| 540.01 Update maximum MCP Level to 54.1 |
| Mon, January 17 2011 |
|
Jampack is now qualified for use on MCP 54.1 and later; this change
prevents the display of MCP warning messages.
|
| 530.05 Abort if consecutive RESERVE requests fail syntax |
| Wed, July 22 2009 |
|
If consecutive JAMPACK initiated RESERVE commands were rejected by the
MCP because of syntax, JAMPACK will now abort instead of trying to
continue,and often issuing the same failed command continuously. The
following message will be displayed:
JAMPACK ABORT:CONSECUTIVE RESERVE JOBS HAVE FAILED SYNTAX
|
| 530.04 CheckerBoard Enhancements |
| Mon, June 2 2008 |
|
JamPack now gives a more informative message when a checkerboard is created. It will also handle larger numbers of rows in the range.
An optimization has reduced JamPack's elapsed time by about 4 seconds
per RESERVE or report.
If a bad title is detected in a header, the CheckerBoard report will now indicate:
>> Bad Title <<
|
| 530.03 CheckerBoard Report |
| Thu, May 15 2008 |
|
JamPack can now produce a readable CheckerBoard report similar to the
one in *SYSTEM/FILEDATA. However, the report is produced typically in
just 5% of the Elapsed time that FileData uses and only 2% of the CPU
time. Also, unlike FileData it reports not only on Permanent file in
the SYSTEMDIRECTORY, but on temporary files as well as many other types
of disk usage.
The syntax is:
<Family> # <Inx> SEG <Address> FOR <length>
For example: DISK #2 SEG 1000000 FOR 10000
If # <inx> is omitted #1 is assumed.
The report is produced in a permanent disk file called CHKBD, which can
be Label-Equated. Each row has a line in the file of the format:
<start> <hyphen> <end> <fixed> <flags> <row type>
where <start>. and <end> are the start and end addresses
<hyphen> is a space for a single row, or a '-' if there are
multiple contiguous rows of the same file.
<fixed> is normally a space, but is 'F' if the row would block or
otherwise give problems to a RESERVE.
<flags> is normally space, but is 'T' if the file is temporary,
and 'O' if the file is in use.
<row type> is either a FILENAME or one of the following:
>> Label << - The unit label area (at address 0 to 27)
>> Olay Row << - An overlay row (only on DL OLAY family)
>> MCP Info << - The Mem Dump workspace (only CM'ed Families)
>> Unavailable << - Areas that are in use, but which are not
known to RESERVE.
>> Job File << - A job in the live JOBDESC (on DL JOBS family)
>> Obsolete Job << - Rows in an obsolete JOBDESC, which should
be removed.
For example:
0 27 F >> Label <<
28-4027 O *SYSTEMDIRECTORY/001
4093 4597 (SHAFT)METALOGIC/MG
|
| 530.02 Simplify Using JamIt and Cleanup |
| Fri, December 21 2007 |
|
Jampack will now use a simpler interface to Magus.
The MCP interfaces have been simplified so Jampack no longer has any
INCLUDEs.
These changes reduce the Jampack symbol by 200 cards and the images
compiled by about 1000.
|
| 530.01 Update for Mcp 5.3 |
| Tue, December 11 2007 |
|
Jampack is now qualified for use on MCP 5.3
|
| 520.02 Internal change |
| Mon, December 3 2007 |
|
Use Include/Meta
|
| 520.01 Increase Max MCP version to 52 |
| Tue, April 10 2007 |
|
Version 510.01 of Jampack is compatible with MCP 5.2 but gave
incompatibility warnings. This change changes the maximum MCP check
to use MCP 5.2.
|
| 510.01 COMPATIBILITY |
| Thu, March 17 2005 |
|
Confirm compatibility with MCP 5.1
|
| 500.02 CHECK FOR OTHER RESERVES AND DETECT SYNTAX |
| Thu, December 9 2004 |
|
Previously, if a manual SQUASH or RES command was input for a disk family that was currently being processed by JAMPACK, JAMPACK could fault with a DIVIDE BY ZERO @ 34380000 or hang indefinitely waiting for a RESERVEDISK to complete that had already failed because of the above. In the latter cases, the JAMPACK RESERVEDISK would fail with SYNTAX and its EOJ not always being detected correctly.
Both these issues have been addressed. If a running JAMPACK detects that a manual RES or SQUASH is already active for same family the JAMPACK will gracefully terminate with:
RESERVE OR SQUASH IS ALREADY RUNNING ON <FAMILYNAME>
If JAMPACK detects a RESERVE that has failed with SYNTAX, the following message will be displayed:
RESERVEDISK JOB NNNNN HAS FAILED WITH SYNTAX
If possible, JAMPACK will attempt another RESERVEDISK but, in most cases, this message will be seen if another SQUASH or RESERVE is already active.
Dnote JAMPACK 500.01 Confirm compatibility with MCP 5.0
|
| 500.01 UPDATE VERSION FOR MCP 49.1 |
| Mon, February 16 2004 |
|
Confirm compatibility with MCP 5.0
|
| 490.01 UPDATE VERSION FOR MCP 49.1 |
| Mon, August 18 2003 |
|
JAMPACK has now been updated for MCP 49.1 to relax the warning messages seen on earlier MCP versions.
|
| 480.03 ONLY WARN FOR TARGETS > 7 DIGITS |
| Tue, July 2 2002 |
|
Previous versions of Jampack would give an error if a target area > 7 digits was requested. Metalogic still recommend that 7 digit targets should not be exceeded, but a warning is now given instead of an error. The new limit is now 549755813887, the maximum size of an integer allowed on MCP systems.
|
| 480.02 ALLOW FAMILY INDEX UP TO MCP MAX OF 255 |
| Wed, March 13 2002 |
|
Previously, JAMPACK would reject requests for individual family indices greater than 99. This restriction has been removed and index values up to 255 (the MCP maximum) are permitted.
|
| 480.01 MCP 48.1 QUALIFICATION |
| Fri., January 18 2002 |
|
Although JAMPACK has been qualified for use on MCP 48.1 for some time now, any JAMPACK run on this level was giving misleading warnings about MCP qualification. This warning has now been removed.
|
| 470.04 MAKE MCP LEVEL CHECKS WORK AGAIN |
| Date: Thu, January 11 2001 |
|
Dnote 470.03 did not quite fix the described problem because the MCP version checking was not correctly including the patch level. This problem is now fixed
|
| 470.03 FIX INTEGER OVERFLOW IN 470.02 |
| Date: Wed, January 10 2001 |
|
Patch 470.02 inadvertently caused JAMPACK to F-DS with integer overflow when checking MCP levels. This fault has now been fixed.
|
| 470.02 DETOUR MCP GETSTATUS BUG |
| Date: Wed, January 3 2001 |
|
MCP 46.189.8633 introduced a bug which would stop Jampack from recognising that ReserveDisk was waiting. This change detours the problem.
|
| 470.01 SUPPORT FOR MCP 4.7 |
| Date: Thu, September 7 2000 |
|
This change supports Jampack on MCP 4.7
|
| 460.02 SUPPORT FOR 5-DIGIT MIXNUMBERS |
| Date: Fri., June 9 2000 |
|
JAMPACK now supports the Unisys MORETASKS implementation allowing the provision of mixnumbers up to 65535. This is achieved by setting the system options MORETASKS (OP+MORETASKS), halt-loading and changing the MAX command.
|
| 460.01 PREVENT INVALID INDEX |
| Date: Mon., February 14 2000 |
|
In very rare circumstances if Jampack missed the waiting notice for RESERVEDISK, it would fail with an invalid index after DSing the waiting RESRVEDISK.
|
| 450.04 HANDLE PACKS WITH > 2**24 SEGS |
| Date: Fri., August 20 1999 |
|
JAMPACK can now handle pack units with up to 2**32 segments, which is the SSR 45.1 MCP limit. The memory usage of JAMPACK will be somewhat larger due to this change, but the CPU consumption should normally be slightly less.
JAMPACK will now correctly report some errors when linking to MAGUS.
This JAMPACK must be used with MAGUS Version 460.02 or later.
|
| 450.03 JAMPACK 46.1 FIELD TEST COMPATIBILITY |
| Date: Thu, July 15 1999 |
|
JAMPACK is now compatible with MCP Field Test version 46.143.
|
| 450.02 LIMIT JAM TO FAMS WITH U SEGS < 2**24 |
| Date: Thu, June 24 1999 |
|
Jampack will now notify the user with an error message if the family contains a pack with more than 2**24 segments.
This restriction will be lifted as soon as possible.
|