Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!mcsun!ukc!pyrltd!pyra.co.uk!tonys
From: tonys@pyra.co.uk (Tony Shaughnessy)
Newsgroups: comp.arch
Subject: Re: isamax and instruction set design
Message-ID: <677323133.AA763@flaccid>
Date: 19 Jun 91 09:18:53 GMT
References: <396@validgh.com> <1991Jun13.234834.22970@neon.Stanford.EDU> <1991Jun14.134338.4673@linus.mitre.org> <1991Jun14.163141.17728@rice.edu> <64739@bbn.BBN.COM>
Sender: news@pyra.co.uk (News system)
Reply-To: tonys@pyra.co.uk (Tony Shaughnessy)
Organization: Pyramid Technology Ltd, Farnborough, ENGLAND
Lines: 23

In article <64739@bbn.BBN.COM> slackey@BBN.COM (Stan Lackey) writes:
>
>I once worked on a machine that had instructions like x=min(y,z) in hardware.
>Also there were vector instructions that would elementwise compare vectors
>and build a vector of logicals which could be applied to a subsequent vector
>op.  So a loop of the form
>
>  where (a(i) .lt. b(i)) c(i) = d(i) + e(i)
>
>could be done with two vector instructions and no branches (other than
>loop control, of course).
>
>-Stan

How many cycles would it take to run each of these instructions? If it would
take many cycles, then wouldn't this dominate any expense incurred in taking
a branch, or are there other reasons for avoiding a branch?

--
Tony Shaughnessy
tonys@pyra.co.uk

	"Pedal away those tag-nut blues"