Date:	Wed, 9 Feb 2000 00:30:24 -0400 (AST)
From:	Peter Cordes <peter AT Cordes DOT Phys DOT Dal DOT Ca>
To:	pgcc AT delorie DOT com
Subject:	Re: Perhaps a stupid question
In-Reply-To:	<20000209003053.DKOU11@thanny>
Message-ID:	<Pine.LNX.4.10.10002090024370.30802-100000@Cordes.Phys.Dal.Ca>
MIME-Version:	1.0
Reply-To:	pgcc AT delorie DOT com
Errors-To:	dj-admin AT delorie DOT com
X-Mailing-List:	pgcc AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

On Tue, 8 Feb 2000, Mike Ruskai wrote:
> Are you suggesting that PGCC does not improve on -mpentium over EGCS?

 No, I'm not suggesting that.  You didn't mention any pentiums, and I
assumed you would be running the code on the 486.  If you distribute the
compiled code, make sure pgcc hasn't used any non-i386 instr.  (so it will
run on all machines, but run fastest on pentiums.)  Obviously, the 486 is
the machine to use for that.

 Give the occasional miscompilation of certain code by pgcc, you should be
careful about distributing binaries compiled with it.  Normal gcc with
-mpentium might be good enough for you code.  Maybe if you ran a benchmark
and found that pgcc was a significant improvement (5 or 10% or more), then
distributing pgcc compiled code would make sense.

 Happy hacking,

#define X(x,y) x##y
DUPS Secretary ; http://is2.dal.ca/~dups/
Peter Cordes ;  e-mail: X(peter AT cordes DOT phys. , dal.ca)

"The gods confound the man who first found out how to distinguish the hours!
 Confound him, too, who in this place set up a sundial, to cut and hack
 my day so wretchedly into small pieces!" -- Plautus, 200 BCE

- Raw text -