caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: Pierre Weis <Pierre.Weis@inria.fr>
To: jjgarcia@ind-cr.uclm.es (Juan Jose Garcia Ripoll)
Cc: caml-list@inria.fr
Subject: Re: Size of arrays
Date: Sun, 24 Jan 1999 23:31:42 +0100 (MET)	[thread overview]
Message-ID: <199901242231.XAA12515@pauillac.inria.fr> (raw)
In-Reply-To: <36A9D5F0.F9215F45@ind-cr.uclm.es> from "Juan Jose Garcia Ripoll" at Jan 23, 99 03:00:16 pm

> I've noticed that OCaml imposes a very small upper limit in the size of
> arrays: 8Mb or so, I think. Can this limit be modified somehow?
[...]

No there is no way to modify this limit: it is wired into the data
structures representation algorithm. However, this limit is machine
dependant, and very high on a 64bits machine
(18014398509481983).

Anyway, it is not difficult to design an ADT that gives you large
arrays on a 32 bits machine:

(* Module Bvect of big vectors (length up-to max_int).
   Access and assigment 2 times slowlier than regular arrays.
   Iteration is as efficient as Array.iter.  *)
(* bvect.mli *)
type 'a bvect;;
val make : int -> 'a -> 'a bvect;;
val length : 'a bvect -> int;;
val item : 'a bvect -> int -> 'a;;
val assign : 'a bvect -> int -> 'a -> unit;;
....

(* bvect.ml *)
type 'a bvect = 'a array * 'a array;;

let make len init =
 if len <= Sys.max_array_length then (Array.make len init, [| |])
 else (Array.make Sys.max_array_length init,
       Array.make (len - Sys.max_array_length) init);;

let length (v1, v2) = Array.length v1 + Array.length v2;;

let item (v1, v2) i =
 if i < Sys.max_array_length then v1.(i)
 else v2.(i - Sys.max_array_length);;

let assign (v1, v2) i item =
 if i < Sys.max_array_length then v1.(i) <- item
 else v2.(i - Sys.max_array_length) <- item;;

let iter f (v1, v2) = Array.iter f v1; Array.iter f v2;;

(* Omitted, sub_vect, map_vect, ..., as in module Array. *)

Hope this helps,

Pierre Weis

INRIA, Projet Cristal, Pierre.Weis@inria.fr, http://cristal.inria.fr/~weis/





      reply	other threads:[~1999-01-24 22:32 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
1999-01-22 19:10 Array interface question Brian Rogoff
1999-01-22 19:21 ` Pierre Weis
1999-01-22 20:05   ` Brian Rogoff
1999-01-23 10:47   ` Anton Moscal
1999-01-23 10:57   ` David Monniaux
1999-01-24 15:44     ` Jerome Vouillon
1999-01-24 21:36       ` Pierre Weis
1999-01-28 19:57         ` Brian Rogoff
1999-01-23 14:00   ` Size of arrays Juan Jose Garcia Ripoll
1999-01-24 22:31     ` Pierre Weis [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=199901242231.XAA12515@pauillac.inria.fr \
    --to=pierre.weis@inria.fr \
    --cc=caml-list@inria.fr \
    --cc=jjgarcia@ind-cr.uclm.es \
    /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).