Subject: Re: Considering port to new processor
To: None <firstname.lastname@example.org, email@example.com>
From: Thor Lancelot Simon <firstname.lastname@example.org>
Date: 03/30/2001 19:27:46
On Fri, Mar 30, 2001 at 07:08:39PM -0500, Thor Lancelot Simon wrote:
> You don't necessarily need page protections. You *do* need the ability
> to protect some region of memory specified *somehow* in order to switch
> contexts and to prevent processes from scribbling on each others' "private"
> view of the system memory.
I don't think I was sufficiently clear. What I was getting at is that we
no longer have the severely constrained address space of the PDP11, so it
may be that we can do more with "bare minimum" hardware support.
Given enough memory, and a way to cause *any* fault on access to a given
physical address, it ought to be possible to use some of your "protected"
(a.k.a.) "kernel" memory as a scratchpad and emulate pretty much all of
a "real" hardware MMU. You may have to relocate things around in memory
quite a bit, but that sure beats swapping entire processes to disk (which
you can still do, perhaps slicing them into "segments", if you have to).
It'll be slow as all hell, but it ought to be possible -- or am I missing
Thor Lancelot Simon email@example.com
And now he couldn't remember when this passion had flown, leaving him so
foolish and bewildered and astray: can any man?