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

Simon Clubley clubley at remove_me.eisner.decus.org-Earth.UFP
Wed Apr 10 12:49:35 EDT 2013


On 2013-04-10, Johnny Billquist <bqt at softjar.se> wrote:
> On 2013-04-10 14:07, Simon Clubley wrote:
>> On 2013-04-10, Paul Sture <nospam at sture.ch> wrote:
>>>
>>> I got the impression in the V2 era that it should be possible to plug
>>> your own file system into VMS, and the hooks were there for your own CLI
>>> (the /CLI= qualifier for the username prompt during login and
>>> AUTHORIZE.EXE).  IIRC a recent discussion here pointed out that the
>>> latter wasn't publicly documented.
>>>
>>
>> Are you thinking about a user written ACP instead of a new filesystem ?
>
> Ignoring more recent things like XQP, the ACP is the thing that 
> implements the file system, unless VMS is very different than RSX...
> I could easily add a new file system in RSX to a running system, but 
> there are caveats. RSX itself expects certain things of files, and some 
> of them might be hard to work around with a file system that have rather 
> different semantics.

The only thing I've ever seen a ACP used for on VMS is for tape drives
and I've never even come across any public documentation which would
allow me to write one of those.

However, I am open to the possibility I missed something, so I did some
digging in case I was wrong and I found this:

From: http://labs.hoffmanlabs.com/node/357

|Foreign file systems and FUSE
|
|OpenVMS does not have particular support for adding a file system into OpenVMS;
|there is no file system API or shim; there are no documented layer interfaces
|within the kernel, nor any analog to the Filesystem in Userspace (FUSE)
|mechanisms available.
|
|There are a few such file systems that have been retrofit into OpenVMS
|including NFS and ISO-9660, but these have involved modifications to various
|OpenVMS components, and particularly either bypassing MOUNT or modifications to
|MOUNT.
|
|Which in aggregate means that there's no documented means to extend the file
|system layer to support a new or different file or disk volume structure, and
|thus the existing NFS client and DFS clients are unique solutions, and there's
|no direct way to retrofit a Samba/SMB/CIFS client or a FAT file system client
|nor simple convenience tools such as transparent zip access directly into
|OpenVMS. Block-level encryption can be feasible with an encrypting disk driver,
|but transparent file-level encryption nor file-level compression isn't.

Simon.

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



More information about the Info-vax mailing list