Rocksolid Light

Welcome to RetroBBS

mail  files  register  newsreader  groups  login

Message-ID:  

We are MicroSoft. You will be assimilated. Resistance is futile. (Attributed to B.G., Gill Bates)


devel / comp.lang.fortran / Optimization options

SubjectAuthor
* Optimization optionsBeliavsky
`- Re: Optimization optionsSteve Lionel

1
Optimization options

<1ab9138a-5594-4cf1-86db-117d6f1b4464n@googlegroups.com>

  copy mid

https://www.rocksolidbbs.com/devel/article-flat.php?id=2190&group=comp.lang.fortran#2190

  copy link   Newsgroups: comp.lang.fortran
X-Received: by 2002:a05:622a:1827:b0:2f3:6d90:1504 with SMTP id t39-20020a05622a182700b002f36d901504mr8443991qtc.268.1651068738537; Wed, 27 Apr 2022 07:12:18 -0700 (PDT)
X-Received: by 2002:a81:9a56:0:b0:2f4:d346:7866 with SMTP id r83-20020a819a56000000b002f4d3467866mr27477405ywg.58.1651068738203; Wed, 27 Apr 2022 07:12:18 -0700 (PDT)
Path: i2pn2.org!i2pn.org!aioe.org!feeder1.feed.usenet.farm!feed.usenet.farm!tr2.eu1.usenetexpress.com!feeder.usenetexpress.com!tr1.iad1.usenetexpress.com!border1.nntp.dca1.giganews.com!nntp.giganews.com!news-out.google.com!nntp.google.com!postnews.google.com!google-groups.googlegroups.com!not-for-mail
Newsgroups: comp.lang.fortran
Date: Wed, 27 Apr 2022 07:12:17 -0700 (PDT)
Injection-Info: google-groups.googlegroups.com; posting-host=173.76.111.99; posting-account=Ac_J3QkAAABih73tf3Yz4sHazwGUM-hW
NNTP-Posting-Host: 173.76.111.99
User-Agent: G2/1.0
MIME-Version: 1.0
Message-ID: <1ab9138a-5594-4cf1-86db-117d6f1b4464n@googlegroups.com>
Subject: Optimization options
From: beliavsky@aol.com (Beliavsky)
Injection-Date: Wed, 27 Apr 2022 14:12:18 +0000
Content-Type: text/plain; charset="UTF-8"
Lines: 7
 by: Beliavsky - Wed, 27 Apr 2022 14:12 UTC

Regarding compiler options to maximize speed I may tweet

The ifort -fast and gfortran -O3 -march=native options can be used to
increase speed. Gfortran -Ofast turns on -ffast-math and can give a
further 2x speedup in some cases, but it can also give incorrect
results and should be used with care. Use gfortran -Og to debug.

That is 274 characters. If there is something wrong, please let me know.

Re: Optimization options

<jcvhi0FospbU1@mid.individual.net>

  copy mid

https://www.rocksolidbbs.com/devel/article-flat.php?id=2191&group=comp.lang.fortran#2191

  copy link   Newsgroups: comp.lang.fortran
Path: i2pn2.org!i2pn.org!news.swapon.de!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail
From: steve@seesignature.invalid (Steve Lionel)
Newsgroups: comp.lang.fortran
Subject: Re: Optimization options
Date: Thu, 28 Apr 2022 09:01:51 -0400
Lines: 28
Message-ID: <jcvhi0FospbU1@mid.individual.net>
References: <1ab9138a-5594-4cf1-86db-117d6f1b4464n@googlegroups.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
X-Trace: individual.net Nm4yiVu9ycNO9WHsnTjsAwIZr2/kIiXv3rWRfN3h3HmblYTaCT
Cancel-Lock: sha1:Ciai4OB+N5kXHpxB37zc000Wv38=
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101
Thunderbird/91.8.1
Content-Language: en-US
In-Reply-To: <1ab9138a-5594-4cf1-86db-117d6f1b4464n@googlegroups.com>
 by: Steve Lionel - Thu, 28 Apr 2022 13:01 UTC

On 4/27/2022 10:12 AM, Beliavsky wrote:
> Regarding compiler options to maximize speed I may tweet
>
> The ifort -fast and gfortran -O3 -march=native options can be used to
> increase speed. Gfortran -Ofast turns on -ffast-math and can give a
> further 2x speedup in some cases, but it can also give incorrect
> results and should be used with care. Use gfortran -Og to debug.
>
> That is 274 characters. If there is something wrong, please let me know.By itself this is mostly OK, but for the ifort users I'll probably add a
comment that -fast also turns on -xHost, and is best used when compiling
and running on the same processor. Like the gfortran option, the things
turned on by -fast can also change results.

The suggestion of using -Og to debug differences caused by high
optimization isn't useful, in my opinion.

I'd also wonder if it's appropriate for these tweets to be about
individual compilers and not the language.

--
Steve Lionel
ISO/IEC JTC1/SC22/WG5 (Fortran) Convenor
Retired Intel Fortran developer/support
Email: firstname at firstnamelastname dot com
Twitter: @DoctorFortran
LinkedIn: https://www.linkedin.com/in/stevelionel
Blog: https://stevelionel.com/drfortran
WG5: https://wg5-fortran.org

1
server_pubkey.txt

rocksolid light 0.9.81
clearnet tor