[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gcc-fortran
Subject:    Re: Comments on gfortran performance
From:       Toon Moene <toon () moene ! indiv ! nluug ! nl>
Date:       2006-05-31 18:28:38
Message-ID: E1FlVQs-0001hT-RB () laptop ! moene ! indiv ! nluug ! nl
[Download RAW message or body]

Tim Prince wrote:

> I tested a 2,000,000 line commercial application on x86-64, using Intel 
> MPI and MKL.
> It could not be supported by g77.

Ditto over here.  I showed off gfortran 4.1.1 on the "All Staff" meeting
of HIRLAM this year (15-18 May in Sofia, Bulgaria) and people were
stunned that I could actually run the operational HIRLAM Weather Forecasting
suite KNMI used until March, 2002, on my laptop *in the same time as
the original* - which ran on a 12 CPU Origin 2000.

This is also an example of code that cannot be compiled by g77 anymore
since around 2000, due to development and replacement of parts by
completely new ones, written to the Fortran 90/95 Standards.

See my signature for the article I would have written for the GCC Summit
if the paper would have been accepted (space kindly provided by Jerry).

PS: HIRLAM web pages at http://hirlam.org (since May this year).

Cheers,

-- 
Toon Moene - e-mail: toon@moene.indiv.nluug.nl - phone: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
A maintainer of GNU Fortran 95: http://gcc.gnu.org/fortran/
Gfortran At Work: http://mysite.verizon.net/serveall/moene.pdf

[prev in list] [next in list] [prev in thread] [next in thread]