From:	"Markus F.X.J. Oberhumer" <markus DOT oberhumer AT jk DOT uni-linz DOT ac DOT at>
Newsgroups:	comp.os.msdos.djgpp
Subject:	Re: gcc optimize performance, V2.7.2.1 versus V2.6.0
Date:	Thu, 03 Apr 1997 20:29:43 +0200
Organization:	Johannes Kepler University Linz
Message-ID:	<3343F717.31B3@jk.uni-linz.ac.at>
References:	<Pine DOT SUN DOT 3 DOT 91 DOT 970331181732 DOT 17577N-100000 AT is>
NNTP-Posting-Host:	c210.edvz.uni-linz.ac.at
Mime-Version:	1.0
CC:	Dan Hirschberg <dan AT ics DOT uci DOT edu>
Lines:	23
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp

Eli Zaretskii wrote:
> 
> On Sat, 29 Mar 1997, Dan Hirschberg wrote:
> 
> > The same program compiled with the newer compiler had much worse
> > performance (about 50% worse) on both my 486 and my Pentium Pro.
> > The performance was even worse with -O2.  I was able to improve
> > the performance somewhat by using -O1 and adding a few of the
> > -f options (found by reading the info file on gcc), but still about
> > 40% worse than under the 2.6.0 version.
> >
> > Is this a common experience?
> 
> No.  People who tested this report that the optimization quality is the
> same, and sometimes slightly better in 2.7.2.1.

One reason that 2.7.2.1 compiled programs might be slower than 2.7.2 is
that the specs do not contain '-fno-strength-reduce' anymore because the
optimizer
bug has been fixed in gcc 2.7.2.1. 
In fact all programs I tested (on the Intel platform) run slower
with '-fstrength-reduce' enabled (read the Linux GCC-FAQ/Howto for more
optimization tips).

- Raw text -