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=-0.3 required=5.0 tests=BAYES_00,
	REPLYTO_WITHOUT_TO_CC autolearn=no autolearn_force=no version=3.4.4
X-Google-Thread: a07f3367d7,3a6a9f1d654285ba
X-Google-Attributes: gida07f3367d7,public,usenet
X-Google-NewGroupId: yes
X-Google-Language: ENGLISH,ASCII-7-bit
Path: 
 g2news2.google.com!news1.google.com!news.glorb.com!feed.xsnews.nl!border-1.ams.xsnews.nl!feeder.news-service.com!newsfeed.eweka.nl!feeder3.eweka.nl!81.171.88.15.MISMATCH!eweka.nl!lightspeed.eweka.nl!npeer.de.kpn-eurorings.net!npeer-ng0.de.kpn-eurorings.net!newsfeed.arcor.de!newsspool1.arcor-online.net!news.arcor.de.POSTED!not-for-mail
Date: Fri, 14 Aug 2009 00:08:03 +0200
From: Georg Bauhaus <rm.tsoh.plus-bug.bauhaus@maps.futureapps.de>
Reply-To: rm.tsoh+bauhaus@maps.futureapps.de
User-Agent: Thunderbird 2.0.0.22 (Windows/20090605)
MIME-Version: 1.0
Newsgroups: comp.lang.ada
Subject: Re: Ada Shootout program for K-Nucleotide (patches)
References: <4a743343$0$32674$9b4e6d93@newsspool2.arcor-online.net>
 	<4a76cd4f$0$31329$9b4e6d93@newsspool4.arcor-online.net>
 <87bpmxdqfo.fsf@nbi.dk>
 	<b6a1f807-3ff9-4bd9-afc8-6c3c97024d2b@y10g2000prf.googlegroups.com>
 	<wccab2gmyiy.fsf@shell01.TheWorld.com>
 <e4f6ff69-b128-4abc-92cf-e4c08dd77efb@y10g2000prf.googlegroups.com>
 	<2bae762e-0d8a-4389-843a-466e87f59fd1@a37g2000prf.googlegroups.com>
 	<4a786b15$0$30230$9b4e6d93@newsspool1.arcor-online.net>
 <67e76046-62d4-4c0e-bdd8-8d00cdf93bca@l35g2000pra.googlegroups.com>
 	<4a79f712$0$31874$9b4e6d93@newsspool3.arcor-online.net>
 <7a5aea9c-1ade-4d73-98ef-08d226823161@z31g2000yqd.googlegroups.com>
 	<4a7a0826$0$31862$9b4e6d93@newsspool3.arcor-online.net>
 <h5d5aa$qqg$1@news.tornevall.net>
 	<4a7a96a2$0$31867$9b4e6d93@newsspool3.arcor-online.net>
 <4a7b4daa$0$31333$9b4e6d93@newsspool4.arcor-online.net>
 <211a9019-6b96-431d-a634-e399075110ec@t13g2000yqt.googlegroups.com>
In-Reply-To: 
 <211a9019-6b96-431d-a634-e399075110ec@t13g2000yqt.googlegroups.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Message-ID: <4a848ec4$0$30231$9b4e6d93@newsspool1.arcor-online.net>
Organization: Arcor
NNTP-Posting-Date: 14 Aug 2009 00:08:04 CEST
NNTP-Posting-Host: 257be563.newsspool1.arcor-online.net
X-Trace: 
 DXC=A]TRkJG6dK=85[]]\]T081ic==]BZ:af>4Fo<]lROoR1^YC2XCjHcb9CW9^O>US0X9A:ho7QcPOV3X9KmO;B8if;nD@GbE4CPY?
X-Complaints-To: usenet-abuse@arcor.de
Xref: g2news2.google.com comp.lang.ada:7748
Date: 2009-08-14T00:08:04+02:00
List-Id: <comp.lang.ada>

jonathan wrote:
> On Aug 6, 10:39 pm, Georg Bauhaus <rm.tsoh.plus-

> FWIW, I finally found time to experiment with compiler flags
> on my x86_64 GNU/Linux pc. I used GNAT GPL  2009 (gcc 4.3.4).
> 
> fasta   running times:
> 
> with -gnatN         get 20% to 25% speedup (NOT -gnatn)

Based on earlier findings regarding the inlining options,
I had suggested, as a followup to the submission,
to add -gnatn and -gnatN to the flags;
this has already been done a few days ago. (AFAICS, the
effects seem to have been less impressive than I would have
hoped.)

Also, while working on the regexdna program, adding -gnatN
caused exception followed by some valid outputs and then
something that hangs. Whatever this might mean in general.

> with -funroll-loops get 4% to 5% speedup (NOT -funroll-all-loops)
> with -O2 same running time (almost) as -O3
> 
> and if you change
> 
>    Last : Natural := 42;
> 
> to
> 
>    type Unsigned_32 is mod 2**32;
>    Last : Unsigned_32 := 42;
> 
> get 10% speedup! (I bet arithmetic on Natural is 64 bit;
> likely same will happen on some of the shootout benchmarks.)

I'll try this on some machines to which I have access;
next week, though.

> So I'ld use
> 
>  gnatmake -O3 -gnatNp  -funroll-loops  -march=native ...
> 
> Jonathan
> 
> 
> 
> 
> 
> 
> 
>