Boston Linux & Unix (BLU) Home | Calendar | Mail Lists | List Archives | Desktop SIG | Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings
Linux Cafe | Meeting Notes | Linux Links | Bling | About BLU

BLU Discuss list archive


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RTFM VMWare question



The IBM VM is based on the use of the "Diagnose" instruction, which is 
not really an instruction at all.  It is an instruction operation code 
that actually generates an invalid instruction interrupt.  
The 'invalid' instruction is examined by software (more of the VM OS) 
to determine what you really meant to do, and it does it for you.  This 
is the heart of IBMs VM.  Historically, "Diagnose" instructions were 
a 'secret' used in "CE Only Code" (Customer Enginner, that did hardware 
support mainly) used for system diagnostics without the customers OS or 
software running.  Intelegent use of the Diag instruction allowed some 
OSes (like VM, MVS, etc) to realize they were running as a 'guest' or 
not as the main OS on the machine, and let the 'first level' VM take 
care of tables like those used for memory virtualization, etc.

I am not sure what an Intel based (style) processor does when it hits 
an 'invalid' instruction or how VMWare implements it.

Does this help, or am I just reminising about former jobs? 
Sorry if it was only reminising. ... Jack


> Actually, the 370 instruction set was never written with VM in mind 
either. 
> Each 370 had a different internal architechture. Most of the 370s had 
> support for VM in the microcode. Additionally, production OSs like 
VS1 and 
> DOS(370 DOS not PC DOS) had additional handles for VM. 
> There are some interesting issues:
> Let's say what heppens when an application page faults. Does the VM 
allow 
> the OS to select another app or does the VM place the entire OS on a 
page 
> wait. 
> On 22 Jul 2002 at 16:48, Nathan Meyers wrote:
> > I hope my answer didn't read like an attempt at a VM tutorial :-).
> > 
> > As I understand it, the Intel architecture was never designed to 
support
> > VMs the way systems like VM370 did, and the VMWare folks had to be 
pretty
> > clever about making it work.
> 
> -- 
> Jerry Feldman <gaf at blu.org>
> Associate Director
> Boston Linux and Unix user group
> http://www.blu.org PGP key id:C5061EA9
> PGP Key fingerprint:053C 73EC 3AC1 5C44 3E14 9245 FB00 3ED5 C506 1EA9
> 
> _______________________________________________
> Discuss mailing list
> Discuss at blu.org
> http://www.blu.org/mailman/listinfo/discuss
> 
> 

-- 
Coats Closet - Faith, Family, Fun, Fortune - http://www.coats.org




BLU is a member of BostonUserGroups
BLU is a member of BostonUserGroups
We also thank MIT for the use of their facilities.

Valid HTML 4.01! Valid CSS!



Boston Linux & Unix / webmaster@blu.org