Re: High Level Language vs Assembly

Toon Moene <toon@moene.indiv.nluug.nl>
12 Mar 2001 02:33:31 -0500

          From comp.compilers

Related articles
[20 earlier articles]
Re: High Level Language vs Assembly tfjellstrom@home.com (Tom Fjellstrom) (2001-03-10)
Re: High Level Language vs Assembly samiam@cisco.com (Scott Moore) (2001-03-10)
Re: High Level Language vs Assembly bobduff@world.std.com (Robert A Duff) (2001-03-10)
Re: High Level Language vs Assembly mr@peakfive.com (Matt) (2001-03-10)
Re: High Level Language vs Assembly joachim_d@gmx.de (Joachim Durchholz) (2001-03-10)
Re: High Level Language vs Assembly joachim_d@gmx.de (Joachim Durchholz) (2001-03-10)
Re: High Level Language vs Assembly toon@moene.indiv.nluug.nl (Toon Moene) (2001-03-12)
Re: High Level Language vs Assembly ts3@ukc.ac.uk (Tom Shackell) (2001-03-14)
Re: High Level Language vs Assembly jthorn@galileo.thp.univie.ac.at (2001-03-14)
Re: High Level Language vs Assembly tfjellstrom@home.com (Tom Fjellstrom) (2001-03-22)
Re: High Level Language vs Assembly anton@mips.complang.tuwien.ac.at (2001-03-22)
| List of all articles for this month |

From: Toon Moene <toon@moene.indiv.nluug.nl>
Newsgroups: comp.compilers
Date: 12 Mar 2001 02:33:31 -0500
Organization: Moene Computational Physics, Maartensdijk, The Netherlands
References: 01-02-094 01-02-101 01-03-074
Keywords: assembler, performance
Posted-Date: 12 Mar 2001 02:33:30 EST

Matt wrote:


> I would think talking about how good compilers are in this news group
> is like singing to the choir, so let me ask this; how much improvement
> is there between non-optimized and highly optimized code?


Days before the EGCS project started in August '97, I sent off an
e-mail to RMS and Richard Kenner (then maintainter of gcc) about my
"improvements" to the gcc-heuristics w.r.t. to loop optimisation.


On the 25 Mhz 68040 system I owned at that time, the changes made a
factor of 3 difference with regard to the standard gcc heuristics on a
simple 750-line Fortran benchmark.


This is not even between "non-optimised" and "optimised" code, it's
between "-O2" (vanilla) and "-O2" (tuned).


Definitely, in the Real World, larger differences are attainable.
--
Toon Moene - mailto:toon@moene.indiv.nluug.nl - phoneto: +31 346 214290
Saturnushof 14, 3738 XG Maartensdijk, The Netherlands
Maintainer, GNU Fortran 77: http://gcc.gnu.org/onlinedocs/g77_news.html
Join GNU Fortran 95: http://g95.sourceforge.net/ (under construction)


Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.