Return to BSD News archive
Xref: sserve comp.os.386bsd.misc:2638 comp.arch:42205
Newsgroups: comp.os.386bsd.misc,comp.arch
Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!msuinfo!agate!howland.reston.ans.net!pipex!uknet!festival!edcogsci!richard
From: richard@cogsci.ed.ac.uk (Richard Tobin)
Subject: Re: FreeBSD platform
Message-ID: <CsExHo.Hs9@cogsci.ed.ac.uk>
Keywords: configuration
Organization: HCRC, University of Edinburgh
References: <Cs19IJ.B07@mozo.cc.purdue.edu> <2uo9ju$12b@reuter.cse.ogi.edu> <2v79ig$s8u@masala.cc.uh.edu>
Date: Mon, 4 Jul 1994 11:05:47 GMT
Lines: 22
In article <2v79ig$s8u@masala.cc.uh.edu> wjin@moocow.cs.uh.edu (Woody Jin) writes:
>What I read from an article some time ago was that the cache does not
>affect any performance on multi-user platforms such as Unix, since
>most PC boards use *direct mapped* cache.
This is certainly not true, as you can immediately discover by turning
the cache off!
I assume the claim is that the cache will be ineffective because
different processes will have the same addresses and compete for the
same cache locations. I think this would only be the case if the
cache were virtually-addressed; I don't know whether {4,5}86s use
virtually- or physically-addressed caches. In any case, each
(processor-bound) process is likely to execute hundreds of thousands
of instructions between switches, so the effect will not be so great.
I'm not an expert on this, but the people in comp.arch are, so I'm
adding that to the newsgroups line.
-- Richard
--
Richard Tobin, HCRC, Edinburgh University R.Tobin@ed.ac.uk