*BSD News Article 19093


Return to BSD News archive

Path: sserve!newshost.anu.edu.au!munnari.oz.au!network.ucsd.edu!dog.ee.lbl.gov!overload.lbl.gov!agate!spool.mu.edu!uwm.edu!cs.utexas.edu!uunet!haven.umd.edu!umd5.umd.edu!roissy.umd.edu!mark
From: mark@roissy.umd.edu (Mark Sienkiewicz)
Newsgroups: comp.os.386bsd.questions
Subject: Re: usefulness of memory-cache in multitasking systems
Date: 2 Aug 1993 22:43:56 GMT
Organization: University of Maryland
Lines: 17
Message-ID: <23k5bc$6ae@umd5.umd.edu>
References: <AgKP52600WB7MsjVgA@andrew.cmu.edu>
NNTP-Posting-Host: roissy.umd.edu

In article <AgKP52600WB7MsjVgA@andrew.cmu.edu> "Alex R.N. Wetmore" <aw2t+@andrew.cmu.edu> writes:
>I've been sort of wondering how useful a cache would be in a
>multitasking systems (specifically NetBSD).  It seems that every time
>there was a process swap all of a sudden there would be a different
>piece of code that would be best off in the cache.

This is true, but after each context switch you quickly fill the cache
with code for the new process.  Most machines have some way to disable
the cache-- mine even lets you disable the cache in the BIOS setup.  Try
it and see what a *big* difference it can make.

b.t.w.  If you have what appear to be timing problems, you can sometimes
make them go away by switching off the cache.  This lets you 1) get some
more evidence that it is a timing problem, and 2) run your machine while
you go about fixing it.