Date: Wed, 3 Mar 1999 00:46:14 +0100
From: Hans-Bernhard Broeker <broeker AT physik DOT rwth-aachen DOT de>
Message-Id: <199903022346.AAA27910@acp3bf.physik.rwth-aachen.de>
To: djgpp AT delorie DOT com
Subject: Re: Need some help with assembler
Newsgroups: comp.os.msdos.djgpp
Organization: RWTH Aachen, III. physikalisches Institut B
X-Newsreader: TIN [version 1.2 PL2]
Reply-To: djgpp AT delorie DOT com

In article <01be64f7$c620f160$LocalHost AT aguia> you wrote:
> Hi there,

>         i have the following function in assembler (but it was for my old
> compiler Borlandc v3.1 16bits).
>         i really need it, and i don't know AT&T syntax.

You do *not* need it, I think. 

>         what thefunction actually does is move double-words from source to
> dest. it is much faster than the memcpy routine from C.

It is most probably not a bit faster than the memcpy in djgpp. That
one is already written in inline assembly, and it does use 32bit
moves. To see, compile a simple program that uses memcpy(), like this:

	gcc -g -S foo.c

Have a look a the resulting foo.s file. If you look at it long enough,
you'll find the 32bit loop 'rep movsl' in there just fine. Or, if you want
to avoid having to look at AT&T style assembly, debug the program with
'fsdb'. It looks remarkably like TurboDebugger, and uses Intel-style
assembly to show the code.

--
Hans-Bernhard Broeker (broeker AT physik DOT rwth-aachen DOT de)
Even if all the snow were burnt, ashes would remain.