Message-Id: <199903242230.RAA25983@indy1.indy.net>
From: "Steve Snyder" <ssnyder AT indy DOT net>
To: "pgcc AT delorie DOT com" <pgcc AT delorie DOT com>
Date: Wed, 24 Mar 1999 17:28:59 -0500 (EST)
X-Mailer: PMMail 2.00.1500 for OS/2 Warp 4.00
In-Reply-To: <Pine.GSO.4.05.9903242031460.6739-100000@legolas.mdh.se>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Subject: Re: optimize
Reply-To: pgcc AT delorie DOT com
X-Mailing-List: pgcc AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

On Wed, 24 Mar 1999 20:38:54 +0100 (MET), Henrik Berglund  SdU wrote:

>Does -ffast-math make the programs any faster?
>
>Using Intel C/C++ on a PentiumII-400 PC Magazine gets 15.0 int in bytemark
>benchmark but i only get 5 int something with pgcc on a amdk6-2 300. How
>can Intel C/C++ optimize so much better? PII-300 only get 4.3 int with
>pgcc.

In my v2.2.x arch/i386/Makefile I've got these lines:

ifdef CONFIG_M686                                             
CFLAGS := $(CFLAGS) -O6 -fno-inline-functions -mcpu=i686 -march=i686 -DCPU=686
endif                                                         

These work great for me.  Note that I'm using pgcc v1.1.1.
        

*** Steve Snyder ***