Newsgroups: comp.arch
Path: utzoo!utgpu!watserv1!watdragon!watsol.waterloo.edu!tbray
From: tbray@watsol.waterloo.edu (Tim Bray)
Subject: Re: Page size and linkers (was: Re: SunMMU history)
Message-ID: <1991Jan27.214522.24408@watdragon.waterloo.edu>
Sender: daemon@watdragon.waterloo.edu (Owner of Many System Processes)
Organization: University of Waterloo
References: <1991Jan19.133914.23871@bellcore.bellcore.com> <PCG.91Jan20191955@teacho.cs.aber.ac.uk> <3981@skye.ed.ac.uk> <MOSS.91Jan21172107@ibis.cs.umass.edu> <45242@mips.mips.COM>
Date: Sun, 27 Jan 91 21:45:22 GMT
Lines: 15

mash@mips.COM (John Mashey) writes:
 (Consider the kind of
 program that's >200MB of code, much of it in a giant single loop,
 leading to a high I-cache miss rate.)

PLEASE tell us that's a thought experiment, John.  Does the world really
contain such appalling creations?  If so, I think a taxonomy would be of
general interest to the readers here...

(Mind you, in our group, we have occasionally executed large Deterministic
Finite Automata very fast by compiling them into a few Mb of branch
statements, which effectively trashes any I-cache with a relatively small
volume of code - but it's still the fastest way to run a big DFA).

Cheers, Tim Bray, Open Text Systems
