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=-1.9 required=5.0 tests=BAYES_00 autolearn=ham
	autolearn_force=no version=3.4.4
X-Google-Thread: a07f3367d7,3737542e2ed2b8e1
X-Google-Attributes: gida07f3367d7,public,usenet
X-Google-NewGroupId: yes
X-Google-Language: ENGLISH,UTF8
Path: g2news2.google.com!news2.google.com!news.glorb.com!aioe.org!nospam
From: "John B. Matthews" <nospam@nospam.invalid>
Newsgroups: comp.lang.ada
Subject: Re: Problem with optimizations
Date: Tue, 28 Apr 2009 15:15:47 -0400
Organization: The Wasteland
Message-ID: <nospam-ACB1DD.15154628042009@news.aioe.org>
References: <49f73730$0$2850$ba620e4c@news.skynet.be> <gt7f1s$t86$1@online.de>
 <49f742d0$0$2851$ba620e4c@news.skynet.be>
NNTP-Posting-Host: ib4TTflHUauJidfWP/+Rjw.user.aioe.org
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
X-Complaints-To: abuse@aioe.org
NNTP-Posting-Date: Tue, 28 Apr 2009 19:16:01 +0000 (UTC)
X-Notice: Filtered by postfilter v. 0.7.7
Cancel-Lock: sha1:l0doQNpRo1NLGT4LojRKYI+MqO4=
User-Agent: MT-NewsWatcher/3.5.3b3 (Intel Mac OS X)
Xref: g2news2.google.com comp.lang.ada:5591
Date: 2009-04-28T15:15:47-04:00
List-Id: <comp.lang.ada>

In article <49f742d0$0$2851$ba620e4c@news.skynet.be>,
 Olivier Scalbert <olivier.scalbert@algosyn.com> wrote:

> Albrecht Käfer wrote:
> 
> > Works fine on Windows (GNAT GPL 2009).
> > However, I can't help but notice that you are doing premature
> > optimization. Did you *test* if it is faster?
> > 
> > 
> > Albrecht
> 
> With -O3 also ?
> 
> It is not premature optimization. The difference between Is_Solved and 
> Is_Solved1 is quite important on my system:
> 
> with -O2 and Is_solved:
> time ./rubikmain
> real    0m2.164s
> user    0m2.160s
> sys     0m0.000s
> 
> with -O2 and Is_solved[1]:
> time ./rubikmain
> real    0m3.604s
> user    0m3.596s
> sys     0m0.000s
> 
> Olivier

Using FSF-GNAT[1], I was able to reproduce the -O3 failure. This version 
of Is_Solved seems to work at -O3, and it is similarly faster:

    function Is_Solved(Cube: Cube_T) return Boolean is
    begin
        for i in Face_Index_T'range loop
            for j in Face_T'range(1) loop
                for k in Face_T'range(2) loop   
                    if Cube(i)(j,k) /= Final_Position(i)(j,k) then
                        return False;
                    end if;
                end loop;
            end loop;
        end loop;
        return True;
    end Is_Solved;

This version of Is_Solved1 was not measurably faster:

    function Is_Solved1(Cube: Cube_T) return Boolean is
    begin
        return Cube = Final_Position;
    end Is_Solved1;

[1] GNAT 4.3.4 20090225 (prerelease) [gcc-4_3-branch revision 143854]
Copyright 1996-2007, Free Software Foundation, Inc.

-- 
John B. Matthews
trashgod at gmail dot com
<http://sites.google.com/site/drjohnbmatthews>