olivier FERNANDEZ PONS
2006-10-13 13:15               ` skaller
2006-10-13 13:36                 ` Luc Maranget
2006-10-13 13:53             ` Gerd Stolpmann
2006-10-13 14:16               ` Luc Maranget

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20061012054939.GA31702@ours.starynkevitch.net \
    --to=basile@starynkevitch.net \
    --cc=caml-list@yquem.inria.fr \
    --cc=cpitadev@yahoo.com.ar \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).