Date:	Sun, 21 Jun 1998 14:33:51 +0300 (IDT)
From:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
To:	Russell Wallace <manorsof AT iol DOT ie>
cc:	djgpp AT delorie DOT com
Subject:	Re: Performance of -lemu
In-Reply-To:	<358ACE23.384B@iol.ie>
Message-ID:	<Pine.SUN.3.91.980621143334.5335U-100000@is>
MIME-Version:	1.0

On Fri, 19 Jun 1998, Russell Wallace wrote:

> -O3		21 megaflops
> -O3 -lemu	13 megaflops
> 
> My understanding had been that -lemu should make no difference at all if
> the program does in fact end up running on a machine with FP
> hardware.

This is how it should work, yes.

> Can anyone explain the reason for this discrepancy, and whether there's
> anything that can be done about it?

Did you try setting 387=y in the environment?  If you did, you will
need to see how exactly does the emulator affect your program.  AFAIK,
it shouldn't.

In general, these issues are somewhat less debugged, since almost
everybody has FP hardware these days.

> (Short of forgetting about working
> as-is and requiring users of 386s to run one of the .exe emulators,
> which is what I'm planning to do as things currently stand.)

Why do you need to ``require'' your users?  The DJGPP startup code
will automatically load the emulator, it just needs to be in the same
directory as the binary.

- Raw text -