HP3000-L Archives

June 1998, Week 4

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:
Christian Lheureux <[log in to unmask]>
Reply To:
Date:
Wed, 24 Jun 1998 13:47:03 +0200
Content-Type:
text/plain
Parts/Attachments:
text/plain (142 lines)
CHRISTIAN LHEUREUX@MOULINEX
24/06/98 13:47

Hi fellow listers !

As far as I remember, MPE version numbering scheme is not THAT cumbersome.
There are some basic rules. We'll deal exclusively with MPE/XL and /iX. MPE
V/E versions were not that different, just less compliant with their own
conventions.

Let's remember that the :SHOWME command will give us more than just one
version number. Here is an example :

MPE/iX  HP31900 B.79.06  Copyright Hewlett-Packard 1987.  All rights
reserved.
ALEHP6:showme
USER: #S119,MANAGER.SYS,AUDIT      (NOT IN BREAK)
RELEASE: C.50.00   MPE/iX HP31900 B.79.06   USER VERSION: C.50.00
CURRENT: WED, JUN 24, 1998, 10:36 AM
LOGON:   WED, JUN 24, 1998, 10:08 AM
CPU SECONDS: 1         CONNECT MINUTES: 28
$STDIN LDEV: 154         $STDLIST LDEV: 154
ALEHP6:

So we have "release", "MPE/iX", and "User" versions. That's three !!!

'Release' is the "commercial" number of whatever FOS (Fundamental Operating
 System) runs on your machine. It is the number
out of which we will (later, later ...) derive the version number per se.

'MPE/iX' is the core OS version number. At that stage, I should stress that
 FOS and MPE/iX (XL at that time) version numbers
were 'de-coupled' as of 2.0, if I'm correct (could anyone still with HP
confirm this, please ?). So, you're bound to have a Core OS
and a FOS version with different- number. I'll provide a cross-reference
table later.

'User' is customizable thru Sysgen, and could (user choice ...) be used to
track your powerpatch number. By default, 'User'
is equal to 'Release'. I vaguely remember some powerpatch installs
modifying the 'User' field, but don't ask me which one, my
memory suddenly betrays me there, sorry ...

Now, how is each version number designed ? For clarity's sake, let's assume
 an X.YY.ZZ layout. In your example, B.30.45, 'B'
would be my X, 30 would be my YY, and 45 would be my ZZ. Now, here are some
 details.

X is what I would call MPE/iX 'generation' level. 'A' was the first
generation. It was changed to 'B' as of 3.0 with the advent of
multi-processor systems, which implied a lot of modifications deep inside
MPE. It was later changed to 'C' with the advent of
Posix in 4.5. Here began what I call the 'MPE Version Numbering Big
Discrepancy'. For several releases, we had a 'C'
release with a 'B' MPE/iX version number. That was to be corrected only
with 5.5.

Some limited distribution releases have been prefixed with an 'X', which I
would assume just means that, 'eXperimental'. Anyone
remember X.50.20 (limited 5.0) and/or X.40.15 (limited 4.0) ? Or some beta
releases for those of us who have been involved in
beta tests ?

YY is the version number by itself. 'Release' uses it to track what we are
all looking for, the "commercial" version number. This
one really is a piece of cake : '40' tracks MPE/iX 4.0, '45' tracks 4.5,
'50' tracks 5.0, and so on... It has not yet been reset, but
what when we reach release 10.0 ?

'MPE/iX' uses the YY field to track internal 'builds' (MPE intermediate
versions) that were used to put MPE together inside the
labs. I would therefore assume that, in B.30.45, build # 30 was the finally
 comercially released for 4.0. 'MPE/iX' resets this
number for each 'generation' level. For example, C.05.09 (5.5) comes AFTER
B.79.06 (5.0).

Then ZZ is the fix level. I have little precise recollection about
'Release' ever using this field, at least since 2.0. However,
'MPE/iX' extensively uses it to track fix levels. For example, we can
safely assume that B.30.45 came after a never released
B.30.44, and so on. Of course, this field is more useful to the MPE/iX
Support Engineers than to the basic 'user-in-the-street'.

Now, here is a reference table :

Commercial version number       RELEASE (:SHOWME)        MPE/iX (:SHOWME)
Remarks

3.0                B.30.00                  B.08.14             Changes to 'B' for
multi-processors
                                                        First to 'follow
the rule'
3.1                B.31.00                  B.09.66             For 9x7s only, first
SCSI release
4.0                B.40.00                  B.30.45             Name changes to MPE/iX,
though
                                                        Posix not yet
implemented
4.5                C.45.00                  B.57.06             Release changes to 'C',
MPE/iX
                                                        remains 'B' : Big
Discrepancy
                                                        Introduces a
limited Posix
5.0 PULL              X.50.20               B.76.22             Introduces a full Posix.
                                                        Offered on a 'PULL'
 basis to willing
                                                        users
5.0                C.50.00                  B.79.06             Pushed to the whole
installed base
                                                        under software
contract
                                                        Introduces
CONTIGVOL and
                                                        SETCLOCK, among
others
5.5                C.55.00                  C.05.09             Ends the Big Discrepancy
                                                        Introduces on-line
config, and soooo
                                                        many others !!!!

Now, an exercise, just for fun. What do you expect 6.0 will be named ? Any
ideas ? I am waiting for your suggestions, either
on the list or privately.

Note : the material in this post is ENTIRELY collected from my own memory
when I used to work at HP as a Response Center
Engineer.I ask you to forgive me if any inaccuracies have inadvertently
slipped here and there. Yes, that's my idea of a
disclaimer !!!

Of course, I am ready to report to the list all the inaccuracies and
corrections that will be privately reported to me. Well, you
know where to reach me, and that's [log in to unmask]

If you have read this post down to here, I just want to thank you for your
patience.

Christian "hopefully finding his way thru the maze" Lheureux
As usual, my own ideas and opinions, 'cos no one else will ever dare defend
 them !!!

ATOM RSS1 RSS2