Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!usc!cs.utexas.edu!uunet!mcrware!davely
From: davely@mcrware.UUCP (Dave Lyons)
Newsgroups: comp.sys.mac
Subject: Re: 68851 PMMU Test?
Message-ID: <1253@mcrware.UUCP>
Date: 20 Jun 89 17:44:59 GMT
References: <134@nisca.ircc.ohio-state.edu>
Reply-To: davely@mcrware.UUCP (Dave Lyons)
Distribution: comp.sys.mac
Organization: Microware Systems Corp., Des Moines, Iowa
Lines: 44

In article <134@nisca.ircc.ohio-state.edu> bobd@hpuxa.ircc.ohio-state.edu 
	(Bob DeBula) writes:
>
>Well, I got my 68851 PMMU installed in my MAC-II, now (without buying
>Virtual or A/UX) does anyone know how to test it?  I'm planning on waiting
>until System 7.0 to actually exploit it, but I would like to assure myself
>that it's properly installed and functioning. Any ideas netters?  

If you have access to an assembler you could just write a little assembly
code to try moving a register from the PMMU into d0, say, and see if it
works or if you get an illegal instruction error (bomb box number 3).
If your assembler won't generate pmmu instructions you can probably just
use "dc.w" and put the object code in directly as follows:

    move.l d0,-(sp)		; save d0
    dc.w $f000,$4200	; moves pmmu tc register into d0
    move.l (sp)+,d0		; restore d0

You could do the same kind of thing with LightSpeed C as follows:

void testpmmu()
{

asm {
	move.l d0,-(sp)
 	dc.w 0xf000,0x4200
	move.l (sp)+,d0
}
}

If you call this function and it works I'd say the pmmu is o.k.

Finally, now that I've wasted all this band width, I would guess
that if your system runs with a PMMU installed the chip must be
working at least partially since the address lines from the '020
have to go through the PMMU on their way to the bus.

I hope this helps (or at least makes sense).

-- 
 Dave Lyons - reach him at ...!sun!mcrware!davely                    
 Ask not from whom the opinions come.  They come from me.            
----------------------------------------------------------------------
 "Gravity is arbitrary" - Calvin