Path: utzoo!attcan!uunet!samsung!dali!uakari.primate.wisc.edu!zaphod.mps.ohio-state.edu!usc!elroy.jpl.nasa.gov!ames!amdcad!proton!davec
From: davec@proton.amd.com (Dave Christie)
Newsgroups: comp.sys.cdc
Subject: Re: VE on a MIPS box
Message-ID: <29849@amdcad.AMD.COM>
Date: 12 Apr 90 14:09:12 GMT
References: <10127.262384ad@vax1.cc.lehigh.edu>
Sender: news@amdcad.AMD.COM
Reply-To: davec@nucleus.amd.com (Dave Christie)
Organization: Advanced Micro Devices, Inc., Austin, Texas
Lines: 28
Summary:
Expires:
Sender:
Followup-To:

In article <10127.262384ad@vax1.cc.lehigh.edu> lusol@vax1.cc.lehigh.edu writes:

| Yes, it will be difficult to smoosh a NOS/VE task's 48 bit PVA (Process Virtual
| Address) space of 8.8 trillion bytes into a 2 billion byte space!  And what
| about the 16 rings...  The 386 at least supported 4 rings, if I remember
| correctly (which I probably do not).
| 
| Steve Lidie
| Lehigh University Computing Center

I have heard at least one VE developer say that porting VE to the
386 would be doable.  (He might have used the word "trivial", but
then, he's not your average programmer. He can also be prone to
hyperbole ;-).  

It does have four rings of protection (although I'm not sure if they
provide the same degree of protection between the user and OS),
not to mention a segmented (in a twisted sort of way) address space.  
Even in a stripped-down 1-6 user version one would probably need 
fairly large real memory (8+ MB ?) plus more than your average amount 
of disk space.  And a 386 might not have quite enough horsepower for 
it to make sense.  A 486 might be a different story, however....

It's a shame CDC hasn't published more about VE & 180.  I don't think
anyone except CDC and VE users has a clue about it.

-------------
Dave Christie        My opinions only