HP3000-L Archives

December 2000, Week 1

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Bill Cadier <[log in to unmask]>
Reply To:
Bill Cadier <[log in to unmask]>
Date:
Tue, 5 Dec 2000 18:53:30 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (62 lines)
Hello Christian,

I am not aware of any 'bugs' in 6.5's new DST enhancement but then I would
not be the definitive answer on that. There are no reports of such that I can
locate in our internal database. And nobody has called me to report any sort of
problem in this area.

We have had a few (maybe two) incidents of running pre-6.5 MPE/iX utilities on
6.5 and causing aborts. Perhaps this is what you had heard of?

In order to increase the number of DST's to 65,535 we had to restructure the
Compatibility Mode Process Control Block (PCB).  These utilities were compiled
using inlined versions of many system functions. Since they were pre-6.5 these
inlined routines would have pre-6.5 definitions of system structures. The problem
resulted from inlined calls to routines to enter or leave system code.

The 'system code' flag in the PCB had been adjacent to the CM Stack DST prior
to 6.5. It was moved to allow the CM Stack field to be 16 bits in width rather than
the 14 bits it had occupied pre-6.5. Non-GETDSEG DST's (or system DST's)
on 6.5 range from 16,384 up to 65,535 and 16,384 = 040000 octal. Bit 1 is on.
This bit happens to wind up where the old 'system code' flag would be in the
pre-6.5 PCB definition. So a program using the old definition of the PCB could
enter system code without altering a non-GETDSEG (system) DST but as soon
as it left system code and set that bit false (zero) any reference to the CM Stack
DST after that would be incorrect and likely to result in a system abort.

Not a bug but rather an unfortunate mistake.

I hope this is of some help.

Best Regards,

Bill
HP/CSY
reply to: [log in to unmask]


-----Original Message-----
From: Christian Lheureux <[log in to unmask]>
To: [log in to unmask] <[log in to unmask]>
Date: Tuesday, December 05, 2000 10:19 AM
Subject: [HP3000-L] 65,536 DSTs


>Hi fellow listers !
>
>I recently got informed by a customer that the 65,535 DSTs enhancements in
>MPE/iX 6.5 has some unwanted side effects, i.e. bugs. But I can't find
>anything on ITRC. Has anyone got any information about that ?
>
>Any help appreciated. Thanks to all.
>
>Christian Lheureux
>Responsable du Département Systèmes et Réseaux
>Head of Systems and Networks Department
>APPIC R.H.
>HPConnect Systems Integrator / HP3000 Expert / HP e-Partner
>Tel : +33-1-69-80-97-22   /   Fax : +33-1-69-80-97-14 / e-mail
>[log in to unmask]
>"Le Groupe APPIC recrute, contactez nous !"
>

ATOM RSS1 RSS2