home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.mac.hardware
- Path: sparky!uunet!usc!sdd.hp.com!ux1.cso.uiuc.edu!csrd.uiuc.edu!s4.csrd.uiuc.edu!bruner
- From: bruner@sp10.csrd.uiuc.edu (John Bruner)
- Subject: Re: Motorola's 68060 for the Mac
- In-Reply-To: erics@infoserv.com's message of Fri, 22 Jan 93 09:00:06 PDT
- Message-ID: <BRUNER.93Jan22144051@sp10.csrd.uiuc.edu>
- Sender: news@csrd.uiuc.edu
- Organization: CSRD, University of Illinois
- References: <D2150056.oaq0me@erics.infoserv.com>
- Distribution: na
- Date: 22 Jan 93 14:40:51
- Lines: 18
-
- There is a very ugly way to implement virtual memory in a 68000-based
- system. I don't have my 68000 handbook handy, but the technique
- involves its "bus retry" feature. If HALT is asserted while BERR is
- asserted, the processor stops, and when HALT is negated the processor
- will retry its last bus access. A virtual memory system is
- constructed by using two 68000's. The primary one runs until it gets
- a page fault, at which point it is frozen in mid-instruction. The
- secondary 68000 handles the fault, sets up the page, and then suspends
- itself and restarts the primary 68000. The primary 68000 retries the
- last bus cycle, which now succeeds.
-
- There are a lot of drawbacks to this approach, but there was at least
- one vendor that used it in the early 68000 days.
- --
- (Dr.) John Bruner, Deputy Director bruner@csrd.uiuc.edu
- Center for Supercomputing Research & Development (217) 244-4476 (voice)
- University of Illinois at Urbana-Champaign (217) 244-1351 (FAX)
- 465 CSRL, MC-264; 1308 West Main St.; Urbana, IL 61801-2307
-