Digital PDFs

AA-H427H-TC

January 1990

116 pages

Original

6.3MB

Document:	RSX-11M-PLUS Release Notes
Order Number:	AA-H427H-TC
Revision:	0
Pages:	116
Original Filename:	http://bitsavers.org/pdf/dec/pdp11/rsx11m_plus/RSX11Mplus_V4.x/1/AA-H427H-TC_RSX-11M_PLUS_V4.3_Release_Notes_Jan90.pdf

OCR Text

RSX-IIM-PLUS

Release Notes

Order Number: AA-H427H-TC

RSX-ll M-PLUS
Release Notes
Order No. AA-H427H-TC

Revision/Update Information:
RSX-ll M-PLUS Version 4.3

Digital Equipment Corporation

Maynard, Massachusetts

First Printing, October 1979
Revised, April 1982
Revised, April 1983
Revised, July 1985
Revised, September 1987
Revised, May 1988
Revised, April 1989
Revised, January 1990

The infonnation in this document is subject to change without notice and should not be construed as
a commitment by Digital Equipment Corporation.
Digital Equipment Corporation assumes no responsibility for any errors that may appear in this
document.
Any software described in this document is furnished under a license and may be used or copied only
in accordance with the tenns of such license. No responsibility is assumed for the use or reliability
of software or equipment that is not supplied by Digital Equipment Corporation or its affiliated
companies.
Restricted Rights: Use, duplication, or disclosure by the U.S. Government is subject to restrictions as
set forth in subparagraph (cX1Xii) of the Rights in Technical Data and Computer Software clause at
DFARS 252.227-7013.
© Digital Equipment Corporation 1979, 1982, 1983, 1985, 1987, 1988, 1989, 1990
All rights reserved. Printed in U.S.A.
The Reader's Comments fonn at the end of this document requests your critical evaluation to assist
in preparing future documentation.
The following are trademarks of Digital Equipment Corporation:
DDCMP
DEC
DEC/CMS
DECfMMS
DEC net
DECUS
DEC writer
DIBOL
LA50
LA75
LN01
LN03
LSI
MicroPDP-11

MicrolRSX
MicroVMS
PDP
Professional
Q-bus
RA70
RA82
RA90
RD31
RD32
RD50
RD54
ReGIS
RK07

RL01
RL02
RSTS
RSX
RSX-11M1S
TK50
TU81
UNIBUS
VAX
VAXrdblVMS
VAXcluster
VAX DOCUMENT
VMS
VT

mamBUm0

S1376

This document was prepared with VAX DOCUMENT, Version 1.2.

Contents

Chapter 1 New Features, Restrictions, and Supplementary
Information
1.1
1.2

1.3

Sum.mary of New Features .............................................. 1-1
New Hardware Support in Version 4.3 ...................................... 1-2
1.2.1
Support for RA70 and RA90 Disk Drives ............................. 1-2
1.2.1.1
ERRLOG Support ........................................ 1-2
1.2.2
Supplementary Hardware Information .............................. 1-2
1.2.2.1
Large Disks ............................................. 1-2
Software Enhancements for Version 4.3 ..................................... 1-2
1.3.1
Task Builder Enhancement ....................................... 1-2
1.3.2
PLAS Directive ................................................ 1-3
1.3.3
Support for PDP-II C Compiler ................................... 1-3
1.3.4
Backup and Restore Utility (BRU) .................................. 1-3
1.3.4.1
Compatibility with VAX-II RSX and VAX CoProcessorlRSX ........ 1-3
1.3.4.2
Enhancement to BRU and the MU Device Driver ................ 1-3
1.3.4.3
Options for BRU Qualifiers /COMPARE and NERIFY ............ 1-4
1.3.4.4
BRU Error Messages ...................................... 1-6
1.3.5
VFY Warning Message ........................................... 1-7
1.3.6
Load Rates for Writes ........................................... 1-7
1.3.7
PIP Wildcard Support ........................................... 1-8
1.3.8
HELLO ...................................................... 1-9
1.3.9
Resource Accounting ...........................
1-9
Software Problems Corrected in Version 4.3 ................................. 1-9
1.4.1
Executive Support for Volume Valid ......., ............. " ............ 1-9
1.4.2
Device Driver ................................................. 1-10
1.4.3
Terminal Driver ............................................... 1-10
1.4.4
BRU ........................................................ 1-10
1.4.5
EXEC ....................................................... 1-10
j

1.4

iii

• • • • • • • • • • • • • • • •

1.5

1.6

Data Caching ................................................. 1-10
1.4.6
Installing EDT ................................................ 1-10
1.4.7
Installing VMR with Deferred Binding ............................. 1-11
1.4.8
SET !INQUIRE Command ....................................... 1-11
1.4.9
1.4.10 SORT Command .............................................. 1-11
1.4.11 Task Builder ................................................. 1-11
1.4.11.1
Bitmask Option ......................................... 1-11
1.4.11.2
Overlaid Libraries ....................................... 1-11
1.4.12 LP: Device ................................................... 1-11
1.4.13 Replace Control Task (RCT) Memory Allocation ...................... 1-12
1.4.14 HELLO and HELP Overlays ..................................... 1-12
1.4.15 Queue Manager (QMG) ......................................... 1-12
1.4.16 Set Password Task (PSW) ....................................... 1-12
1.4.17 VMR LOAD and UNLOAD Commands ............................. 1-12
Software Problems and Restrictions in Version 4.3 ........................... 1-12
1.5.1
Error Log File Reporting ........................................ 1-12
1.5.2
Report Generator Task History Qualifier ............................ 1-13
Supplementary Software Information ..................................... 1-13
1.6.1
Disk Save and Compress Utility .................................. 1-13
1.6.2
File Control Services Changes Supplementary Information .............. 1-13
1.6.3
TKB Restrictions and Supplementary Information .................... 1-14
1.6.3.1
OTS Fast Map Restriction ................................. 1-14

Chapter 2 Corrections to Documentation
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
2.10
2.11
2.12
2.13
2.14
2.15
2.16
2.17

RSX-11M-PLUS Batch and Queue Manual .................................. 2-1
RSX-11M-PLUS Command Language Manual ............................... 2-1
RSX-11M-PLUS Indirect Command Processor Manual ......................... 2-6
RSX-11M-PLUS MCR Operations Manual .................................. 2-7
RSX-11M-PLUS and MicrolRSX Crash Dump Analyzer Reference Manual ........ 2-15
RSX-11M-PLUS and MicrolRSX Debugging Reference Manual ................. 2-16
RSX-11M-PLUS and MicrolRSX Error Logging Manual ....................... 2-16
RSX-11M-PLUS and MicrolRSX Executive Reference Manual .................. 2-17
RSX-11M-PLUS and MicrolRSX Guide to Writing an I/O Driver ................ 2-24
RSX-11M-PLUS and MicrolRSX I/O Drivers Reference Manual ................. 2-25
RSX-11M-PLUS and MicrolRSX System Management Guide ................... 2-34
RSX-11M-PLUS and MicrolRSX System Library Routines Reference Manual ...... 2-35
RSX-11M-PLUS and MicrolRSX Task Builder Manual ........................ 2-36
RSX-11M1M-PLUS and MicrolRSX I/O Operations Reference Manual ............ 2-43
RSX-11MIM-PLUS RMS-11 Macro Programmer's Guide ...................... 2-43
RSX-11M1M-PLUS RMS-11 User's Guide .................................. 2-44
RSX-11M1M-PLUS RMS-11 Utilities ..................................... 2-46

Chapter 3 RMS-11 Version 2.0 Software Modifications, Restrictions,
and Supplementary Information
3.1

3.2

3.3

RMS-11 Corrections .................................................... 3-1
3.1.1
RMS-11 Local Access ............................................ 3-1
3.1.2
RMS-11 Ease of Use ............................................ 3-2
3.1.3
RMS-11 Utilities Enhancements ................................... 3-2
Software Restrictions ................................................... 3-2
3.2.1
RMS-11 Access Methods ......................................... 3-2
3.2.2
RMS-11 Remote Access Methods ................................... 3-2
3.2.3
RMS-11 Utilities ............................................... 3-2
RMS-11 Supplementary Information ....................................... 3-3
3.3.1
Enhancements to RMSDES Utility ................................. 3-3
3.3.2
RMS-11 Installation ............................................ 3-3
3.3.2.1
Location ofRMS-11 Files on the Distribution Kit ................ 3-3
3.3.2.2
Startup Command Procedures ............................... 3-7
3.3.2.3
Utility Configurations ..................................... 3-8
3.3.2.4
Utility Command and ODL Files ............................. 3-8
3.3.2.5
Rebuilding the Utilities ................................... 3-10

Chapter 4
4.1
4.2

Correction Files ....................................................... 4-1
Installing Layered Products Restriction ..................................... 4-2

Appendix A
A.1
A.2
A.3

Layered Product Support

The Digital Equipment Computer Users Society
(DECUS)

DECUS Goals ....................................................... A-1
DECUS Activities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. A-2
DECUS Membership . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. A-2

Appendix B Reporting Problems
Appendix C
C.1
C.2
C.3
C.4

Applying Corrections to Source Files

Updating an Executive Source Module. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Updating an MCR Source Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
Updating a DCL Source Module .........................................
Applying Corrections to the File Systems (F1IACP) ..........................

C-1
C-3
C-5
C-6

C.5
C.S
C.7
CoB
C.9

Applying COlTections to the Reconfiguration Tasks CON and HRC ...
Applying FCS COlTections ...............................
Updating the FCS Sources .................................
Assembling FCS
Replacing the FCS Object Modules ..........
0

••••••••••••

•••••••••

•

••••••••

•

••••

••••••

•

••

•••••

••

•••••••••••

•

•••••••

•

••

C-7
C-8
C-9
C-9
C-12

Index
Tables
2-1
2-2
2-3
2-4
3-1
4-1

Number of Index File Headers
2-2
Default Number of Files-Approximation .
2-2
Number of Index File Headers ..................................... 2-10
Default Number of Files-Approximation ...
2-11
RMS-11 Files on RSX-11M-PLUS Distribution Kit ........
3-4
Layered Product COlTection File References ................
4-1
0

••••

••

•••••••••••

•

••

•

•••••••

•••••

•

•••••

••••••••••••••••

••••••••••••

••••••••••

Preface

Manual Objectives
The RSX-llM-PLUS Release Notes contains important information for using the RSX11M-PLUS Version 4.3 operating system. Read this document before generating and using
your system.
This manual contains the following categories of information:
•

Descriptions of new software features as well as changes to existing software features

•

Descriptions of software and hardware restrictions

•

Supplementary information, which appeared in previous release notes but is still
applicable

•

Corrections to documentation errors and omissions

•

Software modifications, restrictions, and supplementary information for RMS-l1 Version
2.0

•

Correction files and restrictions for layered products

Intended Audience
This manual is intended for all users of the RSX-11M-PLUS Version 4.3 operating system.

Document Structure
Chapter 1 describes software features, software and hardware restrictions, and
supplementary information for the RSX-IIM-PLUS Version 4.3 operating system.
Chapter 2 corrects errors and omissions in the RSX-11M-PLUS documentation set.
Chapter 3 describes software modifications, restrictions, and supplementary information for
RMS-ll Version 2.0.

vii

Chapter 4 contains correction files and restrictions for layered products.
Appendix A describes the Digital Equipment Computer Users Society (DECUS).
Appendix B describes the procedure for submitting a Software Performance Report (SPR).
Appendix C describes the procedure for applying corrections to source files.

Conventions
The following conventions are used in this manual:
Convention

Meaning

A right angle bracket is the default prompt for the Monitor Console
Routine (MCR), which is one of the command interfaces used on
RSX-IIM-PLUS systems. All systems include MCR.

A dollar sign followed by a space is the default prompt of the Digital
Command Language (DCL), which is one of the command interfaces
used on RSX-IIM-PLUS and MicrolRSX systems. Many systems
include DCL.

Three characters followed by a right angle bracket indicate the
explicit prompt for a task, utility, or program on the system.
UPPERCASE

Uppercase letters in a command line indicate letters that must
be entered as they are shown. For example, utility switches must
always be entered as they are shown in format specifications.

command abbreviations

Where short forms of commands are allowed, the shortest form
acceptable is represented by uppercase letters. The following
example shows the minimum abbreviation allowed for the DCL
command DIRECTORY:
$ DIR

lowercase

Any command in lowercase must be substituted for. Usually the
lowercase word identifies the kind of substitution expected, such as
a filespec, which indicates that you should fill in a file specification.
For example:
filename.filetype;version

This command indicates the values that comprise a file specification;
values are substituted for each of these variables as appropriate.
/keyword,
Iqualifier,
or
Iswitch

A command element preceded by a slash (I) is an MCR keyword; a
DCL qualifier, or a task, utility, or program switch.
Keywords, qualifiers, and switches alter the action of the command
they follow.

parameter

Required command fields are generally called parameters. The most
common parameters are file specifications.

viii

Convention

Meaning

Commas are used as separators for command line parameters
and to indicate positional entries on a command line. Positional
entries are those elements that must be in a certain place in the
command line. Although you might omit elements that come before
the desired element, the commas that separate them must still be
included.
[g,m]
[directory]

The convention [g,m] signifies a User Identification Code (UIC).
The g is a group number and the m is a member number. The UIC
identifies a user and is used mainly for controlling access to files
and privileged system functions.
This may also signify a User File Directory (UFD), commonly called
a directory. A directory is the location of files.
Other notations for directories are: [ggg,mmm], [gggmmm], [ufd],
[name], and [directory].
The convention [directory] signifies a directory. Most directories
have 1- to 9-character names, but some are in the same [g,m] form
as the UIC.
Where a UIC, UFD, or directory is required, only one set of brackets
is shown (for example, [g,m]). Where the UIC, UFD, or directory is
optional, two sets of brackets are shown (for example, [[g,m]]).
A vertical ellipsis shows where elements of command input or
statements in an example or figure have been omitted because they
are irrelevant to the point being discussed.

KEYNAME

This typeface denotes one of the keys on the terminal keyboard; for
example, the RETURN key.

"print" and "type"

As these words are used in the text, the system prints and the user
types.

In interactive examples, what the user types is printed in red.
System responses appear in black.

Chapter 1
New Features, Restrictions, and Supplementary
Information
This chapter describes the new software features, lists software problems corrected in this
release, lists software and hardware restrictions, and provides supplementary information
for the RSX-IIM-PLUS Version 4.3 operating system.

1. 1 Summary of New Features
The RSX-IIM-PLUS Version 4.3 operating system supports the following new features:
•

New support for the RA70 and RA90 disk drives (in SA550 and SA650 arrays only).

•

The Task Builder has been enhanced so tasks with memory resident overlays can be
extended at runtime with the EXTK$ directive.

•

The PLAS create region directive has a new option whereby the executive will create
a region if, and only if, it can fit in memory. (The current create region directive will
create it even if it cannot fit and, when the task tries to map this region, it will hang.)

•

Support has been added to DCL to translate commands for the PDP-II C compiler.

•

BRU has been modified to produce more explicit error messages during the double
buffered verify or compare operations.

•

VFY now has an abort AST that displays a warning message.

•

RMD now supports write load with data caching.

•

PIP now supports wildcards as part of the directory specification. PIP is also 5 to 10
times faster doing wildcard lookups than in previous versions of RSX-IIM-PLUS.

•

HELLO now logs invalid login attempts on the console.

New Features, Restrictions, and Supplementary Information 1-1

1.2 New Hardware Support in Version 4.3
1.2. 1 Support for RA70 and RA90 Disk Drives
The following new devices are supported also:
•

RA70 disk drive (in SA550 and SA650 arrays only).

•

RA90 disk drive (in SA550 and SA650 arrays only).

The SA550 is an array of RA70s and RA82s and the SA650 is an array of RA70s and RA90s.
Please note that RA82 disks were already supported.
1.2. 1. 1 ERRLOG Support

Error logging support has been added for RA70 and RA90 disk drives (in SA550 and SA650
arrays only).

1.2.2 Supplementary Hardware Information
For a complete list of hardware supported by RSX-llM-PLUS, refer to the Software
Product Description.
1.2.2. 1 Large Disks

Large disks exceed the previous Files-II Structure Levell restriction on the maximum
number of logical blocks. The new restriction allows a maximum of approximately 16
megablocks.
Please note that even though the RSX-llM-PLUS operating system supports a larger
number of logical blocks, support for a larger maximum number of files has not been
included. The maximum number of files allowed on a disk is still limited to 65,500.

1.3 Software Enhancements for Version 4.3
1.3. 1 Task Builder Enhancement
Previously, the EXTK$ and EXTM$ extend task directives would not attempt a task extend
if the task had memory resident overlays. This restriction has been removed as described
in the following paragraphs. (See Section 5.3.32, page 5~4 in the RSX-11M-PLUS and
Micro / RSX Executive Reference Manual for a description of the EXTK$ directive, and
Section 2-8, page 2-17 of the RSX-11M-PLUS ~rsion 4.3 Release Notes for a description of
the EXTM$ directive.)
The restriction is overridden if the task is built with the new Task Builder /EX task file
switch. Using this switch when performing a task build will permit extension of the task
when a valid EXTK$ or EXTM$ directive is issued. Normal physical allocation of memory
positions the root first, then any memory resident overlays, followed by any disk resident
overlays. When the lEX switch is used, the Task Builder modifies the physical allocation
of memory so that the root of the task is positioned after all memory resident overlays, but
before any disk resident overlays, thus allowing the task to be extended, since any extension
is done to the root. For ID tasks, only the D-space root (along with the D-space stack and
header) is shifted, since only the D-space root is extended.

1-2 New Features, Restrictions, and Supplementary Information

Virtual memory is also reallocated to facilitate run-time task extensions over APR
boundaries as follows: after all allocation of APRs has been carried out (for the task, shared
regions, VSECTS, etc.), the task overlays are shifted up as high as possible, to allow for
run-time task extension. This shift is done in one contiguous block, and applies to both
memory resident and disk overlays. Note that a disk overlay can only occur up-tree from a
memory resident overlay, and will share the same window as the memory resident overlay.
As an example, suppose a task is built which normally uses APR 0 for the root, APRs 1
and 2 for memory resident (and possibly disk resident) overlays, and APR 7 for a library. If
the /EX switch is used, APR 0 will continue to be allocated for the root, and APR 7 will be
allocated for the library as before. However, the task overlays will now be allocated to APRs
5 and 6, leaving room for task extension into APRs 1 through 4.

1.3.2 PLAS Directive
A new facility has been added to the Create Regions (CRRG$) directive.
If RS.RES is set in the region definition block prior to executing the CRRG$ directive, the
region will be created only if it can immediately be made resident (without necessitating
checkpointing of other resident regions). This includes checking to see that it will fit in
the currently available holes. If space is not currently available, the directive will fail and
return a status of IE.ALG (-84.)
This allows an application to create a shared region without the complication that the
region may be permanently or semi-permanently locked out of memory by other regions,
even though its size can be accommodated in the partition requested.

1.3.3 Support for PDP-II C Compiler
DCL will support PDP-II C Version 1.0. When PDP-II C is installed, issuing "CC" along
with any PDP-II C command qualifiers or files will invoke the PDP-II C compiler.

1.3.4 Backup and Restore Utility (BRU)
1.3.4. 1 Compatibility with VAX-11 RSX and VAX CoProcessor/RSX

For compatibility with the VAX-II RSX and VAX CoProcessorlRSX versions of BRU,
BRU was modified to unload (IO.STC with VV$UNL set) disks used in multi-volume disk
operations before prompting the user to mount the next volume.
1.3.4.2 Enhancement to BRU and the MU Device Driver

BRU has been enhanced to improve its method of verifying or comparing data backed up
from a disk to an MU-type tape device, such as the TK50. MUDRV, the driver for the
MU-type devices, has also been enhanced to direct the hardware to perform a compare-host
function.
Previously, BRU used a single buffer to hold the data from the disk and another buffer to
hold the data from the magnetic tape that it was comparing or verifying. This caused the
tape to stop and restart frequently while data was being loaded into BRU's single buffer.

New Features, Restrictions, and Supplementary Information 1-3

Now BRU uses both of its buffers to hold data from the disk, and a buffer created by the
compare-host function in the hardware holds the data from the tape. BRU can now load
data into one buffer while the other is in use, thus requiring fewer starts and stops of the
tape. These changes increase the efficiency of compare and verify operations.
The new behavior is the default for MU-type devices only. You may override this default
by using the SINGLEBUFFER option added to the existing BRU qualifiers ICOMPARE
and NERIFY, which ensures that compare and verify operations are done as they were
previously. For a complete description of the new options, see Section 1.3.4.3.
1.3.4.3 Options for BRU Qualifiers /COMPARE and NERIFY

The BRU qualifiers ICOMPARE and NERIFY have two options: DOUBLEBUFFER and
SINGLEBUFFER. The options can be used only with MU-type devices when verifying or
comparing information from a disk to a tape. The input device must be a disk and the
output device must be a tape or tapes.
Notes

1. BRU does not support the ICOMPARE:DOUBLEBUFFER qualifier for backup
sets that contain more than one tape. Specify the ICOMPARE:SINGLEBUFFER
qualifier if you must compare a backup set that consists of more than one tape.
2.

During a verify operation with both MU-type and non-MU-type output devices,
you must use the NERIFY:SINGLEBUFFER qualifier.

Specifying the DOUBLEBUFFER option may improve the performance of streaming
tape devices such as the TK50 cartridge tape device by double buffering 1/0 operations.
Specifying the SINGLE BUFFER option will cause BRU to single buffer I/O operations,
which was the default action for versions of BRU prior to Version 4.1.
The formats of the ICOMPARE qualifier and the NERIFY qualifier are shown next.
Formats

ICOMPARE[:option]
and
NERIFY[:option]
Parameter
option

Specifies one of the following two options:
SINGLEBUFFER If you use the SINGLEBUFFER option for a compare or a verify operation
from a disk to a tape, BRU will print the first error found and continue
comparing or verifying. It prints any other errors found.
The SINGLEBUFFER option is the default for devices that are not
MU-type devices.

1-4 New Features, Restrictions, and Supplementary Information

DOUBLEBUFFER The DOUBLEBUFFER option can only be used on MU-type devices. For most
system configurations with MU-type devices, a compare or verify operation
from disk to tape will be more efficient if you use the DOUBLEBUFFER
option. However, the error messages that are reported when you use the
DOUBLEBUFFER option are less explicit than the messages reported when
you use the SINGLEBUFFER option. As soon as an error occurs, BRU reports
the actual error. If there is any subsequent data in the block, BRU cannot
accurately compare or verify the integrity of the remainder of that block. All
the remaining file IDs or file names will be listed as potential errors.
You must use an MU-type device drive that supports the compare
host-data function. The DOUBLEBUFFER option is the default for
MU-type devices.

Examples
BRU>

Specifies a compare operation with a list of all actual errors found in the data block or
header block reported.
BRU>

Specifies a compare operation with improved performance. As soon as an error occurs, BRU
reports the actual compare error. If there is any subsequent data in the block, BRU cannot
accurately compare the remainder of that block. All the remaining file IDs or file names
will be listed as potential errors.
BRU>

Specifies a verify operation with a list of all actual errors found in the data block or header
block reported.
BRU>

Specifies a verify operation with improved performance. As soon as an error occurs, BRU
reports the actual verify error. If there is any subsequent data in the block, BRU cannot
verify the integrity of the remainder of that block. All the remaining file IDs or file names
will be listed as potential errors.

New Features, Restrictions, and Supplementary Information 1-5

1.3.4.4 BRU Error Messages

The following error messages may be displayed during a backup, restore, or compare
operation:
BRU-*FATAL*-Device conflict

Explanation: This error message can be caused by the following problems:
•

You specified the ICOMPARE:DOUBLEBUFFER or
NERIFY:DOUBLEBUFFER qualifier without all MU-type output devices.

•

You specified both tape and disk drives as part of either the input device
specification (for a restore operation) or the output device specification (for a backup
operation).

User Action: If you want to mix types of output magnetic tapes (non-MU with
MU), you must specify either the ICOMPARE:SINGLEBUFFER qualifier or the
NERIFY:SINGLEBUFFER qualifier.
BRU-*WARNING*-Potential data record verify error
File ID number LBN number

Explanation: A data block on the input disk device did not match a data block on the
output tape device. As soon as an error occurs, BRU cannot accurately compare or
verify the integrity of the remainder of that data block. All the remaining file IDs will
be listed as potential errors.
User Action: Repeat the backup operation. If the backup operation fails again, repeat
the operation with a different disk or tape device.
To obtain a list of all actual errors in the data block, repeat the compare operation,
specifying the ICOMPARE:SINGLEBUFFER qualifier.
BRU-*WARNING*-Potential file header record verify error (directory) filename.type;verslon

Explanation: A header record on the input disk device did not match a header record on
the output tape device. As soon as an error occurs, BRU cannot accurately compare or
verify the integrity of the remainder of that header block. All the remaining file names
will be listed as potential errors.
User Action: Repeat the backup operation. If the backup operation fails again, repeat
the operation with a different disk or tape device.
To obtain a list of all actual errors in the header block, repeat the verify operation,
specifying the ICOMPARE:SINGLEBUFFER qualifier.
BRU-*WARNING*-Potentlal file ID area or data record verify error
File ID number LBN number

Explanation: The file ID area of a data block on the input disk device did not match
the file ID area of a data block on the output tape device. As soon as an error occurs,

1-6 New Features, Restrictions, and Supplementary Information

BRU is unable to accurately compare or verify the integrity of the remainder of the data
block.

User Action: Repeat the backup operation. If the backup operation fails again, repeat
the operation with a different disk or tape device.
To obtain a list of all actual errors in the file ID area of a data blocK, repeat the verify
operation, specifying the ICOMPARE:SINGLEBUFFER qualifier.
BRU-*FATAL*-Contlnuatlon volumes can not be double buffered

Explanation: BRU does not support the ICOMPARE:DOUBLEBUFFER qualifier for
backup sets that contain more than one tape.
User Action: Specify the ICOMPARE:SINGLEBUFFER qualifier if you must compare a
backup set that consists of more than one tape.
BRU-*WARNING*-Double buffered compare or verify not supported

Explanation: The tape driver does not support
ICOMPARE:DOUBLEBUFFER or NERIFY:DOUBLEBUFFER the qualifier.
User Action: No user action is required. BRU continues the verify or compare operation
in the single buffer mode.

1.3.5 VFY Warning Message
An abort AST has been added to VFY. If the user requests that VFY be aborted while a /UP,
IRE, IDE, or IDH switch is being processed, a message warns the user that disk corruption
may occUr if the abort occurs at this time. If the user issues a second request to abort VFY,
then VFY will be aborted.

1.3.6 Load Rates for Writes
The RMD C and D displays now include percentages on load rates for writes. Also, the
format of these displays have been slightly modified to provide easier reading.

New Features, Restrictions, and Supplementary Information 1-7

>
RSX-11M-PLUS V4.3

BL60

Cache Statistics (General)

Cache Region Name: CACHE

17-JUL-89 12:56:54

Region Size: 47440 (2500. disk blocks)

Device
Name
DUO:
DU3:

Total Hit Fail Load
Reads Rate Rate Rate
304725. 97%
0%
1%
O.
0%
0%
0%

Total Hit Fail Load Defer
Writes Rate Rate Rate Rate
107309. 82%
0%
0%
0%
O.
0%
0%
0%
0%

Total Cache
I/O Ops Used
412034. 95%
O.
0%

Total

304725.

97%

107309.

412034.

RSX-11M-PLUS V4.3

BL60

82%

95%

>
Cache Statistics (Detailed)

Device Name: SYO:
Cache Status: Active

17-JUL-89 12:57:04

Region Name: CACHE
Region Size: 47440
Requests Being Cached: Dir,Ovr,Vir,Rdh

Virtual
Reads
122764.
Read Hit Rate
96%
Read Load Rate
1%

Readahead
O.
0%
0%

Directory
40793.
94%
3%

Logical
O.
0%
0%

Overlay
141176.
99%
0%

Total
304733.
97%
1%

Read Overlap
0%
Extent Too Big
0%
Max Extent Size 127.

0%
0%
127.

2%
0%
3.

0%
0%

0%
0%
127.

49792.
72%
0%

314.
1%
0%

90585.

314.

Writes
57203.
Write Hit Rate
90%
Write Overlap
0%
Total I/O

179967.

Failure Rates (as a % of Total) :
Primary Pool Allocation
0%
Cache Pool Allocation
0%
Read Load
0%

107309.
82%
0%
141176.

Deferred Write Rate
Write Load Rate

412042.

0%
0%

1.3.7 PIP Wildcard Support
PIP has been modified to use the embedded wildcard capability of the F1IACP instead of
processing embedded wildcards internally. This makes PIP 5 to 10 times faster when doing
lookups of files with embedded wildcards. In addition, PIP has been modified to accept
embedded wildcards in a directory specification. For example:
>
Directory lb: [RMS001001]
20-JUL-89 09:28
RMSRESAB.TSK;9
RMSLBL.TSK;9
RMSLBM.TSK;9

48.
19.
31.

C
C
C

30-JUN-89 13:57
30-JUN-89 13:57
30-JUN-89 13:57

Total of 98./98. blocks in 3. files
Directory lb: [RMS001054]
20-JUL-89 09:28

1-8 New Features, Restrictions, and Supplementary Information

RMSCNV.TSK;5
RMSIFL.TSK;5
RMSRST.TSK;5
RMSBCK.TSK;5
RMSDES.TSK;5
RMSDSP.TSK;5
RMSDEF.TSK;5

176.
136.
191187.
248.
183.
139.

C
C
C
C
C
C
C

06-JUL-89
06-JUL-89
06-JUL-89
06-JUL-89
06-JUL-89
06-JUL-89
06-JUL-89

11:33
11:34
11:34
11:34
11:34
11:33
11:33

Total of 1260./1260. blocks in 7. files
Grand total of 1358./1358. blocks in 10. files in 2. directories

1.3.8 HELLO
The HELLO task now displays login attempts on the console. If the account name or mc
is found in the user account file but the failure was due to an invalid password, the failure
will be reported as:
hh:mm:ss

[ggg,mmm] TTnn:

If the failure results from an invalid account name or mc, the report will be:
hh:mm:ss

TTnn:

1.3.9 Resource Accounting
Resource Accounting has been expanded to include a new qualifier to the START and SET
commands. The qualifier is:
/INTRUSION:yes/no

When disabled, invalid login transaction blocks include the user's name and mc, as found
in the user account file, if the login failure results from a password discrepancy. If the user
name or mc is not found in the account file, the invalid login transaction block will simply
list INVALID USER as the user name and the mc field will be blank.
When enabled, invalid login transaction blocks include the user's responses to the Account:
and Password: prompts. The system manager can use this information to determine which
patterns are being used in an intrusion attempt.
The default is IINTRUSION:no

1.4 Software Problems Corrected in Version 4.3
1.4. 1 Executive Support for Volume Valid
The Executive's support for volume valid has been changed to allow the setting of the
hardware and software volume valid bit (a QIO with a function code of IO.STC and a
second parameter value ofVV$SET) on any volume which is mounted foreign without
an ACP. Previously, only hardware volume valid could be set on volumes where software
vol ume valid was already set.
This change is primarily intended to ease the handling of multi-volume data sets on foreign
mounted devices, by allowing an application (such as the RSX backup and restore utility,
BRU) to clear hardware volume valid prior to requesting the change of volumes, and then
to reset it after the volume has been changed.

New Features, Restrictions, and Supplementary Information 1-9

1.4.2 Device Driver
An offline DR: device that has a defined system data structure, for an obscure circumstance,
might generate an Wlsolicited interrupt. The device driver, DRDRV, did not properly
handle the unsolicited interrupt resulting in yet another unsolicited interrupt. The system
would continuously service these Wlsolicited interrupts and hang. This problem has been
corrected.

1.4.3 Terminal Driver
If an input interrupt occurred while executing ACD code called by the terminal driver, a
loss of context would occur. This was caused by the terminal driver's use of the UCB word,
U.AFLG. This problem has been corrected.
The terminal driver used to time-out if a large DMA write request was executed. This
problem has been fixed.

1.4.4 BRU
In previous releases, restoring a backupset to a mounted disk with the /NOINI switch would
occasionally cause the error message "BRU-*WARNING*-Error accessing file." This error
would occur when the internal table of file IDs built by BRU when restoring a save set to
a mounted disk would fill up during the restore. Because this table is allocated one entry
for each retrieval pointer, and zero block files contain no retrieval pointer information, BRU
would zero the counter and then overwrite the file IDs at the beginning of the table. This
restored data to the wrong file ID and resulted in this error.
The restore pass of BRU has been modified to exclude the file IDs of zero block files in the
internal file ID table. This corrects this problem.

1.4.5 EXEC
A bug has been fixed in the EXEC where the use count of virtual terminals would be
incremented based on an APR 6 mapping which was not always the Task Control Block
(TCB). This prevented virtual terminals from being eliminated when they should. This
problem has been fixed.

1.4.6 Data Caching
Data caching would crash the system when readahead support was turned on with a virtual
disk driver. This problem has been fixed.

1.4.7 Installing EDT
If EDT was installed first with an external header (IXHR=YES or
INORESIDENT_HEADER) and then with a pool-resident header (IXHR=NO or
/RESIDENT_HEADER), the task failed by trapping at location 6. This problem has been
corrected.

1-10 New Features, Restrictions, and Supplementary Information

1.4.8 Installing VMR with Deferred Binding
If a task was built against clustered resident libraries and then, at run time, attempted to
create and map an address window using the CRAW$ directive, the directive would return a
W.NSTS word with the WS.UNM bit incorrectly set. This problem occurred only if the task
was installed from VMR with deferred binding enabled. MCR INSTALL did not display this
effect. This problem has been corrected.

1.4.9 SET /INQUIRE Command
The MCR command SET !lNQUIRE=TI: would hang on a VT300-series terminal when
logged in via a SET !HOST command. A carriage return would have to be typed at the
keyboard for the command to complete successfully. If the VT300 terminal was logged in
directly rather than through a SET !HOST, the command worked correctly. This problem
has been corrected.

1.4.10 SORT Command
Previously, DCL did not allow use of the !INDEXED_SEQUENTIAL qualifier to the input
and the output file for the DCL SORT command. DCL has been modified to allow use of the
!INDEXED_SEQUENTIAL qualifier to both the SORT input and output files.

1.4. 11 Task Builder
The Task Builder would issue the fatal error message "Work file I/O error" when a task
with an extremely large number of global symbols in the root was built with one or more
undefined symbols. This problem has been fixed.
1.4. 11. 1 Bitmask Option

The bitmask option on the TKB library option commands (LIBR, RESLIB, and CLSTR) did
not work with a value of 000. This prevented I&D tasks linked to libraries built with an
RSX-11M-PLUS Version 4.1 or later Task Builder from accessing the library data space
APRs for task use. This problem has been fixed.
1.4.11.2 Overlaid Libraries

Previously, manually-loaded overlaid libraries were not supported for use with I&D tasks.
Instead, the overlaid library had to use autoload to load its overlays. If you tried to link
an I&D task to an overlaid library that required manual-loading, the following error would
occur:
TKB

-- *FATAL* -- Module name contains incompatible autoload vectors

Support has been added for manually-loaded overlaid libraries.

1.4. 12 LP: Device
An open request on the LP: device that failed because of insufficient buffer space in the File
Storage Region (FSR) for the allocation of buffers for that file would result in the spooling of
a nonexistent file. This problem has been fixed.

New Features, Restrictions, and Supplementary Information 1-11

1.4. 13 Replace Control Task (RCl) Memory Allocation
The replacement control task, ReT... , was not automatically fixed in memory. This made a
deadlock possible, in the case where ReT... was not resident and ... LDR issued I/O to a DU:
device that could not be completed without intervention by the ReT... task. The request for
ReT... was queued to ...LDR but could not be processed until the outstanding I/O request
completed. The solution was to fix ReT... in memory so it would not require processing by
the loader. This correction has been added to the VMR commands produced by SYSGEN.

1.4.14 HELLO and HELP Overlays
The overlay structure of the HELLOIHELP tasks (HEL.TSK and HELRES. TSK) has been
modified to improve maintainability. Formerly, there was insufficient task address space to
build either version to include the On-Line Debugging Tool (ODT). HELLO now includes
two overlay segments plus a third segment that contains common data and routines, acting
as a "root" for these two. Data and routines needed only by HELLO have been moved from
the task root (HELROT) to one of the HELLO segments. The HELP overlay segment is not
affected by this modification.
If you have added "user-provided" routines to your HELLO task, you will need to rebuild
using the new overlay structure defined in HELBLD.ODL or HELRESBLD.ODL.

1.4. 15 Queue Manager (QMG)
The print job attributes "Print adjacent to prior job" and "Print job should be held" have
been included in [1,20]QMGBLD.BLD as being valid options for $JATDF. This global symbol
determines the selection of default qualifiers for print jobs which are spooled by the PRINT$
macro or the .PRINT subroutine in an application, or by using the /SP switch in a command
such as MAe or TKB.
Note that if QMG is rebuilt to include these options they will apply to all such spooled print
jobs. The PRINT$ macro, .PRINT subroutine, and /SP switch have no provision to modify
the task-built defaults.

1.4. 16 Set Password Task (PSW)
The PSW (set password task) has been fixed to return the correct exit status.

1.4. 17 VMR LOAD and UNLOAD Commands
An invalid VMR LOAD or UNLOAD command would cause all subsequent commands of the
same type to fail. This problem has been fixed.

1.5 Software Problems and Restrictions in Version 4.3
1.5. 1 Error Log File Reporting
RPT reports the version number of the LOG.ERR file in octal instead of decimal in the
"Final Selection" area of the error log report.

1-12 New Features, Restrictions, and Supplementary Information

1.5.2 Report Generator Task History Qualifier
The Report Generator Task /HISTORY qualifier does not always work when a date/time
range is specified. The history summary information within the Report Generator (RPT)
is valid only when there is a MOUNT (or RESET) followed by device activity followed by a
DISMOUNT (or RESET) with no activity outside that interval. It is likely that devices that
are mounted before Error Logging is activated and that remain mounted for the duration,
or are left mounted when Error Logging is turned off, will not be properly displayed in
the history summary. No record will be created for those devices in the history summary
database even if they reported errors during the selection range. This occurs because those
devices were not followed by a RESET record within the specified date/time range to force
updates of the counts. The QIO counts may also be wrong because they were not updated
after the last MOUNT record. This problem may be fixed in a later release.

1.6 Supplementary Software Information
1.6. 1 Disk Save and Compress Utility
The Disk Save and Compress Utility (DSC) will not be included on the software kit after
this release of RSX-11M-PLUS Version 4.3. Any data currently saved on media by the
DSC utility will have to be restored using the DSC utility and then resaved using BRU.
Future releases of RSX-11M-PLUS will require customers to use the save and compression
features of the Backup and Restore Utility (BRU).

1.6.2 File Control Services Changes Supplementary Information
File Control Services (FCS) has been modified to support VMS ancillary control process
(ACP) functions that are needed for compatibility with future RSX products and versions
of VAX-11 RSX. These modifications increase the size of the FCS code that is included in
the task image. Because the increase in FCS code size may affect the building of some
large tasks, FCS routine versions that do not have VMS ACP support are included
in the object library [1,1]NOVACPLIB.OLB and in the concatenated object module
[1,1]FCSNOVACP.OBS.
Some large tasks that have a complicated Overlay Description Language (ODL) may be
affected by the internal reorganization of specific FCS modules. This situation is indicated
by "multiply defined symbol" errors occurring in ODL structures that were previously valid.
You must correct the situation by revising the ODL structure.
The object library [1,1]NOVACPLIB.OLB can be used to build individual tasks that do not
have VMS ACP support. The default routines in the system library remain unchanged.
If you replace the routines in the system library (SYSLIB) with the routines from the
concatenated object module [l,l]FCSNOVACP.OBS, you will affect all the tasks that are
built using the system library, and you will be unable to build tasks with VMS ACP support.
If this support is not desired, it can be removed from the system library. The concatenated
object module [l,l]FCSNOVACP.OBS is included on the kits and contains the affected
modules with the support removed. To remove this support from your system, replace
[1,1]FCSNOVACP.OBS in the system library by using the following Librarian Utility
Program (LBR) command:
LBR>

New Features, Restrictions, and Supplementary Information 1-13

If you also want to remove extended logical name support from the system library
by replacing [l,l]FCSNOLOG.OBS in the system library, you must first replace
FCSNOVACP.OBS. Some of the same modules are affected by both extended logical name
support and VMS ACP support, and the modules are included in both of the concatenated
object modules. If FCSNOLOG.OBS is replaced first, some modules in the system library
will contain extended logical name support when FCSNOVACP is replaced because the
modules in FCSNOVACP.OBS contain extended logical name support.

1.6.3 TKB Restrictions and Supplementary Information
The following section describes TKB restrictions and supplementary information.
1.6.3. 1 OTS Fast Map Restriction

The Object Time System (OTS) Fast Map routine uses the RSX-IIM-PLUS fast-mapping
facility, which means that the task must not use the lOT instruction for any purpose except
fast mapping. (For more information on the fast-mapping facility, see the RSX-11M-PLUS
and Micro / RSX Executive Reference Manual.)
Caution
Be sure to use both TKB and the system library supplied for your system to ensure
that the correct OTS modules are incorporated in your task. If you use the Version
4.1 TKB with an incompatible (older) version of the system library to build an
autoloaded overlaid task, the following new fatal error message is issued:
TKB -- *FATAL* -- Incompatible OTS module

Conversely, if you use the Version 4.1 system library with an older version of TKB,
the OTS module FSTMAP is automatically included in any task built, but the
module will not be accessible. The FSTMAP module will not affect the execution or
performance of your task.

1-14 New Features, Restrictions, and Supplementary Informatit

Chapter 2
Corrections to Documentation
This chapter corrects errors and omissions in the RSX-11M-PLUS and RMS-11
documentation sets. The section titles in this chapter are the titles of manuals that require
correction.

2. 1 RSX-ll M-PLUS Batch and Queue Manual
In Section 2.6.1, page 2-15 of the RSX-11M-PLUS Batch and Queue Operations Manual,
please add the following description of the fALL qualifier to the SHOW QUEUE command:
fALL

Displays information on all entries in all queues.

2.2 RSX-ll M-PLUS Command Language Manual
Please add the following information to the RSX-11M-PLUS Command Language Manual:
•

In Section 3.8.1, page 3-41, please note that the DeL command SET
TERMINAUINQUIRE supports the VT300-series terminals in VT200 mode only.

•

In Section 3.8.1, page 3-42, please note that the DeL command SET TERMINAL/model
supports the VT300-series terminals in VT200 mode only. You cannot specify
VT3xx_SERIES as a terminal type.

•

The following listed changes should be made to the IHEADERS:n and
IMAXIMUM_FILES:n qualifiers in Section 5.11, which is the description of the DeL
INITIALIZE command. Please make the following changes:
•

In place of all the text that follows the table in the explanation of the IHEADERS:n
qualifier, please add the following text and table:
The value used for IHEADERS:n is derived from the maximum number of file
headers specified by IMAXIMUM_FILES:n, with consideration for pre-extending
the index file if the maximum number of file headers will require a multiheader
index file (see the description of IMAXIMUM_FILES:n for more information on

Corrections to Documentation 2-1

multiheader index files). You can use Table 2-1 to determine an appropriate value
to use for the IHEADERS:n parameter.
Table 2-1:

Number of Index File Headers

Maximum Number
of Files

Disk Size

Number of
File Headers

Maximum files less than 25588

Disks less than 209 Mb

Headers=maximuml2

Maximum files greater than 25588
and less than 51176

Disks between
210Mb and 419Mb

Headers=25593

Maximum files greater than 51176
and less than 51693

Disks between
420Mb and 423Mb

Headers=25846

Maximum files greater than 51693
and less than 65500

Disks greater than 423Mb

Headers=51693

•

Please replace the second paragraph and formula after the IMAXIMUM_FILES:n
qualifier with the following paragraph, table, paragraph, and formula:
The maximum number of files varies according to disk size. You may want to use
the values in Table 2-2 as the default approximations:

Table 2-2:

Default Number of Files-Approximation

Disk Size

Approximate Number of Files

Up to 64Mb

Calculate the maximum number of files using the
formula following the table

From 64Mb to 532Mb

Specify the number of blocks divided by 16 as the
maximum number of files

Greater than 532Mb

Specify 65500 as the maximum number of files

The default used for the maximum number of files is the theoretical maximum
possible number of one-block files, divided by eight. The following formula calculates
the theoretical maximum number of files:
MAXIMUM FILES = «N-«(N+4095.)/4096.)+9.»*127.)/258.

•

In Section 5.12.5.1, Section 5.12.5.2, and Chapter 9, pages 5-94, 5-96, and 9-6, the
format of the BACKUP qualifiers ICREATED and /MODIFIED is incorrect. The correct
formats are shown next.
Formats

(\;~

BACKUP/CREATED/arg
~"

*~I)
(~
h

:;~
j";

Lt'

and
BACKUP/MODIFIED/arg

Where arg can be specified as either BEFORE:(dd-mmm-yy hh:mm[:ss]) or
AFTER:(dd-mmm-yy hh:mm[:ss]).

2-2 Corrections to Documentation

•

In Section 5.12.6, page 5-104, the second example is incolTect. The COlTect example is
shown next.
$

This example backs up all files on the fixed disk that were modified after 5:00 P.M. on
May 14, 1987. After all the files have been copied onto the tape, BACKUP verifies the
tape. If files on the tape do not verify, BACKUP returns an elTor message.
•

The following section describes an enhancement and lists restrictions for the Backup
and Restore Utility (BRU).
The behavior of the BRU /ERROR qualifier with restore operations is unchanged.
However, the qualifier can now be used with double-buffered compare and verify
operations on data backed up from a disk to an MU-type device. The default number of
elTors is 25.
The BRU Utility has the following restrictions:

•

The !LENGTH qualifier does not apply to cartridge tape devices, such as the TK25
and TK50, and does not function properly if used on those devices.

•

The /NEW_VERSION and !VERIFY qualifiers are mutually exclusive. They cannot
be specified together in the same command line.

•

You must specify /REWIND/APPEND when placing the first backup set on a
magnetic tape that contains a bootable system image.

•

The !TAPE_LABEL qualifier is ignored during a restore operation if there is a
bootable system image at the beginning of the tape.

In Section 5.13.1 of the manual, on page 5-132 immediately before the Examples
heading, please add the following text:
The DCL commands SHOW ASSIGNMENTS and SHOW LOGICALS can display at
your terminal all logical assignments for a specific logical name.
Formats

SHOW ASSIGNMENTS [logical name] [/qualifier]
and
SHOW LOGICALS [logi calname][lqualifier]
Parameter
loglcalname

Specifies the logical name. You can also specify a portion of a logical name and
either the generic wildcard character asterisk (*) or the specific wildcard character
percent sign ( %). The asterisk can represent any number of characters. The percent
sign can represent only one character.
If you specify quotation marks around the logical name, the SHOW ASSIGNMENTS
command interprets the quotation marks literally, while the SHOW LOGICALS
command removes the quotation marks.

Corrections to Documentation 2-3

Qualifiers
IALL
IGLOBAL
IGROUP[:nG]
/LOCAL
/LOGIN
ISYSTEM
trERMINAL[:ttnn:]

Refer to the RSX-llM-PLUS Command Language Manual for information on qualifiers.
•

The display format of logical assignments has been modified to be more readable. The
display specifies the logical name table and then lists the logical name assignments for
the issuing terminal.
Examples
$
(Session Login Logical Table for VT2:)
"DCM"
"ER"

"GO"
"HOME"
"MAP"

"R"
"REVIEW"
"SYS$LOGIN"
"SYS$PROMPT"

"DU: [CACHE]"
"EDT/RO "
"SET /DEF="
"SYS$LOGIN"
"LB: [1,54] RSXIIM. MAP "
"RECALL"
"DU1: [REVIEWS] "
"DU: [USER]" [Final]
"Jim>"

(Session Local Logical Table for VT2:)
"SYS$CLI"

"DCL"

Displays all the login and local assignments for the issuing terminal.
$
(Session Login Logical Table for VT2:)
"SYS$LOGIN"
"SYS$PROMPT" =

"DU:[USER]" [Final]
"Jim>"

(Session Local Logical Table for VT2:)
"SYS$CLI"

"DCL"

Displays the login and local assignments that begin with the letter S for the issuing
terminal.
$
(Session Local Logical Table for VT2:)
"SYS$CLI"

"DeL"

2-4 Corrections to Documentation

Displays the login and local assignments that match the string SYS$CLI for the issuing
terminal.
$
$

Displays no matching login and local assignments for the string "SYS$CLI" for the
issuing terminal.
$
(System Logical Table)
"LI"
"SS"

"WEe"

"LB:" [Final]
"LB:" [Final]
"LB:" [Final]

(Group 7 Logical Table)
"GROUP"

"DR5: [7,43]"

(Session Login Logical Table for VT2:)
"DCM"
"ER"
"GO"
"HOME"
"MAP"
"R"
"REVIEW"
"SYS$LOGIN"
"SYS$PROMPT"

"DU: [CACHE] "
"EDT/RO "
"SET /DEF="
"SYS$LOGIN"
"LB: [1,54] RSX11M.MAP"
"RECALL"
"DU1: [REVIEWS] "
"DU: [USER]" [Final]
"Jim>"

(Session Local Logical Table for VT2:)
"SYS$CLI"

"DCL"

Displays all the system, group, login, and local assignments for the issuing terminal.
$

(System Logical Table)
"SS"

"LB:" [Final]

(Session Login Logical Table for VT2:)
"SYS$LOGIN"
"SYS$PROMPT" =

"DU: [USER]" [Final]
"Jim>"

(Session Local Logical Table for VT2:)
"SYS$CLI"

"DCL"

Displays the system, group, login, and local assignments that begin with the letter S for
the issuing terminal.
•

Section B.1.3, page 8-5, states that the DCL command SET SYSTEM/CRASH_DEVICE
is valid only for pregenerated operating systems. This is no longer true; the command is
now valid for all RSX-llM-PLUS operating systems.

Corrections to Documentation 2-5

2.3 RSX-ll M-PLUS Indirect Command Processor Manual
Please make the following changes to Chapter 2 of the RSX-llM-PLUS Indirect Command
Processor Manual:
•

Change the final sentence in Section 2.4.2 to read as follows:
See Section 2.4.6.1 for more information.

•

In Section 2.6.16, page 2-61 of the manual, remove the second sentence of the third
paragraph. In its place, please add the following text:
The .FORM directive in the Indirect Command Processor (Indirect) includes support for
VT200-series tenninals in VT200 mode.

•

Delete the following text in Section 2.6.22:
Task not installed in system (XQT, .WAIT)

•

Replace the example in Section 2.6.26 with the following:
The following example is from an interactive terminal session:
>
AT.>
AT.>
AT.>
AT.>
>;1
AT.>
>;2
AT.>

(null substring)

AT.>
AT.>
>;1
AT.>
>;2,
AT.>

>@ <EOF>

>
•

An initialization test has been added in the Indirect Command Processor. This test
assures that the allocation of buffers and file structures does not extend into the internal
string storage region. A failure is reported as
AT. -- Initialization error, oode 6.

Change the explanation of this error code, in Appendix A.2 by replacing the last
sentence. The resulting description reads as:
6. Unable to allocate enough space for command and data I/O buffers.
For privileged I:ndirect tasks, Indirect was not installed with a large enough
increment value. The system manager should remove and reinstall Indirect with a
larger increment or in a larger partition.

2-6 Corrections to Documentation

For the nonprivileged Indirect task, the Executive directive Extend Task failed to
return sufficient space for Indirect to allocate the buffers or Indirect was installed
with an excessively large increment value. If it was installed with an increment
value, the system manager should remove and reinstall Indirect with a smaller
increment. Note that it is unnecessary to install the nonprivileged Indirect with an
increment value, unless checkpointing is disabled or checkpoint space has not been
allocated, because it will extend itself as required to the extent that address space is
available.

2.4 RSX-ll M-PLUS MeR Operations Manual
Please make the following corrections to the RSX-IIM-PLUS MCR Operations Manual:
•

In Section 3.6, page 3-12, the descriptions of the ALT keywords IRPRI and fI'ERM are
incorrect. Delete the second paragraph, which is in parentheses, of the IRPRI keyword
description. Also, the first sentence of the fI'ERM keyword description should read as
follows:
"Alters the priority of a task not initiated from the issuing terminal."

•

In Section 3.7.7 of the manual, on page 3-22, immediately before the heading examples,
please add the following text:
The MeR command ASSIGN (ASN) can display at your terminal all logical assignments
for a specific logical name.
Format

ASN Dogicalname][!keyword]
Parameter
loglcalname

Specifies the logical name. You can also specify a portion of a logical name and
ei ther the generic wildcard character asterisk ( *) or the specific wildcard character
percent sign (%). The asterisk can represent any number of characters. The percent
sign represents only one character.
If you specify quotation marks around the logical name, the ASN command
interprets the quotation marks literally, while the DFL command removes the
quotation marks.
Keywords

IALL
IGBL
IGR[=n]
ISYSTEM
fI'ERM[ =ddnn:]

Refer to the RSX-IIM-PLUS MCR Operations Manual for information on keywords.

Corrections to Documentation 2-7

>
(Session Login Logical Table for VT2:)
"DCM"
"ER"

"GO"
"HOME"
"MAP"

"R"
"REVIEW"
"SYS$LOGIN"
"SYS$PROMPT"

"DO: [CACHE] "
"EDT/RO "
"SET /DEF="
"SYS$LOGIN"
"LB: [1,54] RSX11M. MAP "
"RECALL"
"001: [REVIEWS]"
"DO:[OSER]" [Final]
"Jim>"

(Session Local Logical Table for VT2:)
"SYS$CLI"

"DCL"

Displays all the login and local assignments for the issuing terminal.
>
(Session Login Logical Table for VT2:)
"SYS$LOGIN"
"SYS$PROMPT" =

"DO: [OSER]" [Final]
"Jim>"

(Session Local Logical Table for VT2:)
"SYS$CLI"

"DCL"

Displays the login and local assignments that begin with letter S for the issuing
terminal.

>
>
Displays no matching login and local assignments that match the string "SYS$CLI" for
the issuing terminal.
>
(System Logical Table)
"LI"
"SS"

"WK."

"LB:" [Final]
"LB:" [Final]
"LB:" [Final]

(Group 7 Logical Table)
"GROOP"

"DRS: [7,43]"

(Session Login Logical Table for VT2:)

2-8 Corrections to Documentation

"DCM"
"ER"
"GO"
"HOME"
"MAP"

"R"
"REVIEW"
"SYS$LOGIN"
"SYS$PROMPT"

"DU: [CACHE]"
"EDT/RO "
"SET /DEF="
"SYS$LOGIN"
"LB: [1,54] RSX11M.MAP"
"RECALL"
"DU1: [REVIEWS] "
"DU: [USER]" [Final]
"Jim>"

(Session Local Logical Table for VT2:)
"SYS$CLI"

"DCL"

Displays all the system, group, login, and local assignments for the issuing terminal.
•

In Section 3.22.2 of the manual, on page 3-57 immediately before the heading
Examples, please add the following text:
The MCR command DEFINE LOGICALS (DFL) can display at your terminal all logical
assignments for a specific logical name.
Format

DFL Oogicalnam e] [lkeyword]
Parameter
loglcalname

Specifies the logical name. You can also specify a portion of a logical name and
either the generic wildcard character asterisk (*) or the specific wildcard character
percent sign (%). The asterisk can represent any number of characters. The percent
sign represents only one character.
If you specify quotation marks around the logical name, the ASN command
interprets the quotation marks literally, while the DFL command removes the
quotation marks.
Keywords

IALL
IGBL
IGR[=n]
ISYSTEM
trERM[=ddnn:]
Refer to the RSX-llM-PLUS MCR Operations Manual for information on keywords.
The display format of logical assignments has been modified to be more readable. The
display specifies the logical name table and then lists the logical name assignments for
the issuing terminal.

Corrections to Documentation 2-9

Examples
>
(Session Local Logical Table for VT2:)
"SYS$CLI"

"DCL"

Displays the login and local assignments that match the string SYS$CLI for the issuing
terminal.
>
(System Logical Table)
"SS"

"LB:" [Final]

(Session Login Logical Table for VT2:)
"SYS$LOGIN"
" SYS$PROMPT" =

"DU:[USER]" [Final]
"Jim>"

(Session Local Logical Table for VT2:)
"SYS$CLI"

"DeL"

Displays the system, group, login, and local assignments that begin with the letter S for
the issuing terminal.
•

The following listed changes should be made to the IINF and IMXF keywords in Section
3.29, which is the description of INITIALIZE VOLUME command (INI). Please make
the following changes:
•

In place of the text that explains the IINF keyword, please add the following in place
of the first three paragraphs, table, and fourth paragraph:
Specifies the number of file headers to allocate initially in the index file. The
five system files (INDEXF.SYS, BITMAP.SYS, BADBLK.SYS, CORIMG.SYS, and
OOOOOO.DIR) are not included in the value for INF.
The value used for IINF is derived from the maximum number of file headers
specified by !M.XF, with consideration for pre-extending the index file if the
maximum number of file headers will require a multiheader index file (see the
description of!M.XF for more information on multiheader index files). You can use
Table 2-3 to determine an appropriate value to use for the IINF parameter.

Table 2-3:

Number of Index File Headers

Maximum Number
of Files

Disk Size

Number of
File Headers

Maximum files less than 25588

Disks less than 209 Mb

Headers=maximuml2
(continued on next page)

2-10 Corrections to Documentation

Table 2-3 (Cont.):

Number of Index File Headers

Maximum Number
of Files

Disk Size

Maximum files greater than 25588
and less than 51176

Disks between
210Mb and 419Mb

Headers=25593

Maximum files greater than 51176
and less than 51693

Disks between
420Mb and 423Mb

Headers=25846

Maximum files greater than 51693
and less than 65500

Disks greater than 423Mb

Headers=51693

•

Number of
File Headers

Please replace the second paragraph and formula after the IMXF keyword with the
following paragraph, table, and formula:
The maximum number of files varies according to disk size. You may want to use
the values in Table 2--4 as the default approximations:

Table 2-4:

Default Number of Files-Approximation

Disk Size

Approximate Number of Files

Up to 64Mb

Calculate the maximum number of files using the
formula following the table

From 64Mb to 532Mb

Specify the number of blocks divided by 16 as the
maximum number of files

Greater than 532Mb

Specify 65500 as the maximum number of files

•

In Section 3.31, page 3-103, the description and format of the MeR command LOAD is
incomplete. Please add the following sentence to the end of the second paragraph:
"The Load command also loads the extended Executive partitions into memory."
The correct formats of the Load command are shown next.
Formats

LOA[D] dd:[lkeyword(s)]
and
LOA[D] IEXP=expname[lkeyword(s)]

•

In Section 3.34, page 3-128, the formats of the OPE keywords /TASKD and /TASKl are
incorrect. The correct formats are shown next.

Corrections to Documentation 2-11

Formats

trASKD=taskname
and
trASKI=taskname

•

In Section 3.34, the second sentence in the first paragraph on page 3-129 is incorrect; it
should read as follows:
"You are limited to the specified memory region (TASK or REG)."
Also, the second, third, and fourth examples on page 3-129 are incorrect. The correct
examples are shown next.
>
00000000/50712

This command opens the region TSTREG at location 0 and displays in octal the current
value at the location.
>
00000000/50712 % IEscl MCR

This command opens location 0 of TSTREG, displays the current value in octal, and
then displays the Radix-50 value in ASCII format.
>
00000000/50712
00000002/00000
00000004/14604

IEjCI MCR %DCL IRETI
RET

This command sequence displays the current value at location 0 in octal and then
Radix-50 format, and then changes the value. The new value is also in Radix-50 format.
Pressing the RETURN key enters the new value into location 0 and displays it in octal,
then opens the next location in memory and displays its contents in octal.
•

In Section 3.40, pages 3-138, 3-140, and 3-141, the format of the RUN keyword IUIC is
incorrect in Format 3 and Format 4. The correct format is shown next.
Format

IUIC=[g,m]
•

In Section 3.40, page 3-139, the definition of the keyword IUIC is incorrect. The correct
definition is shown next.
/UIC

(Privileged keyword.) The User Identification Code (UIC) under which the task will
be requested to run. This UIC also determines which files the task can access.
The UIC has the format [g,m], where the variables g and m specify octal numbers
between 1 and 377 that represent the group and member numbers, respectively.
The square brackets are required syntax.
When you specify IUIC, the UIC that you specify becomes the default and protection
UICs for the task.

2-12 Corrections to Documentation

When you do not specify lUIe, the terminal's me becomes the default and
protection mes for the task. This is the default.
•

In Section 3.40, page 3-139, the definitions of the parameters taskname and dtime are
incorrect. The correct definitions are shown next.
taskname

Specifies a 1- to 6-character name of an installed task.
dtlme

Specifies a delta time. Delta time is an increment from the current time.
•

In Section 3.40, pages 3-140 and 3-141, the definition of the parameter task name is
incorrect. The correct definition is shown next.
taskname

Specifies a 1- to 6-character name of an installed task.
•

In Section 3.40, page 3-142, the definition of the parameter $ is incorrect. The correct
definition is shown next.

$
When you specify a dollar sign ($ ) in the command line, the parameter dev defaults
to LB:, and [g,m] defaults to the current library me (usually, [3,54]). If the task
cannot be found in the library me, INSTALL searches the system me on device
LB: (usually, [1,54]).
If you do not specify a dollar sign ($), RUN searches for the task image file in the
me (on device SY) to which the terminal requesting the task is set.
Note
If LB has been reassigned to another device and the system VIC for that
device contains privileged tasks built for another system, your system will
fail.

•

In Section 3.40, pages 3-144 and 3-145, the descriptions of the RUN command
keywords IPRI=number and lUIe are incorrect. The correct description is shown next.
/PRI=number

Specifies the priority of the task. The value range is 1 to 25010, where 250 is the
highest priority. Standard number conventions apply: octal by default, decimal if
followed by a period.
If a nonprivileged user specifies a priority that is greater than 5010, the priority is
set to 5010.
The default is IPRI=50.

Corrections to Documentation 2-13

IUICa(g,m)

(Privileged keyword.) Specifies the User Identification Code (mC) under which the
task will be requested to run. The square brackets are required syntax.
If you specify a mc, the mc becomes the default mc and protection mc for the
task. If you do not specify a mc, your default mc and protection UIC become the
default mc and protection mc for the task.
The default is the mc of the terminal that issues the RUN command.
•

In Section 3.42, page 3-154 and page 3-169, the format of the MCR command SET
!INQUIRE is incorrect. The correct format is shown next.
Format

!INQUIRE=term
•

In Section 3.42, page 3-161, the format of the MCR command SET /CRASHDEV is
incorrect. The correct format is shown next.
Format

SET /CRASHDEV=ddnn:[CSRaddr]
or

SET /CRASH_DEVICE=ddnn:[CSRaddr]
Also, the section states that the MCR commands SET /CRASHDEV and SET
ICRASH_DEVICE are valid only for pregenerated operating systems. This is no longer
true; the commands are now valid for all RSX-11M-PLUS operating systems.
•

In Section 3.42, page 3-163 and page 3-164, the description of the MCR command SET
IDPRO is incorrect. The text incorrectly states that the parameter protection-spec has
two formats. The parameter protection-spec can only be specified in the format shown
next.
Format

[RWED,RWED,RWED,RWED]
In Section 3.42, page 3-164, the first paragraph under the bulleted items is incorrect.
The paragraph should read as follows:
The IDPRO keyword establishes a default file protection for your current session at the
terminal. When you start a new terminal session, it resets your file protection to the
system default protection (or to the default protection in your account file if one was
specified when your account was created). To establish a default protection code for all
of your future sessions at the terminal, specify the code as a command in your login
command file (LOGIN. COM).
•

In Section 3.42, page 3-169, add the following information:
The MCR command SET!INQUIRE supports the VT300-series terminals in VT200 mode
only.

•

In Section 3.42, page 3-185, add the following information:

2-14 Corrections to Documentation

The MCR command SETITERM supports the VT300-series terminals in VT200 mode
only. You cannot specify VT3xx as a terminal type.
•

In Section 3.42, page 3-197, the description of the sixth example is incorrect. The
correct example is shown next.
>
SECPOL=285.:640.:44%

Displays the amount of available secondary pool.
•

Immediately before Section 12.1.4.5 on page 12-11 of the manual, please enter the
following text:
The Peripheral Interchange Program (PIP) qualifier IDD must be used with the
ampersand ( & ) character.

2.5 RSX-ll M-PLUS and Micro/RSX Crash Dump Analyzer Reference
Manual
Please make the following correction to the RSX-11M-PLUS and Micro / RSX Crash Dump
Analyzer Reference Manual:
•

Section 1.1.1 specifies that, for RSX-11M-PLUS operating systems that are not
pregenerated, you must select the crash dump driver during system generation. This is
no longer true; RSX-11M-PLUS now supports loadable crash dump drivers.
You can select loadable crash dump support during system generation and choose a
loadable crash driver during the system startup procedure. Crash dump devices can
be DU-, DL-, MU-, MS-, and MM-type devices. Refer to the RSX-11M-PLUS System
Generation and Installation Guide for more information on including loadable crash
support in your system if it is not pregenerated.
In addition, for systems that have loadable crash dump support included, you no
longer must perform another SYSGEN to change the crash dump device or unit
number; you can use either the DCL command SET SYSTEM/CRASH_DEVICE or
the MCR command SET ICRASHDEY. The previous commands are now valid for all
RSX-11M-PLUS operating systems. Refer to the RSX-11M-PLUS Command Language
Manual and the RSX-11M-PLUS MCR Operations Manual for more information on the
commands.

•

Please add the following text to Section 1.1.2 of the manual:
Crash drivers on RSX-11M-PLUS systems with 4Mb of memory have the following
functions:
The MU and DU crash drivers on 4Mb systems keep track of the amount of memory
dumped and terminate the dump when 2044K words of memory have been dumped.
The DL crash driver on 4Mb systems dumps 2044K words of memory.

•

Section 1.1.2.1 specifies an incorrect format for the MCR command SET ICRASHDEV
(or SET ICRASH_DEVICE). The correct format is shown next.

Corrections to Documentation 2-15

Format

SET /CRASHDEV=ddnn[:CSRaddr]

or
SET /CRASH_DEVICE=ddnn[:CSRaddr]
•

Section 1.2.1 specifies that, for systems that are not pregenerated, transferring processor
control to the crash dump driver depends on whether you built the Executive Debugging
Tool (XDT) into your system during system generation. The information in this section
is generally incorrect because RSX-11M-PLUS Version 4.0 included loadable XDT
support for all RSX-11M-PLUS systems and RSX-11M-PLUS Version 4.1 included
loadable crash dump support for all RSX-11M-PLUS operating systems.
Loadable XDT support enables you to load XDT when you want to use it for debugging.
System performance is improved when XDT is not part of the system. Features
such as instruction decoding and automated searching of symbol addresses listed
in the Executive map are included in the loadable version of XDT. Refer to the
RSX-11M-PLUS and Micro / RSX Crash Dump Analyzer Reference Manual for more
information on loadable XDT support.

2.6 RSX-ll M-PLUS and Micro/RSX Debugging Reference Manual
In the RSX-11M-PLUS and Micro / RSX Debugging Reference Manual, Section 1.2.3.3,
please remove the second paragraph. For information on how to write to supervisor-mode
libraries, refer to the RSX-11M-PLUS Release Notes, Section 1.17.1.

2.7 RSX-ll M-PLUS and Micro/RSX Error Logging Manual
Please make the following corrections to the RSX-11M-PLUS and Micro / RSX Error
Logging Manual:
•

In Table 2-2, page 2-8, the following change must be made to the device entry for the
control file module ETSV05:
TSV05 !I'K25

•

ETSV05

In Section 3.3.3.4, the /HISTORY qualifier does not always work as described in the
following text:
"RPT generates a summary report sorted by device error history. It displays the hard
and soft error count and QIO count for every volume on each device."
The use of a date/time range may make the history summary more prone to error. The
history summary information within the Report Generator (RPT) is valid only when
there is a MOUNT (or RESET) followed by device activity followed by a DISMOUNT (or
RESET) with no activity outside that interval. It is likely that devices that are mounted
before Error Logging is activated and that remain mounted for the duration, or are
left mounted when Error Logging is turned off, will not be properly displayed in the
history summary. No record will be created for those devices in the history summary
database even if they reported errors during the selection range. This occurs because
those devices were not followed by a RESET record within the specified date/time range
to force updates of the counts. The QIO counts may also be wrong because they were
not updated after the last MOUNT record.

2-16 Corrections to Documentation

•

In Example A-1, page A-4, the following change must be made to the list of acceptable
device names:
TSV05 or TK25

2.8 RSX-ll M-PLUS and Micro/RSX Executive Reference Manual
Please make the following corrections to the RSX-11M-PLUS and Micro / RSX Executive
Reference Manual:
•

In Section 2.1, page 2-1, the tenth bullet in the list is incorrect. The following is the
correct text:
The execution of the round-robin scheduling algorithm at the end of a round-robin
scheduling interval if the eligibility of the current task has changed.

•

In Section 3.7.2, page 3-21, the last sentence in the paragraph is incorrect. The
following is the correct text:
RO is returned as the status (IS.SUC or IE.ALG). If a length change was requested
(for example, if the high bit of RO was set) and the new length was set by default (for
example, R2 was set to 0), R2 is returned as the new length of the mapping. In all cases
R2 and R3 are destroyed by the fast-map call.

•

In Section 5.13, page 5-40, the macro call for CMKT$ is incorrect. The correct macro
call is shown next.
Macro Call:
CMKT$

[ [efn] , [ast] ]

where:
efn
ast

=
=

Event flag number
Mark time AST address

Macro Expansion:
CMKT$
. BYTE
. WORD
. WORD

52.,MRKAST
27.,3
52.
MRKAST

;CMTK$ MACRO DIC, DPB SIZE = 3 WORDS
;EVENT FLAG NUMBER 52
;ADDRESS OF MARK TIME REQUEST AST ROUTINE

Note
The above example will cancel only the Mark Time requests that were specified
with efn 52 or the AST address MRKAST. If no ast or efn parameters are
specified, all Mark Time requests issued by the task are cancelled and the DPB
size equals 1.

Corrections to Documentation 2-17

•

In Section 5.36, page 5-95, additional symbols have been added to the FEAT$ directive.
Please add the following symbols to Table 5-1, System Feature Symbols:
Table 5-1: System Feature Symbols

•

Symbol

Value

Meaning

FE$NCT

63.

System has NCT support

FE$LSD

64.

System has LUT scan in LOADR disabled

FE$PC3

65.

System supports Professional3xx series personal computers

FE$DFB

66.

System supports deferred binding

FE$RTB

67.

Run-time binding

FE$ODB

6S.

Deferred binding is overridable

FE$XDJ

69.

XDT is doing I/O via TPRs on KXJ

FE$NSY

70.

No local system disk

FE$NCO

71.

No local console

FE$RTK

72.

Remote task services

FE$RDR

73.

Remote directory storage

FE$RLG

74.

Remote logical support

FE$LDR

75.

Remote load/overlays

FE$VTL

76.

VT logins disabled

FE$ANT

77.

Automatic network startup by save

FE$NRT

7S.

Network remote system

FE$EXE

79.

Task file names default to .EXE

FE$CMO

so.

Resident overlays default to 512. byte alignment

In Section 5.64, page 5-188, the macro expansion for the RLON$ and RLOG$ directives
is incorrect. The correct macro expansion is shown next.
RLON$ MOD,TBMSK,STATUS,LNS,LNSSZ,ENS,ENSSZ,RSIZE,RTBMOD
.BYTE
207.,10.
;RLON$ MACRO DIC, DPB SIZE = 10(10) WORDS
.BYTE
14.
;SUBFUNCTION VALUE (RLOG$ = 10(10»
. BYTE
. WORD

MOD
TBMSK

;LOGICAL NAME MODIFIER
;LOGICAL NAME TABLE INHIBIT MASK

• WORD
• WORD

LNS
LNSSZ

;LOGICAL NAME STRING ARRAY
;SIZE (IN BYTES) OF LOGICAL NAME STRING

• WORD
• WORD

ENS
ENSSZ

;RETURNED EQUIVALENCE NAME ARRAY
;SIZE (IN BYTES) OF EQUIVALENCE NAME

• WORD

RSIZE

;LOCATION OF SIZE FOR RETURNED EQUIVALENCE NAME

2-18 Corrections to Documentation

•

. WORD

RTBMOD

;LOCATION OF LOGICAL TABLE NUMBER (LOWER BYTE) AND
;MODIFIER VALUE OF LOCATED LOGICAL NAME (HIGHER BYTE)

. WORD

STATUS

;LOCATION OF LOGICAL NAME STATUS

In Section 5.98, page 5-281, the reference to the CINT$ section in the second sentence
in the first paragraph is incorrect. Replace that sentence with the following:
For information on mapping the subroutines, see Note 2 for the description of the
CINT$ (Connect to Interrupt) directive.

•

In Section 5.98, page 5-282, the calculation of the addresses in Notes 3 and 4 is
incorrect. The calculation should be as follows:
n = n + 120000 + (base & 177700)

•

In Section 5.99, page 5-284, an additional symbol has been added to the TFEA$
directive. Please add the following symbol to Table 5-2, Task Feature Symbols:
Table 5-2: Task Feature Symbols

•

Symbol

Value

Meaning

T4$DFB

41.

Task has deferred binding

In Section 5.100, page 5-288, the macro expansion for the TLON$ and TLOG$ directives
is incorrect. The correct macro expansion is shown next.
TLON$ MOD,TBMSK,STATUS,LNS,LNSSZ,ENS,ENSSZ,RSIZE,RTBMOD
. BYTE
207.,10.
;TLON$ MACRO DIC, DPB SIZE = 10(10) WORDS
. BYTE
13.
;SUBFUNCTION VALUE (TLOG$ = 9(10»

•

. BYTE
. WORD

MOD
TBMSK

;LOGICAL NAME MODIFIER
;LOGICAL NAME TABLE INHIBIT MASK

. WORD
. WORD

LNS
LNSSZ

;LOGICAL NAME STRING ARRAY
;SIZE (IN BYTES) OF LOGICAL NAME STRING

. WORD
. WORD

ENS
ENSSZ

;RETURNED EQUIVALENCE NAME ARRAY
;SIZE (IN BYTES) OF EQUIVALENCE NAME

. WORD

RSIZE

;LOCATION OF SIZE FOR RETURNED EQUIVALENCE NAME

. WORD

RTBMOD

;LOCATION OF LOGICAL TABLE NUMBER (LOWER BYTE) AND
;MODIFIER VALUE OF LOCATED LOGICAL NAME (HIGHER BYTE)

. WORD

STATUS

;LOCATION OF LOGICAL NAME STATUS

In Sections 5.104, 5.105, and 5.106, the definition of the FORTRAN subroutine call
parameter bufadr is incorrect. The correct definition is as follows:
bufaclr

Specifies an array containing data to be sent (must be word aligned (INTEGER*2)).
Please note that the definition of the macro call parameter bufadr is correct and should
not be changed.
•

The following sections describe an Executive modification, list a restriction, and provide
supplementary information.

Corrections to Documentation 2-19

Modification to the IOSUB.MAC Module

A modification was made to the 10SUB.MAC module to prohibit both UNIBUS Mapping
Registers (UMRs) and error logging processing for 10.ATT and 10.DET functions.
Shared Regions Restriction

You cannot build a shared region that contains 1- and D-space. Shared regions that are
built with the lID switch will not operate correctly. As stated in Section 10.19 of the
RSX-llM-PLUS and Micro / RSX Task Builder Manual, you cannot use the /-HD switch
and lID switch in the same build.
Extend Task Directive

The Extend Task directive with the Active Page Register (APR) protection mask
enhancement instructs the system to modify the size of the issuing task by a positive
or negative increment of 32-word blocks. If the directive does not specify an increment
value or if it specifies an increment value of zero, the Executive makes the issuing task's
size equal to its installed size. The issuing task must be running in a system-controlled
partition and cannot have any outstanding I/O when it issues the directive. The task
must also be checkpointable to increase its size; if necessary, the Executive checkpoints
the task and then returns the task to memory with its size modified as directed.
In a system that supports the memory management directives, the Executive does
not change any current mapping assignments if the task has memory-resident
overlays unless the task was built with the Task Builder lEX switch. Please refer
to Section 1.3.1. However, if the task does not have memory-resident overlays, the
Executive attempts to modify, by the specified number of 32-word blocks, the mapping
of the task to its task region.
If the issuing task is checkpointable but has no preallocated checkpoint space available,
a positive increment may require dynamic memory and extra space in a checkpoint file
sufficient to contain the task.
The Extend Task directive with the APR protection mask enhancement enables you to
specify a mask parameter that will prevent the extension from changing the default
mapping of the task's APR mapping. This enables you to extend the data-space window
of a task without changing the mapping of APRs that default to overmapping a library
in instruction space.
There are several constraints on the size to which a task can extend itself using the
Extend Task directive enhancement. These constraints are as follows:
•

No task can extend itself beyond the maximum size set by the MCR command
SET IMAXEXT or the DCL command SET EXTENSION_LIMIT or the size of the
partition in which it is running. (See the RSX-llM-PLUS MeR Operations Manual
or the RSX-l1M-PLUS Command Language Manual.)

•

A task that does not have memory-resident overlays cannot extend itself beyond
32K minus 32 words.

•

A task that has preallocated checkpoint space in its task image file cannot extend
itself beyond its installed size.

2-20 Corrections to Documentation

•

A task that has memory-resident overlays cannot reduce its size below the highest
window in the task partition.

Format

EXTM$ [inc],mask
Parameters
inc

Specifies a positive or negative number equal to the number of 32-word blocks by
which the task size is to be extended or reduced.
mask

Specifies a mask of APRs to be protected. Bit 0 represents APRO and bit 1
represents APRl. For example, if you specify 340, then APRs 7, 6, and 5 are
protected.
Macro Expansion
EXTM$
• BYTE
• WORD
. WORD

40,340
89.,3
40
340

;EXTK$ MACRO DIC, DPB SIZE = 3 WORDS
;EXTEND INCREMENT, 40(8) BLOCKS (lK WORDS)
;APR PROTECTION MASK

Local Symbol Definition
E.XTIN

Extend increment (2)

DSW Return Codes
IS.SUC

Successful completion.

IE.UPN

Insufficient dynamic memory or insufficient space in a checkpoint file.

IE.ITS

The issuing task is not running in a system-controlled partition.

IE.ALG

The issuing task attempted to reduce its size to less than the size of its task header;
the task tried to increase its size beyond 32K words or beyond the maximum set by
the MCR command SET IMAXEX.T or DCL command SET EXTENSION_LIMIT; the
task tried to increase its size to the extent that one virtual address window would
overlap another; the task has memory-resident overlays and it attempted to reduce
its size below the highest window mapped to the task partition; or the extend would
unmap a protected APR.

IE.RSU

Other tasks are attached to this task partition.

IE.IOP

110 is in progress for this task partition.

IE.CKP

The issuing task is not checkpointable and specified a positive integer.

Corrections to Documentation 2-21

IE.NSW

The task attempted to extend itself to larger than the installed size (when checkpoint
space is allocated in the task).

IE.ADP

Part of the Directive Parameter Block (DPB) is out of the issuing task's address
space.

IE.SDP

Directive Identification Code (DIC) or DPB size is invalid.

General Information Directive

The general information directive function Get Device Information Junior (GI.DVJ)
returns information about a particular device. The device for which information is
returned is determined by first performing a logical assignment (if required) and then
following any redirection assignments. Device assignments are checked if the high bit
in the flag's byte is clear; otherwise, no check of device assignments is made.
Format

GIN$ GI.DVJ, buf, siz, dev, unt
Parameters
GI.DVJ

Specifies GIN$ function code (18).
but

Specifies address of buffer to receive the unit information.
slz

Specifies size of buffer. Buffer size is 1 word.
dev

Specifies device name (if blank, use task's TI:).
unt

Specifies device unit number (if high bit clear, follow assignments).
Buffer Format
-1

Device is not a mass-storage and mountable device.

-2

Device is not mounted for issuer.

Bit 0

READ access is prohibited.

Bit 1

WRITE access is prohibited.

Bit 2

CREATE access is prohibited.

Bit 3

DELETE access is prohibited.

Bit 4

Device is mounted public.

2-22 Corrections to Documentation

Bit 5

Device is mounted private (allocated).

Bit 6

Device is mounted foreign.

Bit 7

Foreign device has ACP.

Macro Expansion
GIN$
. BYTE
. WORD
. WORD
• WORD
. WORD
• WORD

GI.DEV, DVBOF, DVSIZ, "TT, 1
169.,6
GI.DEV
DVBOF
DVSIZ
"TT
1

DSW Return Codes

Is.sue

Successful completion.

IE.ADP

Part of Directive Parameter Block (DPB) is out of task's address space.

IE.IDU

The specified device does not exist, or device is a virtual
terminal and issuing task is not parent or offspring.

IE.SDP

Invalid function code or the DPB size is invalid.

Notes

1. If the task has the "slave" attribute, logical assignments are not checked regardless

of the setting of the high bit in the fourth parameter word.
2. If bit 4 and bit 5 are off, the device may be mounted semiprivate; that is, the device
may be mounted by the user but not allocated.
Creating Region Names

RSX-11M-PLUS Version 4.1 allows a region to create a region with the name GEN. In
addition, RSX-11M-PLUS allows regions in region GEN to be created even if a common
named GEN is installed in any main partition.
The Get Partition Parameters directive (GPRT$) uses the subroutine $SRNAM, which
is used by many parts of the executive to look up region and partition names. GPRT$
looks at the main partition list first; then it looks at the common block directory.
The TKB command shown next performs a GPRT$ directive for the main partition GEN.
>

Any application that uses GPRT$ to get the parameters of a named common region gets
the parameters of a main partition if the name of the common region is not unique with
respect to the main partition name. All applications that use $SRNAM must be changed
to attach to the region by name and then to use the Get Region Parameters directive
(GREG$) to obtain the region parameters.

Corrections to Documentation 2-23

2.9 RSX-ll M-PLUS and Micro/RSX Guide to Writing an I/O Driver
Please make the following corrections to the RSX-llM-PLUS and Micro / RSX Guide to
Writing an I/O Driver:
•

Section 1.5.5, page 1-21, illustrates how to create an executive entry point vector table.
It does not indicate that there are any differences between creating a vector table
for a driver and creating a vector table for a privileged task. However, the length is
calculated differently in each case. Use the following statement when determining the
length of the vector table for a privileged task:
EXEVCL=«.-EXEVEC>/2>

Use the following statement when determining the length of the vector table for a
driver:
EXEVCL=«<.-EXEVEC>/2>-1>

•

In Section 1.5.6, page 1-22, the table entitled Callable Routine for Converting Executive
References in a Driver is incorrect. The correct table is as follows:
MOV
MOV
MOV
MOV
MOV
MOV
CALL
MOV

•

KINAR6,-(SP)
@#112,RO
(RO),RO
(RO),KINAR6
#EXEVEC,R3
#EXEVCL,R2
@#140004
(SP)+,KINAR6

SAVE KINAR6
GET ADDRESS OF TABLE OF ENTRIES
GET ADDRESS OF APR BIAS -- FIRST WORD IN TABLE
MAP COMMON THROUGH I-SPACE APR 6
POINT TO VECTOR
SPECIFY LENGTH OF VECTOR
TRANSLATE THE VECTOR
RESTORE KINAR6

Please add the following information to Section 7.4.16 of the manual:
If you build a vectored driver for the system macro call GTPKT$, the following symbol
must be defined in the driver code:
VC$xx = 0

The parameter xx represents the 2-character device mnemonic.
•

Please add the following information to Section 7.4.19 of the manual:
If you build a vectored driver for the system macro call INTSV$, the following symbol
must be defined in the driver code:
VC$xx = 0

The parameter xx represents the 2-character device mnemonic.
•

Please make the following correction to Section 8.3, page 8-24 and 8-25:
The example of the sample driver BMDRV.MAC is incorrect because the offset I.PRM+16
must be cleared if you cannot do buffered 110. Replace the section of code in the example
with the following code:

2-24 Corrections to Documentation

**************************************************************

*
*
*

*
*

CONVERT TO BUFFERED I/O REQUEST

*
**************************************************************
MOV
R5, R3
; COpy I/O PACKET ADDRESS BACK
+------------------------------------------------------------+
I
I
I
THE INPUT PARAMETERS FOR $ INIBF ARE:
I
I
I
I
R3 = ADDRESS OF THE I/O PACKET TO BUFFER
I
I
I
I
NO OUTPUT PARAMETERS.
I
I
I
+------------------------------------------------------------+
CALL
$INIBF
INITIALIZE BUFFERED I/O

40$:

45$

SKIP CLEARING OF BUFFER ADDRESS

CLR

I.PRM+16(R3}

INDICATE NO BUFFERED I/O

**************************************************************
*
*

*
*

QUEUE THE CLOCK BLOCK

MOV

I.PRM+14(R3},RO ; GET ADDRESS OF CLOCK BLOCK

*
*

**************************************************************
45$:

2.10 RSX-ll M-PLUS and Micro/RSX I/O Drivers Reference Manual
Please make the following cOlTections to the RSX-llM-PLUS and Micro/RSX I/O Drivers
Reference Manual:
The values listed for the RX33 drive in Table 4-1, page 4-2 are incolTect. The correct
characteristics for the RX33 drive are as follows:

Drive

Revolutions
Per Minute

Sectors Tracks

Cylinders

BytesIDrive

Decimal
Blocks

RX.33

3600

1,228,800

2400

Information on message-oriented communication drivers was inadvertently removed from
the RSX-llM-PLUS and Micro / RSX I/O Drivers Reference Manual. This information is
applicable only to RSX-11M-PLUS operating systems. Refer to Appendix B, Section B.3.10
for information on message-oriented 110 function codes.
The information shown next should be added to the manual.
Introduction to Message-Oriented Communication Drivers

Message-oriented communication line interfaces usually link two separate but
complementary computer systems. One system must serve as the transmitting device
and the other as the receiving device. Message-oriented communication line interfaces are
used to transfer large blocks of data.

Corrections to Documentation 2-25

While character-oriented interfaces can only be accessed indirectly through the terminal
driver, the DMC11 and DUP11 synchronous line interfaces allow I/O requests to be queued
directly for them. These devices have drivers of their own and can be accessed by means of
logical device names. You can use these names in assigning Logical Unit Numbers (LUNs)
with the Assign LUN system directive at task build or with the MCR command REASSIGN.
DMCll Synchronous Line Interface

The DMC11 synchronous line interface provides a direct memory access interface
between two PDP-11 computer systems using the DDCMP line protocol, thus delivering
high throughput and reliability while simplifying programming. The DMC11 supports
nonprocessor request (NPR) data transfers of up to BK words at rates of 1,000,000 baud for
local operation (over coaxial cable) and 19,200 baud for remote operation (using modems).
Both full- and half-duplex modes are supported. The DMC11 synchronous line interface also
implements remote load detect, allowing it to reinitialize a halted computer system.
DUPll Synchronous Line Interface

The DUP11 synchronous line interface is a single-line communications device that provides
a program-controlled interface between the PDP-11 and a serial synchronous line. The
PDP-11 can be interfaced with a high-speed line to perform remote batch processing,
remote data collection, and remote concentration applications. Modem control is a standard
feature of the DUP11 and allows using the device in a switched or dedicated configuration.
The DUP11 transmits data at a maximum rate of 9600 baud; this rate is limited by modem
and data set interface level converters.
The DUP11 can be programmed to accept any sync character that you define. The DUP11
incorporates hardware to perform a cyclic redundancy check (CRC).
Get LUN Information Macro

Word 2 of the buffer filled by the Get LUN Information system directive (the
first characteristics word) contains the following information for message-oriented
communication interfaces. A bit setting of 1 indicates that the described characteristic
is true for the interfaces described in this section.
Bit

Setting

Meaning

Record-oriented device

Carriage-control device

Terminal device

File-structured device

Single-directory device

Sequential device

Mass-storage device

User-mode diagnostics supported

2-26 Corrections to Documentation

Bit

Setting

Meaning

Device supports 22-bit direct addressing

Unit software write locked

Input spooled device

Output spooled device

Pseudo device

Device mountable as a communications channel

Device mountable as a Files-11 volume

Device mountable

Words 3 and 4 are undefined, and word 5 has a special meaning for the DUPll interface.
Byte 0 of word 5 contains the number of sync characters to be transmitted before a synching
message (for example, after line turn-around in a half-duplex operation), and byte 1 is a
sync counter.
QIO$ Macro

The following sections summarize the standard and device-specific functions of the QIO$
macro that are valid for the communication interfaces.
Standard QIO$ Functions

The standard functions of the QIO$ macro that are valid for the communication devices are
shown next.
Format

Function

QIO$C IO.ATT, .. .

Attach device 1

QIO$C IO.DET, .. .

Detach device

QIO$C IO.KIL, .. .

Cancel I/O requests

QIO$C IO.RLB, ... ,<stadd,size>

Read logical block (stripping sync)

QIO$C IO.WLB, ... ,<stadd,size>

Write logical block (preceded by syncs)

lOnly unmounted channels may be attached. An attempt to attach a mounted channel results in an
IE.PRI status return in the I/O status doubleword.
stcdd

The starting address of the data buffer (may be on a byte boundary).

Corrections to Documentation 2-27

size

The data buffer size in bytes (must be greater than 0).
Device-Specific QIO$ Functions

The specific functions of the QIO$ macro that are valid for the communication line
interfaces are shown next.
Format

Function

QIO$C IO.FDX

Set device to full-duplex mode

QIO$C IO.HDX, ... ,<stat,mode>

Set device to half-duplex mode

QIO$C IO.INL, .. .

Initialize device and set device characteristics

QIO$C IO.RNS, ... ,<stadd,size>

Read logical block, without stripping sync characters
(transparent mode); for DMCll, treated like IO.RLB.
Not supported on DUPll.

QIO$C IO.SYN, ... ,<syn>

Specify sync character; not applicable to DMCll

QIO$C IO.TRM, .. .

Terminate communication, disconnecting from physical
channel

QIO$C IO.WNS, ... ,<stadd,size>

Write logical block without preceding sync characters
(transparent mode); for DMCll, treated like IO.WLB

stat

Specifies the station assignment (primary or secondary).
mode

Specifies the transmission mode (normal or maintenance).
stadd

Specifies the starting address of the data buffer (may be on a byte boundary),
size

Specifies the data buffer size in bytes (must be greater than 0).

syn
Specifies the sync character, expressed as an octal value.
The device-specific QIO$ functions are described in the following sections.
IO.FDX Function

The QIO$ function IO.FDX sets the mode on a DUPll or DMCll unit to full duplex. The
IO.FDX function code can be combined (ORed) with the IO.SYN function code, if desired, to
set the operational characteristics of the physical device unit.

2-28 Corrections to Documentation

IO.HDX Function

The QIO$ function IO.HDX sets the mode on a DUPll or DMCll unit to half duplex. The
IO.HDX function code can be combined (ORed together) with the IO.SYN function code, if
desired, to set the operational characteristics of the physical device unit.
Setting half-duplex mode on the DMCll also involves setting the station assignment
(primary/secondary) and may include selecting maintenance mode (MOP) as opposed to
normal mode. The station assignment is included in the optional QIO$ parameter pl. A 0
indicates a primary station and a nonzero indicates a secondary station. The DMCll works
properly if both ends are primary stations or if there is one primary and one secondary
station. It does not work if both ends are secondary stations. The optional QIO$ parameter
p2 selects the mode. A 0 selects normal mode and a nonzero selects MOP mode. A DMCll
in MOP mode cannot communicate with a DMC11 in normal mode.
IO.lNL and IO.TRM Functions

The QIO$ functions IO.INL and IO.TRM have the same function code but different modifier
bits.
IO.INL initializes a physical device unit for use as a communications link. It turns the
device on line, sets device characteristics, and ensures that the appropriate data terminal is
ready.
IO.TRM disconnects the device. If the device has a dial-up interface, it also hangs up the
line.
IO.RNS Function

The IO.RNS QIO$ function reads a logical block of data without stripping the sync
characters that may precede the data.
IO.RLB is a similar function, which is nontransparent, in that it causes the sync characters
that precede the data message to be stripped. Use IO.RLB at the start of a segmented data
request, in which the block might have the following layout:
I
I S
I
1

I
I
S I H I H
I
I

I
I H
I

Data

s
Specifies a sync character.
H

Specifies a header character.

Corrections to Documentation 2-29

cs
Specifies a validity check character.
You must strip sync characters from the beginning of a data block in this way. Stripping
only at the beginning of a read operation allows a later character that happens to have the
same binary value as a sync character to be read without stripping. Use IO.RLB to read a
logical block with leading sync characters stripped; use IO.RNS to read the block without
stripping leading sync characters.
10.SYN Function

This QIO$ function allows the programmer to specify the sync character to be recognized
when an IO.RLB or IO.WLB function is performed. IO.SYN can be combined (ORed
together) with IO.HDX or with 10.FDX to set the characteristics of the physical device
unit.
10.WNS Function

This QIO$ function causes a logical block to be written with no preceding sync characters.
To ensure that the two systems involved in a communication are synchronized, two or more
sync characters are transmitted by one system and received by the other before any other
message can be sent.
Use IO.WLB to write a block of data preceded by sync characters.
Use IO.WNS to perform a block transfer without sending sync characters first.
Programming Hints

The following sections contain important information about programming the
message-oriented communication interfaces.
Transmission Validation

Because there is no way for the transmitting device to verify that the data block has
successfully arrived at the receiving device unless the receiver responds, the transmitter
assumes that any message that is clocked out on the line (without line or device outage) has
been successfully transmitted. As soon as the receiver is able to satisfy a read request, it
returns a successful status code (IS.SUC) in the I/O status block. Of course, only the task
receiving the message can determine whether the message has actually been transmitted
accurately.
The receiving device should be ready to receive data (with a read request) at the time the
transmission is sent.
Redundancy Checking

By the nature of message-oriented communications, only the task that receives a
communication can determine whether the message was received successfully. The
transmitter simply transfers data, without validation of any kind. It is therefore the
responsibility of the communicating tasks that use the device to check the accuracy of the
transmission. A simple validity check is a checksum-type longitudinal redundancy check.
A better approach to validating data is the use of a cyclic redundancy check (CRC). A CRC

2-30 Corrections to Documentation

can be computed in software or with a hardware device, such as the KG-II communications
arithmetic option.
The DUPII incorporates hardware to compute a CRC.
Half-Duplex and Full-Duplex Considerations

Because there is a single 1/0 request queue, only one QIO$ request can be performed at a
time. It is therefore not possible, through QIO$s, for a device to send and receive data at
the same time. Also, because timeouts are not set for receive functions, a receive QIO$ is
terminated only by receiving a message from the remote system, or by issuing an 10.KIL
QIO$ for the device. Therefore, if no message is transmitted by the remote system, a receive
does not terminate, and no further 1/0 can be performed on that device until the receive is
killed by issuing an 10.KIL QIO$.
You can use both half-duplex and full-duplex lines with the DMCII and DUPI1. The
mode is settable by using 10.FDX for full-duplex mode and 10.HDX for half-duplex mode.
In half-duplex mode, the modem signal RTS (Request To Send) is cleared after each
"transmit message." In full-duplex mode, this signal is always left on. Using full-duplex
mode eliminates modem delays in transmission, but requires full-duplex hardware and
communication links.
The DMCII Driver maintains both transmit operations and receive operations separately
in its own internal queues. Thus, it is a full-duplex driver. There is no limit on the
number of outstanding 110 requests that can be active at any given time. The DMCII
hardware, however, allows a maximum of only seven transmit operations and seven receive
operations to be active at any time. The driver gives the first seven transmit operations (or
receive operations) directly to the DMCII and queues the eighth and subsequent transmit
operations (or receive operations) internally until the DMCII acknowledges a successful 1/0
request. When running on a MicroPDP-11/70, the driver gives only two transmit operations
(or receive operations) to the DMCII because each request requires a UNIBUS mapping
register (UMR). The DMCII driver is assigned five UMRs: one for base table(s), two for
active transmit operations, and two for active receive operations.
Low-Traffic Sync Character Considerations

If message traffic on a line is low, each message sent from a communications device should
be preceded by a sync train. This enables the controller to resynchronize if a message is
"broken" (that is, part or all of it is lost in transmission). Correspondingly, every message
received by a communications device under low-traffic conditions, when messages are not
contiguous (back-to-back), should be read with an 10.RLB (read, strip sync) function. This
requires that the first character in the data message itself not have the binary value of the
sync character.
Powerfail with DMC 11

The DMCII currently cannot recover after a power failure because the random-access
memory (RAM) in its internal microprocessor is erased when power fails. Any 1/0 requests
outstanding at the time of a power failure return the IE.ABO status. These requests must
be reissued after initializing the DMCII (IO.lNL).

Corrections to Documentation 2-31

Importance of IO.lNL

After the type of communication line has been determined, and after 10.SYN has specified
the sync character, it is extremely important that IO.INL be issued before any transfers
occur. This ensures that appropriate parameters are initialized and that the interface
is properly conditioned. Note that IO.INL provides the only means of setting device
characteristics, such as sync character. For this reason, you should always use IO.INL
immediately prior to the first transfer over a newly activated link.
Tasks sending messages to the DMCll should begin by terminating and reinitializing the
device (IO.TRM, IO.INL). Note that this causes the error IE.CNR to be returned on any 1/0
outstanding on the other end of the line. 10.INL must be issued after each 10.KIL (which
effectively kills the DMCll), after power fail, and upon receipt of any error code.
Programming Example

The following example illustrates the initialization, the setting of device parameters, and
the transmission of a block of data on a message-oriented communication device.
. MCALL

ALUN$S,QIO$S

ALUN$S
QIO$S
QIO$S
QIO$S

1,"XP,0
IO.HDX!IO.SYN,<1"",226>
IO.INL,l
IO.WLB,1",<TXSTS,TXAST,TXBUF,100>

iUSE LUN1 FOR DP11
iSET DEVICE PARAMETERS
iPUT DEVICE ON LINE
iSEND A BLOCK

TXAST:

CMPB

BEQ

10$

iWAS DATA CLOCKED OUT
iSUCCESSFULLY?
iIF SO, SET UP FOR NEXT
iBLOCK

IS.SUC&377,@(SP)+

Please add the following section to the manual:
Section 4.5.6 DU Driver Enhancements

The DU-type device driver (DUDRV) returns more information in the IO.RSN (Read Serial
Number) function. The added information is used by the Bad Block Replacement Control
Task (RCT). In addition to the volume serial number, the information returned includes
the controller identification number, the hardware and software version numbers of the
controller, the unit identifier (device identification number), and the hardware and software
version numbers of the device. RCT sends this information to the Error Logger.
A new subfunction, IQ.SEL (Suppress Error Logging), allows you to prevent error messages
from being sent to the Error Logger.
In addition, DUDRV does not request RCT unless the drive is set volume valid.
Please add the following sections to the manual:

2-32 Corrections to Documentation

Section 4.5.7 Modifications to DUDRV and PUCOM for Digital Storage Architecture
Requirements

The Digital Storage Architecture (DSA) drivers, DUDRV and PUCOM, were modified as
follows:
•

The DSA driver DUDRV and the PUCOM partition now implement a controller specific
I/O sequence number.
The sequence number ensures that the DSA command reference number is unique, and
it is useful in debugging problems related to DSA drivers.

•

Because of the information RCT puts in the system error log file, DUDRV and PUCOM
save information about controller identification in their internal data structures. The
IO.RSN (Read Serial Number) function returns the controller identification number.

•

DUDRV invokes a time delay before it requests the unit status of RASl devices
following a powerfail. Previously, RASl devices did not always spin up because the
driver requested the unit status from the controller too quickly.
The time delay is defined in the module DSAPRE.MAC.

•

DUDRV ensures that error logs are not requested from the DSA controller if error log
support was not generated in the system. This behavior reduces the number of response
packets that the drivers must process on systems without error log support.

•

Information from internal data structures in the DSA driver PUCOM is passed to RCT
because of the information logged by RCT in the system error log file. The information
is returned by the 10.RSN (Read Volume Serial Number) function code. The format of
the information is found in the DSAPRE.MAC module.

•

The DSA driver DUDRV correctly determines, on entry into the powerfail recovery
routine, if the recovery was for a controller or a unit.

•

Offset P.SEQ in the UDADF.MAC module points to a controller-supplied sequence
number and is used for RCT processing.

•

DUDRV uses a bit definition (UU.SEL) in the UCBDF.MAC module to determine if error
packets should be logged during bad block replacement by RCT. In addition, RCT uses
the submodifier IQ.SEL in module QIOMAC.MAC to set and reset UU.SEL. The IQ.SEL
submodifier only is supported by DUDRV.

Section 6.7 Modifications to MUDRV and PUCOM for Digital Storage Architecture
Requirements

The Digital Storage Architecture (DSA) drivers, MUDRV and PUCOM, were modified as
follows:
•

The DSA driver MUDRV and the PUCOM partition now implement a controller specific
1/0 sequence number.
The sequence number ensures that the DSA command reference number is unique, and
it is useful in debugging problems related to DSA drivers.

•

MUDRV ensures that error logs are not requested from the DSA controller if error log
support was not generated in the system. This behavior reduces the number of response
packets that the drivers must process on systems without error log support.

Corrections to Documentation 2--33

•

The DSA driver MUDRV correctly determines on entry into the powerfail recovery
routine if the recovery was for a controller or a unit.

2. 11 RSX-11 M-PLUS and Micro/RSX System Management Guide
•

In Section 5.5.16, page 5-74, of the RSX-llM-PLUS and Micro/RSX System
Management Guide, the third example is incorrect. The correct example is as follows:
VMR>

SECPOL=285.:640.:44%

Displays the amount of available secondary pool.
•

In Section 13.3.4 of the manual, please add the following text:
The Bad Block Replacement Control Task (RCT) supports the latest Mass Storage
Control Protocol (MSCP) disk storage architecture specification. RCT is used with
MSCP controllers, such as the UDA-50, which do not perform automatic revectoring of
bad blocks. Instead, these controllers rely on the RCT task to perform revectoring for
them. Revectoring is the redirection of reference from an unreliable block to a reliable
one.
A controller that performs its own revectoring creates a complete error log report on
the I/O it handles and sends the report to the device driver. The device driver, in turn,
sends a report to the Error Logger. In this way, all messages on bad blocks appear in
the error log file generated when you enter an ANALYZEIERROR_LOG command.

•

In Section 13.4 of the manual, please add the following text:
When RCT performs revectoring, it creates the error log report and sends it to the
Error Logger. A new error message is issued by the Task Termination Notification
program (TKTN), which sends the following message to the console terminal whenever
a nonrecoverable hardware error occurs:

*** ddnn: -- Replace command failure
If you receive this message, Digital recommends that you back up the media and
note any errors that are reported during the backup operation. You can use the error
information to determine if files are corrupted.
In addition, RCT does not produce error log packets if a device is write-protected and
RCT is unable to write to the device.
•

The Resource Monitoring Display (RMD) memory display supports the following two
setup commands:
TOP=n

Specifies the upper limit of the memory display, where n is the limit value in K
words.

BOTTOM=n

Specifies the lower limit of the memory display, where n is the limit value in K
words.

The commands allow you to examine a specific portion of system memory and will
display details of system memory, including tasks and shared regions, that do not
appear in a full display of system memory.

2-34 Corrections to Documentation

You can enter the commands as part of the MCR command line, or you can enter the
commands after you access the setup page. If you specify values for n that are invalid,
RMD will default to a full display of system memory.
•

Please add the following text to Section 15.1.4 of the manual following the
DEFER_WRITES option:
Data caching supports write loading for temporary files.
I/O requests that can be deferred will result, if necessary, in a write load, which creates
a new cache extent. Blocks in a temporary file that have the potential to be deferred
will no longer have to be read before a deferred-write operation can occur. This feature
will increase the number of blocks that are deferred for a task, and it will increase the
performance of the deferred-write support.

•

The Account File Maintenance Program (ACNT) commands ADD and MODIFY prompt
you for confirmation when you press the RETURN key in response to the password field
prompt or the last name (that is, Username) field prompt. Because blank values for
account passwords or user names may result in security problems, Digital recommends
that you do not create accounts with blank values in these fields. However, when you
add or modify account fields, you may inadvertently create blank values for fields if you
press the RETURN key in response to the field prompt instead of the ESCAPE key which
leaves a field unchanged.
Also, the password encryption routine has an improved handling of accounts with blank
values in the user-name field or in the password field. The routine uses the user-name
field and the password field to encrypt the password. Previously, the routine produced
deficient results for accounts with a blank value either in the user-name field or in the
password field.
The enhancement improves security, but you are unable to log in to accounts with blank
user names until you modify the user-name field, which re-encrypts the password.

2.12 RSX-11M-PLUS and Micro/RSX System Library Routines
Reference Manual
Please make the following correction to the RSX-llM-PLUS and Micro / RSX System
Library Routines Reference Manual:
In Table 6-1 on page 6-14, the text explaining file name string conversion using the X
Directive is incorrect. Please replace that text with the following:
Directive

Form

Operation

X
(file name
string
conversion)

Convert Radix-50 filename string in ARGBLK to ASCII string in
format "name.typ"; convert version number, if non-zero, to ASCII
decimal string if decimal version support is selected in your system.
Otherwise the version number is converted to ASCII octal string. If
the version number is zero, no version number is put into OUTBLK.
Store the results in OUTBLK.

Corrections to Documentation 2-35

Directive

Form

Operation

%nX

Convert next n Radix-50 filename strings in ARGBLK to ASCII
strings in format "name.typ"; convert version numbers, if non-zero,
to ASCII decimal strings if decimal version support is selected in
your system. Otherwise version numbers are converted to ASCII
octal strings. If a version number is zero, no version is put into
OUTBLK for that filename string. Store results in OUTBLK and
insert tab between strings.

%VX

Use the value in the next word in ARGBLK as a repeat count,
convert specified number of Radix-50 filename strings to ASCII
strings in format "name.typ"; convert version numbers, if non-zero,
to ASCII decimal strings if the decimal version support is selected
in your system. Otherwise version numbers are converted to ASCII
octal strings. If a version number is zero, no version is put into
OUTBLK for that filename string. Store results in OUTBLK and
insert tab between strings.

KEY

Meaning

ARGBLK=

The argument block containing the binary data to be converted, the addresses of
ASCII and extended ASCII characters or the address of a double precision value.

OUTBLK=

The output block in which $EDMSG is to store output.

2. 13 RSX-11 M-PLUS and MicrojRSX Task Builder Manual
Please make the following changes to the RSX-llM-PLUS and Micro/RSX Task Builder
Manual:
•

Change the fourth line of Figure 4-4 to .NAUTO as shown in the following figure:

2-36 Corrections to Documentation

1
MOV (PC)+,-(SP)
1
1---------------------------------1
1 ADDRESS OF PACKET (D-SPACE)
1
1---------------------------------1
JMP @.NAUTO
1---------------------------------1
PC RELATIVE OFFSET TO .NAUTO
1

I-SPACE PORTION
1

ADDRESS OF SEGMENT DESCRIPTOR

1--------------------------------1
ENTRY POINT ADDRESS
1

D-SPACE PORTION

•

Change all references to FORTRAN IV-PLUS in Chapter 5 to FORTRAN-77.
FORTRAN IV-PLUS is no longer supported.

•

Change all references to F4PRES in Chapter 5 to F7FRES.

•

Change all references to F4P in Chapter 5 to F77.

•

Add the following note to Section 5.2.9.2:
Caution
This command file example will work only for FMS Version 2.0.

•

Replace the TKB command sequence in Section 5.4.2 with the following:
TKB>
TKB>
Enter Options:
TKB>
TKB>
TKB>

Or, if you use the LINK command, use the following command sequence:
$

Option?
Option?
Option?
$

•

Replace FOROTS.OLB with F77FCS.OLB in the second paragraph under the command
sequence in Section 5.4.2.

•

In Section 11.26, page 11-36, the format of the !INCLUDE qualifier is incorrect. The
correct format is shown next.

Corrections to Documentation 2-37

Format

LINKITAS/MAP/SYM inputfile/INCLUDE:(MODl,MOD2, ...MODB)
Also, please remove note number 3. It is no longer applicable.
In note number 5, the example of the /INCLUDE qualifier is incorrect. The correct
example is as follows:
$

•

In Section 11.39, page 11-55, the description of the /SAVE qualifier is incorrect. All
references to the file ATLINKCMD should be changed to ATLINK.TMP.
Also, please remove the clause "and it contains legitimate TKB command syntax" from
the second paragraph.

•

In Section 12.4, page 12-8, the definitions of the device-name and unit-numB
parameters are incorrect. The correct definitions are shown next.
device-name

Specifies a 2-character alphabetic device name followed by a 1- to 3-digit octal unit
number.
unlt-num8

Specifies decimal numbers indicating the logical unit numbers (LUNs). If your task
uses more than six logical units, you should use the UNITS option to specify the
number of logical units that your task will use.
Also, the note is no longer applicable; please remove it.
•

Please add the following text to Section 12.1.25, page 12-34 of the manual and to
Section 12.1.2B, page 12-37 of the manual:
The RESSUP and SUPLIB options have a new parameter code /SW that allows you to
write to data space in a supervisor-mode library when mapped to supervisor D-space
with the MSDS$ directive. Use the /SW parameter code to specify read-write access
when you build a task that links to a supervisor-mode library. You should use the
parameter code ISV to specify read-only access.
Note that, for the /SW parameter code, the supervisor-mode library must be installed
with the IRON=NO switch.
The formats of the /SW parameter code are shown next.
Formats

RESSUP file-specification/[-]SW[apr]
and
SUPLIB file-specification/[-]SW[apr]

2-38 Corrections to Documentation

In addition, the library flag word parameter, R$LFLG, has a new flag in label block o.
The definition of the flag is shown next.

•

Mask

Bit

Flag

Meaning

010000

LD$SMV

Include supervisor-mode vectors (1=NO)

Please add the following text to Section 12.1.34, page 12-44 of the manual:
For an 1- and D-space task build, the task builder automatically assumes the program
section specified in the VSECT option is a data program section.

•

Please add the following information to Section 12.1.5 on page 12-10, Section 12.1.7 on
page 12-13, and Section 12.1.21 on page 12-31 of the manual:
TKB Supplementary Information

TKB now allocates both data space APRs and instruction space APRs for libraries.
At offset 404, label block 0 contains the additional field L$BAPR, which contains the
data space APRs that the task or library requires. (Instruction space APR allocation
information is stored in field $APRMP in psect $$TSKP. For more information on psect
$$TSKP, refer to Appendix E in the RSX-l1M-PLUS and Micro / RSX Thsk Builder
Manual.)
When you build a library by using the TKB switch ILl, the default action is to allocate
the corresponding instruction and data space APRs for the library. However, you can
reserve specific data space APRs by using the new ILl subswitch ILI[:bitmask]. The
appropriate bits in the bit mask should be set to specify the data space APRs that you
want.
If you build a task that links to a library that was built with an earlier version of
TKB, only the instruction space APRs are allocated for the library. Tasks that link
to libraries built with TKB Version 4.1 have the specified library data space APRs
reserved. However, you can override the library data space APR reservations by using
the new bit mask option with the LIBR, RESLIB, and CLSTR options.
The formats for these options are shown next.
Formats

LIBR =name:accesscode[:baseAPR[:bitmask]]
and
RESLIB =fileiaccesscode[:baseAPR[:bi tmask]]
and
CLSTR =lib1,lib2 ...libn:accesscode[:baseAPR[:bitmask]]

The appropriate bits in the bit mask should be set to reserve the desired data space
APRs. If you do not want to reserve data space APRs for your library, you should
specify a bit mask of 000. Note that for clusters any data space APR reservation applies
to all the libraries in the cluster. The bit mask for a position-independent code (PIC)
library is shifted the same amount as the library. For example, if the bit mask for a

Corrections to Documentation 2-39

2-APR PIC library is 200 and the library is placed in APR 5 and APR 6, the bit mask is
changed to 100.
The bit mask uses the same format as the mask for the new EXTM$ directive: bit 0
represents APRO and bit 1 represents APR1 (refer to Section 1.13.3.1). For example, if
you specify 340, the APRs 7, 6, and 5 are reserved.
•

In Appendix B, page B-6, the description of the Label Block Group is incorrect. Please
remove the last sentence in the second paragraph, which states the following: "The
LBLDF$ macro on your system will have the correct offsets."
Also, remove the third paragraph and replace it with the following paragraph:
The LBLDF$ macro defines the label block offsets for a task that is built on a RSX-11M
operating system. Tasks that are built on RSX-11M-PLUS systems have an additional
eight library entries that are inserted in the label block after the entries at offset
L$BLIB. Because of this, the label block offsets from L$BPRI to L$BDMZ must be
adjusted by the size of the additional entries. Label block offset LBLDF$ defines the
symbol $LBXL (label block extra length), which determines the size of the additional
entries. Therefore, for tasks built on RSX-11M-PLUS operating systems (L$BSYS=4),
you must add $LBXL to the offsets from L$BPRI to L$BDMZ in order to determine the
true offset.

•

In Appendix B, Figure B-5: Table Block O-Task and Resident Library Data is incorrect.
The correct figure is shown next.

2-40 Corrections to Documentation

Label

Offset

L$BTSK

Task
Name

L$BPAR

L$BSA

Base Address 01 Task

L$BHGV

Highest Window 0 Virtual Address

L$BMXV

Highest Virtual Address In Task

L$BLDZ

Load Size In 64-Byte Blocks

L$BMXZ

Maximum Size In 64-Byte Blocks

L$BOFF

Task Ollset Into Parlltlon

L$BWND/L$BSYS

System I.D.

L$BSEG

Size 01 Overlay Segment Descriptors

L$BFLG

Task Flag Word

L$BDAT

Task Creation Date --- Year

--- Month

--- Day

Task
Partillon

L$BLlB

Number 01 WindoW Blocks·

Library Entry
Ollset

Library/Common

2 R$LNAM

Name

Base Address 01 Library

6 R$LHGV

Highest Address In First Library Window

10 R$LMXV

Highest Address In Library

12 R$LLDZ

Library Load Size (64-Byte Blocks)

14 R$LMXZ

Library Maximum Size (64-Byte Blocks)

16 R$LOFF

R$LSA

Library Ollset Into Region

20 R$LWND

Number 01 Library Window Blocks

22 R$LSEG

Size 01 Library Segment Descriptors

24 R$LFLG

Library Flag Word

26 R$LDAT

Library Creallon Date --- Year

--- Month

--- Day

344

L$BPRI"

346

T ask Priority

L$BXFR"

350

Task Transler Address

L$BEXT"

352

Task Extension (64-Byte Blocks)

L$BSGL··

354

Block Number 01 Segment Load List

L$BHRB"

356

Block Number 01 Header

L$BBLK"

360

Number 01 Blocks In Label

L$BLUN"

362

Number 01 Logical Units

L$BROB"

364

Relallve Block 01 R-O Image

L$BROL"

366

R/O Load Size

L$BRDL··

370

R/O Data Size in 32-Word Blocks

L$BHDB"

372

Relative Block Number 01 Data Header

L$BDHV"

374

High Virtual Address of Data Window 1

L$BDMV"

376

High Virtual Address 01 Data

L$BDLZ"

400

Load Size 01 Data

L$BDMZ··

402

Maximum Size 01 Data

L$BAPR

404

Library D-Space Reservations

L$BFL2

772

Second Task Flags Word

L$BLRL

774

Label Block Revision Number

776

AME (Must Be 0)

Library Request
(MaXimum 01 Seven
14-Word Entries In
RSX-11M Systems and
Maximum 01 Fifteen
14-Word Entries In
RSX-11 M-PLUS Systems)

• Less library window blocks.
" II parameter L$BSYS Is 4, you must add $LBXL(340 e)
to the ollset to determine the true ollset.
MLO-003DS4

Corrections to Documentation 2-41

•

In Appendix B, Figure B-8: Task Header, Fixed Part is incorrect. The correct figure is
shown next.

Label

Offset

H.CSP

Current Stack Pointer (R6)

H.HDLN

Header Length

I Super D-Space Mask

H. SMAP/H.DMAP

User D-Space Mask

H.FMAP

Pointer to Fast Map Section

H.CUIC

Current UIC

H.DUIC

Default UIC

H.IPS

Initial PS

H.IPC

Initial PC (R7)

H.ISP

Initial Stack Pointer (R6)

H.ODVA

ODT SST Vector Address

H.ODVL

ODT SST Vector Length

H.TKVA

Task SST Vector Address

H.TKVL

Task SST Vector Length

H.PFVA

Power-Fall AST Control Block

H.H.FPVA

Floating-Point AST Control Block

H.RCVA

Receive AST Control Block

H.EFSV

Address of Event Flag Context

H.FPSA

Address of Floating-Point Context

H.WND

Pointer to Number of Window Blocks

H.DSW

Directive Status Word

H.FCS

Address of FCS Impure Storage

H.FORT

Address of FORTRAN Impure Storage

H.OVLY

Address of Overlay Impure Storage

H.VEXT

Address of Impure Vectors

H.SPRI/H.NML

Mailbox LUN

H.RRVA

Receive by Reference AST Control Block

Reserved

Low-Core
Context
}

I Swapping Priority

I H.X25

H.LUTE

Pointer to LUT Extension

H.GARD

Header Guard Word Pointer

H.NLUN

Number of LUNs
MLO-003955

•

In Appendix H, page H-2, the explanation for the "Illegal APR reservation" is not
complete. Please make the following correction:

Illegal APR reservation

Explanation: An Active Page Register (APR) specified either with the ILl switch or in a
COMMON, LIBR, RESCOM, or RESLIB option is outside the range 0 to 7.

2-42 Corrections to Documentation

2.14 RSX-11M/M-PLUS and Micro/RSX I/O Operations Reference
Manual
Please make the following correction to the table of normal error returns and no error
returns in Section 4.1:
Normal Error
Return
(Carry Bit and
F.ERR)

No Error Return

.ASCPP

.RDFDR

.PARSE

.WDFDR

.PRSDV

.RDFUI

.PRSDV

.WDFUI

.PRSFN

.RDFFP

.ASLUN

.WDFFP

.FIND

.RFOWN

.ENTER

.WFOWN

.REMOV

.PPASC

.GTDIR

.MARK

.PGTDID
.POINT
.POSRC
.POSIT
.XQUIO
.RENAM
.EXTND
.TRNCL

2.15 RSX-11M/M-PLUS RMS-11 Macro Programmer's Guide
Please make the following corrections to the RSX-llM / M-PLUS RMS-ll Macro
Programmer's Guide:
•

In Section 2.3, the argument for P$BUF is incorrectly given as "bufcount." The correct
argument is "iopoolsize," as discussed in Section 2.3.4.

•

In Section 5.19, the last paragraph incorrectly describes the use of the FID field in the
NAM block. It should read as follows:
''If this value is nonzero ..."

Corrections to Documentation 2-43

•

In Table 6-2, page 6-14, the value for the symbol XB$DAT is incorrectly given as
000003. The correct value is 000002.

•

In Appendix A, page A-8, please add the following sentence:
An attempt to insert a record that is too small to contain the whole primary key field
may also cause the error ER$KEY.

•

In Section A.1, add the following text to the description of the error ER$MRS:
Or, the sum of the fixed-length record size and the record overhead exceeds the bucket
size. Or, No Span Blocks has been selected with an invalid total record size.

•

In Section A.1, add the following text to the description of the error ER$NOD:
Or, the remote node rejected the operation. (STY contains the Network Services Protocol
(NSP) code and can be found in Appendix C of the DECnet-RSX Programmer's Reference
Manual.)

•

In Section A.1, add the following text to the description of the error ER$FUL:

ER$FUL,

Device or File Allocation Failure
Octal: 176360
Decimal: -784

Explanation: The specified device or directory does not have enough room for file
creation or extension. In the case of a contiguous request, it is also possible that there
is not enough contiguous space on the device.

2. 16 RSX-11 M/M-PLUS RMS-11 User's Guide
Please add the following corrections to the RSX-llM-PLUS RMS-ll User's Guide:
•

In Section 2.2.3.3, please add the following information to the discussion of deadlock:
An application should use multi stream rather than multichannel access to write to the
same indexed file. When RMS-11 updates an RRV in a bucket that is currently locked,
it must wait for that lock to be released. Control will not be returned to the program
until this release occurs. Deadlock will occur when the lock is held on another channel
within the same program; however, RMS-11 can update an RRV in a bucket that is
locked on another stream within the same program. See your programming language
documentation for details on the implementation of multi streaming.

•

In Section 6.2.4, please add the following note to the discussion of writing a record:
In the event that the record includes a partial alternate key but is not large enough to
include space for the full alternate key field, RMS-11 will act as follows:
RMS-11 will treat the alternate key as if it were not present in the record, making no
entry in the alternate key index structure.

•

According to the RSX-llM/M-PLUS Macro Programmer's Guide, RMS-11 cannot
perform an UPDATE operation on an alternate key with the key characteristics
CHANGES and NODUPLICATES. This description is misleading. Th clarify the
description, please add the following information to Chapter 6, Section 6.2.5.2:

2-44 Corrections to Documentation

Although RMS-11 does not support the CHANGES and NODUPLICATES combination,
it does not prevent you from performing an UPDATE operation on an alternate key with
these characteristics. When an update causes a duplicate of an alternate key, RMS-11
returns the completion code ER$DUP. However, it does not terminate the UPDATE
operation. Instead, RMS-11 updates the primary data level for the record without
updating the alternate index. As a result, the file contains duplicates of the alternate
key.

To prevent RMS-11 from creating duplicates when you make changes on alternate keys,
modify your application as follows:
1. Create the file with the key characteristics DUPLICATES and CHANGES.
2. To disallow duplicates, perform a FIND operation on each alternate key. Then,
perform an UPDATE operation on the modified record.
•

In Section 6.3, please add the following information to the discussion of contiguous
areas:
You will gain a small benefit by setting areas to contiguous on a noncontiguous
multiarea file; however, RMS-11 cannot determine if those areas remain contiguous.
Consequently, RMSDSP and RMSDES will display them as noncontiguous. As long as
the areas are preallocated, they will behave like contiguous areas; as soon as they need
to be extended, they will behave like noncontiguous areas.

•

In Section 8.1.2.1, in the discussion of task building against the RMS-11 resident
library, incorrect syntax is documented for the cluster option in the Task Builder
command file. The correct syntax is as follows:
CLSTR = RMSRES,DAPRES:RO

•

In Section 8.1.2.1, change paragraph 2, sentence 3 to read as follows:

To add remote access (DAP) support to a task that is built against the RMSRES in
supervisor mode you must include the modules:
LB: [l,l]RMSLIB/LB:ROEXSY:ROIMPA-LB: [l,l]RMSDAP/LB:ROAULS

and include DAPRES as a LIBR or CLSTR option in the task builder command
sequence.
Note
Unlike RMSRES, the DAPRES cannot be used in supervisor mode.

•

In Appendix B, in the discussion of remote file and record access using the DECnet
package, the documentation states that the RSTS/E file access listener (FAL) does not
support remote record access to indexed files. This is no longer true.

Corrections to Documentation 2-45

2. 17 RSX-11 M/M-PLUS RMS-11 Utilities
Please make the following corrections to the RSX-I1M / M-PLUS RMS-ll Utilities manual:
•

•

In Table 2-1, add the following RMSDES commands:
Command

Format and Function

EXIT_SUPERSEDE

EXIT_S[UPERSEDE] filename[' typ]
Names the description file in which the file design is stored.

SAVE_SUPERSEDE

SAVE_S[UPERSEDE] filename[.typ]
Names the description file in which the file design is saved.

In Section 2.2.5, add the new command EXIT_SUPERSEDE, as follows:
The EXIT_SUPERSEDE command stores the file design in the description file
specified in the command string, superseding any existing file by the same name.
EXIT_SUPERSEDE then terminates RMSDES and returns the system prompt.
The format for the EXIT_SUPERSEDE command is as follows:
EXIT_S[UPERSEDE] filename[.typ]

EXIT_SUPERSEDE names the description file in which the file design is stored. The
default file type is DES. If you do not want to supersede an existing description file, use
the EXIT command.
•

In Section 2.2.5, add the following statement to the description of the EXIT command:

To supersede an existing description file, use the EXIT_SUPERSEDE command.
•

In Section 2.2.9, add the new command SAVE_SUPERSEDE, as follows:
The SAVE_SUPERSEDE command stores the file design in the description file specified
in the command string, superseding any existing file by the same name. If you do not
define areas when you issue the SAVE command, RMSDES prompts you for the areas.
The format for the SAVE_SUPERSEDE command is as follows:
SAVE_S[UPERSEDE] filename[.typ]

SAVE_SUPERSEDE names the description file in which the file design is saved. The
default file type is DES. If you do not want to supersede an existing description file, use
the SAVE command.
If you want to design another file, issue a CLEAR ALL command to restore the attribute
values in the design buffer to their defaults.

2-46 Corrections to Documentation

•

In Section 2.2.9, add the following statement to the discussion of the SAVE command:

To supersede an already existing description file, use the SAVE_SUPERSEDE command.
•

In Section 2.6.2, add the following corrections:
Modify paragraph 3 of the ALLOCATION field discussion to read as follows:
If you intend to create a single-area indexed file and do not require RSX positioning,
RMS-l1 uses the allocation from the file section if no area section exists in your
design buffer. If you are allowing RMSDES to define areas for an indexed file by
default, RMSDES will automatically calculate an allocation value for each area it
defines.
Modify paragraph 4 of the EXTENSION field discussion to read as follows:
If you intend to create a single-area indexed file and do not require RSX positioning,
RMS-l1 uses the extension from the file section if no area section exists in your
design buffer. If you are allowing RMSDES to define areas for an indexed file by
default, RMSDES will automatically calculate an extension value for each area it
defines.
Modify paragraph 5 of the BUCKETSIZE field discussion to read as follows:
If you intend to create a single-area indexed file and do not require RSX positioning,
RMS-l1 will use the bucket size from the file section if no area section exists in your
design buffer. If you are allowing RMSDES to define areas for an indexed file by
default, RMSDES will assign a bucket size value for each area it defines. However,
if you choose to define areas explicitly and specify a bucket size value for each area,
you should accept the default for the file section and set the bucket size value in
each area section.

•

In Section 2.8, add the following correction to the explanation of the error message
"?DES-F-VOR":
You entered a value in response to an attribute prompt that was not in the legal range
of values for that attribute, or the values you entered resulted in a calculation that
caused an overflow for RMSDES. If the value was not within the legal range, the error
message is followed by a display of the incorrect value.

•

In Table 4-1, add the following switch and description to the table ofRMSCNV switches:
IER[::filespec]

Continue processing after encountering an exception record. If a :file speci:fication
is provided, then write the primary keys of exception records into the speci:fied
:file. If no :file speci:fication is provided, then output the exception records to the
terminal.
Default: Stop processing and report RMS error code.

Corrections to Documentation 2-47

•

In Section 4.3, add the following information to the description of RMSCNV switches:
IER(:filespec)

Directs RMSCNV to continue processing when it encounters an exception record
in the input file that cannot be written to the output file (see Section 3.4). If you
specify a file specification, the exception records will be written to that file. If you
do not specify a file specification, the primary key of each exception record will be
issued to the tenninal. RMSCNV also issues exception record codes (see RMSIFL
exception codes, Section 3.3.2).
If you specify an exception file specification, RMSCNV will create the file as an
RMS-11 Variable Fixed Control (VFC) sequential file upon encountering the first
exception record. RMSCNV will then write the exception record with a 4-byte
exception code to the fixed-control area of the record.
By default, if you do not specify the IER switch, RMSCNV will stop processing upon
encountering the first exception record and will issue an error message indicating
the type of exception record.
•

In Table 5-1, add the following information to the description of RMSDSP switches:
IBR

Briefly displays attributes.

/SU

Supersedes existing output file.

•

In Section 5.2, include the asterisk (*) and percent sign (% ) in the description of
wildcard characters permitted in the input file specification.

•

In Section 5.3, add the following information to the discussion of RMSDSP commands:
IBR

Directs RMSDSP to issue basic displays for indexed files (see Section 5.4, Example
5-3) and container files (see Section 5.4, Example 5-6).
ISU

Directs RMSDSP to supersede any existing output file with the same name and
version number as the output file specification. If this switch is not supplied and
the version numbers are the same, RMSDSP will issue the following error message:
?DSP-F_OPNINP, Error opening DDnn:file.dat as output
-RMS-E-ER$FEX, File already exists

•

In Table 6-1, add the following information to the table of RMSBCK switches:
INV

•

Creates a new version of the output file.

In Section 6.2, in the discussion of RMSBCK command line format, include the asterisk
(*) and percent sign (%) in the description of wildcard characters permitted in the
input file specification.

2-48 Corrections to Documentation

•

In Section 6.3.2, add the following information to the description of RMSBCK output
switches:
/NV

Directs RMSBCK to create a new version of the disk output file if a file currently
exists with the same version number as the input file. The current file is not
deleted. If you do not specify this switch and a file currently exists with the same
file name and version number as the input file specification, RMSBCK will issue the
following fatal error message:
?BCK-F-CREOUT, Error opening ddnn:file.dat;n as output
-RMS-E-ER$FEX, File already exists

•

In Table 7-1, add the following information to the table of RMSRST switches:
INV

Creates a new version of output file.

•

In Section 7.2, in the discussion of the RMSRST command line format, include the
asterisk (*) and percent sign (%) in the description of wildcard characters permitted in
the input file specification.

•

In Section 7.3.2, add the following information to the discussion of RMSRST commands:
/NV

Directs RMSRST to create the next higher version number if the expanded input
file has the same version number as an existing output file. If this switch is not
used and the file name and version number are the same, RMSRST will issue the
following error message:
?RST-F-CREOUT, Error opening ddnn:file.dat;n as output
-RMS-E-ER$FEX, File already exists

Corrections to Documentation 2-49

Chapter

RMS-ll Version 2.0 Software Modifications,
Restrictions, and Supplementary Information
This chapter describes RMS-ll software modifications, lists restrictions, and provides
supplementary information for RSX-llM-PLUS Version 4.3.
RMS-ll Version 2.0 has not changed version numbers for this release of RSX-llM-PLUS.
Information in this chapter has not been incorporated into the RMS-ll manuals.

3. 1 RMS-ll Corrections
The following sections describe corrections for RMS-l1. These problems have all been
corrected.

3. 1. 1 RMS-11 Local Access
RMS-ll applications, that performed multi-streaming asynchronous record operations
resulting in a high rate of competition for the same buckets, encountered the following set
of problems:
•

Execution of breakpoint trap in RMS modules RORLSB and RORSET.

•

Looping indefinitely in RMS modules RORSET and RORLCH.

•

Stalled indefinitely while waiting on the RMS event flag.

•

Receiving "Dynamic Memory Exhausted" error (ER$DME) on a $FIND, $GET, $PUT,
$UPDATE or $DELETE operation (operations which do not require new use of dynamic
memory).

•

Receiving "Bucket Header Checkbyte" error (ER$CHK) when in fact the file does not
have this problem.

RMS-11 'Version 2.0 Software Modifications, Restrictions, and Supplementary Information 3-1

3. 1.2 RMS-11 Ease of Use
Two new ODL files have been added to simplify the building of RMS programs that utilize
RMS in supervisor mode.
•

LB:[l,l]RMSSLX.ODL - for programs performing local access only.

•

LB:[l,l]DAPSLX.ODL - for programs performing local and remote access.
Note
The remote access library (DAPRES) cannot be used in supervisor mode.

3. 1.3 RMS-11 Utilities Enhancements
A correction was made in RMSIFL to preserve worst case error on input file read errors.
While the error was reported, if RMSIFL was able to continue, the exit status was
incorrectly reported as success.

3.2 Software Restrictions
This section describes software restrictions that apply to RMS-ll Version 2.0.

3.2. 1 RMS-11 Access Methods
RMS-ll tasks built prior to Version 3.0 of RSX-IIM-PLUS will return the error message
''Directory not found" (ER$DNF) on certain file operations that are executed from an account
set to nonamed directory mode. This will occur if one or both of the following conditions are
true:
•

The tasks were not built against the RMS-ll resident library.

•

A directory is not provided in either the file specification or the default file specification.

You can resolve this problem by first providing a directory in the file specification or by
setting your terminal to named directory mode. If that is not possible, you should rebuild
those tasks by using the new version of RMS-l1.

3.2.2 RMS-11 Remote Access Methods
RMSDAP does not interpret the keysize field value correctly if it is zero. This causes keyed
access to a remote indexed file to fail. The application passes the value zero in the keysize
field for access to a non-string key and the following error message is issued:
ER$RNF (Record not found)

If your program is written in F77 or other high-level languages, you may have to use a
"USER OPEN" in order to be able to set this value.

3.2.3 RMS-11 Utilities
This section describes restrictions to the RMS-ll utilities for Version 2.0.
RMSCNV Restriction

RMSCNV ignores user-provided area extension quantities when loading a file. The values
that it uses are large enough to reduce the number of file extensions in most cases.

3-2 RMS-ll 'Version 2.0 Software Modifications, Restrictions, and Supplementary Information

RMSDEF Restriction

The DCL command DEFINE defines logical names; therefore, when you invoke RMSDEF,
you need to either invoke MCR before typing DEF (to use the MCR mode of DEF instead of
the default DCL mode) or insert the following in the LB:[1,2]STARTUP.CMD file:
INSTALL LB: [3,54]RMSDEF/TASK= ... DFN (or a task name of your choice)

3.3 RMS-ll Supplementary Information
The following sections describe information that has been documented in the previous
release notes, but is still applicable.

3.3. 1 Enhancements to RMSDES Utility
The following enhancements have been made to the RMSDES utility:
•

There are two new RMSDES commands: SAVE_S[UPERSEDE] and
EXIT_S[UPERSEDE].

•

RMSDES issues the warning message "%DES-W-CBK, Continuation buckets will be
allocated for this key" when continuation buckets are required. (Continuation buckets
may result in decreased performance; see the RSX-IIM / M-PLUS RMS-ll User's
Guide.)

•

The informational message "?DES-F-VOR, Value out of legal range" is issued if overflow
occurs during calculations of indexed file area allocations. Also, the "number of
duplicates:" prompt has been expanded to "number of duplicates per record on this
key:" for clarity.

•

The error message "%DES-F-NHF, Help file is not available. Check release notes
for the location of RMSDES.lDX on your installation media" is issued if the help file
LB:[1,2]RMSDES.IDX is not found.
This is a feature for small systems that have limited disk space.

3.3.2 RMS-ll Installation
The following sections describe information for installing RMS-ll on an RSX-IIM-PLUS
Version 4.2 or higher operating system.
On RSX-IIM-PLUS operating systems, all RMS-ll files are automatically on your system.

3.3.2.1 Location of RMS-11 Files on the Distribution Kit
Table 3-1 describes the location of RMS-il Version 2.0 files on the RSX-IIM-PLUS Version
4.3 distribution kit.
Note
Some file names are marked with an asterisk (*). This indicates that the files
are not included on RL02 distribution kits. They are excluded because of space
limitations on the kit, and because they are used only to rebuild components of
RMS-ll Version 2.0.

RMS-ll Version 2.0 Software Modifications, Restrictions, and Supplementary Information 3-3

Table 3-1:

RMS-11 Files on RSX-11M-PLUS Distribution Kit

File Name

Destination

Comments

RMSMAC.MLB

LB:[I,I]

Can be deleted if you are not using MACRO-II RMS-II
programs

RMSLIB.OLB

LB:[I,I]

Object library for RMS-II local access

RMSDAP.OLB

LB:[I,I]

Object library for RMS-II remote access (can be deleted if you
do not need remote access)

RMSFUN.OBJ

LB:[I,I]

RMS function routines for the system library

RMSRESAB.MAP

LB:[1,34]

Map for library root

RMSRESAB.TSK

LB:[I,I]

Installed as the root of the resident library

RMSFAKAB.CMD
RMSROTAB.STB

LB:[1,24]

For rebuilding RMSLBL and RMSLBM

RMSRESAB.CMD
MSRESAB.ODL

LB:[1,24]

For rebuilding RMSRESAB

RMSBCK.TSK

LB:[3,54]

RMSBCK utility; uses RMSRES

RMSRST.TSK

LB:[3,54]

RMSRST utility; uses RMSRES

RMSDEF.TSK

LB:[3,54]

RMSDEF utility; uses RMSRES

RMSDSP.TSK

LB:[3,54]

RMSDSP utility; uses RMSRES

RMSCNY.TSK

LB:[3,54]

RMSCNV utility; uses RMSRES

RMSDES.TSK

LB:[3,54]

RMSDES utility; uses RMSRES

RMSIFL.TSK

LB:[3,54]

RMSIFL utility; uses RMSRES

RMSDES.IDX

LB:[1,2]

Indexed help file used by RMSDES

RMSII.ODL

LB:[I,I]

Prototype ODL file

RORMSI.MAC

LB:[I,I]

For use with the prototype ODL

RMSll8.0DL

LB:[I,I]

ODL file for sequential access

RMS12S.0DL

LB:[I,I]

ODL file for sequential access

RMSl1X.ODL

LB:[I,I]

Standard indexed file ODL

RMSl2X.ODL

LB:[I,I]

Indexed file ODL

RMSRLX.ODL

LB:[I,I]

ODL for use with RMSRES

RMSSLX.ODL

LB:[I,I]

ODL for use with RMSRES in supervisor mode

DAPl1X.ODL

LB:[I,I]

ODL for use with overlaid RMSDAP

DAPRLX.ODL

LB:[I,I]

ODL for use with DAPRES
(continued on next page)

3-4 RMS-ll 'Version 2.0 Software Modifications, Restrictions, and Supplementary Information

Table 3-1 (Cont.):

RMS-11 Files on RSX-11 M-PLUS Distribution Kit

File Name

Destination

Comments

DAPSLX.ODL

LB:[l,l]

ODL for use with RMSRES in supervisor mode and DAPRES
(not in supervisor mode)

RMSRES.TSK

LB:[l,l]

Task image for linking in nonsupervisor mode; not to be
installed as the root of the library

RMSLBL.TSK

LB:[l,l]

Task image for RMSRES segment

RMSLBM.TSK

LB:[l,l]

Task image for RMSRES segment

RMSRES.STB

LB:[l,l]

STB file for RMSRES

DAPRES.TSK

LB:[l,l]

Task image for RMSDAP resident library

DAPRES.STB

LB:[l,l]

STB file for DAPRES

RMSLBL.MAP*
RMSLBM.MAP*
DAPRES.MAP*

LB:[1,34]

Maps for segmented library

RMSRES.TSK

LB:[3,54]

Task image for linking in supervisor mode; not to be installed
as the root of the library

RMSRES.STB

LB:[3,54]

STB file for linking supervisor-mode tasks

RMSLBL.CMD*
RMSLBL.ODL*

LB:[1,24]

For rebuilding RMSLBL

RMSLBM.CMD*
RMSLBM.ODL*

LB:[1,24]

For rebuilding RMSLBM

DAPRES.CMD*
DAPRES.ODL*

LB:[1,24]

For rebuilding DAPRES

BCKNON.CMD*
BCKNON.ODL*

LB:[1,24]

For rebuilding the overlaid version of RMSBCK

BCKNRN.CMD*
BCKNRN.ODL*

LB:[1,24]

For rebuilding the resident-library version of RMSBCK

BCKNSN.CMD*
BCKNSN.ODL*

LB:[1,24]

For rebuilding the supervisor-mode version of RMSBCK

CNVNON.CMD*
CNVNON.ODL*

LB:[1,24]

For rebuilding the overlaid version of RMSCNV

CNVNRN.CMD*
CNVNRN.ODL*

LB:[1,24]

For rebuilding the resident-library version of RMSCNV

CNVNOO.CMD*
CNVNOO.ODL*

LB:[1,24]

For rebuilding the overlaid version of RMSCNV with RMSDAP
(continued on next page)

RMS-ll Version 2.0 Software Modifications, Restrictions, and Supplementary Information 3-5

Table 3-1 (Cont .):

RMS-ll Files on RSX-ll M-PLUS Distribution Kit

File Name

Destination

Comments

CNVNRR.CMD*
CNVNRR.ODL*

LB:[1,24]

For rebuilding the resident-library version of RMSCNV with
RMSDAP

CNVNSN.CMD*
CNVNSN.ODL*

LB:[1,24]

For rebuilding the supervisor-mode version of RMSCNV

DEFNON.CMD*
DEFNON.ODL*

LB:[1,24]

For rebuilding the overlaid version of RMSDEF

DEFNRN.CMD*
DEFNRN.ODL*

LB:[1,24]

For rebuilding the resident-library version of RMSDEF

DEFNSN.CMD*
DEFNSN.ODL*

LB:[1,24]

For rebuilding the supervisor-mode version of RMSDEF

DESNON.CMD*
DESNON.ODL*

LB:[1,24]

For rebuilding the overlaid version of RMSDES

DESNRN.CMD*
DESNRN.ODL*

LB:[1,24]

For rebuilding the resident-library version of RMSDES

DESNSN.CMD*
DESNSN.ODL*

LB:[1,24]

For rebuilding the supervisor-mode version of RMSDES

DSPNON.CMD*
DSPNON.ODL*

LB:[1,24]

For rebuilding the overlaid version of RMSDSP

DSPNON.CMD*
DSPNON.ODL*

LB:[1,24]

For rebuilding the resident-library version of RMSDSP

DSPNSN.CMD*
DSPNSN.ODL*

LB:[1,24]

For rebuilding the supervisor-mode version of RMSDSP

IFLNON.CMD*
IFLNON.ODL*

LB:[1,24]

For rebuilding the overlaid version of RMSIFL

IFLNRN.CMD*
IFLNRN.ODL*

LB:[1,24]

For rebuilding the resident-library version of RMSIFL

IFLNSN.CMD*
IFLNSN.ODL*

LB:[1,24]

For rebuilding the supervisor-mode version of RMSIFL

RSTNON.CMD*
RSTNON.ODL*

LB:[1,24]

For rebuilding the overlaid version of RMSRST

RSTNRN.CMD*
RSTNRN.ODL*

LB:[1,24]

For rebuilding the resident-library version of RMSRST

RSTNSN.CMD*
RSTNSN.ODL*

LB:[1,24]

For rebuilding the supervisor-mode version of RMSRST
(continued on next page)

3-6 RMS-ll ~rsion 2.0 Software Modifications, Restrictions, and Supplementary Information

Table 3-1 (Cont.):

RMS-11 Files on RSX-11 M-PLUS Distribution Kit

File Name

Destination

Comments

RMSUTL.OLB*
RMSODL.ODL*

LB:[1,24]

For rebuilding the utilities

GSA. MAC

LB:[USER]

Demonstration program included as an illustration of how to
extend an RMS-ll task in the event of pool exhaustion

PARSE. MAC
SEARCH. MAC
RENAME.MAC
ERASE. MAC
PARSE.TSK
SEARCH.TSK
RENAME.TSK
ERASE.TSK

LB :[USER]

Demonstration programs for the directory and wildcarding
facilities

Note
All RMSDAP files can be deleted if you are not using RMS-l1 to access files on
remote nodes.

3.3.2.2 Startup Command Procedures

To install RMS-ll, install the resident libraries and RMS-ll utilities at system startup. To
aid you in the installation process, the file LB:[1,2]STARTUP.CMD contains commands and
sample comments that can be edited to become system startup commands. Please note the
following items:
•

On RL02 pregenerated systems, the RMS-ll segmented library (RMSRES, RMSLBL,
RMSLBM) and all the RMS-ll utilities are already installed in the system image. For
these kits, the only installation needed is the optional installation of the DAPRES
resident library if the system has DECnet support and if RMS-ll remote access
facilities are used.

•

On systems other than RL02, the startup file contains commands that install the
segmented resident library in the system image. Note that these are commands, not
comments; if you do not want to install the library, you should edit the file to change
these commands to comments. Please note the following items:
All resident libraries should be installed using the option IRON=YES.
The root of the library, RMSRESAB, must be the task image contained in LB:[l,l]].
The [l,l]RMSRES.TSK should never be installed in the system; it is only used when
tasks are linked to use the non-supervisor-mode version of RMSRES.
Failure to install the root segment RMSRES will cause the error "INS-Common
block not Loaded RMSRES" when a referencing task or utility is invoked. Failure
to install any of the remaining resident library segments will not give an error at
invocation. However, it may cause the error codes ER$LIB or ER$ENv, or a BPT
trap (with RO containing ER$LIB), when the missing segment is needed by RMS-l1.

RMS-ll ~rsion 2.0 Software Modifications, Restrictions, and Supplementary Information 3-7

There are several circumstances involving "inconsistent or incomplete resident
libraries" that can cause a BPT trap to be generated with RO containing the error
code ER$LIB. This can occur if not all the segments of the library are installed or
if the version numbers of one or more segments do not match the root segment,
the RMSDAP code, or the task itself. In particular, this can happen to the RMS-ll
utilities if they are built to use the segmented resident library and the segments are
installed incorrectly.
The resident libraries have been built with the PAR=pamame option in the TKB
command file, where parname is the name of the resident library. This feature
was included for compatibility with RSX-IIM systems. You should install the
resident libraries in the GEN partition. Do not generate individual partitions for
the libraries on the RSX-I1M-PLUS operating system; doing so would negate much
of the benefit of having a "demand-paged" segmented library.
•

If you are using RMSRES in supervisor mode, you do not need to install a different
task. You simply need to link your task.

•

If you want to use the RMS-ll utilities or the RMSDAP resident library (RMSDAP),

you must edit the startup command file, which contains template commands (comments)
that you can edit to install these facilities.
3.3.2.3 Utility Configurations

The RMS-ll utilities that are provided on the distribution kit are built to use the
segmented resident library. Consequently, RMSRES, RMSLBL, and RMSLBM should
be installed before you use any of the RMS-l1 utilities.
3.3.2.4 Utility Command and ODL Files

For each utility, a command file and one or more ODL files are provided to build the utility
using disk-overlaid RMS-l1. The names of the files are as follows:
Utility

Files

RMSBCK

BCKNON.CMD, BCKNON.ODL, and RMSODL.ODL

RMSRST

RSTNON.CMD, RSTNON.ODL, and RMSODL.ODL

RMSCNV

CNVNON.CMD and CNVNON.ODL

RMSDSP

DSPNON.CMD, DSPNON.ODL, and RMSODL.ODL

RMSDESAB

DESNON.CMD, DESNON.ODL, and RMSDESAB.ODL

RMSDEF

DEFNON.CMD, DEFNON.ODL, and RMSODL.ODL

RMSIFL

IFLNON.CMD and IFLNON.ODL

In addition, a command file and ODL file are provided to build each utility with the resident
library RMSRES. The names of the files are as follows:

RMS-ll ~rsion 2.0 Software Modifications, Restrictions, and Supplementary Information

Utility

Files

RMSBCK

BCKNRN.CMD and BCKNRN.ODL

RMSRST

RSTNRN.CMD and RSTNRN.ODL

RMSCNV

CNVNRN.CMD and CNVNRN.ODL

RMSDSP

DSPNON.CMD and DSPNON.ODL

RMSDESAB

DESNRN.CMD and DESNRN.ODL

RMSDEF

DEFNRN.CMD and DEFNRN.ODL

RMSIFL

IFLNRN.CMD and IFLNRN.ODL

Finally, a command file and ODL file are provided to build each utility using the
supervisor-mode library RMSRES. The names of the files are as follows:
Utility

Files

RMSBCK

BCKNSN.CMD and BCKNSN.ODL

RMSRST

RSTNSN.CMD and RSTNSN.ODL

RMSCNV

CNVNSN.CMD and CNVNSN.ODL

RMSDSP

DSPNSN.CMD and DSPNSN.ODL

RMSDES

DESNSN.CMD and DESNSN.ODL

RMSDEF

DEFNSN.CMD and DEFNSN.ODL

RMSIFL

IFLNSN.CMD and IFLNSN.ODL

For RMSCNY, two additional configurations are available if you want to use RMSCNV to
access files on remote nodes. To build RMSCNV using the clustered RMS-ll and RMSDAP
resident libraries, use the files CNVNRR.CMD and CNVNRR.ODL. To build RMSCNV using
disk-overlaid RMS-ll and RMSDAP, use the files CNVNOO.CMD and CNVNOO.ODL.
Note
These command flIes and ODL flIes are not supplied with the RL02 kits.

RMS-ll 'lkrsion 2.0 Software Modifications, Restrictions, and Supplementary Information 3-9

3.3.2.5 Rebuilding the Utilities

To rebuild the utilities, perform the following steps:

1. Log in to a privileged account.
2. Set your default account to directory [1,24] on the system disk.
3. Use the Task Builder (TKB) to build the utility or utilities.
The utilities and the corresponding map files will be built in the current account. You may
then want to put the utility in the system account ([3,54]).
Each command file for a particular utility creates a utility of the correct name. For example,
CNVNRN.CMD and CNVNON.CMD both produce task images called RMSCNY.TSK. One is
built to use the resident library and one is built with disk-overlaid RMS-11.

3-10 RMS-11 version 2.0 Software Modifications, Restrictions, and Supplementary Information

Chapter 4
Layered Product Support
4.1 Correction Files
Table 4-1 lists the layered products and correction files for the RSX-11M-PLUS Version
4.3 operating system. The layered products that have updated correction files for this
release are marked with an asterisk ( *). The application of the updated correction files
to each layered product will result in improved software performance. Table 4-1 also
includes the corresponding directories and documentation files for the layered products. The
documentation files describe the procedure for applying corrections to the specific layered
product.
Table 4-1:

Layered Product Correction File References
Directories
Used

Backup
Set

Documentation
File Name

PDP-II BASIC-PLUS-2
Version 2.5

[211,200]

BP2

[211,200]BP2.DOC

COB0L-81IRSX *
Version 3.0

[215,200]

COBOLSI

[215,200]C81.DOC

DATATRIE~111RSX

[222,200]

DTR

[222,200]DTR.DOC

DCXKMVII-A
Development Tools *
Version 1.0

[224,200]

DCX

[224,200]DCX.DOC

DEC net-lIM-PLUS *
Version 4.0

[230,200]
[225,20]
[225,40]
[230,20]
[230,40]

DECNET

[230,200]DECNET.DOC

Layered Product

Version 3.2

(continued on next page)

Layered Product Support 4-1

Table 4-1 (Cont.):

Layered Product Correction File References
Directories
Used

Backup
Set

Documentation
FileName

RSX-11M-PLUS DIBOL *
Version 1.3

[371,200]

DIBOL

[371,200]DIBOL.DOC

DPM-PLUS
Version 1.1

[236,200]

DPM

[236,200]DPMP.DOC

IP11 Industrial I/O Subsystem
Software Driver
Version 4.1

[340,200]

IP11

[340,200]lPUPINST.DOC

KMV11-A HDLC
Framing Software *
Version 1.0

[253,200]

KMV11

[253,200]KMV11.DOC

KMV11 X25 Link Level *
Version 1.0

[322,200]

KXI

[322,200]KXI.DOC

MIRA *
Version 1.0

[370,200]

MIRA

[253,200]MIRA.DOC

PDP-11 PascallRSX
Version 1.2

[262,200]

PASCAL

[262,200]PASCAL.DOC

PDP-11 SORTIMERGE
Version 3.0

[311,200]

SORT

[311,200]SORT.DOC

PDP-11 Symbolic DebuggerlRSX
Version 2.1

[247,200]

PDPDBG

[247,200]PDPDBG.DOC

RSX-11 2780/3780
Protocol Emulator
Version 4.1

[301,200]
[301,25]

2780

[301,200]2780.DOC

RSX-11 3271 Protocol Emulator·
Version 3.1

[345,200]
[345,25]

3271

[345,200]3271.doc

Layered Product

4.2 Installing Layered Products Restriction
Because of support for named directories, the MCR command SET INONAMED or the DCL
command SET DEFAULTINONAMED should be used before installing layered products.

4-2 Layered Product Support

Appendix A
The Digital Equipment Computer Users Society
(DECUS)
DECUS, the Digital Equipment Computer Users Society, is one of the largest and most
active user groups in the computer industry. It is a nonprofit association, supported and
administered by Digital, but actively controlled by its members. DECU8 headquarters,
located in Marlborough, Massachusetts, administers all international policies and activities.
Members include individuals who have purchased, leased, ordered, or used a Digital
computer, or anyone who has a genuine interest in DECUS. Membership is free and
voluntary.

A. 1 DECUS Goals
The primary goals of DECUS are as follows:

•

To advance the art of computation through mutual education and exchange of ideas and
information

•

To establish standards and to provide channels to facilitate the exchange of computer
programs

•

To provide feedback to Digital regarding hardware and software customer needs

•

To advance the effective use of Digital computers, peripherals, and software by
promoting the interchange of information

To further these goals, DECUS serves its members by holding symposia; by maintaining a
program library; by publishing an association newsletter, technical newsletters, and books;
and by supporting a number of Special Interest Groups (8IGs).

The Digital Equipment Computer Users Society (DECUS) A-l

A.2 DECUS Activities
Local, regional, and national DECUS organizations give members the opportunity to meet
other Digital customers and employees in an informal setting. At local monthly meetings
and national symposia, members can discuss their ideas, learn what others are doing, and
give Digital valuable feedback for future product development.
Often the national meetings in various countries also provide the stage for major new
product announcements as well as a showplace for interesting developments in both
hardware and software technology. At any meeting, members might describe ideas
and programs they have implemented or the fine-tuning of a particular feature or
application. Members present papers, participate in panel discussions, lead workshops and
Pre-Symposium Seminars, and conduct demonstrations for the benefit of other members.
United States DECUS Symposia are held twice per year, in the Spring and Fall, in varying
locations around the country. Events which are currently scheduled include the following:
Spring DECUS

May 7-11, 1990

New Orleans, LA

Fall DECUS

December 10-14, 1990

Las Vegas, NV

Many of the technical papers and presentations from each symposium are published as a
book, the DEGUS Proceedings. Copies of the DEGUS Proceedings are supplied to symposia
attendees and can be purchased by DECUS members. DECUS also publishes newsletters
focusing on special interests and a society newsletter.
DECUS has a Program Library, which contains over 1700 software packages that
are written and submitted by users. A wide range of software is offered, including
languages, editors, numerical functions, utilities, display routines, games, and other
types of application software. Also available are library catalogs, which contain program
descriptions and ordering information. The programs are available for a nominal service
charge that covers the cost of reproduction and media.
Many DECUS members derive additional benefits from joining a DECUS Special Interest
Group (SIG). SIGs often hold smaller meetings while attending regional and national
meetings. They may also meet informally to discuss their special fields, such as operating
systems, languages, processors, and applications. There are at least 25 SIGs, including the
RSX SIG, in the United States alone. Many of the SIGs print newsletters and disseminate
valuable technical information to members.

A.3 DECUS Membership
To obtain a membership form for DECUS, contact a Digital sales representative or the
nearest DECUS chapter office. Membership information may be obtained directly from U.S.
DECUS Headquarters by calling telephone number 508-480-3446.

A-2 The Digital Equipment Gomputer Users Society (DEGUS)

Appendix B
Reporting Problems
This appendix describes the procedure for submitting a Software Performance Report (SPR).
An SPR allows you to report any problems with or questions about your system directly to
Digital.

An SPR can be used for:
•

Software errors

•

Documentation errors (when the Reader's Comments form is not appropriate)

•

Follow-up on a previous SPR

•

Questions

•

Suggestions

An SPR cannot be used for:
•

Software license and price policies

•

Obvious hardware problems

•

Logistical or clerical problems with kits, such as blank media, or failure to receive the
Software Dispatch

•

Problems with user-written software

In general, when you complete an SPR, use the following guidelines:
•

Describe only one problem per form. This will facilitate a more rapid response because
it allows the person answering the problem to concentrate more fully on that particular
problem. One problem per form also helps simplify record-keeping.

•

Define as accurately as possible the state of the system and circumstances when the
problem occurred.

•

Illustrate the problem with specific examples.

Reporting Problems B-1

•

If you report a documentation error, specify the title of the manual, and include the
section and page number where the error occurred. Include a table or figure number if
appropriate.

SPRs are assigned a priority of 1 to 5. An SPR that is assigned a priority of 1 receives the
highest priority. Priorities are described as follows:
1. Most production work cannot be run.

Major system functions are unusable.
You cannot boot the system.
Necessary peripherals cannot be used.
2. Some production work cannot be run.
Certain functions are unusable.
System performance has declined.
Installation does not have excess capacity.
3. All production work can be run with some user impact.
Significant manual intervention is required.
System performance has declined.
Installation has excess capacity.
4. All production work can be run with no significant impact on user.
-

Problem can be patched or easily bypassed.

5. No system modifications are needed to return to normal production.
Suggestions are supplied.
Errors in documentation are noted.
You can submit the following categories of SPRs:
•

ProblemlError SPR
This type of SPR contains a software problem. It is assigned a priority of 1 to 5. You
receive an answer to this report.

•

Suggested Enhancement
This type of SPR contains a suggestion. It is assigned a priority of 5. You do not receive
an answer to this report.

•

Other
This type of SPR contains a question or suggestion. It is assigned a priority of 5. You
mayor may not receive an answer to this report.

B-2 Reporting Problems

Please supply the following information (in machine-readable form where applicable) when
you report a problem:
•

CRASH-Supply a copy of the Executive task-build map, output from the console
terminal, the SYSGEN saved-answer file, the Executive STB file, and the crash dump.
If the crash is reproducible, accurately describe the details and supply a hard copy or
user source code when necessary.

•

DRIVERS-Supply controller/device information, software options, error log output, a
copy of device registers, and a sample program.

•

UTILITIES-Supply a copy of your terminal output, showing setup commands, before
and after effects, and relevant file information.

•

TASK BIDLDER-Supply a copy of your terminal output command files, the task map,
and a dump of the first few blocks of the task image.

•

FILE SYSTEM-For a COITupted volume, supply output from the verification utility and
dump of the volume. For improper results, supply the eITor code, a file header dump,
and a sample program.

•

ERROR LOG REPORT GENERATOR (RPT)-Supply a copy of the report file generated
by RPT, either on a hardcopy listing or on machine-readable media.

If a failure occurs when you are running privileged, add-on software (for example, the
DECnet package), try to reproduce the failure without the additional software. Then, when
you write the SPR, indicate how the system operated with and without the add-on software.
The SPR process is sometimes lengthy. Therefore, if you have a critical problem, contact
your local Digital office. In the meantime, read the Software Dispatch for news on the
RSX-llM-PLUS operating system, which includes changes and problems other people have
found and solved.

Reporting Problems B-3

Appendix C
Applying Corrections to Source Files
Interim changes to the Executive, MCR, and device drivers are made by creating correction
files that are processed by the Source Language Input Program (SLP). SLP generates a new
copy of the modules that contain the errors by applying the corrections to the source file on
the distribution kit. (See the RSX-llM-PLUS Utilities Manual for complete information on
SLP.)
Mter you have applied the corrections and have obtained a new version of the file, do not
delete the original source file. Interim changes that may be distributed later are cumulative
and depend on the availability of the original sources.

c. 1 Updating an Executive Source Module
To update an Executive source file (ABCDEF.MAC, for example), mount the disk on
which you performed your system generation and create a SLP correction file named
ABCDEF.COR in the directory [11,40]. (All the following instructions assume that you are
working on the disk on which you performed your system generation.) Then, while your
system is running under User Identification Code (VIC) [11,10], submit the correction file to
SLP. For example, you could follow this sequence to create REQSB.MAC:
>
>
[Creating new file]
Input

[Exit]

>
>

Applying Corrections to Source Files C-l

If the updated Executive module in your system is not a loadable driver, use the following
procedure:
1. Assemble the new module, using the RSXMC.MAC file for the target system. For

example, type the following commands:
>
>

2. Use the Librarian Utility Program (LBR) to replace the old version of the module in the
RSX11M.OLB file on the target system.
For example:
>
>

3. Perform the following sections of SYSGEN:
•

Building the Executive and Drivers

•

Building the Privileged Tasks

•

Creating the System Image File

If the modified file in your system is a loadable device driver (ZZDRV.MAC, for example), the
updated module can be replaced without rebuilding the Executive. Assemble the updated
module and replace the resulting object file in the RSX11M.OLB file of your target system.
For example, assume that [11,10] contains the RSXMC.MAC file resulting from your system
generation and do the following:

>
>
>
>

Use [200,200]ZZDRVBLD.CMD to rebuild the driver. If necessary, copy ZZDRV.TSK and
ZZDRV.STB into the directory that corresponds to the system VIC, or the library VIC if the
driver is vectored.
For example:

>
>
>
SYSUIC=[g,m]

>
>

Use the Virtual Monitor Console Routine (VMR) to unload the old device driver and to
load the new one. If the new driver is larger than the old one, it may not fit into the same
locations as the old one. It may be necessary to unload and reload all of the loadable drivers
in that partition to create enough room.

C-2 Applying Corrections to Source Files

For example:
>
Enter filename:
VMR>
VMR>
VMR>

If the driver is vectored, you must use the NEC switch with the UNL and LOA commands.
In the preceding example, you would substitute the following VMR commands:
VMR>
VMR>
VMR>

Hardware boot the modified system.

C.2 Updating an MCR Source Module
To update the MCR source file SETOV.MAC, use the following procedure:
1.

Create the SLP correction file [12,40]SETOV.COR and use it to update
[12,10]SETOV.MAC. Assemble SETOV for the target system, as follows:
>
>
[Creating new file]

[Exit]

>
>

2. Assume that directory [11,10] contains the RSXMC.MAC file resulting from your target
system generation and do the following:
>
>

All of the Task Builder command files output a map to the logical device MP, which
must be assigned to NL or another device to avoid a diagnostic error message from the
Task Builder.
3. If it was necessary to modify the MCR Task Builder command file (MCRBLD.CMD)
during the last system generation, it may now be necessary to repeat those changes. To
rebuild the secondary portion ofMCR ( ...MCR) and replace the module SETOV, use the
following procedure:

Applying Corrections to Source Files C-3

>
>
>
>
>
>
>
Enter filename:
VMR>
VMR>

VMR>

To rebuild the MCR dispatcher (MCR ... ) and to replace the module MCRDIS, use the
following procedure:
>
>
>
>
>
>
>
Enter filename:
VMR>
VMR>
VMR>

There is only one procedure for replacing an external MCR task. It involves the
following steps:
1. Create the SLP file, apply it, and create the object file.
2. Incorporate the updated module into the task's object library.
3. Rebuild the task and install it in the system using MCR or VMR. Before using VMR,
you must assign SY and LB to the disk containing the target system.
The following example replaces the module INSLB of the external MCR task INSTALL:
>
>

4. If it was necessary to modify the external task's Task Builder command file during the
last system generation, it may be necessary at this time to repeat those changes:
>
>
>
>
Enter filename:
VMR>
VMR>
VMR>

Hardware boot the system.

C-4 Applying Corrections to Source Files

C.3 Updating a DCl Source Module
The procedure is different for the DCL task. There are two DCL object libraries:
DCLR.OLB for modules in the root segment of the DCL task, and DCLO.OLB for modules
in DCL's overlay segments. (One module, COMMAND, has versions in both the root and
overlay libraries.) These libraries are located in directory [1,24] on your distribution kit.
If you are not sure whether a module belongs in the root or the overlay library, use the
Librarian Utility Program (LBR) to scan the module names in the libraries. For instance,
to view the module names in DCLO.OLB, enter the MCR or DCL command shown next.
>
$

The modules you are most likely to need to modify are the DCL syntax tables, which are in
DCLO.OLB. (All DCL syntax tables are in overlay segments.) See the RSX-llM-PLUS and
Micro / RSX System Management Guide for more information on the structure of the DCL
task and on the DCL task-building process.
File DCL.CMD, located in directory [23,24] on your distribution kit, can be used either
to assemble DCL overlay modules and to rebuild DCL, or as a template for your own
commands. The file contains its own instructions. Note that to use this command file
directly, you must copy various files into directory [23,24] and appropriately modify any
directories that they reference. File DCL.CMD also shows how to assemble the special
module COMMAND.
The following process shows how to reassemble a DCL overlay module and rebuild DCL.
The process is similar for root modules; simply specify DCLR rather than DCLO. See the
DCL.CMD command procedure for details on rebuilding COMMAND.
If your command line interpreter (CLI) is MCR, use the following commands:
>
>
>
>
>

If your CLI is DCL, use the following commands:
$
$

$
$
$

When you have built a new version of DCL.TSK, you must replace the old copy of DCL as
a system CLI. First, any terminals whose CLI is DCL must be logged out or set to another
CLI. Then, the following commands must be executed (note that these commands are for
MCR only):
>
>

>
>

Applying Corrections to Source Files C-5

The REMOVE and INSTALL commands are also valid VMR commands and can be used to
modify your system's image on disk. However, the CLI IINIT command is for MCR only, and
it must be put in your system's startup file (if it is not there already).

C.4 Applying Corrections to the File Systems (F 11 ACP)
All F1IACP updates begin by creating a SLP correction file in directory [13,40]. The
following example patches a module called WTRN1.
1. Boot your system and log in to a privileged account, as follows:
>
Password:

2. If necessary, restore the required files from the distribution tape (the following example
assumes that the files in [13,10] have been deleted, but the files in [1,24] have not):
>
>
>
BRU>
From:
To:
BRU--Completed
BRU>

3. Create the correction file, as follows:
>
>

4. Apply the correction, as follows:
>
>

5. Assemble the corrected module with the Executive macro library, the Executive prefix
file RSX1v.1C.MAC, and the prefix file FI1PRE.MAC, as follows:
>
>

6. Replace the defective module in the file control processor (FCP) library, as follows:
>
>
Modules replaced
WTRNl

All of the task-build command files require that the logical device MP be assigned to the
appropriate device.
In the following example, xxx must be replaced by the 3-character designation for your
desired FCP, that is, MDL or LRG.
1. Task build the new FCP by using the updated library as follows:

>
>

C-6 Applying Corrections to Source Files

2. Install the updated FCP in the system image as follows:
>
Enter filename:
VMR>
VMR>
VMR>

3. Reboot the system to place the new FCP in use.

C.5 Applying Corrections to the Reconfiguration Tasks CON and
HRC
The following example illustrates how to patch module CNCMR for the CON task. Unless
indicated otherwise, use a similar command sequence for correcting module HRONL for the
HRC task.
This example assumes that:
•

You have deleted the source files in [27,10].

•

You have not deleted the object libraries and command files in [1,24] and [1,20].

•

The object library for both CON and HRC is [1,24]OLR.OLB.

•

The disk to which you will apply the patches is mounted Files-lIon drive DBO.

•

The distribution tape is mounted foreign on MMO.

Perform the following steps to patch CNCMR:
1. Boot your system, and log in to a privileged account.

2. If necessary, restore the required files from the distribution tape as follows:
>
>
>
>
BRU>
From:
To:
BRU -- Completed
BRU>

If you were correcting module HRONL, you would substitute the following for the From:
line in the previous command sequence:
From: MMO:

3. Create the SLP correction file in mc [27,40], as follows:
>
>

4. Apply the SLP correction file to CNCMR as follows:
>
>

Applying Corrections to Source Files C-7

5. Assemble the corrected CNCMR module by using the Executive macro library and the
Executive prefix file RSXMC.MAC; you also use these when assembling a corrected
HRC module. In addition, use either the prefix file CNPRE.MAC, if assembling a CON
module, or HRPRE.MAC, if assembling an HRC module.
For example:
>
>

6. Replace the CNCMR object module in the OLR library as follows:
>
>

7. Task build CON by using the updated library as follows:
>
>
>
>
Note
If you do not have the task-build command files [1,24]CONBLD.CMD and
[1,24]CONBLD.ODL, or [1,24]HRCBLD.CMD and [1,24]HRCBLD.ODL, you
must use the section of SYSGEN that rebuilds system-supplied tasks.

8. Install the updated reconfiguration utility in the system image as follows:
For CON, type:

For HRC, type:

Enter filename:
VMR>
VMR>
VMR>

9. In order to use the new reconfiguration tasks, follow these steps:
a. Shut the system down using the SHUTUP utility as follows:
>

b. Reboot the system.

C.6 Applying FCS Corrections
Correcting the File Control Services (FCS) modules on an RSX-IIM-PLUS system can be
done by updating the source files, by assembling them, and by replacing modules in the
system library, usually LB:[l,l]SYSLIB.OLB.
This process is complicated by the fact that there are three kinds of FCS, as follows:
•

ANSI-Supports ANSI-format magnetic tape and big buffers.

•

Non-ANSI-Does not support ANSI tape or big buffers.

•

Multibuffered-Supports ANSI tape, big buffers, and multiple. buffers

C-8 Applying Corrections to Source Files

An FCS source file like CLOSE.MAC contains conditional assembly directives that can
produce three different CLOSE objects, depending on the global symbols defined when
CLOSE.MAC is assembled. These three different CLOSE objects correspond to the three
kinds of FCS. Other FCS source files, like DELETE.MAC, have no such conditional
assembly directives. They are only assembled one way; that is, only one DELETE object
exists.

The SYSLIB.OLB file provided on the kits contains the ANSI FCS. Thus, this system library
contains modules like CLOSE, assembled with the ANSI tape conditionals and big buffer
conditionals enabled, and modules like DELETE) which have no such conditionals and are
the same in any FCS.
An alternate system library called NOANSLIB.OLB is also provided. It contains an FCS
that does not support ANSI tape, big buffers, or multiple buffers. Be sure that you know
whether the system library on your system contains the ANSI FCS, or if it has been
replaced with the non-ANSI or multibuffered FCS.

Details of the correcting procedure follow. MCR syntax is used throughout.

C.7 Updating the FCS Sources
The FCS source files are found in [50,10] on the kits. Updating a source is done by entering
the correction file into [50,10], and by entering "SLP @filename" to apply the correction.

C.8 Assembling FCS
Assembling the updated source or sources can be done in either of two ways. One way is
simple and time-consuming; the other is quick, but it must be done with great care. The
simple way is to set your mc to [50,24] and enter MAC @FCSASM. This will assemble
every FCS variant properly, producing over 100 object files.
The other choice is to assemble only the sources that have been updated. This is not as
simple, but it saves machine time. To assemble only a particular file, look at the five *.CMD
files in [50,24] on the kit. Search the command files for all references to the file you want to
assemble. A description follows of each command file and what you do with its contents.
1. FCSBOTH.CMD assembles files like DELETE.MAC, which contain no code specifically
written to support or deny support to ANSI tape, big buffers, or multiple buffers. If the
source file you have updated is mentioned in FCSBOTH.CMD, then it is assembled the
same way regardless of which kind of FCS is in your system library. For example, the
following lines assemble DELETE:
[50,24]DELETE, [50, 34]DELETE/-SP=,
[50,10]FCSPRE,DELETE
.

(FCSPRE.MAC defines necessary macros and global symbols.)

To assemble DELETE.MAC, you should set your VIC to [50,24] and enter the following
command line:
>

Applying Corrections to Source Files C-9

As an alternative, you could put the command line in a file and enter "MAC @filename".
A file like DELETE.MAC, which has no conditional assembly directives, is mentioned
only in FCSBOTH.CMD.

2. FCSANSl.CMD assembles source files that contain ANSI tape, big buffer, or
multibuffering conditionals. It produces objects for the ANSI kind of FCS. If your
system library contains the ANSI FCS, and the source file you have updated is
mentioned in FCSANSl.CMD, then you should use the command line you found in
FCSANSl.CMD to assemble the source file. For example, the following lines from
FCSANSl.CMD assemble CLOSE.MAC:
[50,24]CLOSE.MTA, [50, 34] CLOSE.MTA/-SP=[50,10]FCSANSI/PA:1,FCSBIGBUF/PA:1,FCSPRE,CLOSE

FCSANSI.MAC enables the ANSI tape conditional assembly directives, and
FCSBIGBUF.MAC enables the big buffer conditionals. Note the sequence of input
file names. It is absolutely imperative that the source file is the last file name
specified, and that FCSPRE is next to it, in every FCS assembly. Otherwise, the proper
conditionals will not be enabled, with potentially confusing and dangerous results. Any
file that is assembled by FCSANSl.CMD is also assembled by FCSNOANSl.CMD and
FCSMULBUF.CMD. Note the MTA file types above, which distinguish between the
three kinds of CLOSE objects.
3. FCSNOANSl.CMD also assembles source files that contain ANSI tape, big buffer, or
multibuffering conditionals. It produces objects for the non-ANSI kind of FCS. If your
system library contains the non-ANSI FCS, and the source file you have updated is
mentioned in FCSNOANSl.CMD, then you should use the command line you found in
FCSNOANSl.CMD to assemble the source file. For example, this command line from
FCSNOANSl.CMD assembles CLOSE.MAC:
[50,24]CLOSE.NMT, [50,34]CLOSE.NMT/-SP=[50,10]FCSPRE,CLOSE

4. FCSMULBUF.CMD assembles source files that contain ANSI tape, big buffer, or
multibuffering conditionals. It produces objects that support all of these features.
For example, this command line from FCSMULBUF.CMD assembles CLOSE.MAC:
[50,24]CLOSE.MBF, [50,34]CLOSE.MBF/-SP=[50,10]FCSANSI/PA:1,FCSMULBUF/PA:1,FCSBIGBUF/PA:1,FCSPRE,CLOSE

FCSMULBUF.MAC enables the multibuffering conditionals.
5. FCSASM.CMD simply causes the other command files to be executed, to assemble FCS
in all ways. If you are assembling single sources, ignore FCSASM.CMD.
More information is necessary to correctly assemble a few FCS sources that require
additional prefix files. These sources are GET.MAC, PUT.MAC, OPEN.MAC, FINIT.MAC,
and RDWRIT.MAC. GET.MAC can be assembled in the normal way (ANSI, non-ANSI,
or multibuffered) to produce GET.MTA, GET.NMT, or GET.MBF. An additional file,
GPSEQ.MAC, will enable conditional assembly directives in GET.MAC to produce GETSQ,
the sequential GET module, for each kind of FCS. For example, this is the command line
from FCSANSl.CMD, which produces GETSQ.MTA:
[50,24]GETSQ.MTA, [50, 34] GETSQ.MTA/-SP=[50, 10] GPSEQ/PA:l,FCSANSI/PA:1,FCSBIGBUF/PA:1,FCSPRE, GE T

C-10 Applying Corrections to Source Files

If you correct GET.MAC, be sure that you assemble it to produce objects GET and GETSQ
for whatever kind of FCS your system library contains.
Similarly, GPSEQ.MAC will produce PUTSQ, the sequential PUT module, for each kind of
FCS. This is the command line from FCSNOANSLCMD, which produces PUTSQ.NMT:
[50,24]PUTSQ.NMT, [50, 34]PUTSQ.NMT/-SP=[50, 10] GPSEQ/PA:1,FCSPRE,PUT

If you correct PUT.MAC, be sure that you assemble it to produce objects named PUT and
PUTSQ.
OPEN.MAC can be assembled in the normal way (ANSI, non-ANSI, or multibuffered)
to produce OPEN.MTA, OPEN.NMT, or OPEN.MBF. Three additional prefix files exist:
OPFID.MTA, OPFNB.NMT, AND OPENR.MAC.
OPFID.MAC produces the open-by-file-id module, OPFID. This is the command line from
FCSANSLCMD, which produces OPFID.MTA:
[50,24]OPFID.MTA, [50, 34]OPFID.MTA/-SP=[50, 10] OPFID/PA:1,FCSANSI/PA: 1, FCSBIGBUF/PA: 1,FCSPRE, OP EN

OPFNB.MAC produces the open-by-file-name block module, OPFNB. This is the command
line from FCSANSLCMD, which produces OPFNB.NMT:
[50,24]OPFNB.NMT, [50, 34]OPFNB.NMT/-SP=[50, 10] OPFNB/PA: 1, FCSPRE, OPEN

OPENR.MAC produces the open module for resident libraries, OPENR. This is the
command line from FCSMULBUF.CMD, which produces OPENR.MBF:
[50,24]OPENR.MBF, [50, 34]OPENR.MBF/-SP=[50, 10] OPENR/PA: 1,FCSANSI/PA: 1, FCSMULBUF/PA: 1, FCSBIGBUF/ PA:1,FCSPRE,OPEN

If you correct OPEN.MAC, be sure that you assemble it to produce objects OPEN, OPFID,

OPFNB, and OPENR for whatever kind of FCS your system library contains.
FINIT.MAC and RDWRIT.MAC can be assembled to produce FINIT.OBJ and RDWRIT.OBJ.
The prefix file FCSSUP.MAC enables conditionals to produce objects FINTSL.SUP and
RDWRSL.SUP to be used in supervisor-mode libraries. These objects are necessary
for correct execution of the user asynchronous system trap (AST) completion routines,
which may be specified for FCS READ$ and WRITE$ functions. This is the line from
FCSANSI.CMD that produces FINTSL.SUP:
[50,24]FINTSL.SUP, [50,34]FINTSL.SUP/-SP=[50,10]FCSSUP/PA:1,FCSPRE,FINIT

Object files FINTSL.SUP and RDWRSL.SUP contain the same entry point names as
FINIT.OBJ and RDWRIT.OBJ. Therefore, they must be replaced in the system library
with their entry points deleted so that they are only used when they are explicitly called
when a supervisor-mode library is task built.

Applying Corrections to Source Files C-ll

C.9 Replacing the FCS Object Modules
Set your mc to [1,1], make a backup copy of the libraries, and use LBR to replace the
corrected FCS objects that you have assembled. For example, suppose you have updated
and assembled DELETE, PUT, and FINIT on a system with an ANSI FCS in the system
library. You would then use the following procedure:
>
>
>
>

If you use NOANSLIB.OLB, also use the following procedure:
>
>
>

If you want to incorporate the corrected modules into an FCSRES resident library or
FCSFSL supervisor-mode library, first rebuild the library, and then rebuild every task that
links to it with SYSGEN. If you rebuild and install a resident library and run a task that
linked to the old resident library, then the task will call routines in the resident library at
the wrong addresses. The results will be uncertain and potentially damaging, especially for
privileged tasks.

C-12 Applying Corrections to Source Files

Index
A
Account Maintenance Program
See ACNT
ACNT
blank account fields, 2-35
ALLOCATION field
RMS-ll , 2-47
ALTcommand
keywords
!RPRI, 2-7
!TERM, 2-7
Alternate record key field
RMS-ll, 2-44
ASSIGN command, 2-7
AST
user completion routines, 0-11
Asynchronous system trap
SeeAST

B
Backup and Restore Utility
See BRU
BACKUP command
qualifiers
/CREATED, 2-2
!ERROR, 2-3
!MODIFIED, 2-2
restrictions
!LENGTH qualifier, 2-3
/NEW_VERSION qualifier, 2-3
!REWIND/APPEND qualifiers, 2-3
!TAPE_LABEL qualifier, 2-3
!VERIFY qualifier, 2-3
Bad Block Replacement Control Task
See RCT
BRU

BRU (Cont.)
compatibility with VAX-II RSX and VAX
CoProcessor!RSx, 1-3
error messages, 1-6
file size allocation, 1-10
!LENGTH qualifier restriction, 2-3
qualifiers
/COMPARE, 1-4
!ERRORS, 2-3
!VERIFY, 1-4
restoring data, 1-10
BUCKETSIZE field
RMS-l1, 2-47

c
C compiler support, 1-3
CLIIINIT command, C-6
COB0L-81 correction file, 4-1
Communication driver
example, 2-32
programming hints, 2-30
Communication mode
setting full duplex (communication driver),
2-28
Communication set operation
IO.SYN function (communication driver),
2-28
/COMPARE qualifier
BRU, 1-5
Contiguous file areas
RMS-ll , 2-45
Continuation buckets
RMSDES utility, 3-3
Controller identification
DSA driver, 2-33
Crash driver
memory dump, 2-15

Index-l

Crash dump
loadable support, 2-15
Create Regions Directive, 1-8

D
Data cache
readahead support, 1-10
write loading temporary files, 2-35
DATATRIEVE-IIIRSX correction file, 4-1
DCL
updating source file, C-5
DCX KMVII-A Development Tools, 4-1
Deadlock
RMS-ll, 2-44
DECnet-llM-PLUS correction file, 4-1
DECUS, A-I
DEFINE LOGICALS command, 2-7
Device
allocation failure, 2-44
disconnecting (communication driver), 2-29
initializing (communication driver), 2-29
large disk restriction, 1-2
LP:, 1-11
RA70 disk drive, 1-2
RA90 disk drive, 1-2
TK25 cartridge tape, 2-3,2-16
TK50 cartridge tape, 2-8
Device driver
changes, 1-10, C-l
updating loadable, C-2
DIGITAL Command Language
See DCL
Digital Equipment Computer Users Society
See DECUS
Disk
maximum files on, 1-2
Disk Save and Compress utility, 1-13
DMCll (communication driver)
message send, 2-82
powerfail, 2-81
synchronous line interface, 2-26
DPM-PLUS correction file, 4-2
DSA driver
controller identification, 2-33
error log file, 2-33
error log support, 2-33
error packets, 2-83
I/O sequence number, 2-83
offset P.SEQ, 2-83
powerfail recovery
type, 2-83, 2-84

Index-2

DSA driver (Cont.)
Read Serial Number function, 2-33
time delay, 2-83
DU driver
Read Serial Number function, 2-32
setting to volume valid, 2-32
Suppress Error Logging subfunction, 2-32
DUPll (communication driver)
sync character, 2-26
synchronous line interface, 2-26

E
$EDMSG directive, 2-35
file name string conversion, 2-35
ER$FUL error, 2-44
ER$KEY error, 2-44
ER$MRS error, 2-44
ER$NOD error, 2-44
Error log file
DSA driver, 2-33
report, 1-12
Error log support
DSA driver, 2-33
RA70 disk drive, 1-2
RA90 disk drive, 1-2
Error packets
DSA driver, 2-33
!ERROR qualifier
BACKUP command, 2-3
BRU Utility, 2-8
!ER switch
RMS-ll, 2-47
Executive
modifying source file, C-1
support for volume valid, 1-9
virtual terminals, 1-10
EXIT_SUPERSEDE command
RMSDES utility, 2-46, 3-8
Extend Task directive
APR protection mask, 2-20
EXTENSION field
RMS-11 , 2-47
EXTK$ directive, 2-20

F
FCS
ANSI, C-8
assembly files, C-9
code size, 1-13
modifying modules, C-8
multibuffered, C-8

FCS (Cont.)
non-ANSI, C-8
READ$ function, C-11
source file
conditional assembly, C-9
updating, C-9
VMS ACP function support, 1-13
WRITE$ function, C-11
File Control Services
See FCS
File loading
RMS-11 restriction, 3-2
File specification
RMS-11, 2-48
wildcard, 2-48
FMS
command file example, 2-37
.FORM directive
terminal support, 2-6
FORTRAN-77 support, 2-37
FORTRAN IV-PLUS, 2-37
Full-duplex mode
considerations (communication driver), 2-31
setting mode (communication driver), 2-28

G
GEN region name, 2-23
Get Device Information Junior function
(GI.DVJ), 2-22
GIN$ directive
GI.DVJ function, 2-22
GLUN$macro
get information
communication driver, 2-26

H
Half-duplex mode
considerations (communication driver), 2-31
set mode (communication driver), 2-29
HELLO, 1-12
HELP, 1-12

110 sequence number
DSA driver, 2-33
Indexed file
RMS-11 , 2-47
INI command
IINF keyword, 2-10
IMXF keyword, 2-10
INITIALIZE command

INITIALIZE command (Cont.)
!HEADERS qualifier, 2-1
!MAXIMUM_FILES qualifier, 2-1
Interface
message-oriented (communication driver),
2-25
IO.FDX function (communication driver), 2-28
IO.HDX function
set mode (communication driver), 2-29
IO.INL function (communication driver), 2-29
after sync, 2-32
IO.RLB function (communication driver), 2-29
IO.RNS function (communication driver), 2-29
IO.SYN function (communication driver), 2-30
set operation, 2-28
with IO.INL, 2-32
IO.TRM function (communication driver), 2-29
IO.WNS function (communication driver), 2-30
IOSUE.MAC module
modification, 2-20
IP11 correction file, 4-2

K
KMVll-A HDLC Framing Software, 4-2
KMVll X25 Link Level, 4-2

L
Layered product
correction files, 4-1
LOAD command, 2-11
Logical unit number
SeeLUN
Low-traffic sync (communication driver), 2-31
LUN
get information
communication driver, 2-26

M
MCR
updating source file, C-4
Message-oriented interface (communication
driver), 2-25
MIRA, 4-2
Mode
maintenance (communication driver), 2-29
set (communication driver), 2-28, 2-29
Monitor Console Routine
SeeMCR
MSCP support
RCT, 2-34

Index-3

o
ODLfile
RMS-11, 3-8
Offset P.SEQ
DSA driver, 2-33
OPENREGISTER command
!TASKD keyword, 2-11
!TASK! keyword, 2-11
OTS Fast Map routine
lOT restriction, 1-14
Overlaid libraries
Manually loading, 1-11

p
PDP-11 BASIC-PLUS-2 con-ection file, 4-1
PDP-11 PascallRSX con-ection file, 4-2
PDP-11 SORTIMERGE con-ection file, 4-2
PDP-11 Symbolic DebuggerIRSX con-ection file,
4-2
PIP
wildcard support, 1-8
Powerfail
DMC11 (communication driver), 2-31
Powerfail recovery
DSA driver, 2-33, 2-34
Programming hints
communication driver, 2-30

Q
QIO$ macro
communication driver, 2-27
device-specific function
communication driver, 2-28
standard function
communication driver, 2-27
Queue Manager, 1-12

R
RCT
memory allocation, 1-12
MSCP support, 2-34
write-protected disks, 2-34
Read operation
logical block (communication driver), 2-29
sync character (communication driver), 2-29
Record Management Services
See RMS-11
Redundancy checking (communication driver),
2-30
Report Generator Task

Index-4

Report Generator Task (Cont.)
/HISTORY qualifier, 1-13, 2-16
Resource Monitoring Display
See RMD
RL02 installation kit, 3-3
RLOG$ macro expansion, 2-18
RLON$ macro expansion, 2-18

RMD
BOTTOM command, 2-34
TOP command, 2-34
Writes Load Rates, 1-7
RMS-l1
accessing files on remote nodes, 3-9
ALLOCATION field, 2-47
alternate record key field, 2-44
BUCKETSIZE field, 2-47
building utilities, 3-8
contiguous file areas, 2-45
creating indexed file, 2-47
deadlock, 2-44
DEF command restriction, 3-3
DEFINE command restriction, 3-3
errors
ER$DNF, 3-2
ER$FUL, 2-44
ER$KEY, 2-44
ER$MRS, 2-44
ER$NOD, 2-44
IER switch, 2-47
EXTENSION field, 2-47
file loading restriction, 3-2
files on the distribution kit, 3-3
file specification, 2-48, 3-2
local access, 3-1
nonamed directory mode restriction, 3-2
ODLfile, 3-8
rebuilding utilities, 3-10
small systems, 3-3
startup command procedure, 3-7
task building against resident library, 2-45
Utilities, 3-2
RMSBCK utility
command line format, 2-48
INV switch, 2-48
RMSCNV utility
IER switch, 2-48
file loading restriction, 3-2
RMSDAP
keysize field, 3-2
RMSDEF utility
DEF command restriction, 3-3
DEFINE command restriction, 3-3

RMSDES utility
continuation buckets, 3-3
EXIT_SUPERSEDE command, 2-46,3-3
SAVE_SUPERSEDE command, 2-46, 3-3
RMSDSP utility
IBR switch, 2-48
ISU switch, 2-48
RMSRST utility
command line format, 2-49
fNV switch, 2-49
RSX-ll 2780/3780 correction file, 4-2
RSX-llM-PLUS DIBOL correction file, 4-2
RUN command
/PRI keyword, 2-13
lUIe keyword, 2-12, 2-13

s
SAVE_SUPERSEDE command
RMSDES utility, 2-46, 3-3
SET ICRASHDEV command, 2-14, 2-15
SET ICRASH_DEVICE command, 2-15
SET IDPRO command, 2-14
SET /INQUIRE command, 1-11, 2-14
SETITERM command, 2-15
SET Command
/INTRUSION qualifier, 1-9
Set mode
half duplex (communication driver), 2-29
IO.HDX function (communication driver),
2-29
SET SYSTEM ICRASH_DEVICE command, 2-5
SET TERMINALIINQUIRE command, 2-1
SET TERMINAL/model command, 2-1
Shared regions
1- and D-space restriction, 2-20
SHOW ASSIGNMENTS command, 2-3
SHOW LOGICALS command, 2-3
SHOW QUEUE command
IALL qualifier, 2-1
SHUTUP utility, C-8
SLP
correcting source files, C-l
SORT command
/INDEXED_SEQUENTIAL qualifier, 1-11
Source file
correcting, C-l
updating, C-l
Source Language Input Program
See SLP
START Command
/INTRUSION qualifier, 1-9

Sync character (communication driver), 2-30
specifying, 2-30
System library
using with TKB, 1-14

T
Task
extending, 2-20
installing, 1-10
Task Builder
SeeTKB
Terminal driver, 1-10
Time delay
DSA driver, 2-33
TKB
allocating library APRs, 2-39
Bitmask option, 1-11
building new FCP, C-6
CLSTR option, 2-39
command file, C-3
compatible library requirement, 1-14, C-12
EXTK$ task directives, 1-2
EXTM$ task directives, 1-2
FMS command file example, 2-37
Illegal APR reservation message, 2-42
LIBR option, 2-39
ILl subswitch, 2-39
RESLIB option, 2-39
RESSUP option, 2-38
RMS-ll resident library, 2-45
SUPLIB option, 2-38
task-build command file, C-6
using system library, 1-14
using undefined symbols, 1-11
VSECT option, 2-39
TLOG$ macro expansion, 2-19
TLON$ macro expansion, 2-19
Transmission
validation (communication driver), 2-30

u
UPDATE operation
RMS-ll correction, 2-44

v
!VERIFY qualifier
BRU, 1-5
VFY
warning message, 1-7
Virtual Monitor Console Routine

Index-5

Virtual Monitor Console Routine (Cont.)
SeeVMR
VMR
INSTALL command, C-6
installing with deferred binding, 1-11
LOAD command, 1-12
REMOVE command, C-6
UNLOAD command, 1-12
updating a loadable driver, 0-2
VMS ACP function support

Index-6

FCS, 1-13

w
Write operation
no sync (communication driver), 2-30

x
XDT
loadable support, 2-16

HOW TO ORDER ADDITIONAL DOCUMENTATION
From

Call

Alaska, Hawaii,
603-884-6660
or New Hampshire
Rest of U.S.A.
and Puerto Rico!

Write

Digital Equipment Corporation
P.O. Box CS2008
Nashua NH 03061

800-DIGITAL

1 Prepaid orders from Puerto Rico, call Digital's local subsidiary (809-754-7575)

Canada

800-267-6219
(for software
documentation)

Digital Equipment of Canada Ltd.
100 Herzberg Road
Kanata, Ontario, Canada K2K 2A6
Attn: Direct Order Desk

613-592-5111
(for hardware
documentation)

Internal orders
(for software
documentation)
Internal orders
(for hardware
documentation)

DTN: 234-4323
508-351-4323

Software Supply Business (SSB)
Digital Equipment Corporation
Westminster MA 01473
Publishing & Circulation Services (P&CS)
NR03-11W3
Digital Equipment Corporation
Northboro MA 01532

Reader's Comments

RSX-11 M-PLUS
Release Notes

AA-H427H-TC

Your comments and suggestions will help us improve the quality of our future documentation. Please note
that this form is for comments on documentation only.

I rate this manual's:
Accuracy (product works as described)
Completeness (enough information)
Clarity (easy to understand)
Organization (structure of subject matter)
Figures (useful)
Examples (useful)
Index (ability to find topic)
Page layout (easy to find information)

Excellent

o
o

Good

Fair

o
o
o
o
o

o
o

o
o
o
o

o
o

What I like least about this manual:

I found the following errors in this manual:
Description

My additional comments or suggestions for improving this manual:

Please indicate the type of user/reader that you most nearly represent:

o Administrative Support
o Computer Operator

o Educator!I'rainer

o Programmer/Analyst

o Sales

Name/TItle

o ScientistiEngineer
o Software Support
o System Manager

o Other (please specify)

Dept.
Date

Company
Mailing Address
Phone
10/87

o
o
o
o
o
o

What I like best about this manual:

Page

Poor

Do Not Tear - Fold Here and Tape

-I

NO POSTAGE
NECESSARY
IF MAILED
IN THE
UNITED STATES

I
I

BUSINESS REPLY MAIL
FIRST CLASS PERMIT NO.33 MAYNARD MASS.
POSTAGE WILL BE PAID BY ADDRESSEE

DIGITAL EQUIPMENT CORPORATION
CORPORATE USER PUBLICATIONS
PK03-1/30D
129 PARKER STREET
MAYNARD, MA 01754-2198

11111'11.11111.1.1.1 •• 11111.1111.111111.1.1.1. 1111.1
Do Not Tear -

Fold Here

-I

I
I

Printed in U.S.A.