[pups] A bootable disk image for a PDP-11/34?

[Ian King]

Again, I'd be glad to do things like this, if I could get an image to run on
an emulator -- that's been a goal.  I've tried various 'processors' (in
emulation), but not been successful at booting the Unix kernel.  Can anyone
say, "I booted image X on emulator Y and had a successfully running Unix"?
If so, please please please share your experience -- I haven't been able to
boot anything out of the PUPS archive on the E11 emulator (held out by some
to be the best).  

And, if/when I have success, I promise to share a field report.  :-)  -- Ian


-----Original Message-----
From: Carl Lowenstein [mailto:cdl at mpl.ucsd.edu]
Sent: Tuesday, January 30, 2001 3:31 PM
To: wkt at cs.adfa.edu.au
Cc: pups at minnie.cs.adfa.edu.au
Subject: Re: [pups] A bootable disk image for a PDP-11/34?


> From wkt at henry.cs.adfa.edu.au Tue Jan 30 14:16 PST 2001
> From: Warren Toomey <wkt at henry.cs.adfa.edu.au>
> Subject: Re: [pups] A bootable disk image for a PDP-11/34?
> To: Carl Lowenstein <cdl at mpl.ucsd.edu>
> Date: Wed, 31 Jan 2001 09:13:30 +1100 (EST)
> CC: pups at minnie.cs.adfa.edu.au
> MIME-Version: 1.0
> Content-Transfer-Encoding: 7bit
> 
> In article by Carl Lowenstein:
> > Just from the historical point of view, note that the first major
> > file on a genuine 6th Edition distribution tape _is_ a bootable
> > RK05 image.  Something like 4000 blocks.
> > 
> >     carl
> 
> Yes, but it doesn't boot on an 11/34 :(
> That's the problem.
> 
> Cheers,
> 	Warren

Why, said he wonderingly.  Is the 11/34 one of those machines that does
not have a directly addressable PSW at 177776 so it must be done with
MPTS and MFPS instructions?  Thats all I can think of immediately.

Pulling out my handy PDP-11 architecture handbook, I find this to be so.
There are some other differences between 11/34 and 11/40-45 but this
one would be a real show-stopper.

The modifications to 6th Ed. Unix to take care of this would be
concentrated in the save/raise/lower processor priority routines.  Just
offhand I don't think one could make a zero-length binary patch that
would work on the 11/34 and remain 11/40-45 compatible.

I guess I will look at locore.s tonight when I get home.  That must
be the place where such things happen.

I wonder what the RT-11 guys did when they had the same problem.  I
seem to remember something involving creative use of stack pushes and
RTI instructions to set the processor priority.

    carl

Received: (from major at localhost)
	by minnie.cs.adfa.edu.au (8.9.3/8.9.3) id SAA27852
	for pups-liszt; Wed, 31 Jan 2001 18:13:33 +1100 (EST)
	(envelope-from owner-pups at minnie.cs.adfa.edu.au)