[Info-vax] Kittson question

[JF Mezei]

On 15-01-01 10:28, terry+googleblog at tmk.com wrote:

> Most of the intentional cases are probably early in operating system initialization,

Yes but VAX-VMS 5.5-2 already had support for greater than 16 megs of
memory because systems with more memory existed. So I have to assume
that the VMS code no longer assume that the high order 8 bits of a 32
bit address was free to use to store other stuff.

> And, of course, an emulator also needs to emulate the peripherals,
> including any undocumented behavior. 

Don't emulators emulate only a subset of devices available on those
systems ? For instance, wouldn't all disks in SIMH appear as "RA" drives
(DUAx:) instead of SCSI (DKAxxx:) or DSSI (DIAx:) ?

Does an application really care if the disk is DUA0: DKA100: or DIA0: ?

At this point in time, if a  VAX emulator is needed on the new VMS
platform, do we really need a gazillion models supported, or just
support one of the last models which allows the greatest configuration
of memory (from the emulator point of view) and then just allow user to
set a lower model number for licensing point of view.

If one is to use a translated image system that runs "natively" on the
new platform, there are some licensing issues because if an app expects
to run on a VAX of a certain class, it is a given that the industry
standard VMS instance will report that license is not valid.

So this would mean that "VEST"ed images would need license upgrades,
whereas when running a VMS instance emulator, that software can run
inside a machine that pretends to be the one it was licensed for.

This is where the grey area between VSI having access to the new
platform, but not Alpha and VAX gets tough. Could it convince HP to
allow it to lt go of licensing for VAX and Alpha so that VSI could then
play "games" to allow VESTed apps from VAX and Alpha to run on the new
industry standard servers with their old licenses ?