Message-ID:	<3600B3C4.9CCDAEF3@mailexcite.com>
Date:	Thu, 17 Sep 1998 03:01:25 -0400
From:	Doug Gale <dgale AT mailexcite DOT com>
MIME-Version:	1.0
Newsgroups:	comp.os.msdos.djgpp
Subject:	Re: djgpp and optimizations
References:	<Pine DOT SUN DOT 3 DOT 91 DOT 980916121635 DOT 5874O-100000 AT is>
NNTP-Posting-Host:	oshawappp11.idirect.com
Organization:	"Usenet User"
Lines:	33
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp

Eli Zaretskii wrote:

> On Tue, 15 Sep 1998, Arthur wrote:
>
> > What does -fstrength-reduce do, exactly, and why do some people report such
> > a huge speed increase when using it?
>
> -fstrength-reduce replaces expensive instruction by less expensive
> ones.  For example, multiplication is replaced with additions or bit
> shifts, etc.
>
> The problem is that this usually needs extra registers, and x86 has a
> few of them to spare.  If a function doesn't use too many registers,
> then -fstrength-reduce can be a big win; but if they are all used,
> -fstrength-reduce in a loop can mean a significant slowdown, since GCC
> will have to constantly reload some of the regsiters.

Oh, -fstrength-reduce does that too? I read somewhere that -fstrength-reduce it
is more of a loop optimization that does stuff like converting

for (i = 0; i < 100; i++)

into

for (temp = 100, i = 0; --temp; i++)

saving a cycle or two at the end of each iteration, and a few bytes of code per
loop.

- Raw text -