From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-0.9 required=5.0 tests=BAYES_00,FORGED_GMAIL_RCVD,
	FREEMAIL_FROM autolearn=no autolearn_force=no version=3.4.4
X-Google-Thread: 103376,5395b8f41548bf11
X-Google-NewGroupId: yes
X-Google-Attributes: gida07f3367d7,domainid0,public,usenet
X-Google-Language: ENGLISH,ASCII
Path: 
 g2news1.google.com!postnews.google.com!dq9g2000vbb.googlegroups.com!not-for-mail
From: David Sauvage <sauvage.david@gmail.com>
Newsgroups: comp.lang.ada
Subject: Re: Interfacing Ada multidimensional arrays with Fortran.
Date: Thu, 9 Jun 2011 00:55:02 -0700 (PDT)
Organization: http://groups.google.com
Message-ID: 
 <9fd67d68-191b-4c05-a144-3d2c02e92d2c@dq9g2000vbb.googlegroups.com>
References: 
 <27d4e9af-ffd0-4c12-80ed-b51deea95506@q32g2000yqn.googlegroups.com>
 <m2tycf5fh4.fsf@pushface.org> <m2pqn26aew.fsf@pushface.org>
NNTP-Posting-Host: 197.224.110.104
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
X-Trace: posting.google.com 1307606102 23937 127.0.0.1 (9 Jun 2011 07:55:02
 GMT)
X-Complaints-To: groups-abuse@google.com
NNTP-Posting-Date: Thu, 9 Jun 2011 07:55:02 +0000 (UTC)
Complaints-To: groups-abuse@google.com
Injection-Info: dq9g2000vbb.googlegroups.com; posting-host=197.224.110.104;
 posting-account=VT4k8QoAAAB0zUvVwUYyaKE5TB63pAuh
User-Agent: G2/1.0
X-Google-Web-Client: true
X-Google-Header-Order: HUALESNKRC
X-HTTP-UserAgent: Mozilla/5.0 (X11; U; Linux i686; fr; rv:1.9.2.15)
 Gecko/20110303 Ubuntu/9.10 (karmic) Firefox/3.6.15,gzip(gfe)
Xref: g2news1.google.com comp.lang.ada:19713
Date: 2011-06-09T00:55:02-07:00
List-Id: <comp.lang.ada>

On May 28, 8:45=A0pm, Simon Wright <si...@pushface.org> wrote:
...
> -O1:
> Transposition: 1200 us
> Assignment: =A0 =A0 300 us
>
> -O2:
> Transposition: =A0500 us
> Assignment: =A0 =A0 300 us

Using this testcase [1], here are my results using Intel Atom CPU N270
@ 1.60GHz / Linux (launched with root privilege) :

 -O1:
 Transposition: 2202 us
 Assignment:    1014 us

 -O2:
 Transposition: 2556 us
 Assignment:     885 us

Transposition (using assignment via pragma Convention) seems to be
quicker than Transposition (without pragma Convention).
The reason why pragma Convention (Fortran, Type) is not used in
Interfaces.Fortran... is unknown and seems to give slower compute
time.

[1]
-- gnatmake -f compare.adb -cargs -gnat05 -O2
-- gnatmake -f compare.adb -cargs -gnat05 -O1
with Interfaces.Fortran.BLAS;

with Ada.Text_IO,
     Ada.Calendar,
     Ada.Numerics.Generic_Real_Arrays;

procedure Compare is
   Start, Stop : Ada.Calendar.Time;
   use type Ada.Calendar.Time;

   type Real_Matrix is
     array (Integer range <>, Integer range <>) of
Interfaces.Fortran.Real;
   pragma Convention (Fortran, Real_Matrix);

   package GRA is new Ada.Numerics.Generic_Real_Arrays (
      Interfaces.Fortran.Real);

   Row, Column : constant Positive :=3D 100;
   Iteration   : constant Positive :=3D 10;

   M : Real_Matrix (1 .. Row, 1 .. Column)      :=3D (others =3D> (others
=3D> 2.0));
   pragma Volatile (M);

   MFA, MFB : GRA.Real_Matrix (1 .. Row, 1 .. Column) :=3D (others =3D>
(others =3D> 2.0));
   pragma Volatile (MFA);
   pragma Volatile (MFB);

   use type Interfaces.Fortran.Real;
begin

   Start :=3D Ada.Calendar.Clock;
   for I in 1 .. Iteration loop
      M :=3D Real_Matrix (MFB);
   end loop;
   Stop :=3D Ada.Calendar.Clock;
   Ada.Text_IO.Put_Line
     ("Assignation (Transposition via pragma Convention)" &
      Duration'Image (Stop - Start));

   Start :=3D Ada.Calendar.Clock;
   for I in 1 .. Iteration loop
      MFA :=3D GRA.Transpose (MFB);
   end loop;
   Stop :=3D Ada.Calendar.Clock;
   Ada.Text_IO.Put_Line
     ("Transposition" & Duration'Image (Stop - Start));

end Compare;