From owner-freebsd-questions  Sun Dec 25 10:29:37 1994
Return-Path: questions-owner
Received: (from root@localhost) by freefall.cdrom.com (8.6.9/8.6.6) id KAA14812 for questions-outgoing; Sun, 25 Dec 1994 10:29:37 -0800
Received: from pelican.pelican.com (pelican.com [134.24.4.62]) by freefall.cdrom.com (8.6.9/8.6.6) with SMTP id SAA14806 for <questions@freebsd.org>; Sun, 25 Dec 1994 18:29:34 GMT
Received: by pelican.pelican.com (Smail3.1.28.1 #5)
	id m0rLxgr-000K2dC; Sun, 25 Dec 94 10:29 WET
Message-Id: <m0rLxgr-000K2dC@pelican.pelican.com>
From: pete@pelican.pelican.com (Pete Carah)
Subject: Cyrix query
To: questions@freebsd.org
Date: Sun, 25 Dec 1994 10:29:09 -0800 (PST)
X-Mailer: ELM [version 2.4 PL23]
Content-Type: text
Content-Length: 548       
Sender: questions-owner@freebsd.org
Precedence: bulk

I wondered if the recent (what date code means recent?) Cyrix 486
substitute chips (the ones that plug into ordinary 386 MBs) handle
freebsd correctly with a 1542B?

They claim some sort of cache coherency trick; snooping would work
if the motherboard is right (mine is a medium old C&T PEAK/DM with
AMI bios).  Does anyone know if 

a) snooping is the cache coherency trick?
b) if the PEAK/DM relays addresses back to the processor on bus-master
   cycles?

(would be nice to get "decent" performance from my old 386.)

Thanks in advance,
-- Pete