Re: Efficient Matrix?

[Bill Findlay <yaldnifw@blueyonder.co.uk>]

On 28/12/02 13:00, in article slrnb0r842.2n7.adi@drcomp.erfurt.thur.de,
"Adrian Knoth" <adi@drcomp.erfurt.thur.de> wrote:

> Robert A Duff <bobduff@shell01.TheWorld.com> wrote:
> 
>> Compilers are required to pack tightly in this case (one bit per Boolean
>> component).  I.e., Matrix'Component_Size will be 1 if there's a pragma
>> Pack.  See RM-13.2(9).
> 
[snip]
> 
>> It's got nothing to do with the fact that the thing is heap allocated,
>> nor the fact that the bounds are not known at compile time.
> 
> For the variable A it is even known, but I was not able to determine
> a valuable difference in memory-comsumption.
> 

I get good results as follows, with GNAT 5.00w for MacOS X.
-----------------------------------------------------------
with Ada.Text_IO; use Ada.Text_IO;

procedure bla is

   type unconstrained_type is
        array (Positive range <>, Positive range <>) of Boolean;
   for unconstrained_type'Component_Size use 1;
   pragma Pack (unconstrained_type);

   subtype constrained_type is unconstrained_type (1 .. 10000, 1 .. 5000);
   type constrained_type_ptr is access constrained_type;

   A : constrained_type_ptr := new constrained_type;

begin
   Put_Line (Integer'Image (constrained_type'Component_Size));
   Put_Line (Integer'Image (unconstrained_type'Component_Size));
   Put_Line (Integer'Image (constrained_type'Size));
   Put_Line (Integer'Image (A.all'Size));
end bla;
----------------------------------------------------------------
% ./bla
 1
 8
 50000000
 50000000
----------------------------------------------------------------
-- 
Bill-Findlay chez blue-yonder.co.uk ("-" => "")