Subject: Re: Is gcc slow? Or is our gcc slow?
To: Niklas Hallqvist <niklas@appli.se>
From: David Gilbert <dgilbert@jaywon.pci.on.ca>
List: current-users
Date: 04/10/1996 10:08:34
>>>>> "Niklas" == Niklas Hallqvist <niklas@appli.se> writes:

	[My description of the sylogics system, deleted]

Niklas> Well the '010 does not have a MMU, but several vendors added
Niklas> custom MMU HW, do you think Sylogics might have done that too?
Niklas> And then you just need to understand hat MMU :-) The '010 had
Niklas> restartable instructions as oppposed to the 68000 so it was
Niklas> easier to make MMU-requiring OSes work for that CPU.  Mind you
Niklas> it is *not* impossible to port NetBSD to a straight 68000, you
Niklas> just need two of them running slightly out of sync, so when a
Niklas> trap occurs the other CPU can be used to restart the failing
Niklas> instruction.  Cute, eh :-) Yes, there existed such machines, I
Niklas> think Apollo did one, possibly Sun too..

	Although I really don't have a lot of documentation on the
boards, I may be able to come up with source code on 8" floppies... if
someone can read those down for me.  I assume that they do have an MMU
since the CPU board appears to have little (if any) memory.  In fact
everything (serial ports, disk, and ethernet) are on other boards in
the system.

	In going through the old HW again, it seems that some of these
were built with a single 68000.  This puzzles me, given your comment.

Dave.

-- 
----------------------------------------------------------------------------
|David Gilbert, PCI, Richmond Hill, Ontario.  | Two things can only be     |
|Mail:      dgilbert@jaywon.pci.on.ca         |  equal if and only if they |
|http://www.pci.on.ca/~dgilbert               |   are precisely opposite.  |
---------------------------------------------------------GLO----------------