[Info-vax] VMS porting/rewrite, was: Re: [OT] Wirth style languages, was: Re: Obscure Ada compiler vendors?

[Simon Clubley]

On 2013-04-09, Keith Parris <keithparris_deletethis at yahoo.com> wrote:
> On 4/9/2013 6:29 AM, Simon Clubley wrote:
>> On 2013-04-08, Keith Parris <keithparris_deletethis at yahoo.com> wrote:
>>> On 4/4/2013 11:01 AM, Stephen Hoffman wrote:
>>>> Rewriting all of OpenVMS? ...
> >
>> This is talking about a rewrite of the code from the ground up, maintaining
>> user level API compatibility, but with different and more modern internals.
>>
>>>> ~Thirty million lines of Bliss and C and Macro32 is a huge project to
>>>> reconstitute.
>>
>> This is talking about a port of the existing code base to a new
>> architecture.
>>
>> These are two separate issues and people are mixing them up.
>>
>> Which one are you thinking of Keith ?
>
> I'm thinking more along the lines of the first. Perhaps some of the 
> existing C code might be something that could be reused or adapted (if 
> HP agrees), but I'd expect the bulk of the code to be new.
>

Thanks for the clarification Keith.

The next question you need to think about then is what level of internal
(kernel level, not user level) compatibility you want in a rewritten OS.
(User level compatibility should be as close as possible.)

Do you want to exactly duplicate the closely bound and linked together
internal architecture or do you want to use more modern OS concepts to
make VMS far more flexible and modular than it currently is ?

To make that more tangible, let me give you a specific example to think
about. Linux has a very flexible filesystem architecture in which new
types of filesystems can be plugged into the existing framework and
these new filesystems can be written by third parties.

VMS OTOH, has a monolithic tightly bound together setup.

A brief pointer on the Linux VFS approach can be found here:

http://www.tldp.org/LDP/khg/HyperNews/get/fs/vfstour.html

and a far more detailed discussion here:

http://www.win.tue.nl/~aeb/linux/lk/lk-8.html

As this second document points out, there have been some issues with the
Linux VFS as some filesystems very different from ext2 have come along,
but the fact remains that because of the VFS layer, there are a large
range of filesystems available on Linux; far, far, more than what VMS
supports.

So the next question you need to ask yourself is: do you want to maintain
the current VMS internal architecture as closely as possible or do you
want to break existing internal compatibility so you can introduce
concepts like, for example, a more flexible Linux style VFS layer ?

Simon.

-- 
Simon Clubley, clubley at remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world