Re: VAX Ada Tasks and VMS Lock Manager

[kilgallen@eisner.decus.org (Larry Kilgallen)]

In article <01bc9579$39aaaf40$83ecced0@jpeterson>, "Jay and Cathy Peterson" <JPeterson@PCBYTE.net> writes:

> 	I am currently debugging a VAX Ada (v1.5, if it makes a difference)

They are well beyond V1.5 now, fixing many bugs, so it may matter to you.

> application which contains roughly four tasks (plus the main task); one of
> which is is used to synchronize access to a file. Additionally, I am using
> the VMS Lock Manager (VMS version 5.4-2 -- I know; but it is what I have to

They are well beyond V5.4-2 now, fixing many bugs, so it may matter to you.

> work with) to assure the tasks don't walk over pertinent data; namely file
> pointers I include in the controlled-access file. All of that to get to my
> question: How do VAX Ada and VMS treat tasks for locking purposes? The
> "traditional" use of SYS$ENQ and SYS$DEQ are to synchronize processes, I
> know. If I issue a TASKING_ENQW, are the other tasks locked out, or do I

The TASKING_ENQW would be the same as any other TASKING_* jacket for the
"Wait" form of a system service -- other Ada tasks will be allowed to
proceed.

Regardless of how you access SYS$ENQ, it is important to realize that
it doesn't have much regard for processes.  You can deadlock a single
process against itself or a single task against itself with the same
ease with which you deadlock two processes in deadly embrace.  The
crucial element of VMS locking is the Lock ID returned by SYS$ENQ[W].

> need find another way of keeping the tasks from walking on each other? The
> VAX Ada documentation at my disposal is silent on this issue, as far as I
> can ascertain. Any insight into this situation would be appreciated. 

They are well beyond V1.5 now, augmenting the documentation, so maybe
it did matter to you :-).

You might do better with the current documentation, even if you have
the older software.

Regards,
Larry Kilgallen