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=ham autolearn_force=no version=3.4.4
X-Google-Thread: 103376,103803355c3db607
X-Google-NewGroupId: yes
X-Google-Attributes: gida07f3367d7,domainid0,public,usenet
X-Google-Language: ENGLISH,ASCII-7-bit
Received: by 10.68.241.162 with SMTP id wj2mr1732303pbc.2.1340968489363;
        Fri, 29 Jun 2012 04:14:49 -0700 (PDT)
Path: 
 l9ni33053pbj.0!nntp.google.com!news2.google.com!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail
From: Keean Schupke <keean.schupke@googlemail.com>
Newsgroups: comp.lang.ada
Subject: Re: GNAT (GCC) Profile Guided Compilation
Date: Fri, 29 Jun 2012 04:14:48 -0700 (PDT)
Organization: http://groups.google.com
Message-ID: <609b9a6d-ec5b-48e4-9d31-342d80700dc7@googlegroups.com>
References: <dac2857a-6f74-4ecb-a5d2-f6b73fbd0ecc@googlegroups.com>
 <m2wr2qxul1.fsf@pushface.org>
NNTP-Posting-Host: 82.44.19.199
Mime-Version: 1.0
X-Trace: posting.google.com 1340968489 27808 127.0.0.1 (29 Jun 2012 11:14:49
 GMT)
X-Complaints-To: groups-abuse@google.com
NNTP-Posting-Date: Fri, 29 Jun 2012 11:14:49 +0000 (UTC)
In-Reply-To: <m2wr2qxul1.fsf@pushface.org>
Complaints-To: groups-abuse@google.com
Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=82.44.19.199;
 posting-account=T5Z2vAoAAAB8ExE3yV3f56dVATtEMNcM
User-Agent: G2/1.0
Content-Type: text/plain; charset=ISO-8859-1
Date: 2012-06-29T04:14:48-07:00
List-Id: <comp.lang.ada>

On Friday, 29 June 2012 11:48:10 UTC+1, Simon Wright  wrote:
> Keean Schupke:
> 
> > I am developing a Monte-Carlo simulation engine, and I have
> > implementations in C++ and Ada. All performance figures come from the
> > same reference hardware. Both are built using GCC, and similar
> > optimisation is applied to each (same optimisation level, cpu arch,
> > inlining etc).
> >
> > My C++ implementation achieves 40,000 simulations per second.
> >
> > My Ada port initially achieves 32,000 iterations per second.
> >
> > I refactored the Ada port to use System.Address to access the main
> > data structure (as in the 'C++' version benchmarking showed a clear
> > performance benefit to *ptr_x compared to array[x])
> >
> > The Ada port using System.Address achieves 40,000 simulations per
> > second - so this is looking good.
> 
> Did you try -gnatp (suppress all checks)? (only to be used when you're
> sure the code is correct!)
> 
> You could also use pragma Suppress at the hot spots.



Here are the build scripts for the C++ and Ada versions:



For normal compilation I am using:

gnatmake -march=native -O3 -gnatp -gnatN xxx
g++ -o yyy -std=gnu++0x -march=native -O3 yyy.cpp

The performance compiled like this of the Ada and C++ versions is virtually identical (if anything Ada is slightly faster) at about 40k simulations per second.



For profile guided compilation I am using:

xxx: xxx.adb
	rm -f *.gcda
	gnatmake -march=native -O3 -fprofile-generate -gnatp -gnatN xxx
	./xxx
	touch xxx.adb
	gnatmake -march=native -O3 -fprofile-use -gnatp -gnatN xxx

yyy : yyy.cpp
    rm -f *.gcda
    g++ -o yyy -std=gnu++0x -march=native -O3 -fprofile-generate yyy.cpp
    ./yyy
    touch yyy.cpp
    g++ -o yyy -std=gnu++0x -march=native -O3 -fprofile-use yyy.cpp

With profile guided compilation, Ada is getting 44.5k simulations per second, to C++ at ~55k per second which makes the C++ 25% faster.



Cheers,
Keean.