Digital PDFs

AA-H427J-TC

January 2000

148 pages

Original

7.0MB

Document:	RSX-11M-PLUS Release Notes Version 4.4
Order Number:	AA-H427J-TC
Revision:	0
Pages:	148
Original Filename:	http://bitsavers.org/pdf/dec/pdp11/rsx11m_plus/RSX11Mplus_V4.x/AA-H427J-TC_RSX-11M-PLUS_V4.4_Release_Notes_Jan93.pdf

OCR Text

RSX-11 M-PLUS

Release Notes
AA-H427 J-Te
February 1993

This revised manual contains technical changes made to the
RSX-11M-PLUS operating system since the previous version,
corrections to the documentation since Version 4.1, a summary of layered
products supported on the operating system, and additional information
on RMS-11.

Revision/Update Information:

This revised manual supersedes the
RSX-llM-PLUS Version 4.3 Release
Notes AA-H427H-TC

Operating System:

RSX-11M-PLUS Version 4.4

Digital Equipment Corporation
Maynard, Massachusetts

1993

The information 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.
The software described in this document is furnished under a license and may be used or copied
only in accordance with the terms of such license.
No responsibility is assumed for the use or reliability of software on 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 (c)(l)(ii) 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, 1991, 1992, and

1993.
All Rights Reserved.
Printed in U.S.A.
The postpaid Reader's Comments form 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, DECnet,
DEC/CMS, DECIMMS, DECUS, DEC writer, DIBOL, LA50, LA75, LN01, LN03, LSI, MicroPDP-ll,
Micro'RSX, MicroVMS, PDP, Professional, Q-bus, RA70, RA82 , RA90, RD31, RD32, RD50, RD54,
ReGIS, RK07, RL01, RL02, RSTS, RSX, RSX-ll, RSX-llM/S, RQZX1, RZ23L, RZ24L, TK50, TU81,
TZ30, UNIBUS, VAX, VAXcluster, VAX DOCUMENT, VMS, VT, and the DIGITAL logo.

This document was prepared using VAX DOCUMENT, Version 2.1.

Contents
Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 New Features, Restrictions, and Supplementary Information
Summary of New Features ................................... .
1.1
New Hardware Support in Version 4.4 ........................... .
1.2
1.2.1
Support for the KDJ11-E Processor .......................... .
RA71, RA72, and RA92 Disk Support ........................ .
1.2.2
Supplementary Hardware Information ....................... .
1.2.3
Software Enhancements for Version 4.4 ......................... .
1.3
Support for QMG after 1992 ............................... .
1.3.1
Queue Manager Changes Affecting DECnet-11M-PLUS and
1.3.1.1
Micro'RSX DECnet .................................... .
New Version of Task Builder ............................... .
1.3.2
Support for Year 2000 in Date Format ........................ .
1.3.3
Changes to the File Exchange Utility (FLX) for Year 2000 and
1.3.3.1
Beyond ............................................. .
1.3.4
Additional Documentation for the GIN$ Subfunctions ............ .
GI.DVJ - Son of Get Device Information ................... .
1.3.4.1
GI.VEC - Translate Executive Entry Point Vector ............ .
1.3.4.2
High-Level Language for the GIN Directive .................... .
1.3.5
1.3.6
Enhancements to the PRINT and SUBMIT Commands ........... .
ASSIGN/QUEUE (QUE /AS: in MCR) Command ............. .
1.3.6.1
PRINT (PRI in MCR) Command ......................... .
1.3.6.2
DEASSIGN/QUEUE (QUE IDEA in MCR) Command ......... .
1.3.6.3
New Directives .......................................... .
1.3.7
Support for TDX Line Continuation .......................... .
1.3.8
Support for Universal Receiver Task ......................... .
1.3.9
Enhancement
to Loadable XDT ............................. .
1.3.10
Software Problems Corrected in Version 4.4 ...................... .
1.4
Data Caching ........................................... .
1.4.1
Data Caching May Have Deadlocked the System ............. .
1.4.1.1
Problems with Write-check or Read-check .................. .
1.4.1.2
File Corruption ...................................... .
1.4.1.3
1.4.1.4
Enhancements to the SET /CACHE Display Command ........ .
1.4.1.5
Correction to CDA .................................... .
1.4.2
Corrections to SYSGEN ................................... .
SYSGEN Entered a Loop if More than 4 Units Existed on a
1.4.2.1
DU-Type Controller ................................... .
Problems with Con Display ............................. .
1.4.2.2
1.4.2.3
Invalid Question on Processor-Type after an Autoconfigure ..... .
1.4.2.4
SYSGEN Terminated Abnormally if a User Chose to Override
ACF Results ......................................... .

1-1
1-1
1-2
1-2
1-2
1-2
1-2
1-3
1-3
1-5

1-7
1-7
1-7
1-8
1-10
1-10
1-11
1-11
1-12
1-12
1-13
1-13
1-13
1-13
1-14
1-14
1-14
1-14
1-14
1-14
1-14

.- .

..,.A

1-15
1-15
1-16

iii

1.4.3
1.4.3.1
1.4.3.2
1.4.4
1.4.4.1
1.4.4.2
1.4.4.3
1.4.5
1.4.5.1
1.4.5.2
1.4.5.3
1.4.5.4
1.4.5.5
1.4.5.6
1.4.5.7
1.4.5.8
1.4.6
1.4.7
1.4.7.1
1.4.7.2
1.4.7.3
1.4.7.4
1.4.7.5
1.4.8
1.4.9
1.4.10
1.4.11
1.4.12
1.4.13
1.4.14
1.4.15
1.4.16
1.4.16.1
1.4.16.2
1.4.16.3
1.4.17
1.4.18
1.4.19
1.4.20
1.4.21
1.4.22
1.4.23
1.4.24
1.4.25
1.4.26
1.4.27
1.4.28

Corrections to ACF ....................................... .
ACF Did Not Correctly Determine the Processor Type and
Memory Size on a PDP-11193 Processor ................... .
ACF Did Not Identify a PDP-11/93 Correctly During an
Installation of MicrdRSX Using the RX50 Kit ............... .
Corrections to Shadow Recording ............................ .
SHA Terminated with a Memory Protect Violation ........... .
SHAERR Reported Invalid LBNs on RA-Type Devices ......... .
User Routine Enhancements for SHAERR .................. .
Backup and Restore Utility (BRU) Corrections ................. .
Error with /EXCLUDE Using Wildcards ................... .
BRU Dismount Problem ................................ .
Invalid Date or Time Error with /REVISED ................ .
Problems with Creating the Master File Directory ........... .
BRU Could Not Abort When Used With a TK50 ............. .
Help File for BRUSYS Updated .......................... .
BRU Error Message Contains Extraneous Text .............. .
BRU Aborted Following a Tape Error ..................... .
Correction to SHUTUP ................................... .
Corrections to Peripheral Interchange Program (PIP) ............ .
Problem with Checking the Directory Specification ........... .
SpecifYing Files Using a Wildcard . . . . . . . . . . . . . . . . . . . . . . . . .
Problem with Using Wildcards and Creating a Directory ....... .
Problem with Issuing the PIPILI Command ................. .
PIP Did Not Give Correct Error for Command [34,477]ILI ...... .
Problems with Submitting a Batch Job with the INa PRINT
Switch ................................................ .
Using a CLI Build with the /CL Switch Correction .............. .
IO.EIO QIO Request ..................................... .
SHOW LOGICALS ....................................... .
KWII-P Clock .......................................... .
FAL .................................................. .
Help Files ............................................. .
Support for RA82 Errors .................................. .
Corrections to Error Logging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ERRLOG Opened Wrong File for Append and/or for Random
Access ............................................. .
Error Logging of Volume Not Mounted Packets for DU Disks ... .
Error Logging Support for PDP-11184 ..................... .
Using the .LIMIT Feature with Tasks ........................ .
Parity Problems with Terminal Multiplexers ................... .
Printing to a Printer Connected to a LAT Port ................. .
UFD Created Directory Files with Embedded Blanks ............ .
Incompatibility Between VMS DIFFERENCES Command and RSX
DCL DIFFERENCES Command ............................ .
Enhancement to the DCL SORT Command .................... .
Correction to FLX ....................................... .
Task Builder (TKB) ...................................... .
Undefined Symbol Error in RPT ............................ .
Changing Crash Device from BOO Request Did Not Pass Information
to Display .............................................. .
DSA Timeout Processing Led to Hung Tasks ................... .
Support for Inputting Lower-Case Characters in the TUNE.CMD and
TUNERL02.Cl\lD Files ................................... .

1-16
1-16
1-17
1-17
1-17
1-17
1-17
1-18
1-18
1-18
1-18
1-18
1-18
1-18
1-19
1-19
1-19
1-19
1-19
1-19
1-19
1-19
1-19
1-20
1-20
1-20
1-20
1-20
1-20
1-20
1-21
1-21
1-21
1-21
1-21
1-21
1-21
1-22
1-22
1-22
1-22
1-22
1-23
1-23
1-23
1-23
1-23

Submitting a File That Did Not Exist Resulted in an Error ....... .
Support for PDP-11194 in the SAVSIZ List of 2044KW CPUs ...... .
Support for the Angle Bracket « » Characters for Directory
Specifications ........................................... .
1.4.32
Support for J-ll UNIBUS Systems in the Crash Driver .......... .
Enhanced Support for RX02 Devices ......................... .
1.4.33
SET TERMINALIINQUIRE Command Did Not Work on a VT420
1.4.34
Terminal .............................................. .
Problems with ODT in an I&D System ....................... .
1.4.35
Intermittent Problem When Getting Data Late Errors on RK06 or
1.4.36
RK07 Drives ............................................ .
Software Problems and Restrictions in Version 4.4 ................. .
1.5
Restrictions in SYSGEN ............
1.5.1
Selecting FCPMDL in SYSGEN .......................... .
1.5.1.1
1.5.1.2
Incorrect Displays for Multiple MM Devices in SYSGEN ....... .
Specifying a Nonexisting Device as the Crash Device During
1.5.1.3
SySGEN ........................................... .
Restrictions on Year 2000 Support ........................... .
1.5.2
1.5.2.1
Special Error Log Display Commands Do Not Use the 4-Digit Year
Field ............................................... .
Error Logging Restrictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.5.3
Support for Error Logging History Summaries .............. .
1.5.3.1
Incorrect Displays of Register Fields for DU and MU Devices . . . .
1.5.3.2
BRU Restrictions ........................................ .
1.5.4
1.5.4.1
Use of Wildcards in BRU ............................... .
Problem with the /NEW_VERSION Qualifier ............... .
1.5.4.2
BRU Does Not Give Correct Error Messages When Backing Up a
1.5.4.3
Named Directory That Does Not Exist ..................... .
Restriction on Incremental Backups ...................... .
1.5.4.4
1.5.5
DCL Command ABORT Taskname/TAS Does Not Translate
Correctly .............................................. .
Tasks Hang When Attaching to Another Terminal Already Attached
1.5.6
to a Task .............................................. .
ELIISH Reports No Errors if ELI/NOLIM Has Been Set .......... .
1.5.7
Problem with Unused Terminal Ports ........................ .
1.5.8
1.5.9
Task Builder (TKB) Restriction ............................. .
1.5.10
RMD Restriction ........................................ .
1.5.11
PIP Restriction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.5.12
Indirect Preprocessor Restriction . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.5.13
Queue Manager Restriction ................................ .
1.6
Supplementary Software Information ........................... .
1.6.1
Disk Save and Compress Utility ............................ .
1.6.2
File Control Services Changes Supplementary Information ........ .
1.6.3
TKB Restrictions and Supplementary Information .............. .
1.6.3.1
OTS Fast Map Restriction .............................. .

1.4.29
1.4.30
1.4.31

••

•••••••••••••••••

1-23
1-24
1-24
1-24
1-24
1-24
1-24
1-25
1-25
1-25
1-25
1-26
1-26
1-26
1-26
1-26
1-26
1-26
1-26
1-26
1-26
1-27
1-27
1-27
1-28
1-28
1-28
1-28
1-28
1-29
1-29
1-29
1-29
1-29
1-29
1-30
1-30

2 Corrections to Documentation
2.1
2.2
2.3
2.4
2.5

naten ana "tueue lV.Lanual . . . . . . . . . . . . . . . . . . . . . . . .

2-1

RSX-IIM-PLUS Command Language Manual .................... .
RSX-IIM-PLUS Indirect Command Processor Manual .............. .
RSX-IIM-PLUS MCR Operations Manual ....................... .
RSX-IIM-PLUS System Generation Guide and Installation Guide .... .

2-1
2-5

~0""'T

........ 1\6

T"\TTTn~

1\.;:;A-l11Vl-rLU~

11"'\

l\/f

2-6
2-15

2.6
2.7
2.8
2.9
2.9.1
2.9.2
2.9.3
2.9.4
2.9.5
2.9.6
2.9.7
2.9.8
2.9.9
2.9.10
2.9.11
2.9.12
2.9.13
2.9.14
2.9.15
2.9.16
2.10
2.11
2.12
2.13
2.14
2.15
2.16
2.17
2.18
2.19

RSX-IIM-PLUS and MicrdRSX Crash Dump Analyzer Reference
Manual ................................................... .
RSX-IIM-PLUS and MicrdRSX Debugging Reference Manual ........ .
RSX-I1M-PLUS and MicrdRSX Error Logging Manual ............. .
RSX-IIM-PLUS and MicrdRSX Executive Reference Manual ........ .
The GINABO Call ....................................... .
The GINAPR Call ....................................... .
The GINDEF Call ....................................... .
The GINDEV Call ....................................... .
The GINDVJ Call ....................................... .
The GINFMK Call ....................................... .
The GINGAS Call ....................................... .
The GINQMC Call ....................................... .
The GINREN Call ....................................... .
The GINSPR Call ........................................ .
The GINTSK Call ....................................... .
The GINUAB Call ....................................... .
The GINUIC Call ........................................ .
The GINUPD Call ....................................... .
The GINVEC Call ....................................... .
Other Corrections to the RSX-IIM-PLUS and MicrdRSX Executive
Reference Manual ....................................... .
RSX-IIM-PLUS and MicrdRSX Guide to Writing an I/O Driver ...... .
RSX-IIM-PLUS and MicrdRSX 1/0 Drivers Reference Manual ....... .
RSX-IIM-PLUS and MicrdRSX System Library Routines Reference
Manual ................................................... .
RSX-IIM-PLUS and MicrdRSX System Management Guide ......... .
RSX-IIM-PLUS and MicrdRSX Task Builder Manual .............. .
RSX-IIM-PLUS and MicrdRSX Utilities Manual .................. .
RSX-IIM1M-PLUS and Micro'RSX I/O Operations Reference Manual .. .
RSX-IIM1M-PLUS RMS-l1 Macro Programmer's Guide ............ .
RSX-IIM1M-PLUS RMS-ll User's Guide ........................ .
RSX-IIM1M-PLUS RMS-ll Utilities ........................... .

2--15
2-16
2-16
2-17
2-18
2-18
2-18
2-18
2-19
2-19
2-20
2-20
2-20
2-21
2-21
2-21
2-22
2-22
2-23
2-23
2-29
2-31
2-39
2-40
2-43
2-48
2-53
2-56
2-57
2-58

3 RMS-11 Version 2.0 Software Modifications, Restrictions, and
Supplementary Information
3.1
RMS-ll Corrections ......................................... .
3.1.1
RMS-ll Local Access ..................................... .
3.1.2
RMS-ll Ease of Use ..................................... .
3.1.3
RMS-ll Utilities Enhancements ............................ .
3.2
Software Restrictions ........................................ .
3.2.1
RMS-l1 Access Methods .................................. .
Restrictions to RMS-l1 Remote Access Methods (RMSDAP) ....... .
3.2.2
3.2.2.1
Incorrect Interpretation of Keysize Field Values ............. .
Using RMSDAP on Your System ......................... .
3.2.2.2
RMS-l1 Utilities ........................................ .
3.2.3
RMS-ll Supplementary Information ............................ .
3.3
3.3.1
Enhancements to RMSDES Utility .......................... .

3-1
3-1
3-1
3-2
3-2
3-2
3-2
3-2
3-2
3-3
3-4
3-4

3.3.2
3.3.2.1
3.3.2.2
3.3.2.3
3.3.2.4
3.3.2.5

RMS-11 Installation ..................................... .
Location of RMS-11 Files on the Distribution Kit ............ .
Startup Command Procedures ........................... .
Utility Configurations ................................. .
Utility Command and ODL Files ......................... .
Rebuilding the Utilities ................................ .

3-4
3-4
3-7
3-8

3-8
3-9

4 Layered Product Support
4.1
4.2
4.3

Correction Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Installing Layered Products Restriction .......................... .
Invalid BASIC-PLUS-2 Installation Warning ..................... .

4-1
4-1
4-1

A Reporting Problems
B Applying Corrections to Source Files
B.1

B.2
B.3
B.4
B.5
B.6
B.7

B.8
B.9

Updating an Executive Source Module ........................... .
Updating an MCR Source Module .............................. .
Updating a DCL Source Module ............................... .
Applying Corrections to the File Systems (F11ACP) ................ .
Applying Corrections to the Reconfiguration Tasks CON and HRC ..... .
Applying FCS Corrections .................................... .
Updating the FCS Sources .................................... .
Assembling FCS ............................................ .
Replacing the FCS Object Modules ............................. .

B-1

B-3
B-4
B-5
B-6
B-8
B-9
B-9
B-11

Index
Tables
1-1

1-2
1-3
1-4
1-5
2-1
2-2
2-3
2-4
2-5
2-6
3-1
4-1

Storage of Dates in RSX Systems after 1990 ................... .
GLDVJ - Son of Get Device Information Subfunction Buffer
Format ................................................ .
GLDVJ - Son of Get Device Information Subfunction DSW Return
Codes and Their Meaning ................................. .
GI.VEC - Translate Executive Entry Point Vector Subfunction Buffer
Format ................................................ .
GI.VEC - Translate Executive Entry Point Vector Subfunction DSW
Return Codes and Their Meaning ........................... .
Number of Index File Headers .............................. .
Default Number of Files - Approximation ..................... .
Number of Index File Headers .............................. .
Default Number of Files-Approximation ..................... .
SET Command Keyword Summary .......................... .
Summary of BRU Command Qualifiers ....................... .
RMS-11 Files on RSX-11M-PLUS Distribution Kit ............. .
Layered Product Correction File References ................... .

1-5
1-8
1-8

1-9
1-10
2-2
2-2
2-10
2-11
2-13
2-49
3-5
4-1

vii

Preface
Manual Objectives
The RSX-llM-PLUS Release Notes contains important information for using
the RSX-IIM-PLUS Version 4.4 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-ll Version 2.0

•

Correction files and restrictions for layered products

Intended Audience
This manual is intended for all users of the RSX-IIM-PLUS Version 4.4
operating system.

Document Structure
Chapter 1 describes software features, software and hardware restrictions,
and supplementary information for the RSX-IIM-PLUS Version 4.4 operating
system.
Chapter 2 corrects errors and omissions in the RSX-IIM-PLUS documentation
set.
Chapter 3 describes software modifications, restrictions, and supplementary
information for RMS-ll Version 2.0.
Chapter 4 contains correction files and restrictions for layered products.
Appendix A describes the procedure for submitting a Software Performance
Report (SPR).
Appendix B 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-llM-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-llM-PLUS and MicrdRSX systems. Many
systems include DCL.

xxx>

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
Is witch

A command element preceded by a slash (I) is an
MCR keyword; a DCL qualifier; or a task, utility, or
program switch.

parameter

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

[option]

Brackets indicate optional entries in a command
line or a file specification. If the brackets include
syntactical elements, such as periods (.) or slashes
(I), those elements are required for the field. If the
field appears in lowercase, you are to substitute
a valid command element if you include the field.
Note that when an option is entered, the brackets
are not included in the command line.

Keywords, qualifiers, and switches alter the action
of the command they follow.

Convention

Meaning

[, ... ]

Brackets around a comma and horizontal ellipsis
points indicate that you can use a series of optional
elements separated by commas. For example,
(argument[, ... ]) means that you can specify
a series of optional arguments by enclosing the
arguments in parentheses and by separating them
with commas.

{ }

Braces indicate a choice among required options.
You must choose one of the options listed.

:argument

Some parameters and qualifiers can be altered by
the inclusion of arguments preceded by a colon. An
argument can be either numerical (COPIES:3) or
alphabetical (NAME:QIX). In DCL, the equal sign
( =) can be substituted for the colon to introduce
arguments. COPIES=3 and COPIES:3 are the
same.

( )

Parentheses are used to enclose more than one
argument in a command line. For example:

SET PROT = (S:RWED,O:RWED)
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 within the
command line. Although you might omit elements
that come before the desired element, the commas
that separate them must still be included.
Periods immediately following a number indicate
a decimal number. For example, 600. would mean
600 10 .
[g,m]
[directory]

The convention [g,m] signifies a V ser Identification
Code (VIC). The g is a group number and the m is
a member number. The VIC identifies a user and
is used mainly for controlling access to files and
privileged system functions.
This may also signify a V ser File Directory (VFD),
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 VIC.
Where a VIC, VFD, or directory is required, only
one set of brackets is shown (for example, [g,m]).
Where the DIC, DFD, or directory is optional, two
sets of brackets are shown (for example, [[g,m]]).
Vertical ellipsis points show where elements of
command input or statements in an example
or figure have been omitted because they are
irrele\Tant to the point being discussed.

Convention

Meaning

filespec

A full file specification includes device, directory,
file name, file type, and version number, as shown
in the following example:

DL2: [46, 63] INDIRECT. TXT; 3
Full file specifications are rarely needed. If you
do not provide a version number, the operating
system uses the highest numbered version. If you
do not provide a directory, the system uses the
default directory. Some system functions default
to particular file types. Many commands accept a
wildcard character ( * ) in place of the file name, file
type, or version number. Some commands accept a
file specification with a DECnet node name.
A period in a file specification separates the file
name and file type. When the file type is not
specified, the period may be omitted from the file
s pecification.
A semicolon in a file specification separates the
file type from the file version. If the version is not
specified, the semicolon may be omitted from the
file specification.
@

The at sign invokes an indirect command file. The
at sign immediately precedes the file specification
for the indirect command file, as follows:

@filenarne[.filetype;version]
Horizontal ellipsis points indicate the following:
•

Additional, optional arguments in a statement
have been omitted.

•

The preceding item or items can be repeated
one or more times.

•

Additional parameters, values, or other
information can be entered.

KEYNAME

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

print and enter

As these words are used in the text, the system
prints and the user enters.
The symbol ICtrl/x I means that you are to press the
ley jarked Ctrl while pressing another key. Thus,
Ctrl/Z indicates that you are to press the Ctrl key
and the Z key together in this fashion. CtrllZ is
echoed on some terminals as I\Z. However, not all
control characters echo.

xii

1
New Features, Restrictions, and
Supplementary Information
This chapter lists the new software features, describes new hardware support,
describes the new software features, lists software problems corrected in this
release, lists software and hardware restrictions, and provides supplementary
information for the RSX-11M-PLUS Version 4.4 operating system.

1.1 Summary of New Features
The RSX-11M-PLUS Version 4.4 operating system supports the following new
features:

•

An additional version of the Task Builder (TKB), known as Hybrid Fast Task
Builder (HFT). HFT uses a fast-mapped VSECT region for virtual memory
table storage. This version of TKB will only be available on I&D systems.

•

Support for the year 2000 and beyond.

•

Support for the RA71 and RA92 devices.

•

Support for the RQZX1 SCSI adaptor.

•

KDJ11-E Processor (used in Micro PDP-11193 and PDP-11/94).

•

Additional documentation for the GI.DVJ and GI.VEC subfunctions of the
G IN$ system directive.

•

New high-level language calls added to the GIN directives.

•

Support for QMG after 1992.

•

Enhancements to the PRINT and SUBMIT commands.

•

Two new Send Data and UN stop directives.

•

Support for TDX line continuation.

•

Support for Universal Receiver Task (URT).

•

An enhancement to loadable XDT.

1.2 New Hardware Support in Version 4.4
The following new devices are supported:
•

RA71

•

RA72

•

RA92

•

RQZX1 SCSI adaptor:
-

RZ23L

New Features, Restrictions, and Supplementary Information

1-1

RZ24L
TZ30
Caution

ANSI tape processing for information interchange is not supported on the
TZ30 cartridge tape drive.

•

KDJ11-E Processor (used in MicroPDP-11/93 and PDP-11194)

1.2.1 Support for the KDJ11-E Processor
Support for the KDJ11-E processor module has been added to Version 4.4. The
following features are supported:
•

The module ID

•

The TOY clock (Time of Year clock)

•

On-board serial lines (as terminal driver option)

Support has been added in SYSGEN, ACF, Error Logging, SAVE, the Executive,
and the MCR SET TIME command for this processor.

1.2.2 RA 71, RA 72, and RA92 Disk Support
Error logging support has been added for the RA71, RA72, and RA92 disk devices.

1.2.3 Supplementary Hardware Information
For a complete list of hardware supported by Version 4.4, refer to the Software
Product Description contained in your media kit.

1.3 Software Enhancements for Version 4.4
This section describes software enhancements added to Version 4.4.

1.3.1 Support for QMG after 1992
Previous versions of the Queue Manager and spooling tasks (QMGCLI) provided
4 bits for storing the year specification in a SUBMIT/AFTER or PRINT/AFTER
job. Also, the data structures passed between the Queue Manager and QMGCLI
that were stored in queue file QUEUE.SYS provided 4 bits for storing year
specification in SUBMIT/AFTER or PRINT/AFTER jobs. The information has
been stored as an increment from 1977.
Now, the year specification is stored as an 8-bit value, as an increment from 1900.
Along with this change, the storage of month, day, hour, and minute have been
changed.
Prior to starting the new Queue Manager, the system manager should assure
that the old QUEUE.SYS file is empty, that is, the results of a SHOW/QUE
instruction shows all queues empty of held, pending, stopped, paused, etc. jobs.
A new QUEUE.SYS file will be created by the Queue Manager when it starts.
The required input format for the date will be consistent with the standard
format for years following the year 2000; for example:

1-2 New Features, Restrictions, and Supplementary Information

Input

Meaning

nn
19nn
20nn

19nn
19nn
20nn

-------

Further, inputs of only one digit or of three digits will be rejected as errors. RSX
assumes that IAFTER specifications on the submitted command line are 1977 or
after; for example:
/AFTER:21-JUN-93 refers to 1993
/AFTER:21-JUN-04 refers to 2004
/AFTER 21-JUN-76 refers to 2076

Although the spooler tasks included with this version have been modified in
accordance with the new formats, application programs may include spoolers that
have not been updated. The former syntax for packets from a spooler task to the
Queue Manager remains acceptable, although certain assumptions are made.
If a packet is received with function QM.OPJ (former Open Job function) or
QM.MDJ (former Modify Job function), the Queue Manager modifies the packet
to the new format before storing it in QUEUE.SYS. The former 4-bit Year
increment value is assumed to be a modulo-16 delta from 1977, specifying a date
subsequent to current system time. This is implemented by adding 16 to the
increment until the year is greater than or equal to that of the current system
time.
Any tasks that directly access the Job Entry structure in QUEUE.SYS will have
to be modified accordingly. Symbolic offset names are maintained so, unless the
task accesses the After Time or Job Time Stamp fields, they probably only require
you to assemble and link them again. Tasks that do make use of the Time fields
should be recoded for consistency with the new format.
1.3.1.1 Queue Manager Changes Affecting DECnet-11M-PLUS and Micro'RSX DECnet
Due to the changes to the Queue Manager, if your system utilizes the DECnet
File Transfer Server (FTS), this component will need to be rebuilt during your
Network Generation. If your system is not a pregenerated RL02 system or
Micro'RSX system, this can also be done by using a "Component mode NETGEN."
If you system is either a pregenerated RL02 system or Micro'RSX, you should
copy the following files to your system's <NETUIC>:

•

[DECNET]FTSRES. TSK

•

[DECNET]FTSFSL. TSK.

1.3.2 New Version of Task Builder
This version of RSX-IIM-PLUS provides an additional version of TKB, known
as Hybrid Fast Task Builder (HFT). HFT uses a fast-mapped VSECT region for
virtuallnenlory table storage. This version of TKB will only be available on I&D
systems.
Currently, TKB uses a disk workfile for virtual memory table storage of symbols.
For tasks with a large number of symbols, link time is greatly increased because
_C./..1.. _ ________ ./.. _C .3:_1_ T/fl ______ : .. _.3 C ___ _______
Ul Llle cUllUUllL Ul UH:i.I:\_ VV l-equlleu lUI ;:;WCll-'l-'l11 b

:_~

__ : .../.. ___ 1 __________ __ : _ _ _ .3 _ .. .IVll LUCll l11el11UlY l-'Clbe;:; 111 ClllU UUl~

of dynamic memory (the free space between the end of TKB's code and its highest
possible virtual address). Using a fast-mapped region will improve performance
by eliminating this disk I/O. Performance improvements will vary depending on
the task's configuration. As an example of possible improvements, test builds
of PDP-II C and PDP-II BP2 done on a standalone Micro'RSX system using
HFT build 35% to 43% faster than with TKB on the same system. Please note
New Features, Restrictions, and Supplementary Information 1-3

however, that for smaller tasks that do minimal amounts of paging, there will be
little or no benefit.
HFT has been optimized to provide faster linking by using a fast-mapped VSECT
region in place of the disk workfile used by TKB, TKBRES, and TKBFSL. Task
images produced by HFT are identical to ones produced by TKB, TKBRES, and
TKBFSL.
HFT can be installed with a default task name of ( ... HFT). The user may also
find it advantageous to install HFT as ... TKB instead of the standard TKB task
image; however, the task image is considerably larger. You can install one, the
other, or both. Digital recommends that you not install both tasks on the system,
as the HFT requires a great deal of memory.
A task image for HFT built with the standard TKB defaults can be found
in LB:[3,54]. If nonstandard defaults are desired, the build command file
(HFTRESBLD.CMD) and .ODL file (HFTRESBLD.ODL) can be found in
LB:[1,24], and HFT can be rebuilt using the section of SYSGEN which permits
rebuilding of supplied system tasks (Building Nonprivileged Tasks). However,
please note that during this section of sysgen, HFT will not automatically
be rebuilt if you specify ALL, ALLRES, or ALLFSL due to the way HFT is
configured. You must specify HFTRES when sysgen prompts for the task to be
rebuilt in sysgen question BN020 in order to rebuild HFT.
All references to and use of the disk workfile have been removed in HFT.
Initialization has been changed to immediately extend dynamic memory to
D-APR 7 (instead of being extended 3100 bytes at a time on demand as TKB
executes). The new HFT has been built with I&D, with D-APRs 0-2 mapped
to D-space used by the task code and D-APRs 3-6 mapped to dynamic memory.
D-APR 7 is used to fast map to the VSECT region. (Note that TKB, TKBRES,
and TKBFSL also provide only 4 APRs worth of dynamic memory, so there is no
reduction in dynamic memory with this implementation.)
The fast-mapped VSECT region will be the same size as the maximum disk
workfile (64KW) and pages will be located in the VSECT region in a fashion
similar to that used with the disk workfile. Virtual memory pages will be
allocated sequentially, with each virtual memory page 256 words long. A page
will be located by means of its page block number (the high byte of its address).
When a page is read in or out of virtual memory, its page block number will be
used to determine the following:
•

Which APR in the VSECT region needs to be fast mapped

•

The correct offset into the APR to locate the page

After fast mapping the appropriate APR, a subroutine copies the 256 words of
the page into (or out of) the VSECT region to (or from) the designated location in
virtual memory.
As a further optimization, FCSFSL is used as a resident library to avoid the
overhead from context switching.
Modified versions of the appropriate TKB modules to support the new HFT have
been added to [31,10] and carry an X suffix (for example, there will now be an
INIDMX used solely by the new HFT, in addition to the original INIDM used by
TKB). The TKB assembly command file(s) have been updated.

1-4 New Features, Restrictions, and Supplementary Information

1.3.3 Support for Year 2000 in Date Format
The RSX operating systems have been changed to support dates beyond the year
2000. Generally, RSX treats the current year as a two-digit field. The following
areas of the RSX system have been changed:
•

Input-All areas of the system that allow input of a date will now accept two
different forms of date input. First, the traditional dates of the dd-mmm-yy
format (or mm/dd/yy for TIM) will still be accepted. These dates will always
be considered as occurring between the years 1900 and 1999. In addition, the
date may now be specified as dd-mmm-yyyy or (mm/dd/yyyy for TIM only), to
allow the system to recognize dates occurring after 1999.

•

Output-In order to preserve compatibility in applications that parse the
output of system commands, the format of date output will not be changed.
All places that output the date as 2 digits will continue to do so, representing
2000 as 00, and so forth. Places that currently output the date as 4 digits will
still output the date as 4 digits.

•

Internal representations-As documented for the GTIM$ directive, the year
field of time vectors is the number of years since 1900. This is still accurate.
This field will reach 100(10) in the year 2000.
For example, Table 1-1 shows how dates between years 1900 and 2209 will
be stored. Dates will be displayed as the lower two digits of the real year or
as a four-digit year.
However, ODS-I only provides a 7-byte ASCII field for creation and other
dates. In order to accommodate this restriction in a way that is consistent
with RSX date handling, the 2-bytes of year in the date attribute will be used
to represent the year since 1900.
The two bytes will be encoded as follows:
Low byte-Low order decimal digit of year, in ASCII.
High byte-Quotient of years since 1900 divided by 10, plus 60s . This
represents the ASCII high digit of the year from 1900 - 1999. In 2000,
however, this will result in ":" being stored for the high digit of the year.
For years 2010 through 2019, ";" will be stored, and so forth. This change
should result in another 200 years or more of file representations.

Table 1-1 Storage of Dates in RSX Systems after 1990
Calendar Year

Year Field Representation

1990 - 1991

90 - 99

2000 - 2009

:0 -:9

2010 - 2019

;0 -;9

2020 - 2029

<0 - <9

2030 - 2039

=0 - =9

2040 - 2049

>0 - >9

'){\h:{\

?O -?9

LJVVV -

'){\h:O

LlVUv

2060 - 2069

@0-@9
(continued on next page)

New Features, Restrictions, and Supplementary Information 1-5

Table 1-1 (Cont.) Storage of Dates in RSX Systems after 1990
Calendar Year

Year Field Representation

2070 - 2079

AO-A9

2080 - 2089

BO-B9

2090 - 2099

CO-C9

2100 - 2109

DO-D9

2110-2119

EO-E9

2120 - 2129

FO-F9

2130 - 2139

GO-G9

2140 - 2149

HO-H9

2150 - 2159

10 - 19

2160 - 2169

JO - J9

2170 - 2179

KO-K9

2180 - 2189

MO-M9

2190 - 2199

NO-N9

2200 - 2209

00-09

The following features have been changed to accept or display the new date
format:
•

BROADCAST utility

•

BRU utility

•

The following DCL commands:
SET TIME
DIRISINCE
BACKUP

•

FIIACP

•

MACRO

•

PIP utility

•

RPT and the /DATE qualifier

•

QMG

•

MCR command line interface

•

The following MCR commands:
BYE command
HEL command
TIM command

1-6 New Features, Restrictions, and Supplementary Information

1.3.3.1 Changes to the File Exchange Utility {FLX} for Year 2000 and Beyond

FLX has been modified to correctly record and display the dates of RT-11 files
created after the year 2003. A FLX directory listing of RT-11 files will now
display the century as part of the year.
The FLX utility correctly displays dates from 1970 to 2001 with the IDO switch
set. A C is displayed when the date is not in the valid range. Additionally, the
dates displayed are invalid. However, the files are intact and correct. FLX will
display the date as:
ACF.TSK

33.C

lS-NOV-98

The following example shows how FLX will display a file with invalid date
information if the year is greater than 2001 (The <233> is the protection code.):
ACF.TSK

33.C

19-70 <233>

1.3.4 Additional Documentation for the GIN$ Subfunctions
Please add the following descriptions of the GI.DVJ and the GI.VEC subfunctions
of the GIN$ system directive to the RSX-llM-PLUS and Micro/ RSX Executive
Reference Manual.
Note

This description supersedes the description of the GI.DVJ in the Version
4.3 Release Notes. The macro expansion shown in Section 2.8 was
incorrect in those release notes.

1.3.4.1 GI.DVJ - Son of Get Device Information

The Son of Get Device Information function returns information about the mount
status and accessibility of a particular mountable mass storage device. The device
on which information is returned is determined by first performing a logical
assignment (if required) and then following any redirects. Device assignments
are checked if the high bit in the flags byte is clear; otherwise, no check of device
assignments is made.
Macro Call
GIN$

GI.DVJ,buf,siz,dev,unt

Parameters
GI.DVJ

GIN$ function code (18.)
buf
Address of the buffer to receive the unit information.
siz

Buffer size in words.
dev

Device name. (If zero, use task's TI.)
unt
Device unit number. (If high bit clear, follow assignments.)

New Features, Restrictions, and Supplementary Information 1-7

Table 1-2 GI.DVJ - Son of Get Device Information Subfunction Buffer Format
Word 0

Device Mount Status

If word 0 equals - 1, then the device is either not mountable or
not a mass storage device.
If word 0 equals - 2, then the device is not accessible to the
issuing task.

If word 0 is greater than zero, then the individual bits have
the following meanings:
Bit 0 - The issuer is not allowed read access to the device.
Bit 1 - The issuer is not allowed write access to the device.
Bit 2 - The issuer is not allowed extend access to the
device.
Bit 3 - The issuer is not allowed delete access to the
device.
Bit 4 - The device is public.
Bit 5 - The device is private (allocated).
Bit 6 - The device is mounted foreign.
Bit 7 - The foreign mounted device has an ACP.

Macro Expansion
GIN$

GI. DVJ, DVBUF, DVSIZ, "DU, 0

. BYTE
. WORD
. WORD
. WORD
. WORD
. WORD

GI.DVJ
DVBUF
DVSIZ
"DU

169.,18 .

Table 1-3 GI.DVJ - Son of Get Device Information Subfunction DSW Return
Codes and Their Meaning
Return Code

Meaning

IS.SUC

Successful completion.

IE.ADP

Part of 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 subfunction code or the DPB size is invalid.

Note

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 (the
device unit number parameter).

1.3.4.2 GI. VEC - Translate Executive Entry Point Vector

The translate executive entry point vector subfunction may be used by privileged
tasks to provide independence from a particular system's symbol table file.
Macro Call

1-8 New Features, Restrictions, and Suppiementary Information

Parameters
GI.VEC

GIN$ function code (15.)
buf

Address of the buffer that contains the offset codes and will contain the returned
entry point addresses.
siz

Buffer size in words.
Table 1-4 GI. VEC - Translate Executive Entry Point Vector Subfunction Buffer
Format

Word 0

Flag word
If word 0 equals 0, the translation is performed.

If word 0 is greater than zero, then the directive completes
with status IS.SUC (success), but no translation is performed.
This prevents successive translates on the same buffer from
resulting in system faults. Word 0 is set to 1 when the
translation is performed.
Word 1 - n Offsets
Each word beyond word 0 in the buffer is an offset into a table.
Each of these offsets represents a symbol in the executive.
The file LB:[3,54]RSXVEC.STB contains these symbols. The
names of these symbols are identical to the represented entry
point. For example, RSXVEC.STB might define $TSKHD to
a value of 1236. The real value of $TSKHD for that system
is at offset 1236 in the executive's vectored entry point table.
For example, a value of 20646. The GI.VEC subfunction will
replace the 1236 in the buffer with the corresponding real
symbol value, 20646.
If the word in the buffer does not represent a valid offset, or
if it represents a symbol that is not included due to SYSGEN
options on the target system, the value 160001 is returned
for that value. No error is returned in this case. Applications
that employ vectoring should test for the existence of optional
features before using any entry points associated with those
features.
For more information on vectoring, see the description of using
vectored executive entry points in the RSX-llM-PLUS and
Micro/ RSX Guide to Writing an I/O Driver.

New Features, Restrictions, and Supplementary Information 1-9

Macro Expansion

GIN$
.BYTE
· WORD
. WORD
. WORD

GI.VEC, EXEBUF, EXES=Z
169.,18.
GI. VEe
EXEBUF
EXESIZ

Table 1-5 GI.VEC - Translate Executive Entry Point Vector Subfunction DSW
Return Codes and Their Meaning
Return Code

Meaning

Is.sue

Successful completion.

IE.ADP

Part of DPB is out of task's address space.

IE.SDP

Invalid subfunction code or the DPB size is invalid.

1.3.5 High-Level Language for the GIN Directive
The following new high-level language calls to the GIN directives may be utilized:

•
•
•
•

CALL GINABO

•
•
•
•
•
•

CALL GINDVJ

•
•

CALLGINUAB

CALL GINAPR
CALL GINDEF
CALL GINDEV

CALL GINFMK
CALL GINGAS
CALL GINQMC
CALL GINSPR
CALL GINTSK

CALL GINUIC

• CALLGINUPD
• CALL GINVEC
See Section 2.9 for more details on these calls.

1.3.6 Enhancements to the PRINT and SUBMIT Commands
The PRINT and SUBMIT commands have been enhanced to have a mnemonic
queue name translated to a processor name without requiring definition and
assignment for the mnemonic queue name.
Previously, the Queue Manager restricted the system to a total of 16 print queues.
By providing the capability to translate the PRINT command mnemonic queue
name to a processor name, you can avoid the need to define and assign the second
queue for each printer, thereby effectively doubling the capacity of the Queue
Manager. The device-specific queue must still be created. As a side benefit, you
also can define a logical name (for instance, in your Session Table) and avoid the
need to remember which IFORM=n to specify when printing to a terminal-like

1-10 New Features, Restrictions, and Supplementary Information

device. Previously, you might have issued a command similar to the following to
print 80 columns, letter quality on an LN03:
PRINT/FORM=4 resume. txt

N ow, you can define a queue similar to the following and then issue the next
command:
DEF QMG$QUE_NICE "TT20/FORM=4"
PRINT/QUE=NICE resume. txt

Rather than requiring the system manager define a mnemonic queue and assign
it to the processor-related queue, the ASSIGN/QUE command will now define
a logical name in the System Table to describe the relationship, if the queue
being assigned has not previously been created. Subsequent use of the PRINT
/QUE=name command attempts to translate the queue name to a processor name,
that is then substituted for the original queue name. The translation repeats
until a failure occurs, at which time the last translated name is submitted to the
Queue Manager.
1.3.6.1 ASSIGN/QUEUE (QUE /AS: in MCR) Command
DCl FORMAT: $ ASSIGN/QUEUE que_name proc_name[lqualifiers]
Description All DCL command qualifiers valid for the PRINT command may be

used.
MCR FORMAT: > QUE proc_name:[lqualifiers]/AS:que_name
Description All MCR Job Switches valid for the PRINT command may be used.
Note

If your system does not have logical name support installed, QMG cannot
use logical names.

If que_name has previously been defined to the Queue Manager, the command
exhibits prior behavior; for example, a path is established from the queue to
the processor. In this case, any qualifiers are ignored and a warning message
is issued to the user. Note that queue names PRINT and BATCH are always
defined in the default system.
If que_name has not been defined to the Queue Manager, a special logical name
is defined in the System Table to represent the translation from que_name to
proc_name, including qualifiers. This is equivalent to issuing the following DCL
command:
DEF QMG$QUE_que_name "proc_name[/qualifiers]"
1.3.6.2 PRINT (PRI in MCR) Command
DCl FORMAT: $ PRINT/QUEUE:que_name [lqualifiers] filespec[lfilequalifiers]
Description All DCL command qualifiers valid for the PRINT command may be
MCR FORMAT: > PRI que_name:[jobname][ljobswitch]=filespec[lfilequalifiers]

New Features, Restrictions, and Supplementary Information

1-11

Description All MCR Job Switches valid for the PRINT command may be used.
Note

Note, however, that any qualifiers that were included in the ASSIGN
IQUEUE (or QUE lAS:) command must not be repeated here. The
resulting multiple qualifiers after que_name translation will cause an
error return from the Queue Manager.

Execution of the PRINT command first iteratively translates que_name. At each
step, the original que_name is replaced by the translation, including qualifiers (if
any), and the new que_name is extracted and submitted for further translation.
When there are no more translations, the last constructed string is submitted to
the Queue Manager for execution.
1.3.6.3 DEASSIGN/QUEUE (QUE /DEA in MCR) Command
DCl FORMAT: $ DEASSIGN/QUEUE que_name proc_name
MCR FORMAT: > QUE proc_name:IDEA:que_name

If que_name has previously been defined to the Queue Manager, the command
exhibits prior behavior; for example, the path from the queue to the processor is
severed.
If que_name has not been defined to the Queue Manager, the special logical name
representing the que_name to proc_name translation is deleted from the System
Table. This is equivalent to issuing the following DCL command:
DEASS/SYSTEM QMG$QUE_que_name

The following commands are unchanged from their prior behavior:
INITIALIZE/QUEUE

(QUE feR)

DELETE/QUEUE

(QUE !DEL)

STOP/QUEUE

(QUE /STO)

START/QUEUE

(QUE /STA)

Note __________________________

Note, however, that they will affect only queues that exist in the
Queue Manager because they were created by the INITIALIZE/QUEUE
command (or by default in the case of PRINT and BATCH queues); they
will not affect mnemonic queues that are mapped to processors by means
of logical names.

1.3.7 New Directives
Two new directives have been added to the system. These are Send Data and
UN stop (SDUN$), and Variable Send and UN stop (VSUN$). Additionally, two
new high-level language interfaces (SDUN and VSUN) have been added. These
directives, and their functional descriptions, are identical to SDAT/SEND and
VSRC, with two exceptions:
•

The DIe for this directive is 179.

1-12 New Features, Restrictions, and Supplementary Information

•

The directives may return a status of IS.CLR(O) or IS.SET (+2). A status of
IS.SET indicates the send was successful but the target task was not active.
A status of IS.CLR indicates the send was successful and the target task was
active but not stopped.

1.3.8 Support for TOX Line Continuation
The TDX utility has been enhanced to accept continuation lines and to pass
command lines greater than 80 characters to spawned utilities. TDX will
now perform multiple GMCR directives if the first command line ends with a
continuation mark ("-"). All segments of the command are assembled before the
parsing routine is called. Additionally, TDX will now spawn external tasks with
command lines that exceed 80 characters. Commands that exceed 132 characters
and are processed by command files will now receive the "AT. tnn - Syntax Error"
error message instead of the "MCR - Line too long" message. Commands that
greatly exceed 132 characters from DCL terminals will receive the "MCR - Iform
buffer full" error message.

1.3.9 Support for Universal Receiver Task
A Universal Receiver Task has been added to RSX. The Universal Receiver Task
receives all SDAT type packets that received IE.INS in previous versions of
RSX-11M-PLUS. This task then routes the packets to the appropriate tasks
and/or systems.
If a task is installed with the name URT ... , it is considered to be a Universal
Receiver Task for all variants of the Send Data (SDAT) executive directive.
SDAT$, VSDA$, SDRC$, SDRP$, and SVSRC$ will not return the status IE.INS
(specified task not installed), but instead the message will be routed to URT ...
If the application already uses a task named URT ... and the user creates a new
task named URT ... , there may be a compatibility problem. If so, clearing the
TCE pointer with an OPEN command will point the application to use the earlier
task.

1.3.10 Enhancement to Loadable XOT
Loadable XDT may now be configured so that at each entry to XDT (by means
of SST, breakpoint, or single step), the next instruction to be executed is
automatically displayed. This is achieved by setting the 10th bit in the /FLAGS=n
qualifier when the LOA IEXP=XDT command is issued.
Additionally, it may be useful to be able to disable this feature at run time, as
well as to enable/disable some of the other XDT features that are controlled by
the /FLAGS= keyword. A new internal register, $X, may be opened to control the
/FLAGS= options. For example, the following sequence will enable the instruction
decode feature:
XDT>SX/
$X 000003 000013 IReturn I

1.4 Software Problems Corrected in Version 4.4
This section describes software problems corrected in Version 4.4.

New Features, Restrictions, and Supplementary Information 1-13

1.4.1 Data Caching
The following corrections have been made to data caching.
1.4.1.1 Data Caching May Have Deadlocked the System
Previously, data caching did not take any special consideration of any 1/0
requests issued by the Replacement Control Task (RCT . . . ) This caused
some I/O requests issued by RCT ... to have the potential for interacting with
data caching including the potential for system deadlock during a replacement
operation. The changes incorporated in data caching ensure that data caching
will not interact in any way with any 1/0 requests issued by RCT ... so system
deadlocks should not occur.

This problem has been fixed.
1.4.1.2 Problems with Write-check or Read-check
Previously, issuing a write with write-check (IO.WLC) QIO request and then
issuing a read (IO.RLC) QIO request sometimes resulted in a loss of data due to
interaction with data caching. This problem has been fixed.
1.4.1.3 File Corruption
Under extremely rare circumstances, a combination of data caching read-ahead
and write operations occurring at the same time may have caused file corruption,
due to certain interactions with the hardware. This problem has only been
seen with RMS indexed files, although it may have been possible with other file
structures. This problem has been fixed.
1.4.1.4 Enhancements to the SET ICACHE Display Command
The SET ICACHE display has been enhanced to be able to display the current
cache settings on a device basis without the need to use RMD to see the current
cache settings.
1.4.1.5 Correction to CDA
Previously, CDA gave an error message and incremented the error count if a data
caching partition was not in the system even if the system was not supposed to
have data caching partition. This problem may have affected command files that
used the exit status from CDA.

Now, CDA no longer includes ICACHE when the IALL qualifier is issued. The
user can reliably determine whether or not the system should return an error if
the system fails to find a cache region and can then design their command files to
rely on the exit status of CDA.

1.4.2 Corrections to SYSGEN
This section describes corrections to SYSGEN.
1.4.2.1 SYSG EN Entered a Loop if More than 4 Units Existed on a DU-Type Controller
When AUTOCONFIGURE was run on a test system that had more than 4
units attached to the same DU controller, all the devices were found. However,
when the information was passed to SYSGEN, SYSGEN did not handle the
extra devices properly. SYSGEN assumed there was a maximum of 4 units per
controller. SYSGEN displayed the following error message:
SGN - Controller A already has 4 units.
You ca~not put JU4 o~ it.

1-14 New Features, Restrictions, and Supplementary Information

SYSGEN then prompted for a physical unit number for DU4:. If a unit number
was given, SYSGEN repeated the above error message, and reprompted for the
unit number. SYSGEN stayed inside this loop. The only way to exit the loop was
to issue a ICtrl/Z I out of SYSGEN.
If there are greater than 4 units per controller, SYSGEN will warn the user they
are creating a non-standard configuration and will update the system rather
than looping without giving an option to change. The problems of overriding
Autoconfigure, lowering the number of controllers, and having the number of
units out of range have also been corrected.
1.4.2.2 Problems with Con Display
A problem occurred after performing a SYSGEN and then running
AUTOCONFIGURE on the host system and then electing to override the results
to add devices. During the override, SYSGEN stepped through the DU devices
and asked for the physical unit number and to what controller (a or b) the device
was attached. The following answers were given: putting the RA60 as DUO on
DUA, and putting the RC25 as DU4/5 on DUB. Logical unit DU7 was set to have
a physical unit number of 200. This was accepted.

After the SYSGEN was done and the system VMRed, the SYSVMR.CMD had
VMR display some of the information in the system using a CON DISP command.
This command listed:
DU7:

DUB177600:

OFL

DRIVER

The unit number in the second column was incorrect. However, after a boot of
the generated system, a CON DISP CONTR ATTR ALL showed the proper unit
number.
This problem has been fixed.
1.4.2.3 Invalid Question on Processor-Type after an Autoconfigure
Previously, the following string of questions may have been asked during a
SYSGEN operation:
>;
>* CE330

Is your system clock programmable (KWll-P)? [YIN D:NJ: n

>;
>* CE350

Is your line frequency 50 Hz? [YIN D:NJ :

>;
>* Is the target processor a PDP-11/70? [YIN D:Nl :

>;
>;
>;

>;=============================================================
>;

Choosing Peripheral Configuration

01-AUG-91 at 03:09

>;=============================================================
Although the systenl was autoconfigured, the installer had elected not to override
the results of auto configure and the system asked if this was a PDP-II170
processor. Also, the system did not have the question in the standard SYSGEN
format. For example, as a CExxx question.
This question was asked as a result of a system memory size being less than
12S.KW. The default range checking has been modified to insure that any default
memory size is in range.

New Features, Restrictions, and Supplementary Information

1-15

1.4.2.4 SYSGEN Terminated Abnormally if a User Chose to Override ACF Results
Previously, the user may have encountered the following series of questions when
generating a system if the user chose to override the ACF results:
>* CE010
>* CE020
>* CE120
>*
>* CE130
>* CE264
>* CE270
>* CE280
>*
>; Note:
>;
>; Note:
>.,
>* CE290
>*

What is your target processor type? [S R:5.-12. D: 11!73"]: 11/93
Do you want the Full-functionality Executive? [YIN D:Y]: y
Do you want support for communications products
(such as DECnet, PSI, and LAT)? [YIN D:N]: y
What is the system name? [S R:0-6 D:"RSXMPL"]: RSXMPl
Do you want IPll industrial 1/0 subsystem support? [YIN D:N]: n
Do you want to include XDT? [YIN D:N]: y
Enter the crash notification device CSR
address [0 R:160000-177700 D:177564]:
Enter both the device and the logical unit number.
For example, MMO: or DL1:.
Enter XX: if you desire loadable DU:, DL:, MU:, MS: or MM:
crash driver support.
On what device and unit do you want crash dumps
to be written? [S R:2-6]: xx:
1

AT.TO
Bad range or default specification
.ASKN ['TEMPN%D.' :2044. :DFLTN] $S$YSZ CE310
Enter memory size (in K words)
.ASKN [256. :2044. :DFLTNj $S$YSZ CE310
Enter memory size (in K words)

This error occurred because the string "DFLTN" was not enclosed in quotes
which allowed substitution. This only occurred if the user elected to override
ACF results. If the user answered N to that question, the SYSGEN proceeded
normally.
This problem was caused by SYSGEN and ACF incorrectly identifying the CPUtype. SYSGEN and ACF were setting the system memory size at 124.K which
was less than the low range for the question. SYSGEN now has better default
checking. Autoconfigure has also been corrected to identify the PDP-ll/93 as a
22-bit processor.

1.4.3 Corrections to ACF
This section describes corrections made to ACF.
1.4.3.1 ACF Did Not Correctly Determine the Processor Type and Memory Size on a
PDP-11/93 Processor
Previously, ACF did not correctly determine the processor type or the memory
size. The highest memory size reported by the ACF task was 1920 10 Kilo Words,
even for systems that supported a maximum memory size of 2048 10 KiloWords.

Autoconfigure has now been corrected to now size memory up to 2044. KW on the
following processors:

•
•
•
•
•
•
•
•

PDP-11123+
PDP-II/53
PDP-11153+
PDP-11173
PDP-11183
PDP-11184
PDP-11193
PDP-11194

1-16 New Features, Restrictions, and Supplementary Information

1.4.3.2 ACF Did Not Identify a PDP-11/93 Correctly During an Installation of Micro'RSX Using
the RX50 Kit
After installing the Micro'RSX floppy disk kit (with all options installed except
the reference disk) and examining ACFPAR.DAT to determine the results of ACF,
the user found the CPU was incorrectly identified. The following is the contents
of the ACFPAR.DAT file that was generated:
CPU=11/00
LIN=60
CON=DLA,CSR=174400,VEC=160
CON=DUA,CSR=172150,VEC=154,UNI=(3,O)
CON=MUA,CSR=174500,VEC=260
CON=YLA,CSR=177560,VEC=60,UNI=(1,O)
CON=YLB,CSR=176500,VEC=300,UNI=(1,O)
CON=YLC,CSR=176510,VEC=310,UNI=(1,O)
CON=YLD,CSR=176520,VEC=320,UNI=(1,O)
CON=YLE,CSR=176530,VEC=330,UNI=(1,O)
CON=YLF,CSR=176540,VEC=340,UNI=(1,O)
CON=YLG,CSR=176550,VEC=350,UNI=(1,O)
CON=YLH,CSR=176560,VEC=360,UNI=(1,O)
CON=MUB,CSR=160404,VEC=370,UNI=(1,O)

Corrections in ACF caused additional bits to be cleared, including the bits that
designated a system as a PDP-11193 or PDP-11194 processor. This should not
affect the user or the system, except in the case of a CPU-based error log packet,
which are extremely rare.

1.4.4 Corrections to Shadow Recording
The following section describes corrections that have been made to shadow
recording.
1.4.4.1 SHA Terminated with a Memory Protect Violation
Previously, shadow recording would terminate with a "Memory protect violation"
if there was no bad block data found on the two disks. Shadow recording will now
report the error as:
SHA -- No bad block data found on input volume.
1.4.4.2 SHAERR Reported Invalid LBNs on RA-Type Devices
When an error occurred on a shadow set, the logical block number printed by
SHE ... was incorrect. This occurred because DUDRV did not convert the LBN
to track/sector/cylinder parameters prior to using the information. This problem
has been fixed.
1.4.4.3 User Routine Enhancements for SHAERR
In order to increase the usefulness of the user action routine in the shadow
recording error task (SHE ... ), the following changes have been included:

•

The internal variable LOST, has been changed to the global symbol $LOST, to
allow access to the number of packets lost due to saturation.

•

The entry point, $SHERR, has been made a local symbol so the variable in
SYSCM can be accessed without conflicting with the local task entry point
label.

New Features, Restrictions, and Supplementary Information 1-17

1.4.5 Backup and Restore Utility (BRU) Corrections
The following corrections have been made to the BRU utility.
1.4.5.1 Error with IEXCLUDE Using Wildcards
Previously, issuing a command line similar to the following caused all files except
the first one in the [named] directory to be copied, even though they should
have been excluded. The file copying also occurred if a numbered directory was
specified.
>BRU/EXC/BAC:name/MOU LB: [*,*]*.LOG, [named] MT1:

The file copying problem has been fixed.
1.4.5.2 BRU Dismount Problem
Dismounting a drive after a multivolume tape backup caused the following error
message:
DMO --

I/O

error

on MUO:

This problem occurred because a change was made to have BRU dismount the
various tapes of a BRU backup in a previous release. This was done to prevent
users from appending a backup set to any tape other than the first tape of a
multivolume set, as BRU was not able to restore a backup set if it did not start
on the first tape of the set. To ensure the tape was not used for appending, BRU
issued a command that rewound and unloaded the tape. This had the effect of
taking the tape unit off line. When DISMOUNT sent its command to the tape
drive, the command was rejected because the drive was not on line.
This problem has been corrected.
1.4.5.3 Invalid Date or Time Error with IREVISED
When using the /REVISED switch, BRU issued the following error message on
files that contained no revision date:
BRU -- *WARNING* -- Invalid date or time (filename)

This problem has been fixed.
1.4.5.4 Problems with Creating the Master File Directory
When BRU/INI and INI created the Master File Directory (OOOOOO.DIR), the
directory was not marked as contiguous even though the file was only one block
long. This did not cause a problem for any RSX system. However, it would cause
a boot failure on VAX 8xxx systems because the boot disks are copied using BRU.
This problem has been corrected.
1.4.5.5 BRU Could Not Abort When Used With a TK50
Previously, BRU could not be aborted when used with a TK50. Issuing a CtrllC
key sequence followed by an ABO BRUTO command resulted in BRU being
marked for abort. However, BRU did not abort. You had to reboot the system to
stop the BRU operation. This problem has been fixed.
1.4.5.6 Help File for BRUSYS Updated
Previously, Help BRU STANDALONE instructed the user to use VMRM42. This
was an outdated version ofVMR. Also, both the VMRM46 and VMRM47 version
of VMS were on the kit in UIC [6,54]. The BRU help text now indicates that
VMRM47 is the correct version of VMR to use. Additionally, VMR47 is the only
version ofVMR in UIC [6,54J.

1-18 New Features, Restrictions. and Suppiementary information

1.4.5.7 BRU Error Message Contains Extraneous Text
While running a BRU backup from one disk to (192,000 blocks) to another
(13,000), you sometimes received the following warning:
This disk will not contain a hardware bootable systew 002300.DIR;l

The last part of this warning indicating a file directory name was misleading.
This problem has been fixed.
1.4.5.8 BRU Aborted Following a Tape Error
Tape errors during the disk-to-tape backup operations occasionally caused BRU
to terminate with an error message of "odd address trap."

This problem has been fixed.

1.4.6 Correction to SHUTUP
SHUTUP.TSK did not complete properly if the CO: device was logged in.
SHUTUP.CMD would not execute and the CPU would not halt. This problem has
been fixed by changing SHUTUP to log out all users before turning off console
logging.

1.4.7 Corrections to Peripheral Interchange Program (PIP)
The following corrections have been made to the PIP utility.
1.4.7.1 Problem with Checking the Directory Specification
PIP was not checking the directory syntax when the IDF switch was used. It
was accepting a directory name greater than 9 characters, and even embedded
hyphens, underscores, and dollar signs. The error was reported only when issuing
a subsequent command without a directory specification. This problem has been
corrected.
1.4.7.2 Specifying Files Using a Wildcard
PIP was ignoring input wildcard file specifications when reading files from an
ANSI tape. PIP was selecting all files instead of just those that matched the
wildcard file specification supplied by the user. This worked correctly prior to
Version 4.3. This problem has been fixed.
1.4.7.3 Problem with Using Wildcards and Creating a Directory
Previously, while doing a file copy, PIP set the filename and type blank if the
command line contained explicit wildcards in the destination field and PIP had to
create the directory for the file. This problem has been fixed.
1.4.7.4 Problem with Issuing the PIP/LI Command
In the previous release, issuing the PIP [l,]ILI command caused PIP to abort
whether it was invoked by MCR or DCL. This problem has been corrected.
1.4.7.5 PIP Did Not Give Correct Error for Command [34,477]/LI
Previously, issuing a command such as PIP [34,477]ILI resulted in the following
incorrect error message, "Cannot find directory file."

N ow, issuing this command will result in the following correct error message,
"Bad directory syntax".

New Features, Restrictions, and Supplementary Information 1-19

1.4.8 Problems with Submitting a Batch Job with the INOPRINT Switch
On RSX-IIM-PLUS Version 4.2, submitting a batch job with a logical device
name in the batch file specification and with the INOPRINT switch caused a "log
file open error" from the batch processor on systems generated without extended
logical name support. For example:
>ASN DB1:=CC25:/GBL
>SUB /NOPRINT=CC25: [240, 1] NORA. BAT
BPR-*FATAL*-Log file open error (BAPO

If the real device name was used or if the INOPRINT switch was omitted, the
SUBMIT command worked. This problem has been fixed.

1.4.9 Using a Cli Build with the ICl Switch Correction
Using the install-run-remove option to run a task which had been built with the
/CL (CLI) switch caused a system crash. Install printed the "INS - Task may not
be run" error message and the system crashed at some point after the message
was printed. This problem has been corrected.

1.4.10 IO.EIO QIO Request
The IO.EIO QIO request to an unattached remote terminal line on a non-I&D
system caused a system crash. This problem has been fixed.

1.4.11 SHOW lOGICAlS
The DCL command SHOW LOGICALS did not accept a qualifier and a parameter
on the same command. For example, the SHOW LOGICALS xxx /SYSTEM
command failed with the following error:
SHOW

Extraneous input - Check for proper command structure.

This problem has been corrected.

1.4.12 KW11-P Clock
When trying to use the KWII-P clock by means of a CINT$ routine, when the
clock was not being used as the system clock, the following DSW error was
returned:
IE.RSU (Vector in Use)

This problem has been corrected.

1.4.13 FAl
On Version 4.3, FAL would hang if a remote node issued a directory command
with a blank file specification. For example:
DIR REMNOD"usernam password": : LB: [1,2]

This problem has been corrected.

1.4.14 Help Files
Previously, HELP INDIRECT ENABLE for the Indirect Command Processor
stated that the .ENABLE CLI and .ENABLE MCR operating modes were local
whereas actual operation showed them to be global, with their effect being passed
when your program chained to a new command file or when a new command
file is invoked. The help files were incorrect in stating that the .ENABLE CLI
and .ENABLE MCR operating modes are local. They are global; their effect
is retained when invoking or chaining to a subsequent command file. The
1-20 New Features. Restrictions, and Supplementary Information

documentation of the fCLI and /MCR switches and of the .ENABLE command
in the RSX-llM-PLUS Indirect Command Processor Manual is correct. Section
2.5 states the switches that are used in initiating indirect processing are used as
defaults when executing subsequent @filespec or .CHAIN commands. Further,
the CLI and MCR operating modes are specified in Section 2.6.12 as being
retained when Indirect passes to a lower-level command file. The help file has
been corrected in RSX-11M-PLUS Version 4.4 to correctly state that the CLI and
MCR modes are global.

1.4.15 Support for RA82 Errors
Previously, RPT trapped on RA82 SDI errors. The MSCP8x error log control
module has been updated to include support for RA82 SDI errors. This problem
has been fixed.

1.4.16 Corrections to Error Logging
This section describes corrections to error logging.
1.4.16.1 ERRLOG Opened Wrong File for Append and/or for Random Access
Previously, ERRLOG would open the wrong file for append (instead of
ERRORS.LOG) and/or for random access (instead of ERRORS.ACC).

This problem has been corrected.
1.4.16.2 Error Logging of Volume Not Mounted Packets for DU Disks
The disk controller error logging has been changed to ignore ''Volume not
mounted" packets. These are not errors and can rapidly clog up error log files.
1.4.16.3 Error Logging Support for PDP-11/84
Previously, the system would not recognize a PDP-11184 as a valid device type
when that device was entered in the error logging tune command file, as shown
in the following example:
UTRPSU$ * Enter CPU type [S]: 11/84
UTRPSU$
UTRPSU$
Unrecognized CPU type 11/84 - Please enter again
UTRPSU$
UTRPSU$ * Enter CPU type [S]: 11/83
UTRPSU$
UTRPSU$
Now enter the devices in your configuration separated by
UTRPSU$
commas

The TUNE.CMD file has been updated to include the PDP-11184 as a valid
CPU-type.

1.4.17 Using the .LIMIT Feature with Tasks
The MACRO .LIMIT feature did not work correctly with a task built with the /EX
feature (extend tasks with memory resident overlays). This problem has been
fixed.

1.4.18 Parity Problems with Terminal Multiplexers
Previously, unsynehronized access to terminal multiplexer CSRs in terminal
driver module TTYHV caused parity to be set on incorrect terminal lines. This
problem has been fixed.

New Features, Restrictions, and Supplementary Information 1-21

1.4.19 Printing to a Printer Connected to a lAT Port
Previously, printing to a shared printer connected to a LAT application port
occasionally produced the console error message:
LAT -- CONNECT FAILURE #-3

(Device not ready)

A LAT connection QIO timing problem exists that may cause an application
to hang or not behave as expected on a heavily loaded Ethernet. This timing
problem exhibits itself most often when an application does a disconnect followed
by a connection in quick succession, as could happen in the print despooler.
Although the despooler does not detach until all relevant jobs have been processed
and cleared from the queue, there can be occasions when it would immediately
attempt to connect again. The print despooler task includes a correction for
this problem. If the "Device Not Ready" status is returned from a LAT printer
connect, the des pooler will wait for one second and retry the connect. After five
failures, the des pooler will give up and report the error as shown above.

1.4.20 UFD Created Directory Files with Embedded Blanks
Previously, UFD did not correctly handle the directory specification when the
default directory was specified. For example:
$ set def/named lb: [61,40J
$ cre/dir [J
$ dir DIR -- Cannot find directory file
DR: [6L 40J

This set of commands resulted in UFD creating "61 40.DIR" rather than
"061040.dir." When the command UFD ddn:[] was entered, the directory
specification should have defaulted to the default directory string (assuming
there was one). In the case where this string existed and was in UIC format (i.e.
[g,m]), UFD created the numbered directory string with leading zeros suppressed,
resulting in the creation of directory files with embedded blanks (e.g. "7 44.dir").
This problem of suppressing leading zeros has been fixed.

1.4.21 Incompatibility Between VMS DIFFERENCES Command and RSX DCl
DIFFERENCES Command
The DCL DIFFERENCES command did not accept DIFF A.TXT, B.TXT but
would accept DIFF A. TXT B. TXT.
The description of the DIFFERENCES command in the RSX-llM-PLUS
Command Language Manual shows only the blank space as a valid file delimiter.
A comma (,) can now be used as a valid file delimiter.

1.4.22 Enhancement to the DCl SORT Command
Previously, DCL did not allow use of the IINDEXED _SEQUENTIAL qualifier
to the input and the output file for the DCL SORT command. DCL has been
modified to allow use of the IINDEXED_SEQUENTIAL qualifier to both the
SORT input and output files.

1.4.23 Correction to FlX
Previously, FLX did not display the maximum number of blocks on a tape if the
number of blocks was too large. This problem has been fixed.

1-22 New Features, Restrictions, and Supplementary Information

1.4.24 Task Builder (TKB)
Support has been added for using the MACRO .LIMIT directive in conjunction
with the EXTM$ system directive.

1.4.25 Undefined Symbol Error in RPT
Previously, RPT encountered an "undefined symbol referenced in expression eval"
error when attempting to process a packet in a DSP4Pl module.
This problem has been corrected.

1.4.26 Changing Crash Device from BOO Request Did Not Pass Information to
Display
Previously, the system was modified to display three digits for the crash unit.
However, this change was not propagated back to INITL. This problem has been
fixed.

1.4.27 DSA Timeout Processing Led to Hung Tasks
DSA drivers should have had a timeout code to prevent the drive/controller from
being hung for more than limited period of time. If a attempt to mount a drive
hung, and the mount was then aborted, the drive was unusable. The only was to
regain use of the drive was to reboot the system.
Timeout processing context should have been kept for the oldest outstanding
command to the DU controller. However, the driver was improperly updating the
context. This problem has been fixed.

1.4.28 Support for Inputting Lower-Case Characters in the TUNE.CMD and
TUNERL02.CMD Files
Previously, the TUNE.CMD and TUNERL02.CMD command files did not allow
the user to enter the device names using lower-case characters.
Now, the user can enter characters from terminals supporting both upper-case
and lower-case characters.

1.4.29 Submitting a File That Did Not Exist Resulted in an Error
Previously, if you tried to submit a file that did did not exist to a BATCH job, you
would receive the following error:
>chu 1, 1
>subm /noprin=do.cmd
22:16:53 Task "SUBTO " terminated
Odd address or other trap four
RO=022135
R1=000030
R2=001062
R3=006022
R4=006060
R5=OOOOOO
SP=000346
PC=020002
PS=170000
>

This problem has been fixed.

New Features, Restrictions, and Supplementary Information 1-23

1.4.30 Support for PDP-11/94 in the SAVSIZ List of 2044KW CPUS
The PDP-11/94 has been added to the SAVSIZ routine, which sizes memory, as
an Unibus CPU that can have a memory size greater than 1920KW.

1.4.31 Support for the Angle Bracket « » Characters for Directory
Specifications
Support has been added to the SET IDEF and the SET IUIC commands to allow
the user to use the angle bracket characters « » for a directory specification.
This is especially useful on the German keyboard which requires the use of the
compose key to produce a square bracket.

1.4.32 Support for J-11 UNIBUS Systems in the Crash Driver
Previously, the crash driver for the TU80 would only write the first 3840Kb
of the memory on PDP-11194 systems (or other J-ll based UNIBUS systems)
that had 4Mb of memory. This created problems when these systems crashed,
especially if the crash was related to DECnet-llM-PLUS or P.S.I.111M-PLUS, as
that software was usually loaded in the top part of memory (above the 3840Kb
border). This problem has been fixed.

1.4.33 Enhanced Support for RX02 Devices
Previously, if you generated a system for a Q-bus PDP-II system with 22-bit
support and you included RX02 devices, the system might crash when you tried
to use one of the RX02s.
The problem occurred because DYCOM was one of the last RSX driver pieces
to be installed in the system during a Virtual Monitor Console Routine (VMR)
operation. DYCOM went into GEN instead of DRVPAR. This sometimes caused
DYCOM to reside outside of the first 124 KW of memory. There was no comment
in SYSVMR.CMD that it should be loaded below 124 KW. Additionally, there was
an error in DYDRV where DYDRV tested for the base address of DYCOM.
These problems have been corrected. Other problems in online processing with
DYDRV and DYCOM have also been corrected.

1.4.34 SET TERft.1INAUINQUIRE Command Did Not Work on a VT420 Terminal
Previously, the SET TERMINAL/INQUIRE did not work properly on a VT420
terminal. The command appeared to work as no error was returned but issuing
the command had no influence on the setup of the terminal characteristics. Now,
the terminal will work the same as a VT3xx terminal.

1.4.35 Problems with ODT in an 1&0 System
Previously, the I&D version of ODT had a problem that appeared if you issued a
sequence of commands similar to the following:
.meall exit$s
.pseet aaeode,i
.blkw 340 .
. pseet zzeoc.e, i
start: .rept 128.
NOP
.endr
ex:tSs
.er:d star':
SOD~DT: (?W,D,LCL,REL,CO~)

OC1304 00:020 08528.

1-24 New Features, Restrictions, and Supplementary Information

UTRPSU$ run odttst
ODT:VT3
1

1554/000240
_d
1554/005020
_s

8B:001556
_$7/001556
_s

8B:$lE
_d
1554/005020
i

1554/005020
_x

The problems included the following:
•

After a register reference ($x1) , when an address was referenced using ODT
that was within the range of the $ODTDT p-sect of ODTID.OBJ, then ODT
did not work properly.

•

The I&D commands no longer worked and all memory references were from
the D-space.

•

The addresses printed were wrong when entering breakpoints or when you
single-stepped. They appeared as BB:$T.

These problems occurred because the address not only has to be within $ODTDT
but it also has to be between labels INTBEG and INTEND. This problem has
been fixed.

1.4.36 Intermittent Problem When Getting Data Late Errors on RK06 or RK07
Drives
Previously, there was an intermittent problem that resulted in getting data late
errors.
Due to potential inaccuracies in the way the RK06 reports word count when a
data late is declared, the retry logic in the driver would restart the read or write
operation in the wrong block.
This problem has been fixed.

1.5 Software Problems and Restrictions in Version 4.4
This section describes software problems and restrictions that have not been
corrected for Version 4.4.

1.5.1 Restrictions in SYSGEN
This section describes restrictions in SYSGEN.
1.5.1.1 Selecting FCPMDL in SYSGEN
If you perform a SYSGEN and select both the nonstandard EXECUTIVE and
FCPIVIDL, the system being generated will fail because FCPMDL is not present
on the distribution and it is not rebuilt automatically. This problem occurs
because SYSGEN was modified to eliminate rebuilding of all vectored privileged
tasks (except SAV) and FCPMDL is the only vectored privileged task not shipped
on the baseline system. You can rebuild FCPMDL by answering Y to the BN010
question in the section on Building Nonprivileged Tasks and entering "FCPMDL"
in response to the following question:
New Features, Restrictions, and Supplementary Information 1-25

* BN020

Enter task name(s)

[S]: FCPMDL

1.5.1.2 Incorrect Displays for Multiple MM Devices in SYSGEN
The RSX-IIM-PLUS SYSGEN RH configuration display lists only one MM
device when there are actually two MM devices. ACF finds two TU77s (and a
TM03 formatter) but SYSGEN prompts for the number of tapes drives (question
CP160B) and formatters (CP1612). SYSGEN does configure correctly for two MM
devices.
1.5.1.3 Specifying a Nonexisting Device as the Crash Device During SYSGEN
If a nonexistent device is specified as the crash device during a SYSGEN
operation, then you cannot change the crash device after the initial boot of
the system. If the system reports an error, you must SYSGEN the system again
and specify an existing device as the crash device.

1.5.2 Restrictions on Year 2000 Support
The following sections describe restrictions on the Year 2000 Support.
1.5.2.1 Special Error Log Display Commands Do Not Use the 4-Digit Year Field
The special error log commands will still display time stamp information in the
following "dd-mmm-yy" format:
ELI /SH/NE
ELI /SH/CU
ELI /SH/HI

These problems will not be fixed.

1.5.3 Error Logging Restrictions
This section describes restrictions to Error Logging.
1.5.3.1 Support for Error Logging History Summaries
Error logging history summaries are no longer supported. See Section 2.B.
1.5.3.2 Incorrect Displays of Register Fields for DU and MU Devices
The errorlog driver internal timeout packet SA register fields for both DU and
MU devices are not being displayed properly.

1.5.4 BRU Restrictions
This section describes restrictions to the Backup and Restore Utility (BRU).
1.5.4.1 Use of Wildcards in BRU
BRU supports the asterisk (*) character only as a replacement of an entire field
in a file specification. BRU will issue the "Invalid value or name" fatal syntax
error when an asterisk is used to replace part of a file specification field.
1.5.4.2 Problem with the INEW_VERSION Qualifier
A problem with the /NEW_VERSION qualifier occurred when used in a BRU
command line during a disk-to-disk backup. From the input device, BRU backed
up multiple versions of a file in the order that they appeared in a directory,
instead of in ascending order by version number. As a consequence, when the
files were restored to the output device, new version numbers were assigned
when each file was restored, and the original order of the files was not preserved.
Thus, the last version of a file on the output device may not have been the last
version created on the input device.

1-26 New Features, Restrictions, and Supplementary Information

The following workarounds are suggested to avoid this problem:
•

Sorting files on the input device.

•

Purging files prior to backing up your files.

1.5.4.3 BRU Does Not Give Correct Error Messages When Backing Up a Named Directory
That Does Not Exist

BRU does not give a correct error message when backing up a named directory
that does not exist. It does not give any error messages during the copy pass.
Also, a number of EOT marker errors occurred before the verify pass began. BRU
also asked for another tape even though there was nothing for BRU to write. The
following example shows the results of backing up a named directory that does
not exist:
>bru /rew/mou/ver:double/bac:test du1: [jjmrel] mu:
BRU - Starting Tape 1 on MUO:
BRU - End of Tape 1 on MUO:
BRU -- *WARNING* -- EOT marker error
I/O error code -4
BRU -- *WARNING* -- EOT marker error
I/O error code -4
BRU -- *WARNING* -- EOT marker error
I/O error code -4
BRU -- *WARNING* -- EOT marker error
I/O error code -4
BRU -- *WARNING* -- EOT marker error
I/O error code -4
BRU - Mount another tape
1.5.4.4 Restriction on Incremental Backups
If BRU performs an incremental backup with the following command and there

are no files after the specified date, then BRU does not properly terminate the
tape:
BRU

/MOU/REW/TAP:label/BAC:name/REV:AFT:date DU1:

MUO:

If you do not specify the NER switch, BRU gives the "No files found" and then
appears to terminate normally. If you then issue a "BRU IREWIDIRlBAC:name
MUO:", BRU responds with the following error messages:
*FATAL * -- Tape read error
I/O error code -10

If you specify the NER switch, BRU returns the following messages after the end
of the writing pass:
*WARNING* -- EOT marker error
I/O error code -4
Mount another tape
Mount Tape 1 on MUO:

BRU repeatedly issues the "Mount Tape 1 on MUO:" message.

1.5.5 DCl Command ABORT TasknamelTAS Does Not Translate Correctly
When an ABORT tasknamelTAS command is entered, it translates to ABORT
tasknameITERM=TAS, which is incorrect. It should be ABORTITAS taskname.
The correct syntax for the DCL ABORT command requires that the ITASK
qualifier follow immediately after the ABORT command. Furthermore, the MCR
ABO command does not allow any of the following (you get an "invalid keyword"
error message):

New Features, Restrictions, and Supplementary Information 1-27

ABO/TAS task
llBO /TAS task
}""BO task/Ti\S
ABO task /TAS

ABO task does work in MCR, which is identical to the action of the DCL ABORT
command.

1.5.6 Tasks Hang When Attaching to Another Terminal Already Attached to a
Task
Tasks would occasionally hang when they attempted to attach a terminal that
was already attached to another task. The attach QIO request never returned
and the task had to be manually aborted.
As a workaround, the Marktime directive (MRKT$) can be used to implement
a timeout function within the application code. If the attach function is issued
as a QIOW, you can precede it with a MRKT$ directive, specifying the same
event flag number using the I/O status block to distinguish between timeouts
(IS.PND) and 1/0 completion (other values). Alternatively, the MRKT$ directive
can specify an AST routine to be entered at the expiration of the timer with a
CMKT immediately following the QIOW. There are other workarounds such as
using WTLO or WTSE with QIO and MRKT$ directives. In these workarounds,
the expiration of the time interval without I/O completion would signal the need
for an I/O kill to be issued.

1.5.7 ELI/SH Reports No Errors if ELI/NOLIM Has Been Set
If error log limiting has been disabled with the ELI INOLIM switch, then
subsequent hard errors are not displayed by the ELI ISH switch even though a
full report generated with RPT shows that device hard errors have been logged.
This restriction still exists.

1.5.8 Problem with Unused Terminal Ports
The system can crash if you have an unused terminal port with a cable connected
to the port. This problem can be eliminated by either removing the unused cables
or setting the unused ports to slaved ports.

1.5.9 Task Builder (TKB) Restriction
To obtain the long version of the memory allocation file (map) when building an
I&D task, you should specify only the I-SH switch (the DCL !LONG qualifier)
after the map file in the command line. The IMA switch (the DCL /MAP qualifier)
should not be specified with the I-SH switch for I&D tasks.

1.5.10 RMD Restriction
The following restriction applies to RMD:
If the RMD C (cache) page is being displayed interactively and the caching
configuration is changed, RMD must be exited and a new RMD C command
issued to display the new caching configuration correctly. RMD cannot refresh
the display correctly for a caching configuration changed during an existing RMD
display.

1-28 New Features, Restrictions, and Supplementary Information

1.5.11 PIP Restriction
When requesting PIP to create a file to be contiguous with a file size of 32767 10
blocks (777778) or less, PIP does not create the file and issues the following error
message when there is not enough contiguous space:
PIP -- Allocation failure - no contiguous space

However, when the requested file size is 32768 10 blocks (1000008) or larger, PIP
creates the file with zero blocks but still issues the same error message when
there is not enough contiguous space.

1.5.12 Indirect Preprocessor Restriction
The Indirect Preprocessor (IPP) does not handle .END statements in an ENABLE
data block correctly.

1.5.13 Queue Manager Restriction
If a printer on an asynchronous terminal line is not functioning due to a fault or
power failure after the queue manager submitted a job to the printer, the queue
manager can go into a deadlock state. The print job is shown as active although
nothing is being printed and the job cannot be aborted, killed or deleted from the
queue.

1.6 Supplementary Software Information
This section provides additional information on the following topics:
•

Changes in the utility used to save and restore data on media

•

Changes to File Control Services

•

Restrictions and additional information on the Taskbuilder utility

1.6.1 Disk Save and Compress Utility
RSX-IIM-PLUS Version 4.3 was the last release to have the Disk Save and
Compress utility (DSC) included on the software kit. Any data currently saved on
media by the DSC utility will have to be restored using the DSC utility and then
resaved using BRU. Beginning with RSX-IIM-PLUS Version 4.4, customers will
have 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-II 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,I]NOVACPLIB.OLB
and in the concatenated object module [1,I]FCSNOVACP.OBS.
Some large tasks that have a complicated Overlay Description Language (ODL)
may be affected by the internal reorganization of specific FeS 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.

New Features, Restrictions, and Supplementary Information 1-29

The object library [l,l]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 [l,l]FCSNOVACP.OBS in the system library by using
the following Librarian Utility Program (LBR) command:
LBR>[l,llSYSLIB/RP=[l,llFCSNOVACP.OBS IReturn I

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-11M-PLUS fastmapping 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-llM-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.

1-30 New Features, Restrictions, and Supplementary Information

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-11 M-PLUS Batch and Queue Manual
In Section 2.6.1, page 2-15 of the RSX-IIM-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-11 M-PLUS Command Language Manual
Please add the following information to the RSX-IIM-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
TERMINAUmodel supports the VT300-series terminals in VT200 mode only.
You cannot specify VT3}ocSERIES 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 multiheader index files).
You can use Table 2-1 to determine an appropriate value to use for the
IHEADERS:n parameter.

Corrections to Documentation 2-1

Table 2-1 Number of Index File Headers
Maximum Number
of Files

Disk Size

Number of
File Headers

Maximum files fewer than 25588

Disks fewer than 209 Mb

Headers=maximuml2

Maximum files greater than
25588 and fewer than 51176

Disks between
210Mb and 419Mb

Headers=25593

Maximum files greater than
51176 and fewer than 51693

Disks between
420Mb and 423Mb

Headers=25846

Maximum files greater than
51693 and fewer than 65500

Disks greater than 423Mb

Headers=51693

Please replace the second paragraph and formula after the
!MAXIMUM_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.

•

Please add the following information to Section 5.11.1, page 5-80 at the end
of the first paragraph:
You must specify the IWRITE:n qualifier when formatting volumes on DL-type
devices.

•

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 command qualifiers /CREATED and
!MODIFIED is incorrect. The correct formats are:
Formats
BACKUP/CREATED/arg
BACKUP/MODIFIED/arg

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

In Section 5.12.6, page 5-104, the second example is incorrect. The correct
example is shown next.
$ [3ACKUP/:vl0J=F=EJ/A?T~R: (:~-r','l.A~r-8j :/:00) /~iE~_=FT~{ L':';O: :',:30: !Return I

2-2 Corrections to Documentation

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 error message.
•

Please add the following text to Section 5.13.1 of the manual:
The DCL commands SHOW ASSIGNMENTS and SHOW LOGICALS have a
new functionality and a new display format. The commands can now display
at your terminal all logical assignments for a specific logical name.
Formats

$ SHOW ASSIGNMENTS [logicalname][/qualifier]
$ SHOW LOGICALS [logicalname][/qualifier]
Parameter
logicalname

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.
Qualifiers

fALL
IGLOBAL
fGROUP[:n]
!LOCAL
!LOGIN
fSYSTEM
ITERMINAL[: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

$ SHOW ASSIGNMENTS

IReturn I

(Session Login Logical Table for VT2:)
"DCM"
"ER"
"GO"
"HOME"
"REVIEW"

"DU: [CACHE] "
"EDT/RO "
"SET /DEF="
"SYS$LOGIN"
"LB: [1, 54] RSXllM. JvI.AP"
"RECALL"
"DU1: [REVIEWS] "

!! ('"'I,TnAT r..r"1'"?ITII

IIT"\TT.

"SYS$PROMPT"

"Jim>"

"MAP"
IIR n
2l12l :;>lJUu H~

rTTnnnl II

lJU; L U2l~I\J

rn..! .............. 11

L I' .lllQ.l J

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

"DCL"

Corrections to Documentation 2-3

Displays all the login and local assignments for the issuing terminal.
$ SHOI>] ASSIGNMENTS S* IReturn I
(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.
IReturn I

$ SHOW LOGICALS "SYS$CLI"

(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.
$ SHOW ASSIGNMENTS "SYS$CLI"

IReturn I

Displays no matching login and local assignments for the string "SYS$CLI"
for the issuing terminal.
$ SHOW ASSIGNMENTS /ALL

IReturn I

(System Logical Table)
"LI"
"SS"
"WK"

"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] RSXllM. 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.

2-4 Corrections to Documentation

$ SHOW LOGICALS S*/ALL

IRffium I

(System Logical Table)
"SS"

"L8:" [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 8.1.3, page 8-5, states that the DCL command SET SYSTEM
ICRASH_DEVICE is valid only for pregenerated operating systems. This is
no longer true; the command is now valid for all RSX-11M-PLUS operating
systems.

•

The following section describes an enhancement and lists restrictions for the
Backup and Restore Utility (BRU).
The behavior of the BRU qualifier /ERROR with restore operations is
unchanged. However, the qualifier can 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 errors 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 it 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 /REWINDIAPPEND 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

2.3 RSX-11M-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 COIInnand Processor (Indirect) includes
support for VT200-series terminals 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:

Corrections to Documentation 2-5

The following example is from an interactive terminal session:
>@ti: ! Return!
AT.>.enable substitution !Return!
A 1 • > . set sa" 1, L,"
I Return I
AT.>.parse a "," b c d !Return!
AT . >; 'b' ! Return !
>;1
AT . >; , C ' ! Return !

>;2
AT . >; 'd' ! Return !
>;
AT.>.parse a "," b C IReturn!
AT.>; 'b' !Return!
>;1
AT . >; , c ' ! Return !

(null substring)

>;2,
AT. > !CtrIlZ!
>@ <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, code 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 110 buffers.
For privileged Indirect 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.
For the non privileged 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-11 M-PLUS MeR Operations Manual
Please make the following corrections to the RSX-llM-PLUS MeR Operations
Manual:
•

In Section 3.6, page 3-12, the descriptions of the ALT keywords IRPRI and
!TERM are incorrect. Delete the second paragraph, which is in parentheses,
of the IRPRI keyword description. Also, the first sentence of the !TERM
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 MCR command ASSIGN (ASN) can display at your terminal all logical
assignments for a specific logical name.

2-6 Corrections to Documentation

Format

>ASN

[logicalname][/keyword]

Parameter
logical name

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
fALL
fGBL

fGR[=n]
fSYSTEM
ITERM[ =ddnn:]
Refer to the RSX-llM-PLUS MeR 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.
Examples
1.

>ASN

IReturn I

(Session Login Logical Table for VT2:)
"DCM"
"ER"
"GO"
"HOME"
"V.tAP"
IIR"

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

"DU: [CACHE] "
"EDT/RO "
"SET IDEF="
"SYS$LOGIN"
"LB: [1,54 J RSXllM.lILl\P"
"RECALL"
"DU1: [REVIEWS]"
"DU: [USER]" [Final]
"Jim>"

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

"DCL"

This example displays all the login and local assignments for the issuing
terminal.
2.

>ASN S* IReturn I
(Session Login Logical Table for VT2:)
"SYS$LOGIN"
"SYS$PROMPT" =

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

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

"DCL"

Corrections to Documentation 2-7

This example displays the login and local assignments that begin with letter
S for the issuing terminal.
3.

>ASN" SYS$CLI"

IReturn I

This example displays no matching login and local assignments that match
the string "SYS$CLI" for the issuing terminal.
4.

>ASN / ALL

IReturn I

(System Logical Table)
"LI"
"SS"
"WK"

"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] RSXllM.MAP"
"RECALL"
"DU1: [REVIEWS]"
"DU: [USER]" [Final]
"Jim>"

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

"DCL"

This example 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

[Iogicalname][/keyword]

Parameter
logical name

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.

2-8 Corrections to Documentation

Keywords

IALL
IGBL
IGR[=n]
ISYSTEM
ITERM[ =ddnn:]
Refer to the RSX-llM-PLUS MeR 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.
Examples
1.

>DFL "SYS$CLI"

IReturn I

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

"DCL"

This example displays the login and local assignments that match the string
SYS$CLI for the issuing terminal.
2.

>DFL S* /ALL

IReturn I

(System Logical Table)
"LB:" [Final]

"SS"

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

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

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

"DCL"

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

Please add the following information to Section 3.26 on page 3-67:
The HELLO task displays login attempts on the console. If the account name
or VIC 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 VIC, the report will be:
hh:mm:ss

TTnn:

The overlay structure of the HELLO 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.

Corrections to Documentation 2-9

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.
•

Please add the following information to Section 3.27:
The overlay structure of the HELLO 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.

•

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
(IND. 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 (lNDEXF.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 IMXF, 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 IMXF 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

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
423lVlb

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:

2-10 Corrections to Documentation

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

The default used for the n1axin1um 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:
MXF = ((N-(( (N+4095.)/4096.)+9.) )*127.)/258.

•

In Section 3.31, page 3-103, the description and format of the MCR 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:[/keyword(s)]

LOA[D]

IEXP=expname[/keyword(s)]

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

ITASKD=taskname
ITASKI=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.
>OPE O/REG=TSTREG IReturn I

00000000/50712

This command opens the region TSTREG at location 0 and displays in octal
the current value at the location.
>OPE O/REG=TSTREG IReturn I

00000000/50712 % ~ MCR

This command opens location 0 of TSTREG, displays the current value in
octal, and then displays the Radix-50 value in ASCII format.
>OPE O/REG=TSTREG IReturn I
00000000/50712 % ~ MCR %DCL IReturn I

00000002/00000 IReturn I
00000004/14604

Corrections to Documentation 2-11

This command sequence displays the current value at location 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 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 RVN
keyword IUIC is incorrect in Format 3 and Format 4. The correct format is
shown next.
Format
/UIC=[g,m]

•

In Section 3.40, page 3-139, the definition of the keyword IUIC is incorrect.
The correct definition is shown next.
IUle
(Privileged keyword.) The Vser Identification Code (VIC) under which the
task will be requested to run. This VIC also determines which files the task
can access.
The VIC 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 VIC that you specify becomes the default and
protection VICs for the task.
When you do not specify IUIC, the terminal's VIC becomes the default and
protection VICs 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.
dtime

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
taskname 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 deu
defaults to LB:, and [g,m] defaults to the current library VIC (usually, [3,54]).
If the task cannot be found in the library VIC, INSTALL searches the system
VIC on device LB: (usually, [1,54]).
If you do not specify a dollar sign ( $), RUN searches for the task image file in
the UIC (on device SY) to which the terminal requesting the task is set.

2-12 Corrections to Documentation

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 RVN
command keywords IPRI=number and IUIC are incorrect. The correct
description is shown next.
IPRI=number
Specifies the priority of the task. The value range is 1 to 25° 10 , 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 5° 10 , the
priority is set to 5010.
The default is IPRI=50.
IUIC=[g,m]
(Privileged keyword.) Specifies the Vser Identification Code (VIC) under
which the task will be requested to run. The square brackets are required
syntax.

If you specify a VIC, the VIC becomes the default VIC and protection UIC for
the task. If you do not specify a VIC, your default VIC and protection VIC
become the default VIC and protection VIC for the task.
The default is the VIC of the terminal that issues the RVN command.
•

Please add the following information to Table 3-4 in Section 3-42:

Table 2-5 SET Command Keyword Summary
Keyword

Description

Ensuring System
Protection

/INTRUSION :yes

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.

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

When disabled, invalid login transaction blocks include the user's name and
VIC, as found in the user account file, if the login failure results from a
password discrepancy. If the user name or VIC is not found in the account
file, the invalid login transaction block will simply list INVALID VSER as the
user name and the VIC 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.

Corrections to Documentation 2-13

The default is IINTRUSION:no
•

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

II NQU IRE=term
•

In Section 3.42, page 3-161, the format of the MCR command SET
ICRASHDEV is incorrect. The correct format is shown next.
Formats
SET ICRASHDEV=ddnn:[CSRaddr]
SET ICRASH_DEVICE=ddnn:[CSRaddr]

Also, the section states that the MCR commands SET ICRASHDEV 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.
•

Please add the following information to Section 3.42, page 3-162:
Format
SET ICTRLC=ttnn:

This qualifier establishes whether a CTRUC causes an abort (/CTRLC=ttnn:)
or causes an explicit MCR> prompt (INOCTRLC=ttnn:).
•

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:
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.
>SET /SECPOL IReturn I
SECPOL=285.:E~O. :44%

2-14 Corrections to Documentation

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-11 M-PLUS System Generation Guide and Installation
Guide
Please add the following information to the text in Section 3.2.2, Choosing
Executive Options (CE), on page 3-26. This text should be added after the
question * CE200 Which FCP do you want? [S R:1-6 D:"FCPLRG']: and just
before the * CE210: Do you want support for file windows in secondary pool?
[Y / N D :N}: question:
If you entered FCPMDL, then you must answer Y to question BN010 in the
section on Creating the System Image File and enter FCPMDL to question
BN020 Enter task name(s) [S}: FCPMDL. Both of these questions appear on page
3-64 of the RSX-llM-PLUS System Generation Guide and Installation Guide.
The paragraph in Section 5.4.7 on page 5-43 is incorrect. Please replace it with
the following paragraph:
RMS-11 Version 2.0 is included on the pre generated kit disk. The RMS-11
segmented library (RMSRES, and RMSLBL through RMSLBM) and all
the RMS-11 utilities are already installed in the system image. No further
installation is needed, unless you install the DECnet package on your system and
you want to use the RMS-11 remote access facilities. See Section 5.4.1 in the
RSX-llM-PLUS System Generation Guide and Installation Guide for information
on installing the RMS-11 remote access package (DAPRES).

2.6 RSX-11M-PLUS and Micro'RSX Crash Dump Analyzer
Reference Manual
Please make the following correction to the RSX-llM-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-llM-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 anoLheI' 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 /CRASHDEV. The previous
commands are now valid for all RSX-11M-PLUS operating systems. Refer to
the RSX-llM-PLUS Command Language Manual and the RSX-llM-PLUS
MCR Operations Manual for more information on the commands.

Corrections to Documentation 2-15

•

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
/CRASHDEV (or SET /CRASH_DEVICE). The correct format is shown next.
Formats
SET /CRASHDEV=ddnn[:CSRaddr]
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-llM-PLUS and Micro/ RSX Crash Dump
Analyzer Reference Manual for more information on loadable XDT support.

2.7 RSX-11M-PLUS and Micra'RSX Debugging Reference Manual
In the RSX-llM-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-llM-PLUS Release Notes Version 4.2,
Section 1.17.1.

2.8 RSX-11M-PLUS and Micra'RSX Error Logging Manual
Please make the following corrections to the RSX-llM-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:
TSV051TK25

•

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

2-16 Corrections to Documentation

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.
•

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

2.9 RSX-11M-PLUS and Micro'RSX Executive Reference Manual
Please make the following corrections to the RSX-llM-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.

•

Please add the following descriptions of the new high-level language calls
to the GIN directives to Chapter 5 of the RSX-llM-PLUS and Micro/ RSX
Executive Reference Manual. The following GIN directives may be called:

• CALL GINABO
• CALL GINAPR
• CALL GINDEF
• CALL GINDEV
• CALL GINDVJ
• CALL GINFMK
• CALL GINGAS
• GLA..LL GINQMC
• CALL GINREN
• CALL GINSPR
• CALL GINTSK
• CALL GINUAB
• CALL GINUIC
• CALL GINUPD

Corrections to Documentation 2-17

•

CALL GINVEC

2.9.1 The GINABO Call
CALL GINABO

(buf, siz [,ids] )

buf
A buffer area containing the name of the device for the aborts.
siz
An integer containing the size of the buffer in words. If the buffer is declared as
an INT*2 array, then the size is equal to the number of elements in the array.
ids
An integer to receive the directive status.

See the description of the GI.ABO subfunction of the GIN directive in the
RSX-llM-PLUS and Micro/RSX Executive Reference Manual for details on the
buffer format and use of this directive.

2.9.2 The GINAPR Call
CALL GINAPR

(buf, siz [,ids] )

buf
A buffer area for the return of the information.
siz
An integer containing the size of the buffer in words. If the buffer is declared as
an INT*2 array, then the size is equal to the number of elements in the array.
ids
An integer to receive the directive status.

See the description of the GI.APR subfunction of the GIN directive in the
RSX-llM-PLUS and Micro/ RSX Executive Reference Manual for details on the
buffer format and use of this directive.

2.9.3 The GINDEF Call
CALL GINDEF

(uic [,ids] )

uic
An integer representing an user identification code (VIC).
ids
An integer to receive the directive status.

See the description of the GI.DEF subfunction of the GIN directive in the
RSX-llM-PLUS and Micro/ RSX Executive Reference Manual for details on the
buffer format and use of this directive.

2.9.4 The GINDEV Call
CALL GINDEV

2-18 Corrections to Documentation

(buf, siz , devnam , unit [,ids] )

but
A buffer area for the return of the information.
siz
An integer containing the size of the buffer in words. If the buffer is declared as
an INT*2 array, then the size is equal to the number of elements in the array.
devnam
A two-character ASCII device mnemonic.
unit
An integer containing the device unit number.

ids
An integer to receive the directive status.

See the description of the GLDEV subfunction of the GIN directive in the
RSX-llM-PLUS and Micro/RSX Executive Reference Manual for details on the
buffer format and use of this directive.

2.9.5 The GINDVJ Call
CALL GINDVJ

(but, siz , devnam , unit [,ids] )

ids
An integer to receive the directive status.

See the description of the GLDVJ subfunction of the GIN directive in
Section 1.3.4.1 and also, later in this section in this document, for details on
the buffer format and use of this directive.

2.9.6 The GINFMK Call
CALL GINFMK

(but, siz Lids] )

but
A buffer area for the return of the information.
siz

An integer containing the size of the buffer in words. If the buffer is declared as
an INT*2 array, then the size is equal to the number of elements in the array.
ids
An integer to receive the directive status.

Corrections to Documentation 2-19

See the description of the GI.FMK subfunction of the GIN directive in the
RSX-llM-PLUS and Micro/ RSX Executive Reference Manual for details on the
buffer format and use of this directive.

2.9.7 The GINGAS Call
CALL GINGAS

(but, siz , devnam , unit, udev , unum Lids] )

buf

A buffer area for the return of the information.
siz

An integer containing the size of the buffer in words. If the buffer is declared as
an INT*2 array, then the size is equal to the number of elements in the array.
devnam

A two-character ASCII device mnemonic.
unit

An integer containing the device unit number.
udev
unum

A device name and unit number pair, as described for devnam and unit. This
device name is used to identify a terminal for the interpretation of logical names.
ids

An integer to receive the directive status.

See the description of the GI.GAS subfunction of the GIN directive in the
RSX-llM-PLUS and Micro/ RSX Executive Reference Manual for details on the
buffer format and use of this directive.

2.9.8 The GINQMC Call
CALL GINQMC

(but, siz Lids] )

buf

A buffer area containing the command line.
siz

An integer containing the size of the buffer in words. If the buffer is declared as
an INT*2 array, then the size is equal to the number of elements in the array.
ids

An integer to receive the directive status.

See the description of the GI.QMC subfunction of the GIN directive in the
RSX-llM-PLUS and Micro/RSX Executive Reference Manual for details on the
buffer format and use of this directive.

2.9.9 The GINREN Call
CALL GINREN

2-20 Corrections to Documentation

(nam1, nam2 [,ids] )

nam1
nam2

Two integers that represent, respectively, the first and second half of a task name
expressed in RAD50.
ids

An integer to receive the directive status.

See the description of the GI.REN subfundion of the GIN directive in the
RSX-IIM-PLUS and Micro/RSX Executive Reference Manual for details on the
buffer format and use of this directive.

2.9.10 The GINSPR Call
CALL GINSPR

( flag [,ids] )

flag

An integer that is interpreted as true if not zero; false if zero.
ids

An integer to receive the directive status.

See the description of the GI.SPR subfunction of the GIN directive in the
RSX-IIM-PLUS and Micro/ RSX Executive Reference Manual for details on the
buffer format and use of this directive.

2.9.11 The GINTSK Call
CALL GINTSK

(buf, siz , nam1 , nam2 [,ids] )

buf

A buffer area for the return of the information.
siz

An integer containing the size of the buffer in words. If the buffer is declared as
an INT*2 array, then the size is equal to the number of elements in the array.
nam1
nam2

Two integers that represent, respectively, the first and second half of a task name
expressed in RAD50.
ids

An integer to receive the directive status.

See the description of the GI.TSK subfunction of the GIN directive in the
RSX-IIM-PLUS and Micro/ RSX Executive Reference Manual for details on the
buffer format and use of this directive.

2.9.12 The GINUAB Call
CALL GINUAB

(buf, siz , devnam , unit [,ids] )

Corrections to Documentation 2-21

See the description of the GI.UAB subfunction of the GIN directive in the
RSX-llM-PLUS and Micro/RSX Executive Reference Manual for details on the
buffer format and use of this directive.

2.9.13 The GINUIC Call
CALL GINUIC

(buf, siz [,ids] )

See the description of the GI.UIC subfunction of the GIN directive in the
RSX-llM-PLUS and Micro/ RSX Executive Reference Manual for details on the
buffer format and use of this directive.

2.9.14 The GINUPD Call
CALL GINUPD

(buf, siz [,ids] )

See the description of the GI.UPD subfunction of the GIN directive in the
RSX-llM-PLUS and Micro/ RSX Executive Reference Manual for details on the
buffer format and use of this directive.

2-22 Corrections to Documentation

2.9.15 The GINVEC Call
CALL GINVEC

(but, siz [,ids] )

buf

A buffer area for the return of the information.
siz

An integer containing the size of the buffer in words. If the buffer is declared as
an INT*2 array, then the size is equal to the number of elements in the array.
ids

An integer to receive the directive status.

See the description of the GLVEC subfunction of the GIN directive in
Section 1.3.4.2 in this document for details on the buffer format and use of
this directive.

2.9.16 Other Corrections to the RSX-11 M-PLUS and Micro'RSX Executive
Reference Manual
This section contains additional corrections to the RSX-llM-PLUS and
Micro/ RSX Executive Reference Manual.
•

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

[[efnl, last] 1

where:
efn

Event flag number

ast

Mark time AST address

Macro Expansion:
CMKT$
. BYTE
. WORD
.WORD

52., MRKAST
27. 3
f

52.
MRKAST

;CMKT$ 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
canceled and the DPB size equals 1.

•

In Section 5.18, page 5-53, please add the following text to the description of
the CRRG$ directive:
A 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 current available holes. If space
is not currently available, the directive will fail and return a status of IE.ALG
(- 84.)
Corrections to Documentation 2-23

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.
•

In Section 5.35, page 5-91, please delete the second bullet in the list. Add the
following text to the description of the EXTK$ directive:
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.
The restriction is overridden if the task is built with the new Task Builder
lEX 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.
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 occur only 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 that 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 lEX 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.
•

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

FE$NCT

63.

System has NCT support

FE$LSD

64.

System has LUT scan in LOADR disabled

FE$PC3

65.

System supports Professional 3xx series
personal computers

FE$DFB

66.

System supports deferred binding

FE$RTB

67.

Run time binding

FE$ODB

68.

Deferred binding is overridable

FE$XDJ

69.

XDT is doing VO via TPRs on KXJ

FE$NSY

70.

No local system disk

2-24 Corrections to Documentation

Meaning

•

Symbol

Value

Meaning

FE$NCO

7l.

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$A..l\TT

77.

Automatic network startup by save

FE$NRT

78.

Network remote system

FE$EXE

79.

Task file names default to .EXE

FE$CMO

80.

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
.WOR;)

MOD
TBMSK

;LOGICAL NAME MODIFIER
;LOGICAL Nfu~E TABLE INH=BIT 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 B~TES) OF EQUIVALENCE NAME

. WORD

RSIZE

; LOCATION OF SIZE FOR RETURNED EQUIVALENCS NfulVIE

.WORD

RTBMOD

;LOCATION OF LOGICAL TABLE NUMBER (LOWER BYTE) AND
;MODIFIER VALUE OF LOCATED LOGICAL N~ME (HIGHER BYTE)

.WOED

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.

Corrections to Documentation 2-25

ILOr\"S ;·10D TBr1SK ST~;I:~S LNS, LNSSZ ENS, ENSSZ F.SIZE RTBlvl0D
f

•

. BYTE
.BYTE

207.,10.
13.

;TLON$ MACRO DIC, DPB SIZE = 10(10) \AIORJS
;SUBFUNCTION VALUE (TLOG$ = 9(10))

.BYTE
. WORD

MOD
IBlvjSK

;LOGICAL NAME MODIFIER
; LOGICA_L NAME 'l'ABLJ;: INHIBIT ~~l\_SK

. WORD
. WORD

LNS
LNSSZ

;LOGICAL NAME STRING ARRAY
; SIZE (IN BYTES) OF LOGICAL NA.TvIE STRING

. WORD
.WORD

ENS
ENSSZ

;RETURNED EQUIVALENCE NA~E ARRAY
; SIZE (IN BYTES) OF EQUIVALENCE NA.TvIE

.\AIORD

RSIZE

;LOCATION OF SIZE FOR RETURNED EQUIVALENCE Nfu~E

.WORD

RTBMOD

;LOCATION OF LOGICAL TABLE NUMBER (LOWER BYTE) AND
; MODIFIER VALUE OF LOCATED LOGICAL N~lLME (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:
bufadr

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.
Modification to the IOSUB.MAC Module

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

You cannot build a shared region that contains I-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 I-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 memoryresident overlays unless the task was built with the Task Builder lEX switch.
(Please see the additional text for Section 5.35 explaining enhancements to
the EXTK$ directive. This text appears earlier in this section.) 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.

2-26 Corrections to Documentation

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 !MAXEXT or the DCL command SET EXTENSION_LIMIT
or the size of the partition in which it is running. (See the
RSX-llM-PLUS MCR Operations Manual or the RSX-llM-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.

•

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 represents APRO and bit 1
represents APRl. For example, if you specify 340, then APRs 7, 6, and 5 are
protected.
Macro Expansion
EXTM$
40,340
.BYTE
89. ,3
40
. WORD
. WORD
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.sue

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.

Corrections to Documentation 2-27

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 IMAXEXT 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. lOP

I/O is in progress for this task partition.

IE.CKP

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

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).
buf

Specifies address of buffer to receive the unit information.
siz
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, follmv 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.

2-28 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$
G1. DEV / DVBUF / DVSIZ / "TT / 1
. BYTE
169./6
. WORD G1. DEV
. WORD DVBUF
. WORD DVSIZ
. WORD
"TT
. WORD
1
DSW Return Codes
IS.SUC

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 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.
>TKB task=object

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.

2.10 RSX-11 M-PLUS and MicrdRSX Guide to Writing an 1/0 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

Corrections to Documentation 2-29

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 callable routine 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 2.4 of the manual:
Error logging support has been added for RA70 and RA90 disk drives (in
SA550 and SA650 arrays only).

•

Please add the following information to Section 4.5.12 of the manual:
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 that
is mounted foreign without an ACP. Previously, only hardware volume valid
could be set on volumes where software volume valid was already set.
This change is primarily intended to ease the handling of multivolume 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.

•

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, pages 8-24 and 8-25:

2-30 Corrections to Documentation

The example of the sample driver BMDRV.MAC is incorrect because the offset
LPRM+16 must be cleared if you cannot do buffered I/O. Replace the section
of code in the example with the following code:
**************************************************************

*
*
*

CONVERT TO BUFFERED I/O REQUEST

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

; COpy I/O PACKET ADDRESS BACK
R5,R3
MOV
+------------------------------------------------------------+
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

1. PRM+16 (R3)

INDICATE NO BUFFERED I/O

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

*
*
*

QUEUE THE CLOCK BLOCK

*
*
*

MOV

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

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

45$:

2.11 RSX-11M-PLUS and Micro'RSX 1/0 Drivers Reference Manual
Please make the following corrections to the RSX-llM-PLUS and Micro/RSX
I/O Drivers Reference Manual:
The format shown in Section 2.4.3 on page 2-23 is incorrect. The correct format
is:
QIO$C

IO.ATA

!TF.ESQ ]
!TF.NOT
,lun,[efn],[pri],[isb],,<[ast1 ],[parameter2],[ast2]>
[
!TF.XCC

The values listed for the RX33 drive in Table 4-1, page 4-2 are incorrect. The
correct characteristics for the RX33 drive are as follows:

Drive

Revolutions Per
Minute

Sectors

Tracks

Cylinders

Bytes/Drive

Decimal
Blocks

RX33

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.I0 for information on 11lessage-oriented
I/O function codes.
The information shown next should be added to the manual.

Corrections to Documentation 2-31

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.
While character-oriented interfaces can only be accessed indirectly through the
terminal driver, the DMCll and DUPll synchronous line interfaces allow 110
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.
DMC11 Synchronous Line Interface

The DMCll synchronous line interface provides a direct memory access interface
between two PDP-II computer systems using the DDCMP line protocol, thus
delivering high throughput and reliability while simplifying programming. The
DMCll supports nonprocessor request (NPR) data transfers of up to 8K 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 DMCll synchronous line interface also implements remote load
detect, allowing it to reinitialize a halted computer system.
DUP11 Synchronous Line Interface

The DUPll synchronous line interface is a single-line communications device
that provides a program-controlled interface between the PDP-II and a
serial synchronous line. The PDP-II 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 DUPll
and allows using the device in a switched or dedicated configuration. The DUPll
transmits data at a maximum rate of 9600 baud; this rate is limited by modem
and data set interface level converters.
The DUPll can be programmed to accept any sync character that you define. The
DUPll 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 messageoriented 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

2-32 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 I

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)

IOnly 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.

stadd

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

Corrections to Documentation 2-33

Format

Function

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.8YN, ... ,<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 UlOrmal 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.FOX 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.
IO.HOX 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.8YN 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 pI. A 0 indicates a primary station and a nonzero
indicates a secondary station. The DMC 11 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 DMC 11 in normal mode.

2-34 Corrections to Documentation

IO.lNL and IO.TRM Functions

The QIO$ functions IO.INL and IO.TRM have the same function code but
different modifier bits.
IO.lNL 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:

Data

S
Specifies a sync character.
H
Specifies a header character.

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.
IO.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 IO.FDX to set the characteristics
of the physical device unit.
IO.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.vVLB 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.

Corrections to Documentation 2-35

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
translnitter 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 checksumtype longitudinal redundancy check. A better approach to validating data is the
use of a cyclic redundancy check (CRC). A CRC 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 I/O 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 IO.KIL QIO$ for the device. Therefore, if
no message is transmitted by the remote system, a receive does not terminate,
and no further I/O can be performed on that device until the receive is killed by
issuing an IO.KIL QIO$.
You can use both half-duplex and full-duplex lines with the DMCII and DUPI1.
The mode is settable by using IO.FDX for full-duplex mode and IO.HDX for halfduplex 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 lIO 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 I/O request. When running
on a MicroPDP-II170, 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.

2-36 Corrections to Documentation

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-toback), should be read with an IO.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 DMC11

The DMCll currently cannot recover after a power failure because the randomaccess memory (RAM) in its internal microprocessor is erased when power fails.
Any I/O requests outstanding at the time of a power failure return the IE.ABO
status. These requests must be reissued after initializing the DMCll (IO.INL).
Importance of IO.lNL

After the type of communication line has been determined, and after IO.8YN
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 I/O outstanding on the other end of the line.
IO.INL must be issued after each IO.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

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

;USE LUNI FOR DPll
;SET DEVICE PARAMETERS
;PUT DEVICE ON LINE
;SEND A BLOCK

TXAST:

CMP3

BEQ

lOS

; Wll"S DATA CLOCKED OUT
; SUCCESSFULLY?
;IF SO, SET UP FOR NEXT
;BLOCK

IS.SDC&377,@(SP)+

Please add the following section to the manual:

Corrections to Documentation 2-37

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:
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 RA81
devices following a powerfail. Previously, RA81 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 IO.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.

2-38 Corrections to Documentation

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 PUC OM 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.

•

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.

•

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

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.
II

%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.
II

Key
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 i~ Lo sLor·e output.

Corrections to Documentation 2-39

Directive

Form

Operation

%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.

2.13 RSX-11M-PLUS and MicraRSX System Management Guide
•

Please add the following information to Section 1.3.3:
The print job attributes "Print adjacent to prior job" and "Print job should be
held" are included in [l,20]QMGBLD.BLD as valid options for $JATDF. This
global symbol determines the selection of default qualifiers for print jobs that
are spooled by the PRINT$ macro or the .PRINT subroutine in an application,
or by using the ISP switch in a command such as MAC 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 ISP
switch have no provision to modify the task-built defaults.

•

In Section 5.5.16, page 5-74, the third example is incorrect. The correct
example is as follows:
VMR>SET /SECPOL IRetum I
SECPOL=285. :640. :44%

Displays the amount of available secondary pool.
•

Please add the following to Section 7.9 and replace Example 7-6 with the
following example:
The RMD C displays now include percentages on load rates for writes. Also,
the format of these displays has been slightly modified to provide easier
reading.
>rmd c
RSX-llM-PLUS V4.4
12:56:54

EL78

Cache Statistics
Reg~on

Cache Region Name: CACHE
Device
Name

•

Total Hit Fail ~o3d
Reads Rate Rate Rate

(Ge~eral)

17-JA~-93

Size: 47440 (2500. disk blocks)
Total Hit Fail ::eo ad Jefe:"I::-ites F.ate Rate ?"ate :\ate

Total Cache
=/J Ops :jseo

DUO:

304725.

0%
0%

~O7309.

97%
0%

DU3:

:1%

82%
0%

0%
0%

C2C34. 95%
O. 0%

Tot.a~

104775.

9'7%

..c-:-3D9

82 %

4~2~'3

95%

Please add the following to Section 7.10 and replace Example 7-7 with the
following example:

2-40 Corrections to Documentation

The RMD D displays now include percentages on load rates for writes. Also,
the format of these displays have been slightly modified to provide easier
reading.
>:::-md d

RSX-11M-PLUS V4.4
12:57:04

BL78

Cache Statistics (Detailed)

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

Device Name: SYO:
Cache Status: Active
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%

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%

•

17 -JA.'J-93

90585.

314.

107309.
82%
0%
141176.

Deferred Write Rate
Write Load Rate

412042.

0%
0%

Please add the following information to Section 10.2.1:
Resource Accounting has been expanded to include a new qualifier to the
START command. The qualifier is:
/INTRUSION:yes/no

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 VDA-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:

Corrections to Documentation

2-41

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 writeprotected 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.
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.
110 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) ADD and MODIFY
commands behave differently. They 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 reencrypts the password.

2-42 Corrections to Documentation

2.14 RSX-11M-PLUS and Micro'RSX 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:
MOV (PC) + , - ( SP)
1---------------------------------1

1 ADDRESS OF PACKET (D-SPACE)

1---------------------------------1

1
JMP @.NAUTO
1
1---------------------------------1
PC RELATIVE OFFSET TO .NAUTO

I-SPACE PORTION
1

ADDRESS OF SEGMENT DESCRIPTOR

1--------------------------------1
1

ENTRY POINT ADDRESS

D-SPACE PORTION

•

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

•

The example on page 4-16 in Section 4.5 is incorrect. Please replace it with
the following correct example. (Note the corrected positioning of the period
(.) in first line of the example.)
MOV @#N.OVPT,RO
BISB #200,N.FAST(RO)

•

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>VSECT,VSECT/-SP=VSECT,LB: [l,l]F77FCS/LB
TKB> / IReturn I
Enter Options:

Return I

TKB >~AJI:JD~AJS =1 IReturn I

TKB>VSECT=MARRAY:160000:20000:200 IReturn I
TKB> / / IReturn I
>

Or, if you use the LINK command, use the following command sequence:
$ LINK/TAS/l'ii\P:VSECT/NOPRINT/OPT VSECT,LB: [1, 1]F77FCS/LIB IReturn I
Option? WNDWS=l IReturn I
Option? VSECT=MARRAY:160000:20000:200 IReturn I
Option? IReturn I
$

Corrections to Documentation

2-43

•

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

•

Replace the Note in Section 7.8.1 with the following text:
Prior to Version 4.3, 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.
•

In Section 8.6.1, page 8-8, the examples are incorrect. Please replace them
with the following correct examples:
TKB>CSM/ -HD/LI/PI, CMS/MA, CMS= IReturn I
TKB>LB: [1, 1] SYSLIB/INCLUDE:CMPAL, SY: [301,55]CSM
TKB>/ IReturn I
Enter Options:
TKB>STACK=O JRffiurnl
TKB>PAR=GEN:O:2000 JReturn I
TKB>CMPRT=$CMPCS JReturn I
TKB>BLGXCL=$SAVAL JReturn I
TKB> / / IReturn I
>

Or, you use the following LINK command sequence to build the same library:
TKB>LINK/TAS:CSM/NOH/SHARE:LIB/CODE:PIC/MAP:CSM/SYS/SYM:CSM/OPT IReturn I
->LB: [1, 1] SYSLIB/INCLUDE:CMPAL, SY: [301,55]CMS IReturn I
Option?STACK=O IReturn I
Option?PAR=GEN:O:2000 JReturnl
Option?CMPRT=$CMPCS JReturn I
Option?GBLXCL=$SAVAL IReturn I
Option? JReturnl
$

Also, add the following sentence to the end of page 8-8:
Note that the SYSLIB module CMPAL contains both the $CMPCS and
$CMPAL routines.
•

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

$ LlNKfTAS/MAP/SYM inputfile/INCLUDE:(MOD1 ,MOD2, ... MOD8)
Also, please remove note number 3. It is no longer applicable.
In note number 5, the example of the IINCLUDE qualifier is incorrect. The
correct example is as follows:
$ LINK/TAS/MAP/SYM INLIB1/INCLUDE: (MOD1,MOD2),- IReturn I
->inputfile2,INLIB1/LIBRARY IReturn I

•

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

2-44 Corrections to Documentation

•

In Section 12.4, page 12-8, the definitions of the device-name and unit-num8
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.
unit-numB

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 12.25, page 12-34 of the manual and
to Section 12.29, 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 supervisormode library. You should use the parameter code /SV 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

RESS UP

file-specification/[ -]SW[apr]

SUPLIB

file-specification/[-]SW[apr]

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
(l=NO)

Please add the following text to Section 12.34, 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 12.5 on page 12-10, Section
12.7 on page 12-13, and Section 12.21 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-IIM-PLUS and Micro! RSX Task Builder Manual.)

Corrections to Documentation 2-45

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[:bitmaskJ. 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 4.2 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

LlBR =name:accesscode[:baseAPR[:bitmask]]
RESLIB =file/accesscode[:baseAPR[:bitmask]]
CLSTR =lib1 ,lib2 ... Iibn: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 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
an 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.

2-46 Corrections to Documentation

Offset

Label
L$BTSK

Task
Name
Task

L$BPAR

Partition
Base Address of Task

L$BSA

L$BHGV

Highest Window 0 Virtual Address

L$BMXV

Highest Virtual Address in Task

L$BLDZ

Load Size in 64-Byte Blocks

L$BMXZ

Maxumum Size in 64-Byte Blocks

L$BOFF

Task Offset into Partition

L$BWNDIL$BSYS

System I.D.

L$BSEG

Size of Overlay Segment Descriptors

L$BFLG

Task Flag Word

L$BDAT

Task Creation Date --- Year

--- Month

--- Day

L$BLlB

I Number of Window Blocks'
Library Entry
Offset

LibrarylCommon

2 R$LNAM

Name

4 R$LSA

Base Address of 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

Library Offset into Region

20 R$LWND

Number of Library Window Blocks

22 R$LSEG

Size of Library Segment Descriptors

24 R$LFLG

Library Flag Word

26 R$LDAT

Library Creation Date --- Year

--- Month

--- Day

344

L$BPRI"

346

Task Priority

L$BXFW*

350

Task Transfer Address

L$BExr*

352

Task Extension (64-Byte Blocks)

L$BSGL**

354

Block Number of Segment Load List

L$BHRW*

356

Block Number of Header

L$BBLK**

360

Number of Blocks in Label

L$BLUW*

362

Number of Logical Units

L$BROB**

364

Relative Block of R-O Image

L$BROL**

366

RIO Load Size

L$BRDL'*

370

RIO Data Size in 32-Word Blocks

L$BHDB"

372

Relative Block Number of Data Header

L$BDHV**

374

High Virtual Address of Data Window 1

L$BDMV"

376

High Virtual Address of Data

L$BDLZ"

400

Load Size of Data

L$BDMZ"

402

Maximum Size of 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 of Seven
14-Word Entries in
RSX-11 M Systems and
Maximum of Fifteen
14-Word Entries in
RSX-11 M-PLUS Systems)

* Less library window blocks.
** If parameter L$BSYS is 4, you must add $LBXL(340 )
s
to the offset to determine the true offset.

MLO-003954

•

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

Corrections to Documentation

2-47

Label

Offset
Current Stack Pointer (R6)

H.CSP
H.HDLN

Header Length

H.SMAP/H.DMAP

User D-Space Mask

H.FMAP

Pointer to Fast Map Section

I Super D-Space Mask

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-Fail 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
Address of FCS Impure Storage

H.FCS

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

}

I Swapping Priority
I H.X25

H.LUTE

POinter to LUT Extension

H.GARD

Header Guard Word Pointer

H.NLUN

Number of LUNs

•

Low-Core
Context

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.

2.15 RSX-11M-PLUS and MicraRSX Utilities Manual
Please make the following corrections to the RSX-llM-PLUS Utilities Manual:
•

Please add the following supplemental information to Chapter 3 in the
description of the Backup and Restore Utility (BRU):
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.

2-48 Corrections to Documentation

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.
•

Please add the following information to Table 3-3:

Table 2-6 Summary of BRU Command Qualifiers
Command Qualifiers

Options

Defaults

/COMPARE

IDOUBLEBUFFER
/SINGLEBUFFER

IDOUBLEBUFFER with MUtype devices only. Otherwise,
/SINGLEBUFFER is the default.

NERIFY

IDOUBLEBUFFER
/SINGLEBUFFER

IDOUBLEBUFFER with MUtype devices only. Otherwise,
/SINGLEBUFFER is the default.

•

Please add the following information to Section 3.3.5 on page 3-13:
The BRU qualifier ICOMPARE has two options: DOUBLE BUFFER and
SINGLE BUFFER. The options are used only with MU-type devices when
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.
Note

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.

Specifying the DOUBLE BUFFER option may improve the performance of
streaming tape devices such as the TK50 cartridge tape device by double
buffering 110 operations. Specifying the SINGLEBUFFER option will cause
BRU to single buffer 110 operations, which was the default action for versions
of BRU prior to Version 4.l.
The format of the ICOMPARE qualifier is shown next.
ICOMPARE[:option]
Parameter
option
Specifies one of the following two options:

Corrections to Documentation 2-49

Option

Meaning

SINGLEBUFFER

If you use the SINGLEBUFFER option for a compare 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.

DOUBLEBUFFER

The DOUBLEBUFFER option can be used only on MUtype devices. For most system configurations with MU-type
devices, a compare 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
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>!COMPARE:SINGLEBUFFER

DUO:

MUD: IReturn I

Specifies a compare operation that reports a list of all actual errors found in
the data block or header block.
BRU>/COMPARE:DOUBLEBUFFER

DUO:

MU1: IReturn I

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.
•

Please add the following information to Section 3.3.33 on page 3-25:
The BRU NERIFY qualifier has two options: DOUBLEBUFFER and
SINGLEBUFFER. The options are used only with MU-type devices when
verifying information from a disk to a tape. The input device must be a disk
and the output device must be a tape or tapes.
Note

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 1/0 operations, which was the default action for versions
of BRU prior to Version 4.1.
The format of the NERIFY qualifier is:
/VERIFY[:option]

2-50 Corrections to Documentation

Parameter
option

Specifies one of the following two options:
Option

Meaning

SINGLEBUFFER

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

DOUBLEBUFFER

The DOUBLEBUFFER option can be used only on MU-type
devices. For most system configurations with MU-type devices,
a 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
SINGLE BUFFER option. As soon as an error occurs, BRU
reports the actual error. If there is any subsequent data in
the block, BRU cannot accurately 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>/VERIFY:SINGLEBUFFER DUO:

MUO: IReturn I

Specifies a verify operation that reports a list of all actual errors found in the
data block or header block.
BRU>/VERIFY:DOUBLEBUFFER DUO:

MU1: IReturn I

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.
•

Please add the following information to Section 3.10.3:

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 ICOMPA_"RE:SINGLEBUFFRR qualifier.

Corrections to Documentation 2-51

BRU-*WARNING*-Potential file header record verify error [directory]
filename. type ;version
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 /COMPARE:SINGLEBUFFER qualifier.
BRU-*WARNING*-Potential 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, 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 /COMPARE:SINGLEBUFFER qualifier.
BRU-*WARNING*-Double buffered compare or verify not supported
Explanation: The tape driver does not support the
/COMPARE:DOUBLEBUFFER or NERIFY:DOUBLEBUFFER qualifier.
User Action: No user action is required. BRU continues the verify or
compare operation in the single buffer mode.
BRU-*FATAL*-Device conflict
Explanation: This error message can be caused by the following problems:
•

You specified the /COMPARE: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 /COMPARE:SINGLEBUFFER qualifier
or the NERIFY:SINGLEBUFFER qualifier.
BRU-*FATAL*-Continuation volumes cannot be double buffered
Explanation: BRU does not support the /COMPARE:DOUBLEBUFFER
qualifier for backup sets that contain more than one tape.
User Action: Specify the /COMPARE:SINGLEBUFFER qualifier if you must
compare a backup set that consists of more than one tape.
•

Please add the following information to Section 12.1.3.1 on page 12-7:
PIP has been modified to use the embedded wildcard capability of the
FIIACP, instead of processing embedded wildcards internally. This makes
PIP 5 to 10 times faster when doing lookups of files with embedded wildcards.

2-52 Corrections to Documentation

In addition, PIP has been modified to accept embedded wildcards in a
directory specification. For example:
>dir lb: [rms*Jrms*.tsk
Directory lb: [RMS001001]
20-JAN-93 09:28
RMSRESAB.TSK;9
RMSLBL.TSK;9
RMSLBM.TSK;9

48.
19.
31.

C 30-JAN-93 13:57
C 30-JAN-93 13:57
C 30-JAN-93 13:57

Total of 98./98. blocks in 3. files
Directory lb: [RMS001054J
20-JAN-93 09:28
RMSCNV.TSK;5
RMSIFL.TSK;5
RMSRST.TSK;5
RMSBCK.TSK;5
RMSDES.TSK;5
RMSDSP.TSK;5
RMSDEF.TSK;5

176.
136.
191.

187.
248.
183.
139.

C
C
C
C
C
C
C

26-JAN-93 11:33
26-JAN-93 11:34
26-JAN-93 11:34
26-JAN-93 11:34
26-JAN-93 11:34
26-JAN-93 11:33
26-JAN-93 11:33

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

•

Please add the following information to Section 14.5 on page 14-9:
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.

2.16 RSX-11 M/M-PLUS and MicrdRSX 1/0 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

Corrections to Documentation

2-53

Normal Error Return
(Carry Bit and F.ERR)

No Error Return

.POSRC
.POSIT
.XQUIO
.RENAM

.EXTND
.TRNCL

The examples in Sections J.2 and J.3 are incorrect. The correct examples are as
follows:
Program CRCOPA

The following sample program is titled, CRCOPA. The CRCOPA program uses a
data-set descriptor instead of the default filename hlock used in CRCOPY.
.TITLE CRCOPA

; Card reader copy routine

. MCALL FDBDF$,FDAT$A,FDRC$A,FDOP$A,NMBLK$,FSRSZ$
. MCALL OPEN$R,OPEN$W,G~~S,pu:S,C~OSE$,EXIT$S
. MCALL FINIT$
,
INLUN=3
OUTLUN=4
;Assign CR or file device
;Assign to output dev~ce
FSRSZ$ 2
FDBOUT: FDBDF$
FDAT$A R.VAR,FD.CR
FDRC$A ,RECBUF,80.
FDOP$A CUTLUN,OFDSP'::"
FDBIN: FDBDF$
FDRC$A ,RECBUF,80.
?DOP$A IKLUN,IFDSP~
RECBUF: .BLKB
80.
OFDSP~:

.WORD
.iVORD
. fJO:<'D

2-54 Corrections to Documentation

O,G
D,G

;Device descriptor
;Cirectory descripto~

OXA~,(SZ, O;;-A~l

;?i~e~ame desc~i9to~

IFDSPT: .WORD
0,0
;Device descriptor
. WORD
0,0
;Directory descriptor
.WORD
INAMSZ,INAM
;Filenarne descriptor
ONAM:
.ASCII /OUTPUT.DAT/
ONAMSZ=.-INAM
. EVEN
INAM:
.ASCII /INPUT.DAT/
INAMSZ=.-INAM
. EVEN
START: FINIT$
;Init file storage region
OPEN$R #FDBIN
;Open the input file
BCS
ERROR
;Branch if error
OPEN$W #FDBOUT
;Open the output file
BCS
ERROR
;Branch if error
GTREC: GET$
#FDBIN
;Note - URBD is all set up
BCS
CKEOF
;Error should be EOF indication
MOV
F.NRBD(RO) ,R1
;R1=size of record read
MOV
#RECBUF,R2
;Strip trailing blanks
ADD
R1,R2
CMPB
#40,-(R2)
10$:
BNE
PTREC
SOB
Rl, 10$
iAt this point, R1 contains the stripped size of the
;record to be written. If the card is blank,
;a zero-length record is written.
PTREC:

PUTS
BCC

#FDBOUT"R1
GTREC

;R1 is needed to specify
;the record size

ERROR:

JMP

ERR SUB

;Jurnp to ERROR code

CKEOF:

DMPB
BNE

#IE.EFO,F.ERR(RO)
ERROR

;End of file?
;Branch if other error

CLOSES

#FDDIN

;Close the input file

BCS
CLOSES
BCS
EXIT$S
. END

ERROR
#FDBOUT
ERROR

;Close the output file
;Issue exit directive

START

Program CRCOPB

The following sample program is titled, CRCOPB. The CRCOPB program uses
run-time initialization of the File Descriptor Block (FDB).
.TITLE CRCOPB

; Card reader copy routine

. MCALL FDBDF$,FDAT$A,FDRC$A,FDOP$A,NMBLK$,FSRSZ$
. MCALL OPEN$R,OPEN$W,GET$,PUT$,CLOSE$,EXIT$S
. MCALL FINIT$
INLUN=3
OUTLUN=4
FSRSZ$ 2
FDBOUT: FDBDF$
FDBIN: FDBDF$
RECBUF: . BLKB
80 .
OFDSPT: . WORD
. WORD
. WORD

0,0
0,0

ONAMSZ,ONAM

;Assign CR or file device

;Device descriptor
;Directory descriptor
;Filenarne descriptor

Corrections to Documentation

2-55

IFDSPT: .WORD
. WORD
. ~AJORD

ONAM:
INAM:
START:

GTREC:

10$:

0,0
0,0
I~tL\t.'1SZ I INAl·1

;Device descriptor
;Directory descriptor
;Filename descriptor

.ASCII IOUTPUT.DATI
ONAMSZ=.-ONAM
. EVEN
. ASCII /INPUT. DAT /
INAMSZ=.-INAM
. EVEN
FINIT$
;Init file storage region
OPEN$R #FDBIN,#INLUN,#IFDSPT, ,#RECBUF,#80.
;Runtime initialization
BCS
ERROR
;Branch if error
FDAT$R #FDBOUT,#R.VAR,#FD.CR
;Runtime initialization
OPEN$W RO,#OUTLUN,#OFDSPT, ,#RECBUF,#80.
BCS
ERROR
;Branch if error
GET$
#FDBIN
;Note - URBD is all set up
BCS
CKEOF
;Error should be EOF indication
MOV
F.NRBD(RO) ,R1
;R1=size of record read
MOV
#RECBUF,R2
;Strip trailing blanks
ADD
R1,R2
CMPB
#40,-(R2)
BNE
PTREC
SOB
R1,10$

;At ttis point, R1 contains the stripped size of the
;record to be written. If the card is blank,
;a zero-length record is written.
PTREC:

PUTS
BCC

#FDBOUT, ,R1
GTREC

;R1 is needed to specify
;the record size

ERROR:

JMP

ERRSUB

; Jump to ERROR code

CKEOF:

CMPB
BNE

#IE.EFO,F.ERR(RO)
ERROR

;End of file?
; Branch if other error

CLOSES

#FDDIN

;Close the input file

BCS
CLOSES
BCS
EXIT$S
.END

ERROR
#FDBOUT
ERROR

;Close the output file
;Issue exit directive

START

2.17 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 ... "

•
•

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

2-56 Corrections to Documentation

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

•

In Appendix A, Section A.I, 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.

2.18 RSX-11M/M-PLUS RMS-11 User's Guide
Please add the following corrections to the RSX-llM / M-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 multistream rather than multichannel access to
write to the same indexed file. When RMS-II 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-II 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-II will act as
follows:
RMS-II 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-lllvI/lvI-PLUS lvlacro Programmer's Guide, R~vfS-ll
cannot perform an UPDATE operation on an alternate key with the key
characteristics CHANGES and NODUPLICATES. This description is
misleading. To clarify the description, please add the following information to
Chapter 6, Section 6.2.5.2:
Although RMS-II 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-II returns the completion code ER$DUP.
However, it does not terminate the UPDATE operation. Instead, RMS-II

Corrections to Documentation 2-57

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: [1,1] RMSLIB/LB: ROEXSY: ROIMPA- LB: [1, 1] 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 RSTSIE file access
listener (FAL) does not support remote record access to indexed files. This is
no longer true.

2.19 RSX-11M/M-PLUS RMS-11 Utilities
Please make the following corrections to the RSX-llM / M-PLUS RMS-ll
Utilities manual:
•

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

Format and Function

EXIT_SUPERSEDE

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

2-58 Corrections to Documentation

Command

Format and Function

SAVE_SUPERSEDE

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

•

In Chapter 2, 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 Chapter 2, 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 Chapter 2, 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.
•

In Chapter 2, 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 Chapter 2, 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-ll 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.

Corrections to Documentation 2-59

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-11 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-11 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 Chapter 2, 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 Chapter 4, Table 4-1, add the following switch and description to the table
of RMSCNV switches:
IER[:filespec]

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

•

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 terminal. 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.

2-60 Corrections to Documentation

•

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:
/NV

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.

•

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

INV
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:
/NV

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
charaeters permitted in the input file specification.

•

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

Corrections to Documentation 2-61

INV
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 nUlllber are the sanle,
RMSRST will display the following error message:
?RST-F-CREOUT,

~rro~

opening ddnn:file.dat;n as output

-RMS-E-ER$FEX, File a~ready exists

2-62 Corrections to Documentation

3
RMS-11 Version 2.0 Software Modifications,
Restrictions, and Supplementary Information
This chapter describes RMS-ll software modifications, lists restrictions, and
supplementary information for RSX-IIM-PLUS.
RMS-ll Version 2.0 has not changed version numbers for this release of
RSX-IIM-PLUS. Information in this chapter has not been incorporated into the
RMS-ll manuals.

3.1 RMS-11 Corrections
The following sections describe corrections for RMS-ll. 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.

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.

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

3.1.3 RMS-ll 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.
RMSIFL incorrectly reported exception records on packed decimal type records
containing negative key values. This problem has been corrected.
Backup sets created on RSX-11M, or octal-based RSX-11M-PLUS systems, were
restored with incorrect version numbers. RMSBCK incorrectly assumed that the
source file came from a decimal-based system. For example: file FOO.DAT;10 was
restored as FOO.DAT;12 even though both the input and output systems used
octal version numbers. When this same file was restored to an RSX-11M-PLUS
system that supported decimal version numbers, the version number remained
;10 rather than being converted to ;8. This problem did not exist if the input
system supports decimal version numbers. This problem has been corrected.

3.2 Software Restrictions
This section describes software restrictions that apply to RMS-11 Version 2.0.

3.2.1 RMS-ll Access Methods
RMS-11 tasks built prior to Version 3.0 of RSX-11M-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-11 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-11.

3.2.2 Restrictions to RMS-ll Remote Access Methods (RMSDAP)
The following sections describe restrictions to RMS-11 Remote Access Methods.
3.2.2.1 Incorrect Interpretation of Keysize Field Values
RMSDAP does not interpret a keysize field value of zero correctly. This causes
keyed access to a remote indexed file to fail. The application passes the value
zero for access to a nonstring key and issues the following error message:
ERSRNF (Record not found)

If your program is written in F77 or other high-level languages, you may have to
use a USEROPEN routine in order to be able to set this value.
3.2.2.2 Using RMSDAP on Your System
The RMSDAP provided on your system uses a 576 byte buffer for transfers
between your program's record/user buffer and your DECnet-RSX system network
buffer. This buffer must be large enough to hold the record and the DAP message
overhead (approximately 36 bytes). In certain rare cases, this buffer length may
be insufficient.

3-2 RMS-l1 Version 2.0 Software Modifications, Restrictions, and Supplementary Information

In particular, if you access a sequential variable file with a Maximum Record Size
(MRS) of zero (used when no maximum has been set), and the sum of the largest
record length (LRL) and the DAP message overhead exceeds the 576 byte length,
your record access will fail with the RMS error ER$MRS. This occurs because
RMSDAP functions at a DAP protocol level that does not allow reconfiguration
of buffersize without actually closing and reopening the file (and that is an
unacceptable alternative for RMSDAP). If it is within your control to create the
remote file, then an explicit MRS should be used. If not, the local RMSDAP buffer
can be lengthened by the system manager (prior to run time) depending on the
following considerations:
•

If your program was built with in-task RMSDAP, consult your map to
determine the location of the variable $BUF.M (current contents 1100 octal)
in the module RONFRT in your application. After saving a copy of your
application, ZAP the desired buffer length (including DAP overhead) into your
application. This change will only affect your application.

•

If your program was built against the DAPRES library, consult the map
(LB:[1,34]DAPRES.MAP) and determine the location of $BUF.M in module
RONFRT. Remove the DAPRES from memory. After saving a copy of
LB:[l,l]DAPRES.TSK, ZAP in the desired buffersize. Reinstall the DAPRES
library. This change will affect all programs built and run against the
DAPRES library.

In both cases the increased size of the buffer will be reflected in the size of your
RMSDAP applications.
Note

Network transfers rely on configuration of local RMSDAP and DECnetRSX network buffers as well as remote FAL and DECnet-RSX network
buffers. In all negotiations, the smaller buffer length is used. If your local
RMSDAP buffer size exceeds the local DECnet-RSX buffer, the remote
FAL, or the remote network buffer size the excess will be unusable, thus
effectively eliminating the workaround.

3.2.3 RMS-11 Utilities
This section describes restrictions to the RMS-11 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.
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)

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

3.3 RMS-11 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
EXlT_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-llM / 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.IDX 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-11 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.
The paragraph in Section 5.4.7 on page 5-43 in the RSX-llM-PLUS System
Generation Guide and Installation Guide is incorrect. Please replace it with the
following paragraph:
RMS-ll Version 2.0 is included on the pregenerated kit disk. The RMS-ll
segmented library (RMSRES, and RMSLBL through RMSLBM) and all
the RMS-ll utilities are already installed in the system image. No further
installation is needed, unless you install the DECnet package on your system and
you want to use the RMS-ll remote access facilities. See Section 5.4.1 in the
RSX-llM-PLUS System Generation Guide and Installation Guide for information
on installing the RMS-ll remote access package (DAPRES).
3.3.2.1

Location of RMS-11 Files on the Distribution Kit

Table 3-1 describes the location of RMS-ll Version 2.0 files on the
RSX-IIM-PLUS Version 4.4 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.

3-4 RMS-11 Version 2.0 Software Modifications, Restrictions, and Supplementary Information

Table 3-1

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

File Name

Destination

Comments

RMSMAC.MLB

LB:[l,l]

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

RMSLIB.OLB

LB:[l,l]

Object library for RMS-lllocal access

RMSDAP.OLB

LB:[l,l]

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

RMSFUN.OBJ

LB:[l,l]

RMS function routines for the system library

RMSRESAB.MAP

LB:[1,34]

Map for library root

RMSRESAB.TSK

LB:[l,l]

Installed as the root of the resident library

RMSFAKAB.CMD
RMSROTAB.STB

LB:[1,24]

For rebuilding RMSLBL and RMSLBM

RMSRESAB.CMD
RMSRESAB.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

RMSCNV.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

RMSll.ODL

LB:[l,l]

Prototype ODL file

RORMSl.MAC

LB:[l,l]

For use with the prototype ODL

RMSllS.ODL

LB:[l,l]

ODL file for sequential access

RMS12S.0DL

LB:[l,l]

ODL file for sequential access

RMSllX.ODL

LB:[l,l]

Standard indexed file ODL

RMS12X.ODL

LB:[l,l]

Indexed file ODL

RMSRLX.ODL

LB:[l,l]

ODL for use with RMSRES

RMSSLX.ODL

LB:[l,l]

ODL for use with RMSRES in supervisor mode

DAPllX.ODL

LB:[l,l]

ODL for use with overlaid RMSDAP

DAPRLX.ODL

LB:[l,l]

ODL for use with DAPRES

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 R:r"v'lSRES 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
(continued on next page)

RMS-11 Version 2.0 Software Modifications, Restrictions, and Supplementary Information 3-5

Table 3-1 (Cont.) RMS-11 Files on RSX-11 M-PLUS Distribution Kit
File Name

Destination

Comments

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

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
(continued on next page)

3-6 RMS-11 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

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:[l,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:[l,24]

For rebuilding the supervisor-mode version of
RMSRST

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-ll 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

RMS-11 Version 2.0 Software Modifications, Restrictions, and Supplementary Information 3-7

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$LIB), when
the missing segment is needed by RMS-ll.
There are several circumstances involving "inconsistent or incomplete
resident libraries" that can cause a BPT trap to be generated with RO
containing the error code ER$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=parname 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-IIM-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-ll utilities.
3.3.2.4 Utility Command and OOL 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

Cl'J-rvNON.CMD and CNVNON.ODL

RMSDSP

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

3-8 RMS-11 Version 2.0 Software Modifications, Restrictions, and Supplementary Information

Utility

Files

RMSDES

DESNON.CMD, DESNON.ODL, and RMSDES.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:
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

RMSDES

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 RMSCNV, 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 files and ODL files are not supplied with the RL02 kits.

3.3.2.5 Rebuilding the Utilities
To rebuild the utilities, perform the following steps:
1.

Set your default account to directory [1,24] on the system disk.

3. Use the Task Builder (TKB) to build the utility or utilities.

RMS-11 Version 2.0 Software Modifications, Restrictions, and Supplementary Information 3-9

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 RMSCNv.TSK. One is built to use the resident library and one is built
with disk-overlaid RMS-l1.

3-10 RMS-! 1 Version 2.0 Software Modifications. Restrictions, and Supplementary Information

4
Layered Product Support
4.1 Correction Files
Table 4-1 lists the layered products and correction files that have been updated
for the RSX-11M-PLUS Version 4.4 operating system. 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-11 BASIC-PLUS-2
Version 2.7

[211,200]

BP2

[211,200JBP2.DOC

DATATRIEVE-11IRSX
Version 3.3

[222,200J

DTR

[222,200JDTRREL.DOC

DCXKMV11-A
Development Tools
Version 1.03

[224,200J

DCX

[224,200]DCX.DOC

DECnet-11M-PLUS
Version 4.4

[230,200]
[225,20J
[230,20J

DECNET

[230,200]DECNET.DOC

KMV11-A HDLC
Framing Software
Version 1.02

[253,200J

KMV11

[253,200 JKMVl1.DOC

KMV11 X25 Link Level
Version 1.03

[322,200J

KXI

[322,200JKXI.DOC

PDP-11 FORTRAN-77
Version 5.4

[246,200J

F77

[246,200JF77.DOC

Layered Product

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

4,,3 Invalid BASIC-PLUS-2 Installation Warnina

BASIC-PLUS-2 Version 2.6 checks for the presence of the slow Task Builder
(STK) even though the installation does not need that utility. You may ignore any
warnings displayed when you install BASIC-PLUS-2 Version 2.6 regarding the
absence of the slow Task Builder. Version 2.7 of BASIC-PLUS-2 does not exhibit
this problem.

Layered Product Support 4-1

A
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.

•

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.

Reporting Problems A-1

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.
Consultations are requested.
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.

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 BUILDER-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 corrupted volume, supply output from the File
Structure Verification Utility (VFY) and dump of the volume. For improper
results, supply the error 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.

A-2 Reporting Problems

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 operating system, which includes changes and problems other people
have found and solved.

Reporting Problems A-3

B
Applying Corrections to Source Files
Interim changes to the Executive, MeR, 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.)
After 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.

B.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
(UIC) [11,10], submit the correction file to SLP. For example, you could follow this
sequence to create REQSB.MAC:
>SET /UIC=[11,40] IRetum!
>EDI REQSB.COR!Retuml
[Creating new file]
Input
REQSB.MAC;2/AU/-BF=REQSB.MAC;1

*EX
[Exi t] IRetum!
>SET /UIC=[11,10J IRetum!
>SLP @[11,40]REQSB.CORIRetum!

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:
>SET /JIC=[11,24] IRetum!
>MAC REQSB, [11,34] REQSB/ -SP= [1,1] EXEMC/ML, [11,10 j RSXMC/ PA: 1, ?EQSB!Retum!

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:
>SET /UIC=[l,24J IRetum!
>~BP. RSX~~M/P.P / - E?= [11, 24] ,?_EQSBIRetum!

Applying Corrections to Source Files 8-1

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 (ZZDRY.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:
>SET /UIC=[ll,24l IRetum!
>MAC ZZDRV= [1,1 1EXEMC/ML, [11,10 1RSXMC/ PA: 1, ZZDRVIRetum!
>SET /UIC=[1,24] IRetum!
>LBR RSX11M/RP/-EP=[ll,24lZZDRVIRetum!

Use [200,200]ZZDRVBLD.CMD to rebuild the driver. If necessary, copy
ZZDRV.TSK and ZZDRY.STB into the directory that corresponds to the system
UIC, or the library UIC if the driver is vectored.
For example:
>ASN SY:=OU:IRetum!
>TKB @[200,200lZZDRVBLDIRetum!
>SET /SYSUICIRetum!
SYSUIC=[g,ml
>SET /UIC=[g,mlIRetum!
>PIP /Wv=[1,54lzZDRV.TSK,ZZDRV.STBIRetuml

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.
For example:
>VMR !Retum !
Enter filename:RSX11MIRetum!
VMR>UNL ZZ:IRetum!
VMR>LOA ZZ:IRetum!
VMR> !CtrI/Z!
>RUN $SHUTUP IRetum!

If the driver is vectored, you must use the !VEC switch with the UNL and LOA
commands. In the preceding example, you would substitute the following VMR
commands:
VMR>UNL ZZ:/VECIRetum!
VMR>LOA ZZ: /VEC IRetum!
\TMR> !CtrI/Z!

Hardware boot the modified system.

8-2 Applying Corrections to Source Fiies

B.2 Updating an MeR 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:
>SET /UIC=[12,40] IRffiuml
>EDI SETOV.CORIRetuml
[Creating new file]
Input
SETOV.MAC;2/AU/-BF=SETOV.MAC;1

*EX IRetum I
[Exit]
>SET /UIC=[12,101 IRetum I
>SLP @[12,40]SETOV.CORIRetuml

2. Assume that directory [11,10] contains the RSXMC.MAC file resulting from
your target system generation and do the following:
>SET /UIC=[12,24] IRetum I
>MAC SETOV=[l,l]EXEMC/ML, [11,10]RSXMC/PA:1, [12, 101 SETOVIRetuml

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 of MCR ( .. . MCR)
and replace the module SETOV, use the following procedure:
>SET /UIC=[l,24] IRetuml
>LBR MCR/RP /NOEP= [12,24] SETOVIRetumI
>PIP SETOV.OBJ;*/DEIRetuml
>ASN SY:=MP:IRetuml
>ASN SY:=IN:IRetuml
>TKB @MCRBLDIRetuml
>VMR IRetum I
Enter filename: RSX11MIRffiumi
VMR> REM ... MCR IRetum I
VMR>INS [3, 54]MCRIRetuml
VMR> ICtrllZI

To rebuild the MCR dispatcher (MCR ... ) and to replace the module
MCRDIS, use the following procedure:
>SET /UIC=[l,24] IRetuml
>LBR MCR/RP /NOEP= [12,24] MCRDIS IRetum I
>PIP MCRDIS.OBJ;*/DSIRetuml
>ASN SY:=MP:IRetuml
>ASN SY:=IN:IRetuml
>TKB @MCDBLDIRetuml
>VMR IRetum I
Enter filename: RSXIIIvIIRelumi
VMR>REM MCR ... IRffiumj
VMR> INS [3, 54] MCD / XHR=NO IRetum I
VMR> ICtrllZI
>RUN $SHUTUPIRetuml

Applying Corrections to Source Files

8-3

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.

Incorporate the updated module into the task's object library.

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:
>SET /UIC=[l,24] IReturn I
>LBR INS/RP=[12,24]INSLBIReturnl

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:
>SET /UIC=[l,24] IRffiurn I
>ASN SY:=MP: IReturn I
>TKB @INSBLDIRffiurnl
>VMR IReturn I
Enter filename: RSXIIMIReturnl
VMR>REM ... INS IReturn I
VMR>INS [3,54] INS/ IOP=NOIReturnl
VMR> ICtrl/zl
>RUN $SHUTUP IReturn I

Hardware boot the system.

8.3 Updating a Del 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.
>LBR [l,24]DCLO/LIIReturnl
$LIBRARY/LIST [l,24]DCLOIReturnl

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 ill.odule COMl\1A.l.\TD.

8-4 Applying Corrections to Source Fiies

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:
>SET /UIC=[23,24] IRetum I
>MAC xXX= [11,10] RSXMC/PA: 1, [23,10] DCLMAC/PA: 1, XXXi Retum I
>SET /UIC= [1, 24]IRetuml
>LBR DCLO/RP=[23,24]xxxIRetuml
>TKB @DCLBLDIRetum!

If your CLI is DCL, use the following commands:
$ SET DEFAULT [23,34] IRetuml
$ MACRO [11,10] RSXMC/PASS: I, [23,10] DCLMAC/PASS: I, xxxi Retum I
$ SET DEFAULT [1,24] IRetuml
$ LIBRARY/REPLACE DCLO [23,24]xxxIRetuml
$ LINK @DCLBLDIRetuml

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):
>CLI /ELIM=DCLIRetuml
>REM DCL IRetum I
>INS $DCL/CLI=YESIRffiuml
>CLI /INIT=DCL IRetum I

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).

8.4 Applying Corrections to the File Systems (F11ACP)
All F11ACP 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:
>HELLO SYSTEMIRetuml
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):
>UF8 DBO: [13,10] IRetuml
>UFD DBO: [13,40J IRetum I
>BRU IRetum I
BRG"> !NOI~ITIALIZE/ SUPERSEDE/NOPRESER'J~/BACKUP _SE'::': MPBL40SRCI Retuml
From: MMO: [13, 10] F 11 PRE. MAC, TN'='RNl . MAC! Retum I
To: DBO: IRetuml
BRU> ICtrl/zl

Create the correction file, as follows:
>SET /UIC=[13,40] IRetum I
>EDT 'Ii'TRNl. CORIRetuml

Applying Corrections to Source Files 8-5

Apply the correction, as follows:
>SET /UIC= [13, 10llReturni
>SLP @[13,401wTRN1.CORIReturnl

5. Assemble the corrected module with the Executive macro library, the
Executive prefix file RSXMC.MAC, and the prefix file FIIPRE.MAC, as
follows:
>SET /UIC=[13,24] IReturn I
>MAC WTRN1=[1,llEXEMC/ML, [ll,lOlRSXMC, [13,lOlFllPRE,WTRNlIReturnl

Replace the defective module in the file control processor (FCP) library, as
follows:
>SET /UIC= [1, 24llReturni
>LBR FCP/RP=[13,24lWTRNlIReturnl
Modules replaced
vJTRNl

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:
>ASN NL:=MP:IRetuml
>TKB @FCPxxxBLD

Install the updated FCP in the system image as follows:
>RUN $VMR/Retuml
Enter filename: RSXllMIReturnl
v~R>REM FllACP/Retum/
VMR>INS [3,54) FCPxxx/PAR=GEN/IOP=NO/CKP=NOIRetum/
VMR> /CtrI/Z/
>RUN $SHUTUPIReturn/

Reboot the system to place the new FCP in use.

8.5 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-II on drive
DBO.

•

The distribution tape is mounted foreign on MMO.

8-6 Applying Corrections to Source Files

Perform the following steps to patch CNCMR:
1.

Boot your system, and log in to a privileged account.

If necessary, restore the required files from the distribution tape as follows:
>UFD DBO: [27,10] IReturn I
>UFD DBO: [27,40J IReturn I
>UFD DBO: [27,24 J IReturn I
>BRU IReturn I
BRU>/NOINITIALIZE/SUPERSEDE/NOPRESERVE/BACKUPSET:MPBL4OSRCIRffiurnl
From:MMO: [27,10JCNPRE.M~C,CNCMR.MACIRffiurnl
To: DBO: IReturn I
BRU -- Completed
BRU> ICTRUZ I

If you were correcting modnle HRONL, you would substitute the following for
the From: line in the previous command sequence:
From:MMO: [27, 10J HRPRE .MAC, HRONL . MAC IReturn I

Create the SLP correction file in VIC [27,40], as follows:
>SET /UIC=[27,40J IRffiurn I
>EDT CNCMR. COR IReturn I

4. Apply the SLP correction file to CNCMR as follows:
>SET /UIC=[27,10J IReturn I
>SLP @[27,40JCNCMR.COR IReturn I

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:
>SET /UIC=[27,24J IRffiurn I
>HAC CNCHR= [1, 1 J EXEMC /ML, [11, 10] RSXMC / PA: 1, [27 , 10 1CNPRE CNCMRI Return I
I

Replace the CNCMR object module in the OLR library as follows:
>SET /UIC=[l,24J IReturn I
>LBR OLR/RP/NOEP=[27,24JCNCMR.OBJIReturnl

Task build CON by using the updated library as follows:
>ASN DBO:=IN:IRffiurnl
>ASN DBO:=OU:IReturnl
>ASN NL:=MP:IReturnl
>TKB @CONBLD IReturn I
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.

Applying Corrections to Source Files 8-7

Install the updated reconfiguration utility in the system image as follows: For
CON, type the following:
>RUN $VMR IReturn I
Enter filename: RSXllMIReturnl
VMR> REM ... CON IReturn I
VMR>INS [3, 54jCONIReturni
VMR> Ictrl/zi

For HRC, type the following:
>RUN $VMRIReturnl
Enter filename: RSXllMIReturnl
VMR>REM HRC ... IReturn I
VMR>INS [3, 54]HRC/IOP=NOIReturnl
VMR> ICtrl/ZI

In order to use the new reconfiguration tasks, follow these steps:
a.

Shut the system down using the SHUTUP utility as follows:
>RUN $SHUTUP IReturn I

b. Reboot the system.

8.6 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

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.

8-8 Applying Corrections to Source Files

B.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.

B.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 UIC 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,lO]FCSPRE,DELETE

(FCSPRE.MAC defines necessary macros and global symbols.)
To assemble DELETE.MAC, you should set your UIC to [50,24] and enter the
following command line:
>MAC [50,24]DELETE, [50, 34]DELETE/-SP= [50, 10] FCSPRE,DELETEIReturnl

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.

FCSANSI.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 FCSANSI.CMD, then you should use the command
line you found in FCSANSI.CMD to assemble the source file. For example,
the following lines from FCSANSI.CMD assemble CLOSE.MAC:
[50,21] CLOSE . M'I}:," , [50,34 J CLOSE. MTA/ -SP=[50,lO]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 FCSANSI.CMD is also assembled by FCSNOANSI.CMD
and FCSMULBUF.CMD. Note the MTA file types above, which distinguish
between the three kinds of CLOSE objects.

Applying Corrections to Source Files 8-9

FCSNOANSI.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 FCSNOANSLCMD, then you
should use the command line you found in FCSNOANSI.CMD to assemble
the source file. For example, this command line from FCSNOANSI.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, 24J CLOSE .MBF, [50,34] CLOSE .MBF / -SP=[50,10]FCSANSI/PA:l,FCSMULBUF/PA:l,FCSBIGBUF/PA:l,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 FCSANSI.CMD,
which produces GETSQ.MTA:
[50,24]GETSQ.MTA, [50,34]GETSQ.MTA/-SP=[50,10]GPSEQ!PA:l,FCSANSI/PA:l,FCSBIGBUF/PA:l,FCSPRE,GET

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 FCSNOANSI.CMD, which
produces PUTSQ.NMT:
[50,24]PUTSQ.NMT, [50,34]PUTSQ.NMT/-SP=[50, 10]GPSEQ/PA:l,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, &~D OPENR.MAC.
OPFID.MAC produces the open-by-file-id module, OPFID. This is the command
line from FCSANSI.CMD, which produces OPFID.MTA:
[50, 24 ~ 0 PFI D. MT A, [5 0, 34 J 0 PF I J . ~n A/ - SP=[50,10: OPFID( PA: I, FCSAr;SI / FA: 1, FCS3IG3C? / PA: 1, ?CSPRE CPE:,T
I

8-10 Applying Corrections to Source Files

OPFNB.MAC produces the open-by-file-name block module, OPFNB. This is the
command line from FCSANSI.CMD, 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 CAST) 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.

8.9 Replacing the FCS Object Modules
Set your UIC 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:
>SET /UIC=[l,l] IReturn \
>PIP /NV=SYSLIB.OLBIReturn\
>LBR SYSLIB/RP= [50,24] DELETE. OBJ, PUT. MTA, PUTSQ .MTAJReturn\
>LBR SYSLIB/RP=[50,24]FINIT.OBJ,FINTSL.SUP/-EPIReturn\

If you use NOANSLIB.OLB, also use the following procedure:
>PIP /NV=NOANSLIB.OLBIReturn\
>LBR NOANSLIB/RP= [50,24] DELETE. OBJ, PUT .NMT, PUTSQ. NMTJReturn\
>LBR NOANSLIB/RP=[50,24]FINIT.OBJ,FINTSL.SUP/-EPIReturn\

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.

Applying Corrections to Source Files 8-11

Index
A
ABO (MCR) command
restriction, 1-27
ABORT (DCL) command
restriction, 1-27
Account Maintenance Program
See ACNT
ACF
corrections, 1-16
processor type and memory size on a
PDP-11/93 processor, 1-16
to installing Micro'RSX on a PDP-11/93,
1-17
ACNT
blank account fields, 2-42
ALLOCATION field
RMS-11 , 2-59
ALT command
keywords
/RPRI, 2-6
!TERM, 2-6
Alternate record key field
RMS-11 , 2-57
ASN (MCR) command, 2-6 to 2-8
ASSIGN (DCL) command
/QUEUE, 1-11
AST
user completion routines, B-11
Asynchronous system trap
See AST
Autoconfigure
See ACF

8
BACKUP (DCL) command
corrected example, 2-2
qualifiers
/CREATED, 2-2 to 2-3
lERROR_LIMIT, 2-5
!MODIFIED, 2-2 to 2-3
restrictions
!LENGTH qualifier, 2-5
/NEW_VERSION qualifier, 2-5
/REWIND/APPEND qualifiers, 2-5

BACKUP (DCL) command
restrictions (cont'd)
!TAPE_LABEL qualifier, 2-5
/VERIFY qualifier, 2-5
year 2000 support, 1-6
Backup and Restore Utility
See BRU
Bad Block Replacement Control Task
See RCT
Batch job submission, 1-20
correction, 1-23
IBR switch
RMS-11, 2-61
RMSDSP utility, 2-61
BROADCAST utility
year 2000 support, 1-6
BRU
corrections, 1-18
on files with no revision date, 1-18
tape errors, 1-19
to aborting with a TK50, 1-18
to creating the Master File Directory, 1-18
to dismount problem, 1-18
to error messages, 1-19
to help, 1-18
to using the /EXCLUDE qualifier with
wildcards, 1-18
to using the /REVISED qualifier, 1-18
to version numbers, 1-18
messages
error, 1-19, 2-51 to 2-52
help, 1-18
version numbers, 1-18
qualifiers
/COMPARE, 2-48 to 2-52
IDOUBLEBUFFER, 2-48 to 2-52
lEXCLUDE, 1-18
/REVISED, 1-18
/SINGLEBUFFER, 2-48 to 2-52
/VERIFY, 2-48 to 2-52
restrictions, 1-26, 1-27, 2-5
/ERROR_LIMIT qualifier, 2-5
/LENGTH qualifier, 2-5
/NEW_VERSION qualifier, 1-26, 2-5
on backing up directories that do not exist,
1-27
/REWIND/APPEND qualifiers, 2-5

Index-1

BRU
restrictions (cont'd)
!TAPE_LABEL qualifier, 2-5
use of wildcards, 1-26
NERIFY qualifier, 2-5
supplemental information, 2-48 to 2-52
year 2000 support, 1-6
BUCKETSIZE field
RMS-11 , 2-60

c
CALL GINABO, 2-18
CALL GINAPR, 2-18
CALL GINDEF, 2-18
CALL GINDEV, 2-18
CALL GINDVJ, 2-19
CALL GINFMK, 2-19
CALL GINGAS, 2-20
CALL GINQMC, 2-20
CALL GINREN, 2-20
CALL GINSPR, 2-21
CALL GINTSK, 2-21
CALL GINUAB, 2-21
CALL GINUIC, 2-22
CALL GINUPD, 2-22
CALL GINVEC, 2-23
Cancel Mark Time Request (CMKT$) directive,
2-23
CDA
enhancements to data caching, 1-14
CINT$ directive, 2-25
CLEAR ALL command
RMSDES utility, 2-59
CLI
using with /CL switch, 1-20
CLIIINIT command, B-5
CMKT$ directive, 2-23
Command line interface
See CLI
Communication driver
example, 2-37
programming hints, 2-35
Communication mode
setting full duplex <communication driver)~
2-34
Communication set operation
IO.8YN function (communication driver), 2-34
/COMPARE qualifier
BRU, 2-50, 2-51
Connect to Interrupt Vector (CINT$) directive,
2-25
Contiguous file areas
RMS-11 , 2-58
Continuation buckets
RMSDES utility, 3-4

Index-2

Controller identification
DSA driver, 2-38
Correction files
for DECnet-llM-PLUS, 4-1
for DATATRIEVE-11IRSX, 4-1
for DCX KMV11-A Development Tools, 4-1
for KMV11 X25 Link Level, 4-1
for KMV11-A HDLC Framing Software, 4-1
for layered products, 4-1
restriction on installing with named
directories, 4-1
for PDP-11 BASIC-PLUS-2, 4-1
invalid installation warning in Version 2.6,
4-1
for PDP-11 FORTRAN-77, 4-1
Crash device
restriction on specifying during SYSGEN, 1-26
Crash Device support, 1-23
Crash driver
corrections
to support for the J-11 UNIBUS systems,
1-24
to support for the TU80, 1-24
memory dump, 2-16
Crash dump
loadable support, 2-15
Crash Dump Analyzer
See CDA
Create Regions (CRRG$) directive, 2-23
Creating directory files
support for embedded blank characters, 1-22
CRRG$ directive, 2-23

D
Data caching, 1-14
corrections
file corruption, 1-14
possible system deadlock, 1-14
SET /CACHE display command, 1-14
to CDA, 1-14
Write-check or Read-check, 1-14
write loading temporary files, 2-42
Data late error correction, 1-25
DATATRIEVE-llIRSX correction file, 4-1
DCL
commands
ABORT restriction, 1-27
ASSIGN/QUEUE, 1-11
BACKUP
/CREATED, 2-2 to 2-3
!MODIFIED, 2-2 to 2-3
qualifiers, 2-5
restrictions, 2-5
year 2000 support, 1-6
DEASSIGN/QUEUE, 1-12
DELETE
/QUEUE, 1-12

DCL
commands (cont'd)
DIFFERENCES, 1-22
DIRECTORY
/SINCE, 1-6
FORMAT
WRITE:n, 2-2
INITIALIZE, 2-1 to 2-2
/HEADERS qualifier, 2-1 to 2-2
IMAXIMUM_FILES:n qualifier, 2-1 to

2-2
/QUEUE, 1-12
PRINT, 1-10, 1-11
SET
DEFAULTINONAMED, 4-1
SYSTEM/CRASH_DEVICE, 2-5,2-15
TERMINAL VT300-series restriction
2-1
'
TERMINAL/INQUIRE, 1-24
TERMINAL/model, 2-1
TIME, 1-6
SHOW
ASSIGNMENTS, 2-3 to 2-5
LOGICALS, 1-20, 2-3 to 2-5
QUEUE/ALL, 2-1
SORTIINDEXED_SEQUENTIAL, 1-22
START/QUEUE, 1-12
STOP/QUEUE, 1-12
updating source file, B--4
DCX KMV11-A Development Tools, 4-1
Deadlock
RMS-11 , 2-57
DEASSIGN/QUEUE (DCL) command 1-12
DECnet
'
changes from QMG, 1-3
DECnet-11M-PLUS correction file, 4-1
DELETE/QUEUE (DCL) command, 1-12
Device
allocation failure, 2-57
disconnecting (communication driver), 2-35
initializing (communication driver), 2-35
KDJ11-E processor, 1-1, 1-2
list of supported devices, 1-2
PDP-11/94, 1-24
RA71 disk drive, 1-1, 1-2
RA72 disk drive, 1-1, 1-2
RA92 disk drive, 1-1, 1-2
RQZXl SCSI adaptor, 1-1
RX02
enhanced support, 1-24
RX33 drive, 2-31
SCSI, 1-1
RQZXl adaptor, 1-1
RZ23L, 1-1
RZ24L, 1-2
TZ30, 1-2
TK25 cartridge tape, 2-16,2-17

Device driver
changes, B-1
updating loadable, B-2
DFL (MCR) command, 2-6 to 2-9
DIFFERENCES (DCL) command, 1-22
DIGITAL Command Language
See DCL
Directives, 2-17 to 2-29
Abort (GINABO), 2-18
Cancel Mark Time Request (CMKT$), 2-23
CINT$, 2-25
CMKT$, 2-23
Connect to Interrupt Vector (CINT$), 2-25
Create Region (CRRG$), 2-23
CRRG$, 2-23
$EDMSG, 2-39
Extend Task (EXTK$), 2-24, 2-26 to 2-28
EXTK$, 2-24, 2-26 to 2-28
EXTM$, 1-23, 2-24
FEAT$, 2-24
Find and Return Task Information (GINTSK),
2-21
Get Assigned Device Name (GINGAS), 2-20
Get Device Information (GINDEV), 2-18
Get Device Information Junior (GLDVJ), 2-28
to 2-29
Get Feature Mask Words (GINFMK), 2-19
Get LUN Information (GLUN$), 2-32
Get Partition Parameters (GPRT$), 2-29
Get Region Parameters (GREG$), 2-29
Get Son of Device (GINDVJ), 2-19
Get System APR (GINAPR), 2-18
Get System UIC Information (GINUIC), 2-22
Get User Account Block (GINUAB), 2-21
GLDVJ, 1-7 to 1-8
GLVEC, 1-8 to 1-10
GIN, 1-10
GINABO call, 2-18
GINAPR call, 2-18
GINDEF call, 2-18
GINDEV call, 2-18
GINDVJ call, 2-19
GINFMK call, 2-19
GIN GAS call, 2-20
GINQMC call, 2-20
GINREN call, 2-20
GINSPR call, 2-21
GINTSK call, 2-21
GINUAB call, 2-21
GINUIC call, 2-22
GINUPD call, 2-22
GINVEC call. 2-23
GIN$, 1-7
'
GLDVJ, 2-28 to 2-29
GLUN$, 2-32
GPRT$, 2-29
GREG$, 2-29

Index-3

Directives (cont'd)
Queue (MCR) Command Line (GINQMC), 2-20
Recursive Translation of Logical Name
(RLOG$), 2-25
Recursive Translation of Logical Name
(RLON$), 2-25
Rename Task (GINREN), 2-20
RLOG$, 2-25
RLON$, 2-25
SDUN$, 1-12
Set Task Default UIC (GINDEF), 2-18
Set Task Privilege (GINSPR), 2-21
Test for Specified System Feature (FEAT$),
2-24
Test for Specified Task Feature (TFEA$), 2-25
TFEA$, 2-25
TLOG$, 2-25
TLON$, 2-25
Translate Executive Entry Point Vector
(GINVEC), 2-23
Translate Logical Name String (TLOG$), 2-25
Translate Logical Name String (TLON$), 2-25
Update UICs and Default Directory (GINUPD),
2-22
VSUN$, 1-12
DIRECTORY/SINCE (DeL) command
year 2000 support, 1-6
Disk Save and Compress utility
See DSC
DMCll (communication driver)
message send, 2-37
powerfail, 2-37
synchronous line interface, 2-32
DSA driver
controller identification, 2-38
correction to hung tasks, 1-23
correction to timeout processing, 1-23
error log file, 2-38
error log support, 2-39
error logging support, 2-38
error packets, 2-38
I/O sequence number, 2-38,2-39
offset P.SEQ, 2-38
powerfail recovery
type, 2-38, 2-39
Read Serial Number function, 2-38
time delay, 2-38
DSC, 1-29
DU driver
Read Serial Number function, 2-38
setting to volume valid, 2-38
Suppress Error Logging subfunction, 2-38
DUPll (communication driver)
sync character, 2-32
synchronous line interface, 2-32

Index-4

E
$EDMSG directive, 2-39
file name string conversion, 2-39
ELI task
restrictions
INOLIM switch, 1-28
ISH switch, 1-28
Entering data
inputting lower-case characters, 1-23
IER switch
RMS-ll, 2-60
RMSCNV utility, 2-60
ER$FUL error, 2-57
ER$KEY error, 2-56
ER$MRS error, 2-56
ER$NOD error, 2-57
Error Logging
corrections, 1-21
to appending files, 1-21
to opening files for random access, 1-21
to volume not mounted packets, 1-21
file for DSA driver, 2-38
restrictions, 1-26
DU devices
improper display of register fields,
1-26
ELI
INOLIM switch, 1-28
ISH switch, 1-28
improper display of register fields, 1-26
in Year 2000 support, 1-26
MU devices
improper display of register fields,
1-26
to history summaries, 1-26
support
for PDP-1l/84, 1-21
for DSA driver, 2-38,2-39
for opening files, 1-21
for RA70 disk drives, 2-30
for RA71 disk drive, 1-2
for RA 72 disk drive, 1-2
for RA82 devices, 1-21
for RA90 disk drives, 2-30
for RA92 disk drive, 1-2
Error packets
DSA driver, 2-38
Errors
data late correction, 1-25
ER$FUL, 2-57
ER$KEY, 2-56
ER$MRS, 2-56
ER$NOD, 2-57
RMSDES utility. 3-4
RPT, 1-23

!ERROR_LIMIT qualifier
BACKUP (DCL) command, 2-5
Executive
modification to IOSUB.MAC, 2-26
modifying source file, B-1
restriction on shared regions, 2-26
support for volume valid, 2-30
Executive Debugging Tool
See XDT
EXIT_SUPERSEDE command
RMSDES utility, 2-58, 2-59, 3-4
Extend Task (EXTK$) directive, 2-24
APR protection mask, 2-26 to 2-28
EXTENSION field
RMS-ll, 2-60
EXTK$ directive, 2-24, 2-26 to 2-28
EXTM$ directive, 1-23,2-24

F
FllACP
year 2000 support, 1-6
FAL, 1-20
support for remote record access to indexed
files, 2-58
FCS
ANSI, B-8
assembly files, B-9
code size, 1-30
modifying modules, B-8
multibuffered, B-8
non-ANSI, B-8
READ$ function, B-ll
source file
conditional assembly, B-8
updating, B-9
VMS ACP function support, 1-29
WRITE$ function, B-ll
FEAT$ directive, 2-24
File Access Listener
See FAL
File Control Services
See FCS
File Exchange Transfer
See FLX
File loading
RIVIS-ll restriction, 3-3
File specification
RMS-ll, 2-61
wildcard, 2-61
RMSBCK utility, 2-61
RMSDSP utility, 2-61

FLX
correction on displaying the maximum number
of blocks on a tape, 1-22
support for year 2000 and beyond, 1-7

FMS
command file example, 2-43
.FORM directive
terminal support, 2-5
FORTRAN IV-PLUS, 2-43
FORTRAN-77 support, 2-43
Full-duplex mode
considerations (communication driver), 2-36
setting mode (communication driver), 2-34

G
GEN region name, 2-29
Get Device Information Junior function (GI.DVJ),
2-28 to 2-29
Get LUN Information (GLUN$) directive, 2-32
Get Partition Parameter (GPRT$) directive, 2-29
Get Region Parameters (GREG$) directive, 2-29
GI.DVJ directive, 1-7 to 1-8
GI.VEC directive, 1-8 to 1-10
GIN directives, 1-10
GINABO call, 2-18
GINAPR call, 2-18
GINDEF call, 2-18
GINDEV call, 2-18
GINDVJ call, 2-19
GINFMK call, 2-19
GINGAS call, 2-20
GINQMC call, 2-20
GINREN call, 2-20
GINSPR call, 2-21
GINTSK call, 2-21
GINUAB call, 2-21
Glr-..TUIC call, 2-22
GINUPD call, 2-22
GINVEC call, 2-23
GIN$ directive, 1-7
GI.DVJ function, 2-28 to 2-29
GI.GVJ subfunction, 1-1
GI.VEC subfunction, 1-1
high-level language calls, 1-1
GLUN$ directive, 2-32
GLUN$ macro, 2-32
get information
communication driver, 2-32
GPRT$ directive, 2-29
GREG$ directive, 2-29

H
Half-duplex mode
considerations (communication driver), 2-36
set mode (communication driver), 2-34
HELLO task, 2-9
HELP, 2-10
Help files, 1-20

Index-S

HFT, 1-1, 1-3
Hybrid Fast Task Builder
See HFT

I/O sequence number
DSA driver, 2-38,2-39
Indexed file
RMS-ll, 2-60
Indirect Preprocessor (lPP)
restriction on .END statements in an ENABLE
data block, 1-29
INI (MCR) command
IINF keyword, 2-10
guidelines, 2-10
IMXF keyword, 2-10
guidelines, 2-10
INITIALIZE (DCL) command
!FORMAT
DL-type device restriction, 2-2
!HEADERS qualifier, 2-1 to 2-2
!MAXIMUM_FILES qualifier, 2-1 to 2-2
guidelines, 2-2
IQUEUE, 1-12
IWRITE:n restriction, 2-2
Inputting lower-case characters, 1-23
Interface
message-oriented (communication driver), 2-32
IO.EIO request, 1-20
IO.FDX function (communication driver), 2-34
IO.HDX function
set mode (communication driver), 2-34
IO.INL function (communication driver), 2-35
after sync, 2-37
IO.RLB function (communication driver), 2-35
IO.RNS function (communication driver), 2-35
IO.SYN function (communication driver), 2-35
set operation, 2-34
with IO.INL, 2-37
IO.TRM function (communication driver), 2-35
IO.WNS function (communication driver), 2-35
IOSUB.MAC module
modification, 2-26

K
KDJll-E processor, 1-2
KMVll X25 Link Level, 4-1
KMVll-A HDLC Framing Software, 4-1
KWll-P clock, 1-20

Index-6

L
Layered products
correction files, 4-1
DATATRIEVE-llIRSX, 4-1
DCX KMVll-A Development Tools, 4-1
DECnet-llM-PLUS, 4-1
KMVll X25 Link Level, 4-1
KMV11-A HDLC Framing Software, 4-1
PDP-11 BASIC-PLUS-2, 4-1
PDP-ll FORTRAN-77, 4-1
restriction on installing with named
directories, 4-1
installing
PDP-11 BASIC-PLUS-2
invalid installation warning in Version
2.6, 4-1
Line continuation support, 1-13
LOA (MCR) command, 2-11
Loadable XDT enhancement, 1-13
Logical unit number
See LUN
Low-traffic sync (communication driver), 2-37
LUN
get information
communication driver, 2-32

M
MACRO
year 2000 support, 1-6
MACRO .LIMIT feature, 1-21
Macros
GLUN$, 2-32
QIO$, 2-33 to 2-35
functions
IO.FDX, 2-34
IO.HDX, 2-34
IO.INL, 2-35
IO.RLB, 2-35
IO.RNS, 2-35
IO.SYN, 2-35
IO.TRM, 2-35
IO.WLB, 2-35
IO.WNS, 2-35
QIO$C, 2-31
MCR
command line interface
year 2000 support, 1-6
commands
ABO restriction, 1-27
ALT
IRPRI keyword, 2-6
/TERM keyword, 2-6
ASSIGN (ASN), 2-6 to 2-8
BYE
year 2000 support, 1-6

MCR
commands (cont'd)
DEFINE LOGICALS (DFL), 2-6 to 2-9
HEL
year 2000 support, 1-6
INITIALIZE VOLUME (IN!)
/INF qualifier, 2-10
/MXF qualifier, 2-10
LOAD (LOA), 2-11
OPENREGISTER (OPE)
/TASKD qualifier, 2-11
/TASKI qualifier, 2-11
PIP, 2-15
PRI, 1-10,1-11
QUE
lAS, 1-11
ICR, 1-12
IDEA, 1-12
IDEL, 1-12
ISTA, 1-12
ISTO, 1-12
RUN
!PRI keyword, 2-13
IUIC keyword, 2-12, 2-13
SET
ICRASHDEV qualifier, 2-14, 2-15,
2-16
ICRASH_DEVICE qualifier, 2-14,
2-16
ICTRLC qualifier, 2-14
IDEF, 1-24
IDPRO qualifier, 2-14
/INQUIRE qualifier, 2-14
IINTRUSION qualifier, 2-13
INONAMED, 4-1
ISECPOL qualifier, 2-14
/TERM qualifier, 2-14
IUIC, 1-24
TIM
year 2000 support, 1-6
updating source file, B-4
Message-oriented interface (communication
driver), 2-32
Mode
maintenance (communication driver), 2-34
set (communication driver), 2-34
Monitor Console Routine
See MCR
MSCP support
RCT, 2-41
Multiplexers
parity correction, 1-21

N
INOPRINT switch, 1-20
/NV switch
RMSBCK utility, 2-61
RMSRST utility, 2-61

o
Object Time System
See OTS
ODL files
RMS-ll, 3-8 to 3-9
ODT
correction with I&D systems, 1-24
Offset P.SEQ
DSA driver, 2-38
OPE (MCR) command
/TASKD keyword, 2-11
/TASKI keyword, 2-11
OTS Fast Map routine
lOT restriction, 1-30
Overlaid libraries
Manually loading, 2-44

p
Parity
with multiplexers, 1-21
PDP-ll BASIC-PLUS-2
invalid installation warning in Version 2.6, 4-1
PDP-11 BASIC-PLUS-2 correction file, 4-1
PDP-11 FORTRAN-77 correction file, 4-1
Peripheral Interchange Program
See PIP
PIP
corrections, 1-19
to checking directory specifications, 1-19
to creating a directory, 1-19
to issuing the PIPILI command, 1-19
to wildcard support, 1-19
qualifiers
IDD, 2-15
ILl, 1-19
restriction
on creating contiguous files, 1-29
support
for creating directories, 1-19
for wildcards, 1-19, 2-52
year 2000 support, 1-6
Powerfail
DMC11 (communication driver), 2-37
Powerfail recovery
DSA driver, 2-38, 2-39

Index-7

Pregenerated kit
RMS-11 , 2-15, 3-4
PRINT command
DCL, 1-11
command enhancements 1-1, 1-10
MCR
'
PRI command enhancements, 1-1, 1-10,
1-11
Printing
with a LAT port, 1-22
Problem reporting, A-I
information to be supplied, A-2 to A-3
Programming hints
communication driver, 2-35

Q
QIO request
IO.EIO, 1-20
QIO$ macro, 2-33 to 2-35
communication driver, 2-33
device-specific function
communication driver, 2-33
functions
IO.FDX, 2-34
IO.HDX, 2-34
IO.INL, 2-35
IO.RLB, 2-35
IO.RNS, 2-35
IO.SYN, 2-35
IO.TRM, 2-35
IO.WLB, 2-35
IO.WNS, 2-35
standard function
communication driver, 2-33
QIO$C macro, 2-31
QMG
See also Queue Manager
changes to DECnet, 1-3
QUE (MCR) command, 1-10
lAS, 1-11
ICR, 1-12
IDEA, 1-12

IDEL, 1-12
ISTA, 1-12
ISTO, 1-12

Queue Manager, 2-40
restriction on printers, 1-29
support for years after 1992, 1-1, 1-2
year 2000 support, 1-6

R
RA71 disk drive, 1-1
RA72 disk drive, 1-1, 1-2
RA92 disk drive, 1-1, 1-2

Index-8

RCT
MSCP support, 2-41
write-protected disks, 2-42
Read operation
logical block (communication driver), 2-35
sync character (communication driver), 2-35
Rebuilding utilities
RMS-ll, 3-9
Record Management Services
See RMS-11
Recursive Translation of Logical Name (RLOG$)
directive, 2-25
Recursive Translation of Logical Name (RLON$)
directive, 2-25
Redundancy checking (communication driver),
2-36
Report Generator Task
/HISTORY qualifier, 2-16
Reporting problems, A-I
information to be supplied, A-2 to A-3
Resource Monitoring Display
See RMD
RL02 installation kit, 3-4
RLOG$ directive, 2-25
RLOG$ macro expansion, 2-25
RLON$ directive, 2-25
RLON$ macro expansion, 2-25
RMD
BOTTOM command, 2-42
display restriction, 1-28
TOP command, 2-42
Writes Load Rates, 2-40
RMS-ll
accessing files on remote nodes, 3-9
alternate record key field, 2-57
building utilities, 3-9
contiguous file areas, 2-58
creating a single-area indexed file, 2-60
creating indexed file, 2-60
deadlock, 2-57
errors
ER$DNF, 3-2
ER$FUL, 2-57
ER$KEY, 2-56
ER$MRS, 2-56
ER$NOD, 2-57
FAL support for remote record access to indexed
files, 2-58
fields
ALLOCATION, 2-59
BUCKETSIZE, 2-60
EXTENSION, 2-60
file loading restriction, 3-3
file specification, 2-61, 3-1
files on the distribution kit, 3-4 to 3-7
installing, 3-4
local access, 3-1

RMS-U (cont'd)
ODL files, 3-8 to 3-9
pre generated kit, 2-15, 3--4
rebuilding utilities, 3-9
remote access, 3-2
restrictions
nonamed directory mode, 3-2
remote access methods, 3-2 to 3-3
RMSCNV utilities, 3-3
RMSDEF utilities, 3-3
to file loading; 3-3
to the DEF command, 3-3
to the DEFINE command, 3-3
RMSDAP utility
configuring for your system, 3-2
small systems, 3--4
startup command procedure, 3-7 to 3-8
task building against resident library, 2-58
TKB cluster option syntax, 2-58
UPDATE operation, 2-57
utilities, 3-2
command files, 3-8 to 3-9
configurations, 3-8
ODL files, 3-8 to 3-9
rebuilding, 3-9
RMSBCK
file specifications, 2-61
INV switch, 2-61
RMSCNV
fER switch, 2-60
restrictions, 3-3
RMSDAP
configuring for your system, 3-2
keysize field, 3-2
RMSDEF
restrictions, 3-3
RMSDES
CLEAR ALL command, 2-59
continuation buckets, 3--4
enhancements, 3--4
error messages, 3--4
EXIT_SUPERSEDE command, 2-58,
2-59, 3--4
SAVE command, 2-59
SAVE_SUPERSEDE command, 2-59,
3--4
RMSDSP
IlHt switch, 2-61
command line format, 2-61
file specification, 2-61
/SU switch, 2-61
RMSIFL
enhancements, 3-2
RMSRST
command line format, 2-61
INV switch, 2-61

RMSBCK utility
file specification, 2-61
INV switch, 2-61
RMSCNV utility
fER switch, 2-60
file loading restriction, 3-3
RMSDAP utility
keysize field, 3-2
RMSDEF utility
DEF command restriction, 3-3
DEFINE command restriction, 3-3
RMSDES utility
CLEAR ALL command, 2-59
continuation buckets, 3--4
error messages, 3-4
EXIT_SUPERSEDE command, 2-58, 2-59, 3-4
SAVE command, 2-59
SAVE_SUPERSEDE command, 2-59,3--4
RMSDSP utility
IBR switch, 2-61
command line format, 2-61
file specification, 2-61
/SU switch, 2-61
RMSIFL utility
enhancements, 3-2
RMSRST utility
command line format, 2-61
INV switch, 2-61
RPT
corrections to undefined symbol error, 1-23
/DATE qualifier
year 2000 support, 1-6
/HISTORY qualifier, 2-16
RUN (MCR) command
!PRI keyword, 2-13
IUIC keyword, 2-12, 2-13
RX33 drive, 2-31

s
SAVE command
RMSDES utility, 2-59
SAVE_SUPERSEDE command
RMSDES utility, 2-59, 3-4
SCSI
support, 1-1
supported devices, 1-1
RQZXl adaptor, 1-1
RZ23L, 1-1
RZ24L, 1-2
TZ30, 1-2
SDUN$ directives, 1-12
Send directive, 1-12
SET (DCL) command
DE FAULTINONAMED, 4-1
SYSTEM/CRASH_DEVICE command, 2-5,
2-15
TERMINAL

Index-9

SET (DCL) command
TERMINAL (cont'd)
/INQUIRE
VT420 support, 1-24
/INQUIRE command
VT300-series support, 2-1
I model, 2-1
VT300-series support, 2-1
TIME
year 2000 support, 1-6
SET (MCR) command
/CRASHDEV, 2-14,2-15,2-16
/CRASH_DEVICE, 2-14,2-16
/CTRLC, 2-14
IDEF
support for angle bracket characters, 1-24
IDPRO, 2-14
/INQUIRE, 2-14
/INTRUSION, 2-13, 2-41
INONAMED, 4-1
/SECPOL, 2-14
!TERM, 2-14
IUIC
support for angle bracket characters, 1-24
Set mode
half duplex (communication driver), 2-34
IO.HDX function (communication driver), 2-34
Shadow Recording
corrections, 1-17
to invalid logical block numbers, 1-17
to memory protect violations, 1-17
to user routines, 1-17
Shared regions
1- and D-space restriction, 2-26
SHOW (DCL) command
ASSIGNMENTS, 2-3 to 2-5
LOGICALS, 1-20, 2-3 to 2-5
QUEUE
/ALL qualifier, 2-1
SHUTUP utility, B-8
correction, 1-19
SLP
correcting source files, B-1
Software Performance Report
See SPR
SORT command
/INDEXED_SEQUENTIAL qualifier, 1-22
Source file
correcting, B-1
updating, B-1
Source Language Input Program
See SLP
SPR
category, A-2
completing, A-I
priority, A-I

Index-10

START/QUEUE (DCL) command, 1-12
STOP/QUEUE (DCL) command, 1-12
/SU switch
RMS-ll, 2-61
RMSDSP utility, 2-61
SUBMIT (DCL) command
enhancements, 1-1, 1-10
Switch
INOPRINT, 1-20
Sync character (communication driver), 2-35
specifying, 2-35
SYSGEN
corrections, 1-14, 2-15
abnormal terminations of SYSGEN, 1-16
con displays, 1-15
invalid questions on processor-type, 1-15
looping if more than 4 units existed on a
DU-type controller, 1-14
overriding ACF results, 1-16
to RMS-ll pregenerated kit, 2-15
to selecting FCPMDL, 2-15
restrictions, 1-25
incorrect displays for multiple MM devices,
1-26
to selecting FCPMDL, 1-25
specifying nonexisting devices as the crash
device, 1-26
System library
using with TKB, 1-30

T
Task
extending, 2-26 to 2-28
hanging when attaching to a terminal, 1-28
Universal Receiver Task, 1-13
URT, 1-13
Task Builder
See also Hybrid Fast Task Builder
See TKB
TDX
support for line continuation, 1-1, 1-13
support for passing command lines, 1-1, 1-13
Terminals
unused ports, 1-28
Test for Specified System Feature (FEAT$)
directive, 2-24
Test for Specified Task Feature (TFEA$) directive,
2-25
TFEA$ directive, 2-25
Time delay
DSA driver, 2-38
TKB
See also Hybrid Fast Task Builder
/-SH restriction, 1-28
allocating library APRs, 2-45
building new FCP, B-6

TKB (cont'd)
CLSTR option, 2-46
cluster option syntax, 2-58
command file, B-3
compatible library requirement, 1-30, B-11
corrections
to error message, 2-48
to the /INCLUDE qualifier, 2-44
to the /SAVE qualifier, 2-44
EXTK$ task directives, 2-24
EXTM$ task directives, 2-24
FMS command file example, 2-43
illegal APR reservation message, 2-48
ILl subswitch, 2-46
LIBR option, 2-46
RESLIB option, 2-46
RESSUP option
new /SW parameter code, 2-45
restrictions, 1-30
OTS Fast Map routine, 1-30
RMS-ll resident library, 2-58
SUPLIB option
new /SW parameter code, 2-45
support for the EXTM$ directive, 1-23
support for the MACRO .LIMIT directive, 1-23
task-build command file, B-6
using system library, 1-30
VSECT option, 2-45
TLOG$ directive, 2-25
TLOG$ macro expansion, 2-25
TLON$ directive, 2-25
TLON$ macro expansion, 2-25
Translate Logical Name String (TLOG$) directive,
2-25
Translate Logical Name String (TLON$) directive,
2-25
Transmission
validation (communication driver), 2-36

u
UFD
correction, 1-22
corrections
to creating directory files, 1-22
support for embedded blank characters, 1-22
Universal Receiver Task, 1-13
support, 1-1
Unstop directive, 1-12
UPDATE operation
RMS-ll correction, 2-57
URT, 1-1, 1-13

v
/VERIFY qualifier
BRU, 2-50,2-51
VFY
aborting, 2-53
switches
IDE, 2-53
IDR, 2-53
IRE, 2-53
!UP, 2-53
warning message, 2-53
Virtual Monitor Console Routine
SeeVMR
VMR
INSTALL command, B-5
REMOVE command, B-5
updating a loadable driver, B-2
VMS ACP function support
FCS, 1-29
VSUN$ directives, 1-12
VT420 terminal
enhanced support, 1-24

w
Write operation
no sync (communication driver), 2-35

x
XDT
loadable enhancement, 1-1, 1-13
loadable support, 2-16

v
Year 2000 support, 1-1
BROADCAST utility, 1-6
BRU, 1-6
date format, 1-5
DCL commands
BACKUP, 1-6
DIRECTORY/SINCE, 1-6
SET TIME, 1-6
FllACP, 1-6
FLX, 1-7
MACRO, 1-6
MCR
command line interface, 1-6
commands
"RVJ<'
LJ ..1.. .......... ,

1-6

REL, 1-6
TIM, 1-6
PIP, 1-6
Queue Manager, 1-6
restrictions, 1-26

Index-11

Year 2000 support
restrictions (cont'd)
error logging, 1-26

RPT
!DATE qualifier, 1-6

Index-12

How to Order Additional Documentation

Technical Support
If you need help deciding which documentation best meets your needs, call 800-DIGITAL (800-344-4825)
and press 2 for technical assistance.

Electronic Orders
If you wish to place an order through your account at the Electronic Store, dial 800-234-1998, using a
modem set to 2400- or 9600-baud. You must be using a VT terminal or terminal emulator set at 8 bits, no
parity. If you need assistance using the Electronic Store, call 800-DIGITAL (800-344-4825) and ask for an
Electronic Store specialist.

Telephone and Direct Mail Orders
From

Call

Write

U.S.A.

DECdirect
Phone: 800-DIGITAL
(800-344-4825 )
FAX: (603) 884-5597

Digital Equipment Corporation
P.O. Box CS2008
Nashua, New Hampshire 03061

Puerto Rico

Phone: (809) 781-0505
FAX: (809) 749-8377

Digital Equipment Caribbean, Inc.
3 Digital Plaza, 1st Street
Suite 200
Metro Office Park
San Juan, Puerto Rico 00920

Canada

Phone: 800-267-6215
FAX: (613) 592-1946

Digital Equipment of Canada Ltd.
100 Herzberg Road
Kanata, Ontario, Canada K2K 2A6
Attn: DECdirect Sales
Local Digital subsidiary or
approved distributor

International
Internal Orders l
(for software
documentation)

DTN: 241-3023
(508) 874-3023

Software Supply Business (SSB)
Digital Equipment Corporation
1 Digital Drive
Westminster, MA 01473

Internal Orders
(for hardware
documentation)

DTN: 234-4325
(508) 351-4325
FAX: (508) 351-4467

Publishing & Circulation Services
Digital Equipment Corporation
NR02-2
444 Whitney Street
Northboro, MA 01532

leall to request an Internal Software Order Form (EN-01740-07).

Reader's Comments

RSX-11 M-PLUS
Release Notes

AA-H427 J-TC

Your comments and suggestions help us improve the quality of our publications.
Thank you for your assistance.

I rate this manual's:
Accuracy (product works as manual says)
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

Good

Fair

Poor

0
0
0
0
0
0
0
0

I would like to see morelless

What I like best about this manual is

What I like least about this manual is

I found the following errors in this manual:
Page

Description

Additional comments or suggestions to improve this manual:

For software manuals, please indicate which version of the software you are using:

Nametritle

Dept.

Company

Date

Mailing Address
Phone

Do Not Tear - Fold Here and Tape
No Postage
Necessary
if Mailed
in the
United States

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

DIGITAL EQUIPMENT CORPORATION
Information Design and Consulting
NU01-1/G10
55 NORTHEASTERN BLVD.
NASHUA, NH 03062-3187

111 ••••• 11.11 •••• 11 •••• 1.1" 11 •••• 111 •• 1.1 ••• 111 ••• 1
Do Not Tear - Fold Here - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -