Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!shadooby!samsung!cs.utexas.edu!tut.cis.ohio-state.edu!rutgers!texbell!sugar!ficc!peter
From: peter@ficc.uu.net (Peter da Silva)
Newsgroups: comp.arch
Subject: Re: Software modularity vs. instruction locality
Message-ID: <6984@ficc.uu.net>
Date: 15 Nov 89 22:45:09 GMT
References: <17707@watdragon.waterloo.edu> <23604@cup.portal.com> <6374@dime.cs.umass.edu>
Reply-To: peter@ficc.uu.net (Peter da Silva)
Distribution: na
Organization: Xenix Support, FICC
Lines: 16

In article <6374@dime.cs.umass.edu> shri@ccs1.cs.umass.edu (H.Shrikumar{shri@ncst.in}) writes:
> In forth there is a subroutine call every five or so instructions
> I would guess.

Yeh, but most of these calls would be to the same place, so the subroutine
would be in the cache. You'd expect that { @ ! (colon) (exit) + - / * } and
a few others would be in the cache most of the time.

> Wonder if some better cache philosophy can help here.

I'd worry more about pipeline flushes.
-- 
`-_-' Peter da Silva <peter@ficc.uu.net> <peter@sugar.hackercorp.com>.
 'U`  --------------  +1 713 274 5180.
"*Real* wizards don't whine about how they paid their dues"
	-- Quentin Johnson quent@atanasoff.cs.iastate.edu