caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: Xavier Leroy <xavier.leroy@inria.fr>
To: Richard Cole <rcole@itee.uq.edu.au>
Cc: caml-list@inria.fr
Subject: Re: [Caml-list] Indeterminate Initialization: Is it a Bug?
Date: Thu, 1 Apr 2004 15:37:46 +0200	[thread overview]
Message-ID: <20040401153746.B6842@pauillac.inria.fr> (raw)
In-Reply-To: <406BDB2E.2000807@itee.uq.edu.au>; from rcole@itee.uq.edu.au on Thu, Apr 01, 2004 at 07:04:46PM +1000

> There seems to be a difference in initialization behaviour using the 
> native compiler depending on whether a module contains non "external" 
> calls. [...]

> The output doesn't contain "Funky.Init". The problem is that the 
> initialization code in the funky module didn't get called. If I wrap the 
> functions as in:
> 
> #!/bin/bash
> echo '
>   external ml_funky: int -> int = "ml_funky" ;;
>   let funky x = ml_funky x ;;
>   Printf.fprintf stderr "Init.\n"; flush stderr ;;
> ' > funky.ml

A simpler solution is to define "funky" as an external in the .ml file
but declare it as a regular value in the .mli file:

.ml:  external funky: int -> int = "ml_funky"
.mli: val funky: int -> int

> So I have the following question: Is there some way to guarrentee that 
> modules are initialized exactly once? Does the development team see the 
> current situation, whereby intialization depends on wheher there is a 
> non external method in the module that is called, as a problem? Can it 
> be fixed?

The behavior you observe is a consequence of the link-time elimination
of unreferenced .cma or .cmxa members.  An "external" declaration in a
module interface is like a type declaration: using it elsewhere
doesn't create a reference to the implementation of the defining module.

You can turn off this link-time elimination using the -linkall flag,
either at library creation time (ocamlmklib), or at link-time (ocamlopt).
Then, all modules will be initialized exactly once and in the order
given, like you expect.

- Xavier Leroy

-------------------
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


      reply	other threads:[~2004-04-01 13:39 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-04-01  9:04 Richard Cole
2004-04-01 13:37 ` Xavier Leroy [this message]

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=20040401153746.B6842@pauillac.inria.fr \
    --to=xavier.leroy@inria.fr \
    --cc=caml-list@inria.fr \
    --cc=rcole@itee.uq.edu.au \
    /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).