ObjectAda V7.0.291 compilation dies: Help!

ObjectAda V7.0.291 compilation dies: Help!

[Brian & Karen Bell]

This posting refers to a problem we are seeing with ObjectAda
    V7.0.291 running on Windows 95. I'm posting it here because thus far
    we've had no success, despite repeated attempts, in eliciting a
    response from Thomson (or is it Aonix?); I personally view this
    failure to get back to a customer with fully paid-up support to be
    very troubling; while it may well turn out to be a transient
    side-effect of that company's recent reorganization, the brutal
    fact is that we need a solution or workaround to the problem RIGHT
    NOW, not once their dust has settled!

    The code in question is being ported from VADScross (SunOS->68K),
    an Ada (83) environment in which it compiles and runs without any
    problem.

    If anyone has seen something similar -- or if an Aonix support
    person is listening -- and can suggest a workaround we'd surely
    appreciate hearing from them.

    Rgds,
    Brian Bell
    The Dow Chemical Co
    Global Process Control Dept.

    bgbell@dow.com
    Work: 517-638-7945

        ---------------------------------------------------

      ...
      with GGenericTimer;

      package body GCID is

         ...
         QuestionCIDTimeout : Duration := (60.0);
         ...
 
Line 50  package CIDTimer is new GGenericTimer (
               RequestType           => GCIDDcl.CIDType,
               SignalTimerExpiration => FreeCID );

         task CIDTask is
         ...
         end CIDTask;


         task body CIDTask is

Line 71     type MCBArray is array (GCIDDcl.CIDType range <>) of
                               
GMessageControlBlock.MessageControlBlock;

         ...
         end CIDTask;

      end GCID;

--------------------Target: Win32 (Intel) Debug--------------------
gcid2.ada: Internal: line 50 col 12 Slot vector null in set_nth_slot,
Continuing
    (In instance of generic GGenericTimer in ggener~2.ada)

gcid2.ada: Internal: line 71 col 7 Fatal internal memory access failure.

Tool execution failed.

Re: ObjectAda V7.0.291 compilation dies: Help!

[Brian & Karen Bell]

Just a quick note to update the c.l.a readership following my posting of
a couple of nights ago; since then I=92ve received not one but _two_
offers of help from Aonix; the first, received within hours, offered a
workaround (albeit not a very pretty one) to what may turn out to be a
known compiler problem; the other offered to try to make a repaired
version available if we would supply the minimal source code needed to
reproduce the problem consistently. It looks like we may be out of the
woods soon!

Our experience with Thomson to this point has been basically positive,
and it=92s looking more and more like their failure to respond promptly t=
o
this recent problem resulted from some confusion following their recent
corporate reorganization =96 confusion which I am sure will be cleared up=

very soon. I look forward to continuing to do business with Aonix in the
future.

Brian Bell
The Dow Chemical Co.
Global Process Control

Brian Bell wrote:
> =

> This posting refers to a problem we are seeing with ObjectAda
>     V7.0.291 running on Windows 95.

[snip]