Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!cis.ohio-state.edu!ucbvax!hplabs!hpfcso!mjs
From: mjs@hpfcso.FC.HP.COM (Marc Sabatella)
Newsgroups: comp.arch
Subject: Re: Branchless conditionals (Was: More on Linpack pivoting)
Message-ID: <8840039@hpfcso.FC.HP.COM>
Date: 25 Jun 91 16:32:26 GMT
References: <1991Jun14.173510.22510@dg-rtp.dg.com>
Organization: Hewlett-Packard, Fort Collins, CO, USA
Lines: 14

>HP's RISC CPU (and many others) use conditionals to NULLIFY the next
>instruction.  Here is how the HP compiler 'could' translate the above
>statement:
>...
>Of course some may argue that this is really a branch...

No, I think part of the point is that a nullify is not a branch - in
particular, it never interferes with prefetch, nor does it alter locality.

--------------
Marc Sabatella (marc@hpmonk.fc.hp.com)
Disclaimers:
	2 + 2 = 3, for suitably small values of 2
	Bill and Dave may not always agree with me