Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site alice.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!alice!ark
From: ark@alice.UUCP (Andrew Koenig)
Newsgroups: net.lang.c
Subject: Re: fast string copy
Message-ID: <3505@alice.UUCP>
Date: Fri, 29-Mar-85 10:13:01 EST
Article-I.D.: alice.3505
Posted: Fri Mar 29 10:13:01 1985
Date-Received: Sat, 30-Mar-85 01:19:15 EST
References: <1049@gloria.UUCP>
Organization: Bell Labs, Murray Hill
Lines: 40

> I think this is cute, how VAX/VMS beats Unix at its own game.  The VMS C
> compiler generates code as good as or better than anything I have seen posted
> so far!
> 
> 	char *s,*t;
> 	while (*s++ = *t++);
> 
> generates (on a VAX 780):
> 
> 		movb	(r2)+,(r1)+
> 		beql	sym.2
> 	sym.1:
> 		movb	(r2)+,(r1)+
> 		bneq	sym.1
> 	sym.2:
> 
> This is as fast as you can get.

Ummm... I hate to be a spoilsport, but I do not understand why
the code above is any faster than

	sym.1:
		movb	(r2)+,(r1)+
		bneq	sym.1

which is essentially what the Unix compilers generate.
Apparently, the VMS compiler is trying to be clever by
rewriting

	while (cond) stmt;

into

	if (cond) { do stmt; while (cond);}

This rewriting generally buys speed at the expense of space,
but in this particular case the speed gain is 0 and the space
cost is a factor of two.

By the way, Colonel, this loop is not improved by unrolling.