Date: Tue, 18 Feb 1997 02:32:39 +0800 (GMT)
From: Orlando Andico <orly AT gibson DOT eee DOT upd DOT edu DOT ph>
Reply-To: Orlando Andico <orly AT gibson DOT eee DOT upd DOT edu DOT ph>
To: Nigel Taylor <taylorn AT syntegra DOT bt DOT co DOT uk>
cc: djgpp AT delorie DOT com
Subject: Re: DJGPP Optimisations...
In-Reply-To: <9702171652.AA06657@syntegra.bt.co.uk>
Message-ID: <Pine.SGI.3.93.970218023129.12532A-100000@gibson.eee.upd.edu.ph>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Mon, 17 Feb 1997, Nigel Taylor wrote:

> > > Where am I going wrong, no matter what optimizations I use my EXE
> > > always stays the same length, I use
> > > gcc test.c -O<x> -m486 -o test.exe
> > What about the .exe file contents?  Does it change?  If it doesn't,
> > maybe -m486 in itself implies -O2 or something. 
> 
> > 
> 
> As far as I can remember, its the same with or without the -m486 
> also, I will try this out this evening anyway.
> 
> Nigel...
> 

Really? the Linux kernel howto says that if you use -m486, the compiler
will align stuff on 4-byte boundaries (or something like that), which
makes things a bit (5-10%) faster on 486 and up, but the binaries will
still run on 386. using -m386 makes the binaries a bit smaller because
2-byte packing is used. Or at least that's how I understand it.

.-----------------------------------------------------------------.
| Orlando Andico                email: orly AT gibson DOT eee DOT upd DOT edu DOT ph |
| IRC Lab/EE Dept/UP Diliman   http://gibson.eee.upd.edu.ph/~orly |
|  "through adventure we are not adventuresome" -- 10000 Maniacs  |
`-----------------------------------------------------------------'