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,FREEMAIL_FROM
	autolearn=unavailable autolearn_force=no version=3.4.4
X-Received: by 2002:a24:ce41:: with SMTP id v62mr2542531itg.96.1559871726199;
        Thu, 06 Jun 2019 18:42:06 -0700 (PDT)
X-Received: by 2002:a9d:7741:: with SMTP id t1mr8802057otl.178.1559871725979;
 Thu, 06 Jun 2019 18:42:05 -0700 (PDT)
Path: 
 eternal-september.org!reader01.eternal-september.org!feeder.eternal-september.org!news.gegeweb.eu!gegeweb.org!usenet-fr.net!proxad.net!feeder1-2.proxad.net!209.85.166.216.MISMATCH!g15no129607itd.0!news-out.google.com!l126ni201itl.0!nntp.google.com!g15no129605itd.0!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail
Newsgroups: comp.lang.ada
Date: Thu, 6 Jun 2019 18:42:05 -0700 (PDT)
In-Reply-To: <55b14350-e255-406c-ab11-b824da77995b@googlegroups.com>
Complaints-To: groups-abuse@google.com
Injection-Info: glegroupsg2000goo.googlegroups.com;
 posting-host=87.112.169.186;
 posting-account=Jzt5lQoAAAB4PhTgRLOPGuTLd_K1LY-C
NNTP-Posting-Host: 87.112.169.186
References: <55b14350-e255-406c-ab11-b824da77995b@googlegroups.com>
User-Agent: G2/1.0
MIME-Version: 1.0
Message-ID: <ed2964f8-ce39-4307-9740-28ac60ae3836@googlegroups.com>
Subject: Re: Toy computational "benchmark" in Ada (new blog post)
From: johnscpg@googlemail.com
Injection-Date: Fri, 07 Jun 2019 01:42:06 +0000
Content-Type: text/plain; charset="UTF-8"
Xref: reader01.eternal-september.org comp.lang.ada:56516
Date: 2019-06-06T18:42:05-07:00
List-Id: <comp.lang.ada>

On Thursday, June 6, 2019 at 12:05:02 PM UTC+1, David Trudgett wrote:
> For what it's worth, you may exercise your laughing gear perusing my toy example just for fun at: 
> 
> http://www.eclecticse.com.au/2019/06/procedural-map-reduce-and-parallelism.html
> 
> Any tips or suggestions from Ada old hands out there for other ideas to try would be gratefully considered! :-)
> 
> Enjoy!

On my machine I get a nice improvement over -O3 when I
take the arrays off the heap, and then use the following 2 flags:

   -march=native -funroll-loops

Modifying the programs is easy:

 --Values_Array : Values_Array_Access := new Values_Array_Type;
   Values_Array : Values_Array_Type;

In the parallel version, change the loop in the task body:

--       declare
--          Val : Float64 renames Values_Array (Idx);
--       begin
            My_Sum := My_Sum + Values_Array (Idx) ** 2;
--       end;

The  -funroll-loops  gave me a nice improvement on the parallel
program, less so on the serial version. (Makes no sense to me
at all!) If you are running in a Unix shell, you usually need
to tell the system if you're going to put giant arrays on the
stack. I type this on the command line: ulimit -s unlimited.

Jonathan