Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!caip!princeton!allegra!ulysses!mhuxr!mhuxt!houxm!ihnp4!fortune!wall
From: wall@fortune.UUCP (Jim Wall)
Newsgroups: net.micro.68k
Subject: Lockup in Read-Modify-Write with a PMMU
Message-ID: <6009@fortune.UUCP>
Date: Wed, 13-Aug-86 16:59:55 EDT
Article-I.D.: fortune.6009
Posted: Wed Aug 13 16:59:55 1986
Date-Received: Thu, 14-Aug-86 21:13:06 EDT
Organization: Fortune Systems Corp., Redwood City CA
Lines: 20

Is Motorola still on the net?

Probably, if I'm still at Fortune, they're still on the net,
somethings never change.

When you do a read-modify-write instruction on a systen that
uses a PMMU or the infamous MMB from Motorola the odds are taht
a table walk will be required, unfortunately, the bus request from
the PMMU will be ignored by the 68020 because the RMC is an 
indivisable cycle and so bus requests are ignored, the upshot is
that the system hangs. 

This is ture for the MMB, nothing I see will fix this for the PMMU
chip in its final form. Any one care to comment? I'm assuming this 
to be the case, and since I have no time to pursue it, software 
will work around it. THis is partially for peoples edification, and
partially to see if I'm understanding the situation correctly.

						-Jim Wall
						Fortune Systems Corp.