Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.hawaii.edu!ames!usenet.kornet.nm.kr!news.postech.ac.kr!news.dacom.co.kr!arclight.uoregon.edu!dispatch.news.demon.net!demon!jraynard.demon.co.uk!not-for-mail From: james@jraynard.demon.co.uk (James Raynard) Newsgroups: comp.unix.bsd.freebsd.misc,comp.os.linux.hardware Subject: Re: Does cache memory size 512K over 256K matter? Followup-To: comp.unix.bsd.freebsd.misc Date: 24 Jul 1996 17:28:02 -0000 Organization: A FreeBSD Box Lines: 23 Message-ID: <4t5mf2$9lg@jraynard.demon.co.uk> References: <Pine.A32.3.93.960716180913.29739A-100000@r2d3.sbac.edu> <31F17B64.EDF@www.play-hookey.com> <4t2pe4$9o@anorak.coverform.lan> <4t4klb$q7m@baygull.rtd.com> NNTP-Posting-Host: localhost X-NNTP-Posting-Host: jraynard.demon.co.uk Xref: euryale.cc.adfa.oz.au comp.unix.bsd.freebsd.misc:24448 comp.os.linux.hardware:45692 In article <4t4klb$q7m@baygull.rtd.com>, Don Yuniskis <dgy@rtd.com> wrote: >In article <4t2pe4$9o@anorak.coverform.lan>, > >Seriously, I think you're getting to the point of diminishing returns >when cache size gets up there. I don't *think* (OK, flame me *here* :>) >that any of these OS's particularly try to load the kernel and lock >the cache (etc.) or other similar *potential* optimizations (it's This is getting dangerously off-topic, but the QNX microkernel fits neatly into the cache on a 486 (all 8k of it). Obviously, trying to trim the FreeBSD kernel down to 8kB is perhaps a little extreme, but 256k or 512k may not be totally impossible and might be of interest to the embedded controller world. Has anyone ever tried this? (I've taken comp.os.linux.hardware out of follow-ups to reduce the risk of a "mine's smaller than yours" flamewar ;-) -- James Raynard, Edinburgh, Scotland james@jraynard.demon.co.uk http://www.freebsd.org/~jraynard/