*BSD News Article 35437


Return to BSD News archive

Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!msuinfo!agate!ames!decwrl!pa.dec.com!nntpd.lkg.dec.com!lkg.dec.com!thomas
From: thomas@lkg.dec.com (Matt Thomas)
Newsgroups: comp.os.386bsd.bugs
Subject: Re: FreeBSD: ASUS SP3G and writeback caches
Date: 4 Sep 1994 15:44:17 GMT
Organization: Digital Equipment Corporation
Lines: 27
Sender: thomas@netrix.lkg.dec.com (Matt Thomas)
Distribution: world
Message-ID: <34cq0h$ei3@nntpd.lkg.dec.com>
References: <344vevINN2t7j@rs1.rrz.Uni-Koeln.DE>
Reply-To: thomas@lkg.dec.com
NNTP-Posting-Host: netrix.lkg.dec.com
Keywords: FreeBSD
X-Newsreader: mxrn 6.18-16


In article <344vevINN2t7j@rs1.rrz.Uni-Koeln.DE>, se@fileserv1.MI.Uni-Koeln.DE (Stefan Esser) writes:
|>Maybe something got broken in going from FreeBSD-1.1 to 1.1.5 ?
|>
|>There have been some reports on problems with the ASUS SP3G
|>and the PCI SCSI driver, if the secondary cache has been 
|>configured for writeback operation. The PCI driver makes 
|>the NCR do lots of bus master operations to a data area that
|>is written to by the CPU as well (command queues), and is the 
|>only FreeBSD driver so far, that depends on cache coherency 
|>for PCI to memory operations.

Not quite the only one.  I also have a driver for the DEC
DE435 Ethernet PCI NIC.  It also uses in-memory structures that
are R/W for both the CPU and DC21040 chip.

How ever, since my only disk controller is the NCR 53C810 I've
never been able to tell if my driver has the same problems with
writeback as the NCR 53C810.

My 486 66dx2 PC always panics on startup when the writeback cache
is enabled.
-- 
Matt Thomas                          Internet:   thomas@lkg.dec.com
U*X Networking                       WWW URL:    http://ftp.dec.com/%7Ethomas/
Digital Equipment Corporation        Disclaimer: This message reflects my
Littleton, MA                                    own warped views, etc.