Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!uwm.edu!bionet!agate!ucbvax!bloom-beacon!eru!kth.se!sunic!mcsun!ukc!mucs!jdg
From: jdg@cs.man.ac.uk (Jim Garside)
Newsgroups: comp.arch
Subject: Re: What about loop instructions?
Message-ID: <2413@m1.cs.man.ac.uk>
Date: 16 Apr 91 09:28:37 GMT
Sender: news@cs.man.ac.uk
Reply-To: jdg@cs.man.ac.uk (Jim Garside)
Organization: Department of Computer Science, University of Manchester UK
Lines: 21

[ ldo@waikato.ac.nz (Lawrence D'Oliveiro, Waikato University) writes:- ]

>                      ... I saw some code which showed me the *right*
> way to use the DBRA instruction on a 68000. I was impressed! This
> instruction lets you handle an initial loop count of zero without a
> separate check.

My question: what is the *right* way to use a DBcc instruction and (why does
it not have deleterious effects on prefetching)?

Seriously:
680x0	DBcc	Decrement a (16-bit) register and terminate if it becomes
		FFFF or if a specified condition is met.
80x86	LOOP	Decrement (16-bit) CX register and terminate if 0000
Z80	DJNZ	Decrement (8-bit) B register and terminate if 00
Transputer
	LEND	Increment one memory location (16-/32-bit), decrement an
		adjacent one and terminate if the latter is 0000(0000)
...
Jim Garside
jdg@cs.man.ac.uk