Digital PDFs

AA-L412A-TB

May 1982

121 pages

Original

4.0MB

Document:	AA L412A Tape Processing Manual Jul82
Order Number:	AA-L412A-TB
Revision:	0
Pages:	121
Original Filename:	AA_L412A_Tape_Processing_Manual_Jul82.pdf

OCR Text

TOPS-10
Tape Processing Manual
AA-L412A-TB

July 1982
This manual describes the user interface to the TOPS-10 tape
subsystem. The use of labeled and unlabeled tapes is
addressed.
This is a new manual.

OPERATING SYSTEM:

TOPS-10 V7.01
GALAXY V4.1

SOFTWARE:

Software and manuals should be ordered by title and order number. In the United States, send orders
to the nearest distribution center. Outside the United States, orders should be directed to the nearest
DIGITAL Field Sales Office or representative.

Northeast/Mid-Atlantic Region

Centr·al Region

Digital Equipment Corporation
PO Box CS2008
Nashua, New Hampshire 03061
Telephone:(603)884-6660

Digital Equipment Corporation
Digital Equipment Corporation
Accessories and Supplies Center Accessories and Supplies Center
1050 East Remington Road
632 Caribbean Drive
Schaumburg, Illinois 60195
Sunnyvale, California 94086
Telephone:(312)64Q-5612
Telephone:(408)734-4915

Western Region

digital equipment corporation. marlboro. massachusetts
I

First Printing, July 1982
Copyright ©, 1982, Digital Equipment Corporation. All Rights Reserved.

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
only be used or copied 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 or its affiliated companies.
The following are trademarks of Digital Equipment Corporation:
DEC
DECUS
DECSYSTEM-20
DECwriter
DIBOL
EduSystem

~D~DDmD

DECnet
DECsystem-10
PDT
RSTS
RSX
VMS
RT

lAS
MASSBUS
PDP
UNIBUS
VAX
VT

The postage-prepaid READER'S COMMENTS form on the last page of this
document requests the user's critical evaluation to assist us in preparing future
documentation.

Contents

Preface

Chapter 1 Introduction
1.1

1.2

Magnetic Tape Characteristics .

1-2

1.1.1
1.1.2
1.1.3
1.1.4
1.1.5

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

Records...
Blocks.......
Volume Sets . . . .
Recording Density.
Track Status.

Labeled Tapes .

1-3

Labels.

1-4

1.2.1

System Labels
User Labels

1-6
1-6

Label Processing. . . .

1-6

1.2.2.1
1.2.2.2
1.2.2.3
1.2.2.4

1-7
1-7
1-7
1-7

1.2.1.1
1.2.1.2
1.2.2

TOPS-10 Implementation Level

1-8

1.2.3.1
1.2.3.2

ANSI Labels.
OS/VS Labels

1-8
1-8

Unlabeled Tapes . . . . . . . .

1-8

1.3.1

Bypass Label Processing.

1-8

The System Administrator . . . .

1-8

1.4.1
1.4.2
1.4.3

1-9
1-9
1-9

1.2.3

1.3
1.4

Automatic Volume Recognition (AVR)
Label Verification . . . . . .
Volume Switching . . . . . .
Label Creation and Updating.

Volume Identifiers . . . .
Visual Identification Labels
Unlabeled to Labeled Tape Conversion.

iii

1.5

The Operator . . . . . . . . . . . . . . .
1.5.1
1.5.2
1.5.3
1.5.4

Initializing Magnetic Tapes . . .
Mounting and Dismounting Tapes.
Handling Errors. . . . . . . . . .
Maintaining the Visual Identification Labels.

1-10
1-10
1-10
1-11
1-11

Chapter 2 Tape Initialization
2.1
2.2
2.3

Chapter 3

Initializing Labeled Tapes . . . . . . . . . . .
Initializing Unlabeled Tapes. . . . . . . . . .
The SET TAPE-DRIVE INITIALIZE Command

Using Labeled Tapes
3.1
3.2
3.3
3.4

The Terminal User
The Programmer ..
Accessing Tape Data.
The Mount Command
3.4.1
3.4.2
3.4.3

3.5
3.6

3.7
3.8

The ILABEL-TYPE: Switch
The IVOLID: Switch .
The INOTIFY Switch

3-2
3-3
3-3
3-3

3.5.1

The Accessibility Characters.

3-5

Writing to a Magnetic Tape . . . . .

3-6

3.6.1
3.6.2
3.6.3
3.6.4

3-7
3-7
3-7
3-7

Updating Files in a Volume Set
Specifying End-of-Volume Processing
Writing Label Parameters .
Closing a File . . . . . . .

Reading from a Magnetic Tape . . .

3-7

3.7.1

Reading Label Parameters.

3-8

Positioning the Tape. . . . . . . .

3-8

Using Monitor Commands.
Using Monitor Calls. . . .
Using Language Statements.

3.9
Running COBOL or FORTRAN Programs
3.10 Using the BACKUP Utility . . . .
3.11 Using the DIRECTORY Command.

3-8
3-8
3-8
3-9
3-9
3-9

Using Unlabeled Tapes
4.1

4.2
4.3

Volume Switching . . . . . . . . . . . .

4-1

4.1.1
4.1.2

4-1
4-1

Volume Switching with COBOL .
Volume Switching with BACKUP

Reading from an Unlabeled Tape. . . .
Writing to an Unlabeled Tape . . . . .

4-2
4-2

Updating Files in a Volume Set

4-3

4.3.1

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

Protection Codes . . . . . . .

3.8.1
3.8.2
3.8.3

Chapter 4

2-1
2-2
2-3

Chapter 5

Magnetic Tape Structure
5.1
5.2

5.3

Introduction......
The Physical Structure.

5-1
5-1

5.2.1
5.2.2
5.2.3
5.2.4

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

The Logical Structure . . .
5.3.1
5.3.2

Chapter 6

Volume Sets, Files, Records
Labels...........

5-3
5-4

5.3.2.1
5.3.2.2

5-5
5-6

Classification of Labels.
Tape Label Sequences

DEC-Compatible Core Dump Mode (7- or 9-Track)
Eight-Bit Mode . .
Sixbit Mode . . . .
ANSI-ASCII Mode.

6-1
6-3
6-3
6-4

Record Formats
7.1
7.2
7.3
7.4
7.5

Chapter 8

5-3

Magnetic Tape Data Modes
6.1
6.2
6.3
6.4

Chapter 7

Blocks.....
Tape Marks and Interrecord Gaps
Tape Density
Data Modes . . . .

Blocking of Records . . . . . . . . .
Fixed-Length Records (F Format) . .
Variable-Length Records (D Format).
Spanned Records (S Format). .
Undefined Records (U Format) .

7-1
7-2
7-2
7-4
7-6

Label Formats and Contents

Chapter 9 The Mount Process for Labeled Tapes
9.1
9.2
9.3
9.4

Introduction...............
Mounting the First Tape of a Volume Set
Mounting the ~Next' Tape of a Volume Set.
Automatic Volume Recognition . . .

9-1
9-2
9-3
9-4

Hardware . . . . . . . . . .

9-5

9.4.1.1

9-5

9.4.1
9.4.2
9.4.3
9.4.4

Ignored Interrupts

Volume Recognition .
Volume Assignment.
Premounted Volumes

9-5
9-5
9-6

Chapter 10 The Label Processor's Search Algorithm
10.1 When Neither a Filename nor a Sequence Number Is
Specified . . . . . . . . . . . . . . . .
10.2 When a Filename Is Specified . . . . . . . . . . . .
10.3 When a Sequence Number Is Specified . . . . . . . .
lOA When a Filename and a Sequence Number Are both
Specified . . . . . . . . . . . . . . . . . . . . . . .

Chapter 11

The TAPOP. Monitor Cali

Chapter 12

Crash Procedures
12.1 Recovery Procedures.
12.2 Crash Analysis . . .

10-1
10-1
10-2
10-2

12-1
12-2

Appendix A Label-Processing Stopcodes

Appendix B Tape Label Implementation Levels

Appendix C Glossary

Figures
1-1
1-2
1-3
2-1
3-1

5-1
5-2

5-3
5-4
5-5
5-6
5-7

5-8
5-9
7-1

7-2
7-3
7-4
7-5
7-6

The Tape Processing System. . . .
A Labeled Tape with One File . . .
Continuation Files in a Volume Set
An Initialized Labeled Tape ..
Access Protection Code. . . . . .
A Volume Set of Labeled Tapes . ~.
Volume Sets of Unlabeled Tapes.
Label Groupings. . . . . . . . .
Single File on a Single Volume .
Single File on Multiple Volumes .
Multiple Files on a Single Volume.
Multiple Files on Multiple Volumes
Coincidence of Beginning-of-File Group and
End-of-Tape Marker . . . . . . .
Coincidence of Last Data Block and
End-of-Tape Marker . . . . . . .
Fixed-Length Records (F), Unblocked
Fixed-Length Records (F), Blocked. .
Variable-Length Records CD), Unblocked.
Variable-Length Records CD), Blocked.
Unblocked Variable-Length Records.
Spanned Records (S), Unblocked . . . .

1-2

1-5
1-5
2-2

3-5
5-3
5-4

5-5
5-6
5-7
5-7

5-8
5-8
5-9

7-2
7-2
7-3
7-3
7-3
7-4

7-7 Spanned Records (S), Blocked
7-8 Unblocked Spanned Record
7-9 Blocked Spanned Records
8-1 File Section Numbers
8-2 File Sequence Numbers
9-1 The MOUNT Process for the First Tape of a Volume Set.
9-2 Volume Switching.
12-1 A Crash Block.

7-5
7-5
7-6
8-9
8-9
9-2
9-3
12-2

SET TAPE-DRIVE INITIALIZE Switches
Protection Codes for Field 1
Owner Protection Codes .
3-2 Protection Codes for Fields 2 and 3
Project-Members and Other User Protections
8-1 Machine Codes.
8-2 System Codes .
10-1 File Positioning Summary.

2-3

Tables
2-1
3-1

3-4
3-5
8-4
8-11
10-3

Vll

Preface

Magnetic tape labels are reserved records on the tape used to hold information pertaining to that reel of tape and the files it contains. This manual
discusses label-processing operations in addition to the use and formatting
of magnetic tape labels. It also discusses unlabeled tape organization and
processing. You may not need to read the entire manual. Scan the table of
contents to determine the portions of the manual that concern you.
This manual has been divided into fields of interest for the following
people:

• Terminal users utilizing magnetic tapes. Refer to Chapters 1, 3, and 4.
• Operators mounting and initializing magnetic tapes as well as handling
label-processing errors. Refer to Chapters 1, 2, and 9.
• Programmers and other users creating tapes to interchange between systems and interchange between users. Refer to Chapters 1, 5, 6, 7, 8, 10,
and 11.
• System administrators making judgments concerning magnetic tape use
and operation. Refer to Chapter 1.
• System programmers analyzing label-processing crashes. Refer to
Chapter 12.

Chapter 1
Introduction

This manual discusses labeled and unlabeled tapes that are under the control of the PULSAR and QUASAR programs. When the operator gives the
command SET. TAPE-DRIVE AVAILABLE, these programs are given control of the specified device. When they have control, the following automatic features can be used:
• Automatic Volume Recognition (labeled tapes only)
• Label processing (labeled tapes only)
• Volume switching
Automatic Volume Recognition and label processing are discussed in
Section 1.2, Labeled Tapes. Volume switching is discussed below.
((Volume switching" refers to the internal and external operations involved
in dismounting a volume and mounting the next one of the set. These
operations are required during the reading or writing of a file that is large
enough to span volumes. In these situations, the software that was activated with the original user MOUNT command is automatically reinvoked. For example, the operator receives a directive to mount the new
volume; the new volume is assigned; and the user receives notification (if
he specified INOTIFY with the MOUNT command).
This automatic activity and the operator's response constitute volume
switching. Chapters 3, Using Labeled Tapes; 4, Using Unlabeled Tapes;
and 9, The Mount Process for Labeled Tapes, further describe this topic.
PULSAR and QUASAR are the chief components of the tape-processing
system. aPR and ORION, the software components serving as the operator

1-1

interface to the system, also playa part. All these components communicate with each other through IPCF, the Inter-Process Communication
Facility (described in the TOPS-J 0 Monitor Calls Manual).
Figure 1-1 shows the relationship among these components.

Figure 1-1: The Tape Processing System

r---------------~
PULSAR

USER

MOUNT
DISMOUNT
ALLOCATE
DEALLOCATE

I
QUASAR

L________ ______ _

ORION

MR-S-2250-82

1.1

Magnetic Tape Characteristics
Sections 1.1.1 through 1.1.5 describe some of the logical and physical properties of magnetic tapes. Chapter 5, Magnetic Tape Structure, describes
these properties in detail.

1.1.1

Records

A record consists of related data that a program treats as a unit of information. When you output information to a magnetic tape, the system writes
the information in sequential records on the tape. The default size for data
records is 128 words. You can modify this value with the TAPOP. monitor
call. (Chapter 11 discusses this monitor call.) Note that the label records
are always 80 characters in length, regardless of the data record size.

1.1.2 Blocks
A block is a collection of data that PULSAR reads or writes as a unit. Each
physical tape block may comprise one or more complete records or spanned
record segments. (Chapter 7, Record Formats, discusses records and blocks.)
1-2

Introduction

Where labels are concerned, however, a block contains only one record. The
default block size is 128 words, which is the equivalent of a data record. But
you can change the block size with the TAPOP. monitor call or the SET
BLOCKSIZE monitor command. (Refer to the TOPS-10 Operating System
Commands Manual for details on this command.) Blocks are separated
from each other by spaces called interrecord gaps.

1.1.3 Volume Sets
A magnetic tape may contain one or more files or file sections. If the
end-of-tape marker is reached before a file is completely written, that file
is continued onto another tape and is known as a multivolume file. As with
the first tape, this and all subsequent continuation tapes (up to 60) may
contain files and/or file sections. The files contained on one magnetic tape
or spanning multiple magnetic tapes make up a file set. And the tapes
holding the files make up a volume set C(volume" being an equivalent term
for magnetic tape reel).

1.1.4 Recording Density
Tape densities of 200, 556, 800, 1600, and 6250 bits per inch are supported
by TOPS-10. Not all tape drives support all densities, however. Refer to
Section 5.2.3, Tape Density, for details.

1.1.5 Track Status
The PULSAR-QUASAR system supports 7- and 9-track tapes. However,
only 9-track tapes can be used with label processing. All 7-track tapes are
considered to be unlabeled. (This is not to say that all 9-track tapes are
considered to be labeled.)

1.2 Labeled Tapes
Magnetic tape is a widely used communications medium for the exchange
of information between users, user programs, and computer systems. This
communication may call for processing of tapes on foreign (non-TOPS-10)
systems and accessing of tapes by nonowners of the tapes. All the information required for these interchanges is written directly on the magnetic
tape in the form of distinct identifiable records called tape labels. Tape
labels also serve useful purposes when no interchange is involved; for example, when a TOPS-10 user accesses his own TOPS-10-generated'tapes.
Labels function in the following ways:
• Eliminating problems caused by the mounting of incorrect tapes. The
system reads labels to verify that the correct tape has been mounted .
• Providing data protection by preventing premature overwriting of tape
data and unauthorized access to tape files. The expiration date and
accessibility-related fields contained in the tape labels provide this
protection.
Introduction

1-3

• Identifying and delimiting files for selective processing by user programs.
• Permitting users minimal involvement in reading from and writing to
tapes (volumes). One task a user might ordinarily perform is
end-of-volume processing; that is, reading or writing labels at the end of
the tape and then determining whether another tape is required to continue processing a file. Using tape labels, the system automatically accomplishes this task for the user. (Refer to the discussion on volume
switching in Section 1.2.2.3.)
• Identifying the system on which each file on the tape was created.
• Defining the magnetic tape record and file formats.
• Allowing user programs to perform error checking during tape reads.
Such label fields as block count, file section number, and file number can
serve as audit tools.
• Offering on-line mechanisms for controlling and maintaining the magnetic tape resources at an installation. For example, such label fields as
the volume identifier and project-programmer number can facilitate tape
library maintenance.
Chapter 8, Label Formats and Contents, describes all the label fields mentioned above.

1.2.1

Labels

Magnetic tape labels are initially generated when the tape is initialized.
(Refer to Chapter 2, Tape Initialization, for details.) As you output files to a
volume set, the system writes additional labels as required to define each
new file and each new volume in the set.
A labeled tape contains the following types of labels:
• Volume header labels.
• File header labels.
• File trailer or end-of-file labels.
• End-of-volume labels.
Volume header labels are the two labels at the beginning of the tape that
describe the magnetic tape volume. The two labels that precede a file are
called file header labels. These labels define the file that immediately follows. The two labels that follow a file are called file trailer or end-of-file
labels. Figure 1-2 illustrates a labeled tape containing one file.

1-4

Introduction

Figure 1-2: A Labeled Tape with One File

Volume Header Labels

File Header Labels

File Data

File Trailer Labels

MR-S-2251-82

For multivolume files, the two end-of-volume labels are written at the end
of the tape. These labels, written instead of the file trailer labels, indicate
that the last file on the tape is not a complete file. See Figure 1-3.

Figure 1-3: Continuation Files in a Volume Set
1st Tape

I----------~ } ~~~d~~

Volume
Header
Labels

Labels

File
Header
Labels

}

labels {

~~~der

Labels

Continuation
of File 1
Data

File 1
Data

~~~u~~ ~

Continuation Tape

I----------f

File
Trailer
Labels
MR-S-2252-82

Introduction

1-5

Chapter 5, Magnetic Tape Structure, discusses the sequence of the tape
labels in greater detail, and describes the other items that are included on a
tape (such as tape marks and interrecord gaps). Chapter 8, Label Formats
and Contents, describes the contents of each tape label.
1.2.1.1 System Labels Labels that are specified in the ANSI standard
and that are processed by PULSAR are called system labels. System labels
include the volume and file header and trailer labels. Their 3-character
identifiers are:

• VOL

vol ume header

• UVL

2nd volume header

• HDR

file header

• EOV

end-of-volume

• EOF

file trailer

An identifier followed by a sequence number (1 or 2) constitutes a label
name and is part of the information contained within a label.
1.2.1.2 User Labels - Labels that are specified in the ANSI standard and
that are processed by user programs are called user labels. The user label
identifiers are:

• UVL

user volume header

• UHL

user file header

• UTL

user trailer

The TOPS-IO system does not support user labels. When you transfer tapes
that contain these types of labels from other systems, the label processor
ignores them. UVLl is considered to be a system label, however. It is written by the system and reserved for its use.

1.2.2 Label Processing
~~Label processing" refers to the reading or writing of tape labels by
PULSAR, the TOPS-IO label processor. Data files are not touched during
label processing. And normally, users are unaware of this activity and of
the labels' presence.

In performing its tasks, PULSAR interacts heavily with the QUASAR program. Through QUASAR, it sends messages to and receives them from
other system components. For example, all operator requests for label processing (RECOGNIZE and SET TAPE-DRIVE INITIALIZE, for example)
pass through QUASAR.
PULSAR also receives commands directly from QUASAR. Automatic
Volume Recognition is an example of this. With AVR, QUASAR detects an
on-line interrupt, then signals for PULSAR to verify labels.
1-6

Introduction

Label processing takes place at various times as discussed in the following
subsections.
AVR is the facility that recognizes that a tape has been mounted, verifies its volume header labels,
and then assigns the tape to a waiting user. If the tape has been premounted, that is, mounted before a program has asked for it, assignment
takes place when it is requested.
1.2.2.1 Automatic Volume Recognition (AVR) -

AVR may be enabled or disabled (using operator commands) for a particular drive. When it is enabled, the operator need do nothing more than
mount the correct labeled tape in response to a request; no type-in is required. The operator may even premount a volume in anticipation of a
request. (Refer to Section 9.4.4, Premounted Volumes, for details.) Just the
same, AVR recognizes a premounted volume and assigns the volume when
it is needed.
As stated above, the label processor verifies
labels when a tape is mounted. It also verifies labels when a user accesses a
particular file. It checks the file header labels to see, for example, if the
user is allowed access to the file or if the expiration date allows output to
the file.
1.2.2.2 Label Verification -

1.2.2.3 Volume Switching On input, if PULSAR encounters the EOV
label group, indicating a multivolume file, it requests mounting of the next
volume of the set. This action is repeated until it encounters an EOF label
group.

Similarly, if during output operations PULSAR encounters the end-of-tape
marker, it requests mounting of the next volume so that operations can
continue.
1.2.2.4 Label Creation and Updating -

PULSAR writes labels:

• Before the first data is written on a tape
• As the second and subsequent output files in a set are opened and closed
• As the end of the first volume, and beginning and end of all subsequent
volumes in a set are reached during output operations
The first case refers to volume initialization, a procedure required of all
tapes intended for label processing. Initialization sets up the
beginning-of-volume label set and all the file-related label sets required
for the first output file. Chapter 2, Tape Initialization, describes this
procedure.
The rest of the labels in a volume set are created or updated as needed
during output operations.
Programmers can write label information by executing various TAPOP.
monitor call functions. Refer to Chapter 11, The TAPOP. Monitor Call.

Introduction

1-7

1.2.3 TOPS-10 Implementation Level
The TOPS-I0 magnetic tape label processor partially implements the
ANSI x3.27 - 1977 standard. The system reads and writes labels according
to the standard, but does not ensure that the file data conforms to label
specifications. In this respect, it does not fully conform to the ANSI standard. For example, the actual record format, block size, and record size
could contradict the associated label parameters. User programs must enforce conformance.
1.2.3.1 ANSI Labels - The TOPS-I0 label processor reads and writes level
4 labels. (See Appendix B for a definition of levels.)

The TOPS-I0 label processor implements IBM tape
labels as defined for VSl Release 3, and VS2 Release 2. (See IBM publication GC26-3795-2 for a complete description.)
1.2.3.2 OS/VS Labels -

1.3 Unlabeled Tapes
Unlabeled tapes are defined as tapes having no labeling information that
describes the tape and its contents. Thus, use of unlabeled tapes puts extra
responsibility on system administrators, users, and operators, who must be
vigilant in handling the tapes. They must take care to see that the tapes
are well catalogued, that correct tapes are mounted, and that unauthorized
users do not gain access to tape data.

1.3.1 Bypass Label Processing
The system can process labeled tapes in unlabeled mode. This action is
known as bypass label processing. Although tapes designated for this mode
of processing contain labels, users may want to process the labels themselves rather than rely upon the label processor. Or, there may be other
reasons why they would not want the system to process labels. For example, the tapes might contain nonstandard labels. The system usually requires that the user have privileges for this type of processing.

1.4 The System Administrator
There are several decisions that the system administrator must make regarding tapes. Some of these decisions are:
• Who should use magnetic tapes at the installation
• How many tapes should be assigned to a particular user or a particular
project
• Who should be responsible for initializing magnetic tapes (if not the
operator)
• Who should be responsible for cataloging magnetic tapes, that is, for
maintaining the magnetic tape library

1-8

Introduction

• How should the magnetic tapes be catalogued
• During the GALGEN procedure, what should the system defaults be for
label type, track type, and recording density
• What information should the visual identification labels hold (refer to
Section 1.4.2)

1.4.1 Volume Identifiers
The system administrator should verify that there is no duplication of volume identifiers (VOLIDs - specified at tape initialization) among the
installation's labeled tapes. Switches to the SET TAPE-DRIVE
INITIALIZE command provide automatic incrementation and assignment
of alphanumeric VOLIDs to tapes.

1.4.2 Visual Identification Labels
Visual identification labels (VIDs) give an external indication of the contents of magnetic tapes. These labels are not machine-readable; they
adhere to the outside of the tape. Although the information written on
these labels is of the installation's choosing and need not conform to any
standard, it should unambiguously identify the tape. For example, a visual
identification label could contain the following information:
• File names
• File sequence numbers
• File creation dates
• Number of volumes in set if a multivolume set
• Owner of files
• Volume set name
• VOLID of next/previous volume in a multivolume volume set.

1.4.3 Unlabeled to Labeled Tape Conversion
To convert unlabeled tapes to labeled ones, the system administrator
should follow these steps:
1.

Initialize new or old scratch tapes. Follow the initialization procedures
required for labeled tapes, as specified in Chapter 2, Tape Initialization,
and in the TOPS-J 0 Operator Command Language Reference Manual.

Copy the contents of a designated unlabeled tape to one of the initialized tapes. This process is to be repeated for each unlabeled tape intended for conversion. The data transfer can be accomplished with the
COpy command. Note that COPY assumes a blocksize of 128
characters.

Introduction

1-9

Other than handling the issues above, the system administrator can leave
tape labeling to the operator, the user, and the label processor. The rest of
this manual discusses the user and operator interfaces to the
tape-processing system.

1.5 The Operator
Several activities fall under the operator's responsibility. Some of them are:
• Initializing magnetic tapes
• Mounting and dismounting magnetic tapes
• Handling label-processing and tape mount errors
• Maintaining the visual identification labels
The operator should refer to the TOPS-10 Operator's Guide and to the
TOPS-10 Operator Command Language Reference Manual for details on
these and other topics.

1.5.1

Initializing Magnetic Tapes

In order to use label processing, the operator, or someone else with privileges, must have initialized all tapes by using the aPR command, SET
TAPE-DRIVE INITIALIZE. This command writes the initial labels that
identify the volume and that ready it for the first file to be written.
Chapter 2, Tape Initialization, describes this process.

NOTE
New unlabeled tapes that have never been used must also
be initialized. No labels are written during this process, however. An empty record followed by two tape marks is written
on the tapes. These indicate that the tapes will be used for
unlabeled purposes.

1.5.2 Mounting and Dismounting Tapes
When the operator receives a mount request for a labeled tape, he/she
examines the various magnetic tape reels contained in the installation's
tape library. After finding and mounting the requested tape, the operator
waits for the system's response. This response is an acknowledgement that
the tape has been mounted and that all is well, or it is an error message.
Chapter 9, The Mount Process for Labeled Tapes, describes the internal
aspects of the mounting process. The external aspects are covered in the
operator-specific manuals.
For unlabeled tapes, the operator mounts a tape and issues the appropriate
command to let the system know that the requested tape has been
mounted. No message is then issued to indicate that the correct tape has
been mounted.

1-10

Introduction

1.5.3 Handling Errors
Any errors that occur during label processing or mounting result in messages to the operator, who then rectifies the situation or notifies users that
the tape operation has been aborted. For example, if, in reading the volume
header labels, the label processor finds that an incorrect tape has been
mounted, it sends .an error message to the operator. Then, the operator
searches for the correct tape and mounts it.
Or, the label processor might find that the user is not allowed access to the
specific tape that helshe has requested. Again, the operator receives notification. In this case, the operator informs the user, who takes the next
action. The user might issue a mount request for another volume or try to
learn why helshe was denied access to the volume. (Chapter 3, Using
Labeled Tapes, describes protection codes.)
Also, if a labeled tape is mounted when an unlabeled one is expected, or
vice versa, the operator receives an error message.

1.5.4 Maintaining the Visual Identification Labels
The operator is also responsible for the visual identification labels affixed
to magnetic tapes. These labels should be updated when the tape is first
used, overwritten, or returned to the installation's scratch pool.

Introduction

1-11

Chapter 2
Tape Initialization

Chapter 2 discusses initialization for labeled and unlabeled tapes.

2.1

Initializing Labeled Tapes
All magnetic tapes intended for label processing must be initialized before
any data can be written on them (with the exception of bypass
label-processing tapes). During initialization, the system writes:
• The beginning-of-volume label group.
• Two tape marks framing an empty file.
• The end-of-file label group.
• A double tape mark - logical EOT.
The format of an initialized labeled tape is represented in Figure 2-1:
The newly written labels contain sufficient information to establish a
system-accessible single-file volume. But because initialization neither requires nor allows any file data to be written, the file is empty.
When the tapes are later used for output, the system updates the fields
within these labels and creates additional labels as needed.. Such fields as
the file creation date and expiration date are updated when files are created on initialized tapes, and the end-of-volume label set is written when
needed.
The operator command, SET TAPE-DRIVE INITIALIZE, is used to initialize magnetic tapes. With this command, the operator specifies such information as the following to be assigned to the labels: project-programmer

2-1

number, expiration date, and protection code for the entire volume (different from the protection code for individual files). The system augments
these inputs with all other information required to create the initial labels.
If the operator chooses not to specify the various values, the system takes
such default action as the following: it designates [0,0] as the owner, assigns a protection code of 000, and uses the current date as the expiration
date. This type of tape is recognizable to the label processor as a fully
labeled but unprotected tape. Refer to the list of SET TAPE-DRIVE
INITIALIZE switches in Table 2-1 for complete default information.

Figure 2-1:

An Initialized Labeled Tape

VOL1

UVL1

HDR1

HDR2

****

f - - Tape Mark

Tape Mark

EOF1

EOF2

****

----

Tape Mark

***.

Tape Mark
MR-S-2253-82

2.2 Initializing Unlabeled Tapes
Unlabeled tapes contain only files and tape marks. Before the first file can
be written however, the operator must initialize these tapes with the SET
TAPE-DRIVE INITIALIZE command.
Initialization is performed on tapes intended for unlabeled use to establish
this intent in writing. (The system makes sure that these tapes are not
used when labeled ones are required.) Thus, when the operator initializes a
tape as unlabeled, the system writes a record of 80 blank frames at the
beginning of the tape, followed by two tape marks (logical EOT). The system recognizes this as an unlabeled tape.
When a user issues a write instruction with the tape positioned at the
beginning point, the blank frames and tape marks are overwritten by the
user's data.
2-2

Tape Initialization

2.3 The SET TAPE-DRIVE INITIALIZE Command
Volume initialization can be performed in a few minutes following a user
request for a new tape.
To initialize one or more tapes, the operator gives the following command
to the OPR prompt:
OPR>SET TAPE-DRIVE device: INITIALIZE/switchl,/switch2, ... m
where ((device" is the physical name of the tape drive.
Table 2-1 describes the valid switches for the SET TAPE-DRIVE
INITIALIZE command. Refer to the TOPS-10 Operator Command
Language Reference Manual for details on initialization.

Table 2-1: SET TAPE-DRIVE INITIALIZE Switches
Switch

Meaning

ICOUNT:

specifies the number of tapes to be initialized. The
ICOUNT: switch, when specified, provides a mechanism
for initializing multiple tapes having the same attributes. After initialization, the tape is automatically unloaded and the tape drive is ready.to accept the next tape
to initialize.

IDENSITY:

specifies the density of the tapes to be initialized. If the
IDENSITY: switch is not specified, the density defaults to
that specified during GALGEN (usually 1600 bits/inch).

IINCREMENT:

specifies the numeric value by which to increment the
tape volids. This switch is valid only when specified with
the ICOUNT: and IVOLUME-ID: switches. If this switch
is not included, the default numeric value is 1. For example, to initialize a set of 10 tapes with VOLIDs from
442000 through 442900, the switches ICOUNT:10,
IINCREMENT:100, and NOLUME-ID:442000 are given.
Every tape mounted on the tape drive is initialized with
the next sequence number as specified with the IINCREMENT: switch.

ILABEL-TYPE:type

specifies the type of label to be written on the tape. The
type can be ANSI, EBCDIC, or UNLABELED. The default label is ANSI when this switch is specified. If the
ILABEL-TYPE: switch is not specified, the default is
UNLABELED.

IOVERRIDE-EXPIRATION:
(YES or NO)

specifies whether to check the expiration date of the data
on the tape. If NO is specified and there is an attempt to
reinitialize a labeled tape whose first file has not expired,
the system issues an error message and does not reinitialize the tape. If YES is specified, each tape is reinitialized
unconditionally. When initializing virgin tapes, the operator specifies YES to prevent the tape drive from "running away" when the label processor tries to read a label
from the tape. If this switch is not specified, the default
is NO.

Tape Initialization

2-3

Table 2-1: SET TAPE-DRIVE INITIALIZE Switches (Cont.)

2-4

Switch

Meaning

fOWNER:[proj,prog]

specifies the project-programmer number for the user
who owns the tape(s) to be initialized. If the fOWNER:
switch is not specified, the tape is initialized as owner
[0,0].

fPROTECTION :nnn

specifies a 3-digit octal number as the protection code of
the tape. The number is usually from 000 (where anyone
can use the tape) through 777 (where no one can use the
tape). If the fPROTECTION: switch is not specified, the
default is 000.

ITAPE-DISPOSITION:
condition

specifies what is to be done to the tape after it is initialized. The condition can be either HOLD or UNLOAD. If
HOLD is specified, the tape is initialized and is not unloaded. A user requesting the tape can use it without the
operator's having to reload it. If UNLOAD is specified,
the tape is initialized and is unloaded from the tape
drive. The default is UNLOAD. If a value greater than 1
is specified with the fCOUNT: switch, the
fTAPE-DISPOSITION: switch is ignored and UNLOAD
is assumed.

NOLUME-ID:"volid"

specifies a volume identification for the tape. The VOLID
is from one to six characters long. If the VOLID is to
contain any non alphanumeric characters, it must be enclosed in double quotes ("). This switch is ignored if the
tape is being initialized as an unlabeled tape.

Tape Initialization

Chapter 3
Using Labeled Tapes

This chapter discusses the user interface to the tape-labeling system.

3.1 The Terminal User
When sitting at a terminal writing to a magnetic tape and reading from a
magnetic tape, you need not do anything different with labeled tapes than
with unlabeled tapes. In either case, you issue the MOUNT command to
assign a magnetic tape unit to your job and to alert the operator to mount a
tape for you.
Under most conditions you need not worry about magnetic tape labels. The
label-processing software reads and writes tape labels without your intervention.

3.2 The Programmer
As a programmer, you can control many label-processing operations with
the various TAPOP. monitor call functions. For example, you can use the
.TFLPR function to position the tape to the beginning of any file or to the
end of the last file of the volume set. Using this function, you can also
supply or change the data for many of the tape label fields. Several other
TAPOP. functions are discussed in the remainder of this chapter.
Refer to Chapters 5 through 12 for more complete programmer information.

3.3 Accessing Tape Data
You can access a tape only when you, the operator, and the label processor
can uniquely identify the desired magnetic tape reel.

3-1

Primary identification is the unique name assigned to each tape. This name
is called a VOLID (volume identifier), and it is assigned to a tape during
initialization. You specify the VOLID(s) when you issue the MOUNT command, letting the system and the operator know which tape(s) you wish to
use. For input, all VOLIDs must be specified for the tapes in a volume set.
The label processor then directs the operator to mount and dismount these
tapes as you process files in the set.
The visual identification label also provides identification. This identification is on a label that is glued to the magnetic-tape reel; it is not machine
readable. The system administrator determines the VID label content. The
operator uses this information to select tapes that you request through the
MOUNT command.

3.4 The MOUNT Command
The MOUNT command causes a magnetic tape to be physically mounted by
the operator and assigned to your job.
Command format:
MOUNT resource:logical-name:/switches
where resource can be the:
• Physical device name
• Logical-name previously associated with a resource
• Volume set definition, for example: BILLS-80(NOV,DEC)
Logical-name is the name your program will use to refer to the resource.
The default logical name is taken from the first six characters of the volume set name, or from fewer characters if a nonalphanumeric is encountered. For example, the volume set name above would produce this logical
name: BILLS.
The MOUNT command is described in detail in the TOPS-10 Operating
System Commands Manual.
The switches that you specify in the MOUNT command that relate directly
to label processing are described below.

3.4.1

The ILABEL-TVPE: Switch

The ILABEL-TYPE switch identifies the format of the labels. The parameters for this switch are:
• ANSI
• IBM
• EBCDIC
• NOLABELS
3-2

Using Labeled Tapes

• NONE
• UNLABELED
• USER-EOT
• BYPASS
• BLP
ANSI indicates that the labels conform to ANSI standards. IBM and
EBCDIC are equivalent parameters that indicate IBM-formatted labels.
NOLABELS, NONE, UNLABELED, and USER-EOT are equivalent
parameters indicating that 'the tape does not contain standard labels. With
NOLABELS, NONE, and UNLABELED, the label processor performs automatic volume switching; it does not perform it with USER-EOT. BYPASS
and BLP are equivalent parameters that result in the label processor performing no processing. These two parameters are usually reserved for pri vileged users. Chapter 4 further discusses the parameters for unlabeled and
bypass-labeled tapes.

3.4.2 The IVOLID: Switch
The IVOLID switch must be specified in your MOUNT command unless
ISCRATCH is specified. This switch specifies which volume (VOLID) or
volumes (VOLID(l), ... VOLID(n)) you want an operator to mount. If you
specify fewer volumes on output than are needed, the operator can use
scratch volumes as the extra volumes. For multivolume files, you must
enter (on input) all volume names associated with the file when you request
mounting of the first reel.

3.4.3 The INOTIFY Switch
The INOTIFY switch is important. This switch causes the system to notify
you of automatic volume switches and to give you the VOLID of the new
volume. (Refer to Section 9.3, Mounting the ~Next' Tape of a Volume Set,
for information on volume switching.) By this action, you can know the
VOLIDs of scratch volumes that the operator may have had to mount for
your output operations.
The INOTIFY action is performed if you specify the INOW AIT switch in the
MOUNT command.

3.5 Protection Codes
There are two types of protection codes associated with labeled tapes. The
volume protection code, written in the UVLl label during initialization,
determines who can access the volume. The file protection code, written in
the HDR2 label for each new file, determines who can access the file.
Programmers can write the file protection code with the .TFLPR function of
the TAPOP. monitor call. But the volume protection code is written only

U sing Labeled Tapes

3-3

when the tape is initialized. Note that the protection codes correspond to
the standard disk file protection codes discussed in the TOPS-10 Operating
System Commands Manual.
The protection code consists of nine bits, divided into the following three
classes:
1. The first three bits refer to the owner of the file or volume.

2. The second three bits refer to users with the same project number as
the owner (that is, any user logged in under the same project number as
the owner, regardless of his programmer number).
3. The last three bits refer to all other users.
The protection codes and their meanings are listed in Tables 3-1 and 3-2.

Table 3-1: Protection Codes for Field 1
Owner Protection Codes
Code

Accessibility by Owner

You (the owner) have no access privileges. However, you can rename the file to
change the protection code, using the RENAME command. If another user tries
to access the file, the File Daemon* is called.

You can execute the file. If another user tries to write or rename the file, the File
Daemon is called.

You can read and execute the file. If another user tries to write or rename the
file, the File Daemon is called.

You can change the protection code of the file, rename, write, read, execute,
update, and append to the file. If another user attempts to write, rename, or
supersede the file, the File Daemon is called.

You can only rename the file.

You can write, read, execute, update, and append to the file.

You can rename, write, read, execute, update, and append to the file.

You can change the protection code of the file. You can also rename, write,
execute, update, and append to the file.

* For complete information on the File Daemon, refer to the TOPS-10 Monitor Calls
Manual.

3-4

Using Labeled Tapes

Table 3-2: Protection Codes for Fields 2 and 3
Project-members and Other User Protections
Code

Access Privileges

Users other than the owner cannot access the file.

The user can execute the file. If the user attempts to delete, read, change, rename, append-to, or run the file with the RUN command, the File Daemon is
called.

The user can read and execute the file. If the user attempts to write or rename
the file, the File Daemon is called.

The user can read, execute, and append-to the file. If the user attempts to write,
rename, or supersede the file, the File Daemon is called.

The user can read, execute, append-to, and update the file.

The user can write, read, execute, append-to, and update the file.

The user can rename, write, read, execute, append-to, and update the file.

The user can change the protection-code of the file. He can also rename, write,
read, execute, change, append-to and update the file. This code allows all users
full access privileges.

Figure 3-1 illustrates the 9-bit protection code field of a file or volume
having a protection code of 057. This protection code indicates three things:
1. The owner has full privileges (code 0).
2.

The project members have read and execute privileges (code 5).

All other users have no access privileges (code 7).

Figure 3-1:

Access Protection Code
Owner

Project

Others

~~~

0 1 0 1 0 \1\0\1\11 1 \11
o

7
MR-S-2254-82

3.5.1 The Accessibility Characters
There are two types of accessibility characters. The first type, written in
the VOL1label, indicates to the system whether it should check the volume
protection code in UVL1. If the character is zero, anyone can use the tape
and no checking is performed. If it is one, the system checks the protection
code. During initialization, the system writes this character in accordance
with the volume protection code.

U sing Labeled Tapes

3-5

Likewise, the second type of character, written in HDRI for each new file,
determines whether the file will be protection checked according to the
protection code in the file's HDR2 label. Thus, if this field contains a space,
anyone is allowed access to the file. If it contains a one, the protection code
is checked before access is allowed.
Refer to the descriptions of the Volume Header Label and of the first File
Header Label in Chapter 8, Label Formats and Contents, for further details
on the accessibility characters.

3.6 Writing to a Magnetic Tape
After you have been notified of a successful mount, you can open the magnetic tape file and record data.
Your first output attempt causes the label processor to reverify the
volume-header label set. This action duplicates the action taken during
mounting. (Refer to Chapter 9, The Mount Process for Labeled Tapes.)
Here, the system double-checks to ensure that no one has tampered with
the tape since it was mounted. Again, the system checks your
project-programmer number and other label parameters. If you are updating a file, it checks the protection code of the current file to ensure that you
are allowed to modify it.
In performing' output operations, the system first updates fields in the labels that were written during initialization, then positions the tape by
filename or file position if necessary. (Refer to Chapter 10, The Label
Processor's Search Algorithm, for positioning information.)
You can write one or more files on a magnetic tape, or you can write just
one segment of a file on a tape. In the latter case, the remainder of the file
is placed on additional magnetic tapes. When a file spans more than one
magnetic tape, it is called a multivolume file. A multivolume file can occupy up to 60 volumes.
As you add files to a volume set, the system overwrites the last of the
logical EOT tape marks, leaving one tape mark between the end-of-file
label group and the following beginning-of-file label group.
If, as is the case with multivolume files, your file is large enough to reach
the physical end-of-tape mark before you have closed the file, the label
processor will write an end-of-volume label set, indicating that your file is
to be continued onto another volume. In addition, the label processor will
automatically alert the operator to mount the next volume for the continued section of the file. (Refer to Section 9.3, Mounting the (Next' Tape of a
Volume Set, for details on volume switching.) Thus, you need not be concerned with end-of-tape operations while processing a file. If your data
requires more tapes than were specified in the MOUNT or ALLOCATE
command, the system requests the necessary scratch tapes for your job.

3-6

U sing Labeled Tapes

3.6.1

Updating Files in a Volume Set

When you update a file on a tape, all files beyond this file in the volume set
become lost. You lose these files because when you close a file, the system
closes the volume set. (Refer to Section 3.6.4, Closing a File, for details.)
Therefore, you should use some alternative method for updating tape files.
Perhaps you could begin copying the file set to another set of tapes. When
the designated file has been copied, you could update it on the new tape.
After you have updated the file, you could continue copying the remaining
files in the set.
Or, you could copy the files to disk; update the designated file; then rewrite
the files to tape.

3.6.2 Specitying End-ot-Volume Processing
You can force the label processor to write an end-of-volume label set if you
wish to establish the point at which the continued section of a multivolume
file is to begin on the next volume. Use the .TFFEV function of the TAPOP.
monitor call to force the end-of-volume label set to be written.

3.6.3 Writing Label Parameters
The .TFLPR function of the TAPOP. monitor call allows you to override the
system defaults for file-specific data that is contained in the
beginning-of-file, end-of-volume, and end-of-file label sets. To name a
few, you can specify record length, block length, expiration date, and
filename. Refer to the full description of this function in Chapter 11, The
TAPOP. Monitor Call.

3.6.4 Closing a File
When you close the file (with the CLOSE monitor call, for example), the
label processor writes the end-of-file label set. It then writes two tape
marks. These marks indicate the logical EOT as well as the end of the
volume set.

3.7 Reading from a Magnetic Tape
When you open a file for input, the system reverifies the tape labels as it
does when you write to a file. It similarly positions the tape to the desired
location.
The label processor directs the operator to mount and dismount volumes as
required as you read through the file or the file set. Your job receives
notification when end-of-file is reached.

Using Labeled Tapes

3-7

3.7.1

Reading Label Parameters

The .TFLPR function of the TAPOP. monitor call allows you to read
file-specific information contained in the file header and file trailer labels.
Refer to Chapter 11, The TAPOP. Monitor Call, for details.

3.8 Positioning the Tape
You can position the tape to access specific data by using monitor commands, monitor calls, and language statements.

3.8.1

Using Monitor Commands

You can position the tape with the following monitor commands:
BACKSPACE, SKIP, EOF, REWIND, and UNLOAD. In releases previous
to GALAXY 4.1, the monitor was solely responsible for these commands.
But with GALAXY Version 4.1, this arrangement is changed. If you issue
the MOUNT command (as opposed to the ASSIGN command), and do not
specify bypass label-processing mode, PULSAR controls the operation of
these commands. Otherwise, the monitor handles these commands as it did
previously.
PULSAR performs as follows for each command:
• BACKSPACE and SKIP - These commands cause PULSAR to perform
volume switching when the tape reaches the beginning- or end-of-tape
markers. PULSAR also positions the tape to the previous/next file if a
file-header label set is encountered.
• EOF - This command skips the remaining data in the file and positions
the tape at the beginning of the next file. For the last file of the set, the
tape is positioned at the end of the data area.
• REWIND - This command positions the tape at the beginning of user
data, just after the beginnipg-of-volume label group.
• UNLOAD - This command only rewinds the tape. You should use the
DISMOUNT command to relinquish a tape drive.

3.8.2 Using Monitor Calls
You can access files by name or by position within the volume set. Use
various functions of the TAPOP. monitor call to accomplish this. Chapter
10, The Label Processor's Search Algorithm, describes the operations that
result when you specify a filename or sequence number to access data.
You can also access specific records using the TAPOP. monitor call.

3.8.3 Using Language Statements
You can also select records using language statements. Refer to the appropriate language reference manual or users guide for the specifics of processing tape records.
3-8

Using Labeled Tapes

3.9 Running COBOL or FORTRAN Programs
No special procedures are required for reading and writing tapes with
COBOL or FORTRAN programs. The label processor reads/writes labels
and performs volume switching for your job as described above. This volume switching requires no intervention by the COBOL or FORTRAN
run-time systems. Thus, the trailer records that COBOL ordinarily writes
at the end of a volume are not written. Refer to Chapter 4, Using Unlabeled
Tapes, for a description of differences with unlabeled tapes.

3.10 Using the BACKUP Utility
With BACKUP, the label processor performs as described in Section 3.9.
BACKUP is not involved in volume switching and thus does not write
trailer records. Refer to Chapter 4 for differences with unlabeled tapes.

3.11

Using the DIRECTORY Command
The DIRECTORY command can provide information on labeled tapes. It
can read both ANSI and EBCDIC labels. An example is given below:
.MOUNT T(T,B):T:/WRITE/LABEL:ANSI/NOTIFY
[7:21:51]
[Mount request T queued, request #150]
[Ma!;ftape T ITlounted on MTA2G3 I"lith lo!;fical nalTle T]
• D IR T:
Read Densit}':lGOO Parit}':Odd

8-Trac~~

Write enabled REELID:T

Labeled tape file inforlTlation:
File NaITle:"FILE.001
Protection:OOO, Creation Date:1a-Ma}'-81, Expiration Date:1a-May-81
Bloc~~ size:GaO, Record size:GaO
Record ForlTlat:Undefined, ForlTl control:None

U sing Labeled Tapes

3-9

Chapter 4
Using Unlabeled Tapes

When you choose not to use labeled tapes, you must specify the label type
as either ((NOLABELS", ((NONE", ((UNLABELED", or ((USER-EaT" in the
MOUNT command. These label types have particular significance in regard to volume switching.

4.1

Volume Switching
With unlabeled tapes, volume switching is performed for tapes specified
with the NOLABELS, NONE, and UNLABELED parameters. For
USER-EaT, your program sees the end-of-tape condition and has to then
issue the .TFFEV function of the TAPOP. monitor call in order to switch to
the next reel. It is important to specify the INOTIFY switch in the MOUNT
command if you anticipate volume switching on output. Section 3.4, The
MOUNT Command, discusses this switch.

4.1.1

Volume Switching with COBOL

When using COBOL programs to read or write tape data, you specify the
USER-EaT parameter. This parameter gives the COBOL run-time system
the responsibility for handling end-of-volume operations. Thus, when you
reach the end of a tape while reading or writing files, your job stops and
waits for the next action to come from the COBOL run-time system (rather
than from PULSAR). The COBOL system responds by writing the
COBOL-specific trailer records and issuing the .TFFEV function of the
TAPOP. monitor call. This function calls upon PULSAR to perform the
volume switch.

4.1.2 Volume Switching with BACKUP
Follow the same procedure with BACKUP as with COBOL. This causes
vol ume switching to occur in the same manner.

4-1

NOTE
If you write tapes in USER-EOT mode, you must also read
them in this mode. Likewise, any tapes written in
non-USER-EOT mode must be read in this mode.

4.2 Reading from an Unlabeled Tape
When an unlabeled tape is mounted and verified that it contains no standard volume labels, the system positions the tape before the first record. If
the tape has more than one file and you want to use other than the first file
on the tape, you can position the tape to the specific file by giving the SKIP
command.
You can also position a tape to a specific file by giving the BACKSPACE
command. When you give either command, the tape is spaced forward or
backward the number of files you specified. You should use these commands only if you know the location of the file relative to the tape's current
position. Refer to Section 3.8.1 for a complete discussion of positioning the
tape using monitor commands.
You can also access tape data using the TAPOP. monitor call or language
statements.
When your program encounters an end-of-file condition (reads a tape
mark), the system informs your program and leaves the tape positioned
after the tape mark and before the next file on the tape. For multivolume
files, either you or the system can handle the required volume switch. As
discussed earlier, control over volume switching depends upon the parameters you specify with the ILABEL---TYPE switch in the MOUNT command.
Automatic volume switching takes place when PULSAR reaches the logical
EOT (two tape marks).

4.3 Writing to an Unlabeled Tape
When an unlabeled tape is mounted for output, the system positions the
tape before the first record on the tape. If the tape has been initialized
(refer to Chapter 2) but never used, it contains a record of 80 blank frames
followed by two consecutive tape marks. When you issue the first write
instruction, the system overwrites the current contents of the tape with
your data record. When you close the file (with the CLOSE monitor call, for
example), the system writes two tape marks. These marks indicate the
logical EOT as well as the end of the volume set.
If you want to write the new data following an existing file, you can position the tape beyond an existing file with the SKIP command. Your first
output to the new file causes the last tape mark to be overwritten, leaving
just one tape mark between the files.
You can also use other monitor commands to position the tape on output.
(Refer to Section 3.8.1.) In addition, as mentioned above, monitor calls and
language statements allow you to access tape data.

4-2

Using Unlabeled Tapes

Volume switching is handled the same as when reading a multivolume file.

4.3.1

Updating Files in a Volume Set

When you update a file on a tape, all files beyond this file in the volume set
become lost. You lose these files because two tape marks are written when
you close a file. These marks indicate the end of the volume set. Therefore,
you should use some alternative method for updating tape files. Perhaps
you could begin copying the file set to another set of tapes. When the designated file has been copied, you could update it on the new tape. After you
have updated the file, you could continue copying the remaining files in
the set.
Or, you could copy the files to disk; update the designated file; then rewrite
the files to tape.

Using Unlabeled Tapes

4-3

Chapter 5
Magnetic Tape Structure

5.1

Introduction
This chapter discusses the physical and logical structure of magnetic tapes.
In the discussions, structure is linked to the two tape-data segmentation
modes: physical and logical.
Physical segmentation refers to the separation of data by marks or interrecord gaps (gaps that are lengths of magnetic tape containing no
software-accessible data). Blocks, marks, and gaps relate to a tape's physical structure.
With logical segmentation, data separation is established by the software.
The software accesses logical segments of data and is made aware of the
marks and gaps only through detection of these physical delimiters by the
tape unit. Volume sets, files, and records relate to a tape's logical structure.
Often physical and logical segmentations of tape data coincide. For instance, when block length equals record length, the physical record (block)
is the same length as the software-accessible record. In this case a gap
separates each logical record.

5.2 The Physical Structure
Sections 5.2.1 through 5.2.4 discuss the physical structure of magnetic
tapes.

5.2.1

Blocks

A block is a physical record, a collection of characters that the system reads
or writes as a unit. The default length of a block is 128 words. You can reset
the default to any value up to 4094 words with the .TFBSZ function of the
TAPOP. monitor call. (Refer to Chapter 11, The TAPOP. Monitor Call, for
information.) You can also reset this value with the SET BLOCKSIZE monitor command.
5-1

5.2.2 Tape Marks and Interrecord Gaps
Tape marks and gaps are physical delimiters that are recorded on a magnetic tape by the tape unit. Exactly what is recorded is of concern only to
the hardware. Figure 5-2 and the figures in Section 5.3.2.2, Tape Label
Sequences, show the placement of marks and gaps in the nondata areas of a
volume.
With labeled tapes, tape marks are recorded before and after every data file
to separate a data file from its preceding and following tape labels. They
are also recorded between the end-of-file label group and the following
beginning-of-file label group. Two tape marks are written after the last
label group (EOV or EOF) on the volume.
With unlabeled tapes, a tape mark is recorded after every file. Two tape
marks are recorded at the end of meaningful information on the tape. Also,
if the last file continues onto another volume, only one tape mark is written
at the end of the volume.
Gaps are used to separate physical records (blocks) on the tape.

5.2.3 Tape Density
Density refers to the number of characters that can be recorded on one inch
of magnetic tape. The possible densities for tapes processed under TOPS-I0
are 200, 556, 800, 1600, and 6250. These numbers, which represent
bits-per-inch, are specified with the MOUNT command. The rate of
transferring data from a tape unit to memory, or vice versa, depends on the
tape density and the speed of the unit in use. If you try to record at too high
a density you may lose data.
Note that not all devices can process tapes at all densities. See your system
operator for the valid densities of the tape drives on your system.
Files and labels are written at the same density for all volumes in the set.

5.2.4 Data Modes
Magnetic tape data can be recorded in one of the following data modes:
• DIGITAL-compatible core dump mode (7- and 9-track)
• 8-bit mode, 4 bytes per word
• 6-bit mode, 6 bytes per word (9-track, TU70 only)
• 7-bit mode, 5 bytes per word (ANSI-ASCII mode, TU70 only)
These modes govern the transfer of data between memory and the magnetic
tape and determine how the monitor records the bits of a 36-bit word onto
tape. Chapter 6, Magnetic Tape Data Modes, discusses this topic in detail.

5-2

Magnetic Tape Structure

5.3 The Logical Structure
Magnetic tapes consist of files and records and may contain labels coded in
either industry-compatible ASCII or IBM-compatible EBCDIC. (The information in both types of labels is recorded in 8-bit bytes.)
With labeled tapes, sets of labels preceding and following each tape file
indicate, for example, the record length (in characters) and the file length
(in blocks). Also, the first two records of each volume are volume labels that
contain information common to all files on the volume. They are called the
Volume Header Label and the" Extended Volume Label. The Extended
Volume Label is present only for those volumes created on a TOPS-lO
system. This does not present a problem when transferring tapes between
systems.

5.3.1

Volume Sets, Files, Records

Data recorded by the operating system on magnetic tape is logically divided
into volume sets, files, and records.

Volume Sets
A volume set comprises one or more (up to 60) magnetic tape reels logically
treated as one volume. That is, to user programs, a volume set is a collection of files on one volume.
A tape within a volume set may contain one or more files, or it may contain
one section of a multivolume file. A volume set may consist of a number of
files, any of which may contain multiple sections. For example, a volume
set could contain three volumes, as represented in Figure 5-1.

Figure 5-1:

Volume Set of Labeled Tapes

Volume 1

Volume 2

Volume 3

Labels

File 1

Continuation
of File 2
(Section 2)

File 4

Labels

File 2
(Section 1)

File 3

File 5

Labels

Labels
MR-S-2255-82

Volumes land 2 contain a complete file and a file section. Volume 3 contains two complete files. Note that all sections of a file are contiguous.
Refer to Section 5.3.2.2, Tape Label Sequences, for illustrations of a single
file and multiple files on single- and multiple-tape volume sets.
Magnetic Tape Structure

5-3

Each volume within the set must have a unique VOLID, specified during
initialization.
The organization of files on one or more unlabeled tape volumes is shown in
Figure 5-2.

Figure 5-2:

Volume Sets of Unlabeled Tapes
Single File
Multi-Volume

Single File
Single Volume

Multi-File
Single Volume

File A
File

File

File
Cont.

File B

TM
MR-S-2256-82

Files
The files in a volume set are known as a file set. Each file can consist of one
or more records. When there is only a section of a file on a magnetic tape
volume, the :file is said to be a multiple-volume or multivolume file. Tape
labels define the position of each file/file section of a file set. Each file
within the set must have a unique filename.

Records
A logical record is usually a segment of a file.
Record format refers tp the relation of blocks to records. By default, the
record and block lengths are equal and are established by the operating
system. If you use the system default value when you create a file, the
labels correctly indicate the file format. However, if you choose a different
record format, you must override the system default values before you initiate the first output to the file. You can change the record format with the
.TFLPR function of the TAPOP. monitor call. Using this function, you can
indicate the presence of variable-length records, spanned records, or
multiple-record blocks. (Refer to Chapter 7, Record Formats.)

5.3.2 Labels
Each label is an 80-character record. With only one label per block, the
labels are separated from each other by interrecord gaps. Labels are separated from file data by tape marks. The record size for all labels generated
by the system remains at 80 characters regardless of the logical record and
block sizes selected by the system or the user program for file data.
5-4

Magnetic Tape Structure

Although a label generated by the system is 80 characters in length, longer
labels generated by other systems are accepted, but only the first 80 characters are processed.
Section 5.3.2.2, Tape Label Sequences, shows how labels are placed
throughout a volume set. Various volume set configurations are compared.
By way of introductIon, the section below (5.3.2.1) explains the
label-grouping scheme.
5.3.2.1 Classification of Labels As mentioned in Chapter 1, there are
several types of labels (HDR, EOV, and others). Each type usually includes
more than one label. This is because a single 80-character label is sometimes not long enough to contain all the information necessary to describe
the file data. Together, the labels within a type constitute a label set. Label
set members carry the same label identifier but differ in respect to label
number. (The label identifier and label number are among the information
fields in a label - refer to Chapter 8, Label Formats and Contents.) For
example, there are two file header labels, named HDR1 and HDR2; and
there are two file trailer labels, called EOF1 and EOF2.

A label group contains all the adjacent label sets required to define the
following file or the preceding file. Each label group is separated from the
file data by a tape mark. A tape mark also separates an end-of-file label
group from the following beginning-of-file label group. The label groups
along with the labels and label sets they contain are shown in Figure 5-3.

Figure 5-3:

Label Groupings
Volume Label 1
(VOL1)
Extended Volume
Label 1 (UVL 1)

Beginning of
Volume Label
Group

Header Label 1
(HDR1)

File Header
Label Set

Header Label 2
(HDR2)

•

File 1 Data

••
•

End-of-File
Label 1 (EOF1)
End-of-File
Label 2 (EOF2)
Header Label 1
(HDR1)
Header Label 2
(HDR2)

••
•

First Section
of File 2

End of File
Label Set

End of File or of
Last File Section
Label Group

File Header
Label Set

Beginning of File
or of File Section
Label Group

E~ge~~~flume

End of First or of
Intermediate File
Section Label Group

•••

End-of-Volume
Label 1 (EOV1)

....----------1
End-of-Volume
Label 2 (EOV2)

I
I
I I
I I
I
I
Beginning of
Volume Label
Set

MR-S-2257-82

Magnetic Tape Structure

5--5

Note that at the beginning of a volume, the file-header label set belongs to
the beginning-of-volume label group.
Also note that in most cases a label set corresponds to a label group. The
nonsupport of user-label sets on TOPs-"IO systems causes this situation.
For example, the ANSI-specified user-header label set, if implemented
with TOPS-IO, would belong to. the beginning-of-file or
beginning-of-file-section label group along with the file-header label set.
The sequence of the various tape labels
differs according to the number of files contained on a volume and whether
a file is continued onto another tape. The volume set variations, which
determine the label sequences, are listed below:

5.3.2.2 Tape Label Sequences -

• Single file on a single volume
• Single file continued onto one or more volumes.
• Multiple files on a single volume.
• Multiple files on multiple volumes.
Figures 5-4 through 5-7 illustrate the tape label sequencing possibilities
based on these variations. In the figures, IRG stands for interrecord gap.
Note that a tape ends with either an end-of-file tape label or an
end-of-volume tape label; but never both. If it ends with an end-of-file
label, the last file on the tape is not continued onto another tape, and this
volume is either the last volume or the only volume in a set. If the last label
on a tape is an end-of-volume label, the last file on the tape is continued
onto another volume in the set.

Figure 5-4:

Single File on a Single Volume
Beginning of Tape
Volume Header
Label (VOL 1)

IRG
Extended
Volume Label
IRG
Header Label 1
(HDR1)
IRG
Header Label 2
(HDR2)
Tap e Mark (EOF)

··•

File Data
(File 1)

•
••

Tape Mark (EOF) --~~--------4
End-of-File
Label 1 (EOF1)
IRG-----------.~------------~

End-of-File
Label 2 (EOF2)
"--_
_ _ _ _ _........._--<1 Tape Mark (EOF)

I Tape Mark (EOF)
MR-S-2258-82·

5-6

Magnetic Tape Structure

Figure ~5:

Single File on Multiple Volumes
1st Volume

2nd Volume

Beginning
ot Tape

Volume Label 1
(VOL1)

IRG
Extended Volume
Label (UVL 1)

Extended Volume
Label (UVL 1)
IRG
Header Label 1
(HDR1)
IRG
Header Label 2
(HDR2)

•
••

File 1 Data
(Section 1)

••
•

End-ot-Volume
Label 1 (EOV1)

Header Label 2
(HDR2)

Ta pe Mark
(EOF)

r - - - - - - - - -..............-

Header Label 1
(HDR1)

..
••
•

•
••

File 1 Data
(Section 2)

Tape Mark - -.......- - - - - - - - - - - - i
End-ot-File
Label 1 (EOF1)

......----IRG---~r---------,
End-ot-Volume
End-ot-File
Label 2 (EOF2)
Label 2 (EOV2)
f4---- Tape Mark ~
' - - - - - - - - - -.............Tape Mark - -.....- - - - - - - - - - - '
~--------~

MR-S-2259-82

Figure ~6:

Multiple Files on a Single Volume
~ Beginning ot Tape

Phy sical Record
Deli miter

Volume Label 1

IRG
Extended Volume
Label (UVL 1)
IRG
Header Label 1
(HDR1)
IRG
Header Label 2
(HDR2)
Tape Mark (EOF)

Tap e Mark (EOF)

·••

File Data
(File 1)
End-ot-File
Label 1 (EOF1)

IRG
End-ot-File
Label 2 (EOF2)
Tap e Mark (EOF)
Header Label 1
(HDR1)
IRG
Header Label 2
(HDR2)
Tap e Mark (EOF)

·••

File Data
(File 2)

··•

Tape Mark (EOF) ----.......----------------i
End-ot-File
Label 1 (EOF1)
IRG------~~----------~

End-ot-File
Label 2 (EOF2)
'--_ _ _ _ _ _ _ _....._-JfTape Mark (EOF)
~
"I Tape Mark (EOF)
MR-S-2260-82

Magnetic Tape Structure

5-7

Figure 5-7: Multiple Files on Multiple Volumes
1st Volume

••• Last Volume

Beginning
ot Tape

Volume Label 1
(VOL1)

Volume Label 1
(VOL1)
IRG

Extended Volume
Label (UVL 1)

Extended Volume
Label (UVL 1)
IRG

Header Label 1
(HDR1)

Header Label 1
(HDR1)
IRG
Header Label 2
(HDR2)

•

··

Header Label 2
(HDR2)

·••

File Data
(File 1)

Ta pe Mark

••
•

Ta pe Mark

••
•

File Data
(Cont. File 2)
End·ot·File
Label 1 (EOF1)

End·ot·File
Label 1 (EOF1)
IRG
End·ot·File
Label 2 (EOF2)

End·ot·File
Label 2 (EOF2)
Ta pe Mark

Header Label 1
(HDR1)

Header Label 1
(HDR1)
IRG

--'"

Header Label 2
(HDR2)

•
••

Header Label 2
(HDR2)
Ta pe Mark

••
•

File Data
(File 2)
End·ot· Volume
Label 1 (EOV1)

·••

••
•

File Data
(File 3)

Tape Mark ----+

End·ot·File
Label 1 (EOF1)
IRG

End·ot·Volume
Label 2 (EOV2)

End·ot·File
Label 2 (EOF2)
Tape M a r k Tape M a r k MR-S-2261 -82

Several situations cause tape labels to be written in sequences that differ
from those shown in Figures 5-4 through 5-7:
• If the system senses the' end-of-tape marker while it is writing the
beginning-of-file label group, it finishes the beginning-of-file label
group but does not write any data blocks in this file section. Instead, it
closes the volume and ~~continues" the file on the next volume. The file
section number field ofHDRl is 1 on the first volume for this file and 2 on
the continuation volume. (See Figure 5-8.)
I

Figure 5-8:

Coincidence of Beginning-of-File Group and
End-of-Tape Marker
·····Last Section of File A···*EOF1, EOF2*HDR1, (EOT)HDR2**EOV1, EOV2**
(A)
(B)
(B)

VOL 1 HDR1, HDR2*······First Section ot File B Containing Data······
(B)
MR-S-2262-82

5-8

Magnetic Tape Structure

• If the system senses the end-of-tape marker while it is writing the last
data block of a file, it completes the data block and closes the volume.
Then it continues the file onto the next volume even though there are no
more data blocks to be written for this file. An end-of-file label group
follows the empty file section on the new volume. (See Figure 5-9.)
Figure 5-9:

Coincidence of Last Data Block and
End-of.,....Tape Marker

······Last section of File A Containing Data·····*EOV1. EOV2**
(A)

VOL 1 HDR1, HDR2**EOF1, EOF2*HDR1, HDR2*····First Section of File B·····
(A)
(A)
(B)
MR-S-2263-82

• If the system senses the end-of-tape marker while it is writing the
end-of-file label group and the file is not the last file of the set, it finishes
the end-of-file label group and writes the beginning-of-file-section label
group for the next file. It then closes the volume without writing any data
blocks for the new file. The file is ~~continued" onto the next volume. The
result of this situation is the same as the first situation described above.

Magnetic Tape Structure

5-9

Chapter 6
Magnetic Tape Data Modes
Chapter 6 describes how data is stored on magnetic tapes in each of the
data modes. These are hardware data modes that relate to a tape's physical
structure. They can be set with the .TFMOD function of the TAPOP.
monitor call (Code 1007/2007). (Refer to Chapter 11, The TAPOP. Monitor
Call, for information on how to set the desired data mode.)
Sections 6.1 through 6.4 describe the placement of data bits on tape tracks
and indicate the number of frames r~quired to store a word of data. The
diagrams in these sections are logical representations of data on a magnetic
tape. Actually, the parity frame is in the center of the tape, and the order of
the other frames is not necessarily as pictured.

6.1

DEC-Compatible Core Dump Mode (7- or 9-Track)
Code 0 (.TFMDD) sets DEC-compatible core dump mode for either 7-track
or 9-track magnetic tape (MTAPE 100 also sets this mode.) The choice of
types depends on the magnetic tape unit used. This data mode function code
is the equivalent of code 1 (.TFMID) for 9-track tapes and code 5 (.TFM7T)
for 7-track tapes. On a 9-track tape, DEC-compatible core dump mode
stores one 36-bit word of data in five frames. Note that the last frame is
only half used. One word actually requires four and one-half frames. Each
data word in 9-track DEC-compatible core dump mode is formatted as
shown below.

o
1st
frame

2nd
frame

3rd
frame

4th
frame

5th
frame

6--1

For each data word in memory, there are five tape bytes per 36-bit word,
with parity bits unavailable to the user. Bytes are read and written on the
tape as shown below.
TRACKS
9

BO
B8
B16
B24
0

B1
B9
B17
B25
0

B2
B10
B18
B26
0

B3
Bll
B19
B27
0

B4
B12
B20
B28
B32

B5
B13
B21
B29
B33

B6
B14
B22
B30
B34

B7
B15
B23
B31
B35

FRAMES

When writing on TM10s, bits 30 and 31 are written twice. First they are
written as shown in the diagram above, and then they are copied in tracks
7 and 6 of byte 5. Tracks 9 and 8 of byte 5 remain zero. When writing with a
DX10, TM02-C, or TM10-C, bits 30 and 31 are written only once. Frames
9, 8, 7, and 6 contain zero.
When reading from a TM10, parity bits 9 and 8 of frame 5 are ignored.
Frame 4, tracks 2 and 3 are ORed with Frame 5, tracks 6 and 7. These are
bits 30 and 31 of the data word. On a DX10, TM02-C, and TC10-C, the
parity bits along with frames 9, 8, 7, and 6 are ignored.
In 7-track core dump mode, one 6-bit byte is stored in each frame of the
tape. Six frames are required to store one 36-bit word. This mode is the
only hardware mode supported for 7-track magnetic tapes.
Each data word in 7-track DEC-compatible core dump mode is formatted
as shown below.

17 18

23 24

1st
frame

2nd
frame

4th
frame

3rd
frame

29 30

5th
frame

6th
frame

Bits are read and written on the tape as shown below.
TRACKS

6--2

Magnetic Tape Data Modes

BO
B6
B12
B18
B24
B30

Bl
B7
B13
B19
B25
B31

B2
B8
B14
B20
B26
B32

B3
B9
B15
B21
B27
B33

B4
B10
B16
B22
B28
B34

B5
B11
B17
B23
B29
B35

FRAMES

6.2 Eight-Bit Mode
Code 2 (.TFM8B) sets 8-bit mode for 9-track tapes. This mode can also be
set with MTAPE channel, 101. This mode is compatible with any machine
that reads and writes 8-bit bytes (for example, System 360/370 and
PDP-11s). When a read operation is performed, four frames (that is, 8-bit
bytes) are read into each word (left justified), with the remaining four bits
of the word containing zeroes or copies of the parity bits. The TM02 and
TM10-C copy the parity bits. The DX10 and TC10C return zeroes in these
four bits. The information read into bits 32 through 35 is not user data. On
a write operation, the leftmost four 8-bit bytes of each word are written out
in four frames; the remaining 4 rightmost bits (32 through 35) are ignored.
Only bits 0 through 31 are written onto the tape.
Each data word in 9-track industry-compatible core dump mode is formatted as shown below.
15 16
1
2nd
frame

1st
frame

23 24

4th

3rd
frame

frame

Bits are read and written on the tape as shown below.
TRACKS
7

BI0
B18
B26

B3
Bll

B12

B5
B13

B19

B20

B21

B27

B28

B29

BO
B8
B16

B17
B25

B24

B7
B15

B14
B22
B30

~~ ~ FRAMES
~

B23
B31

When using industry-compatible mode and buffered I/O, your program
must insert the byte size in the byte pointer before issuing the first IN,
INPUT, OUT, or OUTPUT monitor call.

6.3 Sixbit Mode
CODE 3 (.TFM6B) sets 6-bit (SIXBIT) mode for 9-track magnetic tapes.
This mode is available only on TU70s. The format of the data word is shown
below.
23 24

29 30

1st
frame

2nd
frame

3rd
frame

4th

5th

6th

frame

Magnetic Tape Data Modes

&-3

Bits are read and written on the tape as shown below.
TRACKS
9

0
0
0
0
0
0

BO
B6
B12
B18
B24
B30

B1
B7
B13
B19
B25
B31

B2
B8
B14
B20
B26
B32

B3
B9
B15
B21
B27
B33

B4
B10
B16
B22
B28
B34

B5
B11
B17
B23
B29
B35

~~
~j

FRAMES

6.4 ANSI-ASCII Mode
Code 4 (.TFM7B) sets 7-bit mode, called ANSI-ASCII (not available on
TMIOs and TCIO-Cs). This mode stores one 7-bit ASCII byte in each frame
of the tape. This mode is useful for transferring ASCII data from TOPS-IO
systems to 8-bit byte-oriented machines (for example, PDP-lIs, System
360/370). Data must be stored in ANSI-ASCII mode to meet ANSI standards. Five left-justified (in core) 7-bit bytes are stored in five frames on
the magnetic tape. Bit 35 must be zero to conform to ANSI standards. Bit
35 is written into the high-order bit of the last frame of each word. The
other high-order bits are set to zero on write operations. When the tape is
read, all five high-order bits are ORed, and the result is stored in bit 35.
Each data word in ANSI-ASCII mode is formatted as shown below.
13 14

20 21

2nd
frame

1st
frame

4th
frame

3rd
frame

5th
frame

Data is read and written on the tape in the following format.
TRACKS

6-4

0
0
0
0
B35

BO
B7
B14
B21
B28

B1
B8
B15
B22
B29

B2
B9
B16
B23
B30

B3
B10
B17
B24
B31

B4
B11
B18
B25
B32

B5
B12
B19
B26
B33

B6
B13
B20
B27
B34

Magnetic Tape Data Modes

FRAMES

Chapter 7
Record Formats

The record formats used on a labeled tape conform to the ANSI standard.
There are three formats available to meet the diverse needs of your applications and hardware. Sections 7.2 through 7.5 describe the various record
formats. Note that records of only one format type can be recorded in the
same file.

7.1

Blocking of Records
Logical records may be blocked; that is, more than one logical record may
reside in a physical tape block. Records are usually blocked when the physical block is at least twice the logical block (record) size, and the record type
is either fixed or variable. Spanned records are also likely to be blocked, no
matter what the ratio is between block size and record size. In short, a
blocked record is a record that is not equal in size to the block. An unblocked record is equal in size to the block.
The ANSI standard allows you to distinguish between the demands of the
hardware and those of your program by allowing you to specify the physical
block size independent of the logical record structure. The block size you
select should be optimal for the recording density, the drive type, and the
hardware interface necessary to transmit the data. You should select the
logical record size and type so that your program can process the data as
efficiently as possible.

NOTE
Your program must perform its own blocking, deblocking,
and data translation operations. The system does not automatically perform these functions.

7-1

Also note that neither the TOPS-I0 monitor nor the label
processor enforces the characteristics specified in the tape
labels. Thus, for example, if your program attempts to read
or write a logical record of a size different from that specified,
it receives no error. You are free to read or write data in any
format.

7.2 Fixed-Length Records (F Format)
With fixed-length records, the logical record length is the same for all
records. The block length is also the same for all blocks. In addition, the
block length must be an integral multiple of the record length. The blocking factor for fixed format is 1 (for unblocked records) or greater than 1 (for
blocked records).
Fixed-length record blocks may contain fewer records than the possible
maximum. Thus, even though the block length is the same for all blocks,
you do not need to completely fill all blocks. For example your program
might write its last record in the middle of a block.
J

Figures 7-1 and 7-2 illustrate the formats of fixed-length records.
Figure 7-1:

Fixed-Length Records (F), Unblocked

Record

Block

Block
MR-S-2264-82

Figure 7-2:

Fixed-Length Records (F), Blocked

REC

Block

REC

Block
MR-S-2265-82

7.3 Variable-Length Records (0 Format)
Variable-length records provide an increased level of sophistication over
fixed-length records. With variable-length records, your program can
make a decision to block or not block as each record is produced. You should
use variable length when the record size is less than or equal to the block
size, or when records are not of uniform length.

7-2

Record Formats

With variable-length records, a record control word (RCW) precedes each
record. The RCW consists of four characters. The record length (in characters), including the length of the RCW, is expressed as a decimal number
occupying the entire RCW. The RCW is necessary so that a program reading the data can find out the length of each record. If your file is blocked,
each block may contain multiple records, each of which is preceded by its
own RCW. Because variable-length records are not of uniform length, the
number of records that are in each block can vary from one block to another. Also, the block sizes can vary from one block to another. Figures 7-3,
7-4, and 7-5 illustrate the formats of variable-length records.

Figure 7-3: Variable-Length Records (D), Unblocked

Record
RCW

Rei cord

RCW

Block

Block
MR-S-2266-82

Figure 7-4: Variable-Length Records (D), Blocked

Record
RCW

Record

Record
RCW

RCW

Block
MR-S-2267-82

Figure 7-5: Unblocked Variable-Length Records
1780

Data 1776 Characters

1988

Data 1984 Characters

~---.;>

RCW
Each line represents a block.

MR-S-2268-82

Record Formats

7-3

7.4 Spanned Records (S Format)
The logical record size that you wish to use could be larger than the ideal
physical block size for the device. To accommodate this situation, you could
use spanned records. A spanned record is one that may occupy more than
one physical tape block, and may also occupy tape blocks on different volumes. Varying length blocks are permitted for segments of spanned
records.
With spanned records, the first five characters of each segment make up
the segment control word (SeW). The first character of the sew is called
the Spanning Indicator. It has the following meaning:
0- record begins and ends in this segment
1- record begins but does not end in this segment

2- record neither begins nor ends in this segment
3- record ends but does not begin in this segment

The next four characters of the sew indicate the segment length (in characters), including the sew. This value is expressed as a decimal number.
For spanned records, there is no explicit record control word that expresses
the total record length.
These records may span volumes. The record length is unbounded in that
there is no limit to the number of segments in one record. But, there is only
one segment of the same record in a block.
The segments of a record are written in consecutive order. That is, segment
n + 1 is written immediately following segment n. No segments from other
records are interspersed.
Figures 7-6, 7-7, 7-8, and 7-9 illustrate spanned record formats.
Figure 7-6:

Spanned Records (S), Unblocked

Record

sew

__B_IO_Ck__

~11~___

I I
Record

" '"

Record Formats

1 1____ ~

B_IO_Ck____

The first block shows the maximum blocksize.

7-4

I
I

B_IO_Ck__

MR-S-2269-82

Figure 7-7:

Spanned Records (S), Blocked

Record

r. . . . . . . . .
I

........ - .....

--------~---+----~

Record

------_,

sc~1

SCW

Segment

11~_________B_I_OC_k__________~11~__________B_IO_Ck________~

________B_IO_C_k__________

All blocks show the maximum blocksize;
the last record continues In the next subsequent block.

Figure 7-8:

MR~S~2270~B2

Unblocked Spanned Record

I 2048

2043 Data Characters

2048

1~3-,1_0_1_6_0---,,_ _ _ _ _ _~1----- 155 Data Characters
~

Segment
Control
Word
The record lenght is 4241 characters;
each line represents a block.

MR-S-2271-B2

Record Formats

7-5

Figure 7-9: Blocked Spanned Records

2043 Data Characters
SCW

121

2048

2043 Data Characters

SCW

145 Chars.
~

1893 chars.1
~

SCW

2043 Data Characters
~

SCW

2000 Characters
~

SCW
Record 1: Length of record is 4231 characters.
Record 2: Length of record is 5936 characters.
Each line represents a block.
MR-S-2272-82

7.5 Undefined Records (U Fo.rmat)
When records do not meet the specifications described in Sections 7.2, 7.3,
or 7.4, the records are undefined in format.

7-6

Record Formats

Chapter 8
Label Formats and Contents

This chapter describes DEC- (ANSI-) and IBM-formatted tape labels. All
DEC-formatted labels follow the ANSI Tape Label Standard; therefore, the
terms DEC-formatted and ANSI-formatted are equivalent. There are some
fields defined in the ANSI Tape Label Standard as being reserved for operating system information. The DEC-formatted tape labels illustrated and
described in this chapter define the contents of these fields to agree with
the contents of the DIGITAL Tape Label Standard. Any differences between DEC-formatted tape labels and IBM-formatted tape labels are described in the text of this chapter.
The types of tape labels are listed below. These are described in the remainder of this chapter.
• Volume header label.
• Extended volume header label.
• File header label 1.
• File header label 2.
• End-of-file label 1.
• End-of-file label 2.
• End-of-volume label 1.
• End-of-volume label 2.
Each label is an 80-character record on the magnetic tape. On the following pages, each type of label is illustrated with the 80 characters divided
into information fields. Solid lines separate these fields from one another.
The field name appears on the right side of the illustration. The numbers
on the left side of the illustration indicate the first character position of

8--1

that information field within the tape label. And the numbers within
square brackets on the right side of the illustration correspond to the
textual description following each tape label illustration.
Note that fields in the EOF and EOV label sets match those in the HDR
label set. This allows identical label processing when reading a file
backward.
All labels contain a label identifier and label number in addition to the
various control information fields. The first three (1-3) characters form the
identifier field; and the fourth character, a numeric, indicates the position
of the label within the set. (Refer to Section 5.3.2.1, Classification of Labels,
for a description of label sets.) Characters 5 through 80 make up the control
information fields.

8-2

Label Formats and Contents

Volume Header Label (VOLl)
--------------~ [1]

Label Identifier

[2]
4
5 I--------------~ [3]

Label Number
Volume Identifier

11 1------------------1 [4]
12
[5]

Accessibility Character
Reserved

38 ~----------------t [6]

Owner Identifier

51 I--____________~ [7]
52
[8]

DEC Standard Version
Reserved

80 ~_ _ _ _ _ _ _ _ _ _ _ _ _ [9]

Label Standard Version
MR-S-2275-82

Label Formats and Contents

8-3

The VOL1 label is always the first label on a magnetic tape reel. Its main
function is to facilitate the interchange of tapes among installations by
identifying the sender.
1.

The Label Identifier field always contains the three ASCII characters
VOL.

2. The Label Number field always contains the character l.
3. The Volume Identifier field consists of a combination of legal characters
(A through Z and 0 through 9), left-justified. There is a restriction that
all of the character positions cannot be zeros. You specify this field with
the VOLID switch in the MOUNT command. The operator initially sets
this field when initializing the tape; refer to Chapter 2 for a description
of initialization. Each VOLID should be unique, but there is no check
for uniqueness.
4. The Accessibility Character field contains a space or the character 1,
and indicates whether the volume is protected. It should contain a space
for any tapes that are to he interchanged between systems. Note that if
the operator specifies a protection code of 0 during initialization, the
accessibility character is zero or blank. For DEC-formatted tape labels,
this field is always 1 if the operator specifies a protection code. For
IBM-formatted tape labels, this field can be 0, 1, or 3. A 0 indicates
that the volume is not protected; a 1 indicates that the volume is
write-protected; a 3 indicates that the volume is read-protected.
5. Reserved fields in the tape label consist of spaces.
6.

The Owner Identifier field is ignored when reading IBM-formatted
tape labels and is not defined in the ANSI standard. For
DEC-formatted tape labels, the Owner Identifier field is divided into
the following parts:
character 38

is always D.

character 39

is always %.

character 40

is A, meaning the DECsystem-10. For other digital systems, character 40 is as follows:
Table 8-1

8-4

Label Formats and Contents

Machine Codes

Code

Machine

PDP-8

PDP-11

VAX-11

PDP-15

DECSYSTEM-20

characters 41-45

are system-dependent; with TOPS-IO they make
up the operating system code, which is TIO for
TOPS-IO.

characters 46-50

are system-dependent; with TOPS-IO they make
up the system serial number, which the system
administrator sets at monitor generation time
through MONGEN. This number is decimal and is
right-justified.

The label processor ignores this field if the first two characters are
not D%.
7. The DEC Standard Version field is not used for IBM- and ANSIformatted tape labels. For DEC-formatted tape labels it is HI".
8.

Reserved fields in the tape label consist of spaces.

9. The Label Standard Version field is the version number (currently 3) of
the ANSI label standard. This field is not used for IBM-formatted tape
labels.

Label Formats and Contents

8-5

Extended Volume Header Label (UVLl)
[1]

Label Identifier

4
5

~---------------4 [2]

[3]

Label Number
Access Code

[4]

Project-Programmer Number

[5]

Owner's Name

[6]

Reserved

80 '--_ _ _ _ _ _ _ _ _ _ ___
MR-S-2276-82

S-6

Label Formats and Contents

The UVLI label is always the second label on the magnetic tape reel for
tapes created on the TOPS-IO system. It supplements the VOLI label,
providing additional access information for the volume. This label is processed on input only if the Owner Identifier field in the VOLI label contains D%A. UVLI is processed on output for all labeled tapes.
On TOPS-IO systems, this label is treated as the second volume label, not
as a user label. Because the system does not support user labels, any UVL
label beyond the first one of the set is ignored.
1. The Label Identifier field always contains the ASCII characters UVL.

2. The Label Number field always contains the character 1.
3. The Access Code field contains the protection code for this volume. The
operator specifies the volume protection code during initialization.
Note that if this field contains zero, the label processor sets the
Accessibility Character in the Volume Header Label (VOLI) to space
for DEC-formatted tapes or zero for IBM-formatted tapes. Also, if this
field contains zero, the label processor does not protection-check any of
the files; the label processor allows access to all the files. You must set
this field to zero if you are planning to interchange tapes between
systems.
4.

The Project-programmer Number field contains the
project-programmer number associated with the owner of the files on
this magnetic tape. The operator sets this field when initializing the
tape.

5. The Owner's Name field contains the name associated with the file
owner when he/she logged onto the system. The operator initially sets
this field when initializing the tape.
6.

Reserved fields in magnetic tape labels consist of spaces.

Label Formats and Contents

8-7

First File Header Label (HDRl)
[1]
4

Label Identifier

1 - - - - - - - - - - - - - - - 1 [2] Label Number
[3]

File Identifier

[4]

File-Set Identifier

[5]

File Section Number

[6]

File Sequence Number

[7]

Generation Number

[8] Generation Version

[9]

[10] Expiration Date

Creation Date

54 I--_ _ _ _ _ _ _ _ _ _ _~ [11] Accessibility Character
55
[12] Block Count

[13] System Code

[14] Reserved

80~

__________________

MR-S-2277-82

The HDRI label is always the third label on a magnetic tape reel. It helps
the system identify, protect, and audit a file. There is one HDRI label for
each file or each section of a file contained on a reel. (Refer to Section
5.3.2.2 for information on tape label sequences for multiple files on a single
reel or on multiple reels.)
8-8

Label Formats and Contents

1. The Label Identifier
characters HDR.

field

always

contains

the

three

ASCII

2. The Label Number always contains the character 1.
3. The File Identifier field (file name) is supplied by the label processor on
output and is ignored on input. It can be set with the .TFLPR function
of the TAPOP. monitor call. This field is written as a 17-character
ASCII string.
4. The File Set Identifier is the VOLID of the first volume of the set. For
example, if this field is the same as the VOLID in the VOL1 label of
this reel, this reel is the first volume of the set. For IBM-formatted
labels, this field is ignored on input, but written on output.
5. The File Section Identifier indicates whether a file is continued from
another volume. A file always begins with a section number of 0001. If
that file is continued onto another volume, the file header block for that
section of the file has 0002 in the File Section field. For example, see
Figure 8-1 below. There are two files on one volume. The section numbers of the files are written above each segment of the file.

Figure S-1: File Section Numbers
File Section

0001

-"-

Volume 11'-_L_ab_e_ls_.L...-_F_ile_1_-'--_L_a_be_ls.............L--_Fi_le_2_-L-_La_b_e_ls---'
MR-S-2273-82

6. The File Sequence Identifier field specifies the relative position of this
file in the file set. See Figure 8-2, where there are three files on two
volumes. The File Sequence Identifier field is incremented for each successive file in the set.

Figure S-2:

Volume

File Sequence Numbers

11 Labels

0001

File 1

Labels

0001

Volume 2

File 2

File Section #

0002

File Sequence #

0002

0001

File Section #

File 2
Cont.

Labels

File 3

0002

0003

Labels

File Sequence #
MR-S-2274-82

Label Formats and Contents

8-9

7. The Generation Number field always contains the character 1. The
label processor ignores this field on input, but writes it on output.
8. The Generation Version field always contains o. The label processor
ignores this field on input but writes it on output.
9. The Creation Date field contains the Julian date on which the file was
created, in the form yyddd. The label processor ignores this field on
input, but writes it on output.
10. The Expiration Date field specifies the date after which this file and all
subsequent files in the set can be overwritten. You can specify this date
with the TAPOP. monitor call. If you do not specify the expiration date,
the label processor assumes the creation date is the expiration date. If
the current date is equal to or later than the expiration date, you can
overwrite this and all following files.
11. The Accessibility Character field specifies whether this file will be
protection-checked. For DEC-formatted labels, this character can be a
space or a 1. A space indicates that all accesses will be allowed to the
file; there will be no protection checking. A 1 indicates that the file will
be protection-checked according to the protection code in the HDR2
label. For interchanging tapes between systems, this field should be
blank and there sh~uld be a 0 protection code.
For IBM-formatted tapes, this character can be 0 or 1. If it is 0, all
accesses are allowed to the file; there is no protection checking. The
code is checked only when the tape is written on a TOPS-10 system. If
this field is 1, no access is allowed to the file. If interchanging tapes
between systems, this field must be o.
12. In HDR1, the Block Count field always contains O. But in EOF1 and
EOV1 this field is the number of blocks recorded since the preceding
beginning-of-file or beginning-of-file-section label group.
When reading backward, you can use the EOVlIEOF1 Block Count to
establish a count that is decremented by one for each block read, and is
compared with the HDR1 Block Count (zero) at the beginning of the file
section. By this process, you can determine whether any blocks were
skipped or whether any spurious blocks were inserted during the read.

8-10

Label Formats and Contents

13. The System Code field is DECSYSTEM10. Codes for other DIGITAL
systems are:

Table 8-2:

System Codes

Code

System

DECFILEIIA

RSX, lAS

DECRTIIA

RT-ll

DECRSTS/E

RSTS/E

DECFILEl12

VMSRMS

DECSYSTEM20

TOPS-20

14. Reserved fields in labels consist of spaces.

Label Formats and Contents

8-11

Second File Header Label (HDR2)
[1 ]

Label Identifier

4
5
6

[2]
[3]
[4]

Label Number
Record Format
Block Length

[5]

Record Length

[6]

File Access Code

[7]

Owner's Directory

[8]

System Dependent

37 ~_ _ _ _ _ _ _ _ _ _ _ _--I [9] Form Control Character
38
[10] Reserved

[12] Reserved

________________________

[11] Buffer Offset Length

MR-S-227B-B2

8--12

Label Formats and Contents

The HDR2 label is always the fourth tape label from the beginning of a
magnetic tape. Also, it is always the label immediately preceding a file.
HDR2 specifies the user-assigned file attributes.
1. The Label Identifier field always contains the three ASCII characters
HDR.
2. The Label Number field always contains the character 2.
3. The Record Format field can contain F (fixed), D (variable), S
(spanned), or U (undefined). The default format is U for both DEC and
IBlV[ records. Refer to Chapter 7 for more information on record
formats.
You can change the contents of this field with the .TFLPR function of
the TAPOP. monitor call. Refer to Chapter 11 for a description of
TAPOP.
4. The Block Length field contains the number of characters within a
block. For DEC-formatted labels, this value can range from 0 to 99999.
For IBM-formatted labels, the maximum is 32,760 characters per
block.
5. The Record Length field specifies the number of characters per record.
By default, this value is the same as the block length field and can
range from 1 to 99999.
6. The File Access Code field contains the protection code for the file.
Refer to Section 3.5 for a description of protection codes.
7. The Owner's Directory field is the project-programmer number of the
file owner.
8. The System-Dependent field is unused in DEC-formatted labels. For
IBM-formatted labels, it contains file usage data, which is ignored on
input.
9. The Forms Control Character field contains a space, an ASCII A, or an
ASCII M. A space indicates that the file contains no forms control information. That is, if carriage returns and/or linefeeds are present in the
file, the label processor ignores them. An A indicates that the first
character in each record of the file contains a control character. An M
indicates that the file's records contain all necessary format control
information. This field is processed as a space and is ignored on input if
it is not a space.
10. Reserved fields in tape labels consist of spaces.
11. The Buffer Offset Length field contains the length of the information
that prefixes each data block. For DEC-formatted labels, this field is
written as 0, and ignored on input.
12. Reserved fields in tape labels consist of spaces.

Label Formats and Contents

S-13

First End-·of-File Label (EOFl)
[1]

Label Identifier

4
5

[2]
I - - - - - - - - - - - - - - - - - - t [3]

Label Number
File Identifier

[4]

File-Set Identifier

[5]

File Section Number

[6] File Sequence Number

[7]

[8] Generation Version

[9]

[10] Expiration Date

~~~~~~~~~~~~~~

Generation Number

Creation Date

[11] Accessibility Character
[12] Block Count

[13] System Code

[14] Reserved

80 '--_ _ _ _ _ _ _ _ _ _ _ _---'
MR-S-2279-82

S-14

Label Formats and Contents

The EOFI label is always the first tape label following a tape mark (EOF)
at the end of a file. Refer to Section 5.3.2.2, Tape Label Sequences, for
diagrams showing the layout of labels, marks, and data on tapes.
The EOFI label is essentially the same as the HDRI label. Their differences are listed below. The information fields shown in the EOFI illustration that differ from those in HDRI are shaded with slashes (/111).
1. The Label Identifier always contains the three ASCII characters EOF.
2.

The Block Count field contains the number of blocks recorded for this
file section, that is, the length of this file section in blocks. This value
can range from 0 to 99999.
You can use the EOFI Block Count field to determine, when reading a
file section, whether any blocks were skipped or whether any spurious
blocks were inserted. Maintain a count of blocks read and compare this
count with the EOFI Block Count.

Label Formats and Contents

8-15

Second End-of-File Label (EOF2)
[1 ]

Label Identifier

5
6

[2]
[3]
[4]

Label Number
Record Format
Block Length

[5]

Record length

[6]

File Access Code

[7]

Owner's Directory

[8]

System-Dependent

37 r----~---------t [9] Form Control Character
38
[10] Reserved

[11] Buffer Offset Length

[12] Reserved

80 ~----------.........
MR-S-22BO-B2

The EOF2 label always immediately follows an EOFI label.
The EOF2 label is essentially identical to the HDR2 label. There is only
one difference: the Label Identifier field (shaded above, 11111) contains EOF
in the second end-of-file label (EOF2).

8-16

Label Formats and Contents

First End-of-Volume Label (EOVl)
~~~~~~~~~~~~~ [1]

4
5

Labelldentifier

Label Number
1--------------4 [2]
[3] File Identifier

[4]

File Set Identifier

[5]

File Section Number

[6] File Sequence Number

[7]

Generation Number

40
42

,8]

Generation Version

[9]

Creation Date

[10] Expiration Date

54 1<--------------4 [11] Accessibility Character
55
[12] Block Count

[13] System Code

[14] Reserved

80 ~--------------------~

MR-S-2282-82

The EOVl label is always the first label following a file section at the end
of the magnetic tape reel. That is, if end-of-file has occurred at the end of
the tape, there is no EOVl label; the EOFl label is there instead.
The EOVl label is essentially the same as the EOFl label. There is only
one difference: the Label Identifier field (shaded above, 11111) contains EOV
in the first end-of-volume label.
Label Formats and Contents

8-17

Second End-of-Volume Label (EOV2)
[1]

Label Identifier

4
5
6

[2]

l~1

Label Number
Record Format
Block Length

[5]

Record Length

[6]

File Access Code

[7]

Owner's Directory

[8]

System-Dependent

37 ~_ _ _ _ _ _~_ _ _ _--1 [9] Form Control Character
38
[10] Reserved

51
53

[11] Buffer Offset Length
~-------------~

80~

______________________

[12] Reserved

MR-S-22B2-B2

The EOV2 label is essentially identical to the EOF2label. There is only one
difference: the Label Identifier field (shaded above, 11111) contains EOV in
the second end-of-volume label (EOV2).

S-18

Label Formats and Contents

Chapter 9
The Mount Process for Labeled Tapes

9.1

Introduction
This chapter describes internal operations that are invoked by the MOUNT
command and by the operator's response to the command. Automatic
Volume Recognition, volume switching, and prestaging are among the
topics discussed. For details on operator procedures, refer to the TOPS-10
Operator's Guide and to the TOPS-10 Operator Command Language
Reference Manual.
Mount requests originate from user terminals, from batch control files, and
from PULSAR. In the first two instances, the first (or only) tape of a volume
set is mounted. With PULSAR requests, succeeding tapes in the volume set
are mounted as needed. Sections 9.2 through 9.4 discuss the mount process
in all three of these contexts.

9-1

9.2 Mounting the First Tape of a Volume Set
This section discusses the process that begins when a user issues the
MOUNT command from a terminal or from a control file. Figure 9-1 gives
an overview.

Figure 9-1: The MOUNT Process for the First Tape of a Volume Set

MOUNT
Command

User

Notify
User

Request

,...

MOUNT
Program

Acknowledge
Request

OPR

ORION

Ask Operator
to Mount

f--

Request

,..

QUASAR

Request
Label
Check

Receive
Label
Information
'---

PULSAR
Read
Labels

Tape
Drive
MR-S-2283-82

Assuming that the operator did not mount the volume before the MOUNT
command was issued and that Automatic Volume Recognition is enabled,
the following action takes place.
The MOUNT command sets off a cycle of events beginning with the
MOUNT program's request for devices and ending with the program's notification to the user that the mount was successful. Any errors that occur
break the cycle and cause appropriate messages to be sent to the operator.
The operator then informs the user of the error condition.
When the MOUNT command is issued, the MOUNT program gains control.
This program sends an IPCF message requesting the devices to the
QUASAR program. QUASAR, in turn, sends an IPCF message to ORION,
who notifies aPR. Upon receipt of ORION's message, aPR types information similar to the following on the operator's console:

9-2

The Mount Process for Labeled Tapes

10:55:18

-- Ma!:ttape !flount request #138-User: TUCKERtB [Z7t53£1Z] job# £17
VoluMe-set-naMe: T
Wr i t e

VoluMe-ID

Enabled

Labels
ANSI

TracK

Density
1800

The operator then mounts the tape on an available drive and places the
drive on line, at which point QUASAR, receiving notification from the monitor of an on-line interrupt, sends a message to PULSAR. This message
requests PULSAR to read the volume labels.
After the labels have been read, QUASAR ensures that the mounted tape is
the one that the user requested with the MOUNT command.
QUASAR also checks for other errors at this time. For example, it checks
the write-ring status of the tape unit. If the user requested write-lock
status, but the tape is write-enabled, or vice versa, QUASAR sends an
appropriate message to the operator.
QUASAR reassigns the tape unit to the user if it foresees no conflicts then
notifies the MOUNT program. Following this, a message similar to the
following appears on the user's terminal (or log file):
[Ma!:ttape T !flounted on MTAZ81 ",lith lo!:tical na!fle T]

9.3 Mounting the 'Next' Tape of a Volume Set
Volume switching is the term that describes the action of changing from
one mounted tape volume to the next one in the volume set. The operator
must perform this action when a multivolume file is being read or written.
This section describes what happens internally during volume switching.
Figure 9-2 gives an overview of the process.
As in Section 9.2, it is assumed that the operator has not premounted the
next volume and that AVR is enabled on the tape drive.

Figure 9-2:

Volume Switching

Read/Write
Tape
Drive

User
Program

-----

EOV
Reached
Release
Job

PULSAR

Requestf
Receive
Label
Information

Label
Check

-'"

Request
Next
Volume

QUASAR

Ask Operator
to Mount

ORION

----

OPR
MR-S-2284-82

The Mount Process for Labeled Tapes

9-3

When PULSAR encounters the end-of-volume label set during a tape read,
it knows that this is a multivolume file. After putting the job into the
event-wait (EW) state, PULSAR sends an IPCF message to QUASAR requesting the next volume; and it rewinds the recently processed volume.
The operator, by way of QUASAR, ORION, and OPR, receives notification.
The message received by the operator is of the same type as that issued for
the first volume of the set. After the operator mounts the requested tape on
a drive, PULSAR takes the job· out of the event wait state and proceeds as
with the first volume of the set.
If the INOTIFY switch was specified in the MOUNT command, a message
similar to the following is sent to the user's terminal or log file:
Frolll S}'steITl:
Lo~ical nallle T sl,dtched to \,10 1 Ullle B on MTA261

When PULSAR encounters the end-of-tape marker during a tape write, it
writes the end-of-volume label set; asks QUASAR for the next volume as it
does when reading a file; and updates fields in the beginning-of-volume
label group on the newly mounted tape (and writes new labels if necessary).
After this, the writing of file data can continue.

9.4 Automatic Volume Recognition
Automatic Volume Recognition (AVR) performs the following functions:
• Detects that a tape unit has come on line
• Verifies the tape on the unit by reading the labels
• Automatically assigns the newly verified tape to a waiting job
Section 9.2, Mounting the First Tape of a Volume Set, and Section 9.3,
Mounting the ~Next' Tape of a Volume Set, introduced the discussion of
these AVR functions. This section discusses A VR in more depth. It describes operations for premounted tapes and for tapes that satisfy multiple
pending requests. The hardware aspects of AVR are also described in this
section.
With A VR enabled, no operator intervention is required beyond mounting
a labeled tape.
The operator enables and disables AVR with the commands:
ENABLE
DISABLE

VOLUME-RECOGNITION arg

These commands can be given for all tape drives or for a specific tape drive.
AVR should remain enabled except during initialization of new tapes.
Disabling at this time (and specifying the /OVERRIDEEXPIRATION:YES switch) prevents the tapes from ~running away' when
PULSAR tries to verify labels.

9-4

The Mount Process for Labeled Tapes

9.4.1

Hardware

AVR can be fully utilized only on tape drives that generate on-line interrupts. The AVR process begins when the operator mounts a tape on an
AVR-enabled drive, causing the hardware to generate an interrupt to the
monitor when the tape is positioned at load point. The system is thereby
informed that the new tape unit is on line. (The TU70 series tape subsystem generates on-line interrupts.) Label-processing operations and tape
assignment then take place automatically.
For tape units that do not generate on-line interrupts, AVR can be made to
operate in semiautomatic mode. That is, the operator command,
RECOGNIZE, provides a way for the operator to generate the on-line interrupt.
The on-line tape unit interrupt is serviced by the physical device driver
(TXlKON) contained within the TOPS-lO monitor. This interrupt is translated into an IPCF message packet and sent to QUASAR.
9.4.1.1 Ignored Interrupts A VR acknowledges only those on-line interrupts for 9-track magnetic tape units. Since all 7-track tapes, by definition, are unlabeled, they are ignored by AVR. Also, tape units that are
assigned to user jobs in nonlabel or bypass label mode are assumed to be
under direct control of the operator; therefore no attempt is made to verify
any labels.

9.4.2 Volume Recognition
When QUASAR receives notification of an on-line interrupt, it requests
PULSAR to perform label-processing operations. PULSAR then tries to
read the volume-header label. If the volume does not contain an acceptable
VOLllabel as the first record, the operator is informed: ~~drive: Contains an
UNLABELED tape". The operator then takes appropriate action.
If PULSAR detects a properly labeled volume, it sends the following information to QUASAR to be stored in QUASAR's data base: VOLID, drive
name, label type (for example, ANSI), density, and tracks. QUASAR then
tries to reassign the volume to a user.

9.4.3 Volume Assignment
After QUASAR receives the label information from PULSAR, it tries to
associate the volume with a pending mount request. Three cases exist:
1.

Only one user is waiting for the volume. In this case, QUASAR immediately reassigns the device to the user.

More than one user is waiting for the volume. Here, the oldest request
that can be serviced using available resources while not causing a conflict with other requests is serviced.

The Mount Process for Labeled Tapes

9-5

No one is waiting for the volume. In this case, the operator has premounted the tape in anticipation of a reel switch or of a batch stream
start-up. The tape is automatically assigned when a request for it is
issued. Refer to Section 9.4.4 for information on premounted volumes.

9.4.4 Premounted Volumes
The tape-processing system allows the operator to mount labeled volumes
before requests for them have been issued. This action is known as premounting, or prestaging. The operator might premount successive volumes
of a set to prevent job wait during rewind of the current tape and mounting
of the next tape.
As mentioned previously, premounted volumes are initially processed like
other volumes; however, they cannot be immediately assigned to a user.
Instead, the volume information that PULSAR passes to QUASAR remains
stored in QUASAR's data base. When a mount request comes in, QUASAR
searches its data base looking for volume characteristics that match those
specified in the request. If it finds a match, it assigns the tape unit and
volume to the user. From there, processing continues without operator
intervention.

9-6

The Mount Process for Labeled Tapes

Chapter 10
The Label Processor's Search Algorithm

If you are a programmer, you should be aware of the label processor's
search algorithm when you specify the filename and file sequence parameters with the .TFLPR function of the TAPOP. monitor call. This chapter
describes file-positioning operations that are affected by your use of these
parameters. Refer to Table 10-1 for a summary of file-positioning
possi bili ties.

10.1 When Neither a Filename nor a Sequence Number Is
Specified
When neither a filename nor sequence number is specified, the following
action is taken:

Input
The volume set is opened at the current file position. No checking is done
on the file name.

Output
As with input, the volume set is opened at the current file position. A new
file is written using the default file name.

10.2 When a Filename Is Specified
If the filename is specified but the sequence number is not specified or is
zero, the following action is taken:

Input
The volume set is rewound and a search through the volume set is made for
a matching file. If the file is found, the file is opened for input. If the file is
not found, the ~~File Not Found" error code is returned to the user program.
10-1

Output
As with input, the volume set is rewound, and a search through the volume
set is made for a matching file. If the file is found, the file is opened for
output. If the file is not found, the file is opened for output at the end of the
volume set.

10.3 When a Sequence Number Is Specified
If the sequence number is specified but the filename is not, the following
action is taken:
Input
The volume set is positioned at the location indicated by the sequence number. If a file exists there, it is opened for input. If a file does not exist at that
location, such as when a file is beyond the logical end of the volume set, the
positioning error return is given.

Output
As with input, the volume set is positioned at the location indicated by the
sequence number. If a file exists there, it is opened for output. A default
filename is written. All subsequent files in the set become useless.

If a file does not exist at that location but the file sequence number is equal
to 99,999 or equal to the number of files contained in the volume set plus
one, the file is opened at the logical end of tape. A default file name is
written.

10.4 When a Filename and a Sequence Number Are both Specified
When the filename and sequence number are both specified, the following
action is taken:

Input
The volume set is positioned at the location indicated by the sequence number. If a file exists, a comparison is made between the filename you specified and the filename on the label. If they match, the file is opened for
input; otherwise the positioning error is returned.

Output
The volume set is positioned at the location indicated by the sequence number. No check is performed for a matching file name. The file is opened for
output, and the filename you specified supersedes any filename that may
exist.

10-2

The Label Processor's Search Algorithm

Table 10-1: File Positioning Summary
Is File
Name
Given

Is Sequence
Number
Given*

Does Filename
Exist Where
it Should
be Found **

YES

Result
for Input

Result
for Output

YES

The requested file
is opened for input.

The requested file
is opened for
output.

ERROR
(File not
found)

The requested file
is opened. The position follows the
last file of the set.

YES

The requested file
is opened for input.

The requested file
is opened for
output.

YES

ERROR
(Positioning
Error)

The requested file
is opened for output,
thereby
overwriting an old
file. *

The "current" file
is opened for input.

The "current" file
is opened for output. Default file
label is written.

YES

The requested file
(defined by pos) is
opened for input.

The requested file
is opened for output. Default file
label is written. *

* If a position is greater than the number of files on the volume set (+ 1 for output), the
"File Not Found" error is set. To clear tape-labeling errors, you must use function code
1 of the .TFURQ TAPOP. function.

** Sequence number takes precedence over file name. If sequence number is given, the
volume set is positioned first. If no sequence number is given but a filename is given, the
whole volume set is searched.

The Label Processor's Search Algorithm

10-3

Chapter 11
The TAPOP. Monitor Call

Function
The TAPOP. monitor call controls many tape-processing operations. This
chapter defines the various TAPOP. functions.
Several TAPOP. functions are identical to extensions of other monitor calls,
such as MTAPE and MTCHR. All TAPOP. functions assume that the specified device has been assigned to your job by the ASSIGN command or the
OPEN/INIT monitor call, or that the calling job has SPY privileges.
Calling Sequence

arglst:

MOVE
TAPOP.
error return
normal return

ac,[XWD arglength,arglst]
ac,

EXP

fcncode

SIXBIT/devicel
EXP
EXP

channo
udx

last argument
Where:

arglength is the length of the argument list.
arglist is the address of the argument list.
fcncode is one of the function codes described below.
device is the SIXBIT physical or logical name of a device.

11-1

channo is the number of an initialized channel. For information
on initializing devices, refer to the TOPS-JO Monitor Calls
Manual.
udx is the universal device index for a device. The words up
through last argument are arguments for the given function.
On an error return, an error code is ret~rned in the ac. The TAPOP. error
codes are listed at the end of this chapter.
The function codes fall into four groups:

o - 777

Perform specific actions.

1000 - 1777

Read parameters.

2000 - 2777

Set parameters. These function codes are not explicitly listed in the
descriptions below. To set a parameter, use the corresponding read
mnemonic plus the offset .TFSET (= 1000). For example, to set the
density indicator, use the read density indicator mnemonic plus
.TFSET:
.TFDEN + .TFSET

3000 - 3777

Reserved for customer-defined functions.

The function codes and their meanings are:

11-2

Code

Symbol

Function

.TFWAT

Waits for I/O to be completed .

.TFREW

Rewinds tape to load point .

.TFUNL

Rewinds and unloads tape .

.TFFSB

Skips forward one block .

.TFFSF

Skips forward one file.

.TFSLE

Skips to logical end-of-tape.

.TFBSB

Skips backward one block.

.TFBSF

Skips backward one file.

.TFWTM

Writes a tape mark.

.TFWLG

Writes three inches of blank tape.

.TFDSE

Erases entire tape (data security, TXOllTX02 on DXI0lDX20
only). You cannot use this function for tapes that are under the
control of PULSAR and QUASAR.

.TFWLE

Writes logical end-of-tape (two tape marks for unlabeled tapes;
one tape mark followed by the end-of-file label set followed by
two tape marks; for labeled tapes) .

.TFLBG

Gets the tape label device data block. Returns the name in the ac.
This is a privileged function for use by the label processor.

.TFLRL

Releases the tape label device data block. This is a privileged
function for use by the label processor.

.TFLSU

Swaps units. This is a privileged function for use by the label
processor.

.TFLDD

Destroys the tape label data base. This is a privileged function for
use by the label processor.

The TAPOP. Monitor Call

.TFFEV

Forces end-of-volume processing. This function allows the label
processor to write the end-of-volume (EOV) label set before it
finds the end-of-tape mark. The label processor then assumes a
multivolume file and automatically issues a MOUNT request for
the next volume.

.TFURQ

Function code 1 (.TFCLE) clears the tape-label error bits from a
previous label error. You retrieve the extended error bits with the
.DFRES function of the DEVOP. monitor call. Function 1 causes
the tape to be positioned at BOT.

.TFSMM

Sets maintenance mode on the tape controller. This is a privileged
function.

.TFCMM

Clears maintenance mode on the tape controller. This is a privileged function.

.TFCEC

Clears error counters. This is a privileged function for use by the
label processor.

1000

.TFTRY

Returns in the ac the number of retries on the last error .

1001

.TFDEN

Returns in the ac the density code for the tape. To set the density
code, use .TFDEN + .TFSET; the monitor reads the new density
code from arglst+2. Note that in order to set the density with this
function code, IO.DEN must be zero. The density codes and their
meanings are:

1002

.TFKTP

Code

Symbol

Density

0
1
2
3
4
5
6--17

.TFDOO
.TFD20
.TFD55
.TFDBO
.TFD16
.TFD62
Reserved

Unit default .
200 bits/inch (B.1 rows/mm).
556 bits/inch (22.5 rows/mm) .
BOO bits/inch (32.2 rows/mm) .
1600 bits/inch (65.3 rows/mm).
6250 bits/inch (255.5 rows/mm) .
To DIGITAL.

Returns in the ac the controller type code for the tape. To set the
controller type code, use .TFKTP + .TFSET; the monitor reads the
new controller type code from arglst + 2. The controller type codes
and their meanings are:
Code

Symbol

Controller Type

0
1
2
3
4

.TFKTA
.TFKTB
.TFKTC
.TFKTX
.TFKTM

5
6

.TFKRH
.TFKD2

TM10A (TU10/20/30/40/41).
TM10B (TU10/20/30/40/41).
TM10C (TU43).
TXOllTX02 on DX10 (TU70,71,72).
TM02/TM03 on RH10/RH20
(TU16,TU45) .
TM02/TM03 on RH11 (TU45).
TX02 on DX20 (TU70,71,72).

1003

.TFRDB

Returns in the ac the read-backwards bit (TM02/TXOllTX02
only). The bit is on (1) if the tape is set for read-backwards, or off
(0) if not. Refer to the TOPS-IO Monitor Calls Manual for information on reading backwards. To set the read-backwards bit, use
.TFRDB+ .TFSET; the monitor reads the bit from arglst+2.

1004

TFLTH

Returns in the ac the bit for read next record at low threshold
(TM10A/B only). The bit is on (1) if the tape is set for low threshold, or off (0) if not. To set the bit ,use .TFLTH + .TFSET; the
monitor reads the bit from arglst + 2.

The TAPOP. Monitor Call

11-3

11-4

1005

.TFPAR

Returns in the ac the status of the even parity bit (for 7-track
tapes only). To set the status of the even parity bit, use
.TFPAR+ .TFSET; the monitor reads the status from arglst+2.

1006

.TFBSZ

Returns in the ac the block size for the tape. The returned value is
one greater than the number of data words per record. To set the
block size, use .TFBSZ + .TFSET; the monitor reads the block size
from arglst + 2.

1007

.TFMOD

Returns in the ac the data mode code for the tape. To set the data
mode code, use .TFMOD + .TFSET; the monitor reads the data
mode code from arglst + 2. The data mode codes and their meanings are:
Code

Symbol

Data Mode

.TFMDD

.TFMID

2
3

.TFM8B
.TFM6B

.TFM7B

.TFM7T

DIGITAL-compatible core dump mode
(7-track and 9-track).
Industry-compatible core dump mode
(9-track) .
8-bit mode, 4 bytes per word.
6-bit mode, 6 bytes per word (9-track,
TU70 only) .
7-bit mode, 5 bytes per word (TU70
only).
DIGITAL-compatible 7-track core
dump mode (SIXBIT).

Returns in the ac the track status bit for the tape (0 for 9-track, 1
for 7-track). To set the track status bit, use .TFTRK + .TFSET; the
monitor reads the bit from arglst + 2.

1010

.TFTRK

1011

.TFWLK

1012

.TFCNT

Returns in the ac the character count of the last record (the actual
record length).

1013

.TFRID

Returns in the ac the SIXBIT reel identification for the tape. To
set the reel identification, use .TFRID + .TFSET; the monitor
reads the SIXBIT reel identification from arglst + 2.

1014

.TFCRC

Returns in the ac the last cyclic redundancy character (9-track
NRZIonly).

1015

.TFSTS

Returns in the ac the unit status flags for the tape. To set the
flags, use .TFSTS+ .TFSET; the monitor reads the flags from
arglst + 2. The unit status flags and their meanings are:

The TAPOP. Monitor Call

Returns in the ac the write-lock bit for the tape (1 if write-locked,

o if not).

Flag

Symbol

Meaning

18
19

TF.UNS
TF.BOT

20
21
22-32
33
34

TF.WLK
TF.REW
TF.8TA
TF.8EL

Unit is not schedulable.
Unit is positioned at beginning-of-tape
mark.
Unit is write-locked.
Unit is rewinding.
Reserved.
Unit is started.
Unit is selected.

'fF.OFL

Unit is off-line.

1016

TFSTA

Returns unit statistics for the tape device. Your program supplies
the function code and device at arglst and arglst+ 1. (These values are identical to those returned for the MTCHR. monitor call.)
The monitor returns the device statistics at arglst in the format:
Offset

Symbol

Contents

o
2
3

.TSFUN
.TSDEV
.TSRID
.TSFIL

.TSREC

.TSCRD

.TSCWR

.TSSRE
.TSHRE
.TSSWE
,TSHWE
.TSTME
.TSTDE

Function code (user-supplied) .
Device (user-supplied) .
SIXBIT reel identifier .
Number of files since beginning of tape
(current file number).
Number of records since last end-of-file
mark (current record number).
Number of characters read since system
reload.
Number of characters written since system reload.
Soft read errors since system reload .
Hard read errors since system reload .
Soft write errors since system reload .
Hard write errors since system reload.
Total errors since MOUNT .
Total device errors since system
start-up.
Total unloads since system reload .
Number of retries to resolve last error .
Character count of last record read or
written.
Position before last error: file number
(in left halD; record number (in right
halD.
Final error state. See SYSERR documentation.

10
11
12
13
14

15
16
17

.TSTUN
.TSRTY
.TSCCR

.TSPBE

.TSFES

1017

.TFIEP

Returns in the ac the initial error pointer .

1020

.TFFEP

Returns in the ac the final error pointer .

NOTE
Function codes 1021 and 1022 return the blocks
pointed to by 1017 and 1020. These blocks are for
communication of errors to DAEMON and may
change without notice.
1021

.TFIER

Returns in the ac the initial error status .

1022

.TFFER

Returns in the ac the final error status .

1023

.TFFED

Returns in the ac the final error disposition .

The TAPOP. Monitor Call

11-5

1024

.TFLBL

Returns in the ac the label-processing type code. To set the
label-processing type code, use .TFLBL+ .TFSET; the monitor
reads the new code from arglst + 2. The set function requires
JP.POK, [1,2], or JACCT privileges. The label-processing type
codes and their meanings are:
Code

Symbol

Label-Processing Type

.TFLBP
.TFLAL
.TFLAU
.TFLIL
.TFLIU
.TFLTM
.TFLNS
.TFLNL

Bypass label processing .
ANSI labels .
ANSI labels with user labels .
IBM labels.
IBM labels with user labels .
Leading tape mark .
Nonstandard labels.
No labels. When tapes are processed
with no labels, the label processor is
used only to verify that the tape does
not contain a tape label. Unlabeled
tapes can be copied to create a labeled
tape.
DEC COBOL ASCII labels .
DEC COBOL SIXBIT labels .
Same as .TFLNL except that user program is responsible for dealing with an
EOT. This type is the default. To switch
reels at EOT, use TAPOP. function
.TFFEV.

2
3

4
5
6
7

10
11

.TFCBA
.TFCBS
.TFLNV

1025

.TFPLT

Performs same function as .TFLBL (1024) .

1026

.TFLTC

Returns last tape-labeling return code. The return codes and their
meanings are:
Code

Symbol

Termination Code

.TFTCP
.TFTRE
.TFTLT
.TFTHL
.TFTTL
.TFTVL
.TFTDV
.TFTDE
.TFTWL
.TFPSE
.TFBOT
.TFIOP
.TFFNF
.TFCAN
.TFTMV

Continue processing
Return EOF
Label type error
Header label error
Trailer label error
Volume label error
Device error
Data error
Write lock error
Positioning error
BOT
Illegal operation
File not found
OPR cancelled request
Too many volumes requested

2
3

4
5
6
7
10
11

12
13
14
15
16
17

Refer also to the extended I/O error code.

11-6

1027

.TFDMS

Returns in the ac the diagnostic mode set bit (TXOllTX02 on
DXI0 only). This bit is 1 for diagnostic mode, otherwise O. To set
this bit, use .TFDMS + .TFSET; the monitor reads the new bit
from arglst + 2.

1030

.TFFSO

Returns in the ac the bit showing whether a forced sense command will be issued to the controller (TXOllTX02 on DXI0lDX20
only) after the completion of every operation. To set the bit, use

The TAPOP. Monitor Call

.TFFSO + .TFSET; the monitor reads the new bit from arglst + 2.
This bit should be set by diagnostic programs only, since it slows
down tape operation considerably.
1031

.TFMFC

Returns in the ac the maximum frame count. To set the count, use
.TFMFC + .TFSET; the monitor reads the new count from
arglst+2. Use this function to speed tape throughput for a TU16,
or TU45, TU70, TU71, or TU72 that does not have an integral
number of bytes per word. The count stays in effect until your
program performs a RESET with another TAPOP. monitor call or
until the tape is RELEASed (if the device was ASSIGNed). This
function allows a TU70 or a TU16 to read and write tapes that do
not have an integral number of bytes per word. This function provides tape compatibility with other systems.

1032

.TFPDN

Returns in the ac flags showing the possible densities for a tape .
The flags and their meanings are:

1033

.TFLPR

Flag

Symbol

Density

31
32
33
34
35

TF.DN5
TF.DN4
TF.DN3
TF.DN2
TF.DN1

6250 bits/inch (255.5 rows/mm)
1600 bits/inch (65.3 rows/mm)
800 bits/inch (32.2 rows/mm)
556 bits/inch (22.5 rows/mm)
200 bits/inch (8.1 rows/mm)

Returns at arglst + 2 the following tape label parameters. Function code 1033 always returns numbers. This function causes the
first input label processing if there is not file open for input. To set
the parameters, use .TFLPR+ .TFSET; the monitor reads the
parameters beginning at arglst + 2. This set function is legal only
if there is no file open for output on the given channel. The parameters given apply to the next file to be written. The format of the
parameters at arglst is:
Offset

Symbol

Contents

0
1
2

.TPFUN
.TPDEV
.TPREC

Function code (user-supplied) .
Device (user-supplied) .
Record format and form control.

Bits

SymbQ!

Meaning

18-35

TR.RFM

Record format byte.

Code

Symbol

Meaning

0
1
3
4

TRF.DX
TRF.FX
TRF.VR
TRF.SP
TRF.UN

Default (fixed).
Fixed (F).
Variable (D).
Spanned (S).
Undefined (U).

Bits

Symbol

Meaning

0-17

TR.FCT

Forms Control byte.

Code

Symbol

Meaning

TFC.NO

Records on tape do not
contain form control
characters.

The TAPOP. Monitor Call

11-7

3
4
5

6
7
10
14

Code

Symbol

Meaning

TFC.AS

TFC.AM

First character of each
record is a form control
character.
Records on tape contain
all required form control
characters.

.TPRSZ
.TPBSZ
.TPEXP

Record size in characters .
Block size in characters .
Expiration date in 15-bit format .

Bits

Symbol

Meaning

18-35
0-17

TPE.EX
TPE.CR

Expiration date.
Creation date.

.TPPRO
.TPSEQ
.TPFNM
.TPGEN

Protection Code .
File sequence number .
Filename (17 chars ASCII).
Generation and version numbers .

Bits

Symbol

Meaning

0-17
18-35

TP.GEN
TP.VER

Generation number.
Generation version
number.

.TPLEN

.TPGEN + 1

Length of block

Normal Return
The function is performed.

Error Return
One of the following error codes is returned in the ac:
Code

Symbol

Error

-1

TPACS%
TPIFC%
TPPRV%
TPNMT%
TPVOR%
TPACR%
TPCBS%
TPNIA%
TPNLP%
TPETC%
TPIJN%
TPLRF%
TPLSI%
TPLOE%
TPDNC%
TPWWL%

Address check storing answer.
Illegal function code.
Not enough privilege.
Not a magtape device.
Specified value out of range.
Address check reading arguments.
Parameter cannot be set.
Tape not initialized or assigned.
No label process.
Termination code error.
Illegal job number.
Label release function required.
Label parameter set illegal after first output.
Label DDB owned by someone else on label get.
Drive not capable of specified density.
Write attempted to write-locked tape.

o
1
2
3

4
5
6
7

10
11

12
13
14
15
16

11-8

The TAPOP. Monitor Call

Example
The following example uses TAPOP. functions to generate a map of a labeled tape. A tape map is similar to a disk directory in that it lists all the
files stored on the tape along with their attributes. A sample run of the
program follows the listing. Note that the program makes use of the
GALAXY library (GLXLIB).
TITLE
SUBTTL

TAPMAP - Labeled tape mapper program
D. Cornelius 21-Nov-79

SEARCH
PROLOG

GLXMAC
(TAPMAP)

SUBTTL

Data storage

TAPDEV = = SIXBIT/STRI
OCHN= =10
OPNBLK:
DVNAM:

EXP
EXP
XWD

.IODMP
TAPDEV
0,0

VOLID:
FILCNT:
PDL:

BLOCK
BLOCK
BLOCK

1
1

ERTAB:

[ASCIZ/No errors/]
[ASCIZ/Continue processing/]

;Logical name to read from
;Channel upon which to do I/O
;logical drive name

100

[ASCIZ/EOF/]

[ASCIZ/Label type error/]
[ASCIZ/Header label error/]
[ASCIZ/Trailer label error/]
[ASCIZ/Volume label error/]
[ASCIZIDevice error/]
[ASCIZIData error/]
[ASCIZ/Write-Iock error/]
[ASCIZ/Positioning error/]
[ASCIZ/BOT/]

[ASCIZ/Illegal operation/]
[ASCIZ/File not foundl]
RECFMT:

[ASCIZIDef/]

[ASCIZ/Fixl]
[ASCIZ/Var/]
[ASCIZ/Span/]
[ASCIZ/U ndf/]
FORCON:

[ASCIZI????/]
[ASCIZ/N one/]
[ASCIZI ANSI/]
[ASCIZ/Embdl]

LPBLK:

EXP
.TFLPR
EXP
TAPDEV
BLOCK
.TPLEN-2
LPLEN = = .-LPBLK
SUB TTL

START:

RETRY:

;Get label parameters
;Device name
;Label parameters come back here

Main Routine

RESET
P,[IOWD 100,PDL]
MOVE
S1,
SETZ
P,I%INIT
PUSHJ
;Get the generic device we use
S1,DVNAM
MOVE
;Does the user have it around?
DEVNAM S1,
;No, complain
SKIPA
DEVOK
;Yes, keep going
JRST
$TEXT (,<Please assign your tape as logical name 'W/DVNAM/:, and type CONTINUE»
1,
EXIT
JRST
RETRY

The TAPOP. Monitor Call

11-9

DEVOK:

;Point at the arg block
MOVE
S1,[2"T1]
;Code to get label type
T1,.TFLBL
MOVX
;Get the drive name
T2,DVNAM
MOVE
;Find out the label type
S1,
TAPOP.
;Can't, so complain
SKIPA
LABELD
;Got it, so use it
JRST
$TEXT (,<Device -'W/DVNAM/: is not a tape!»
EXIT
1,
RETRY
JRST

LABELD:

CAlL
CAlLE
SKIPA
JRST
$TEXT
EXIT
JRST

ANSLBL:

MOVE
S1,[2"T1]
;Aim at the arguments
MOVX
T1,.TFREW
;Code to rewind the tape
TAPOP.
S1,
;Do it
SKIPA
;Can't
JRST
REWDON
;Got it
$TEXT (,<Can't REWIND -'W/DVNAMI:»
EXIT
1,
JRST
RETRY

;Is it ANSI?
;Or EBCDIC (IBM)
;No, complain
ANSLBL
;Yes
(,<Device -'WIDVNAM/: is not a labeled tape!»
1,
RETRY
S1,.TFLAL
S1,.TFLIU

;Aim at the arguments
REWDON: MOVE
S1,[2"Tl]
T1,.TFRID
;Code to get the reelid
MOVX
S1,
;Ask the monitor
TAPOP.
;Can't
SKIPA
GOTRID
;Wins
JRST
$TEXT (,<Can't get volume id from 'W/DVNAM/:»
1,
EXIT
JRST
RETRY
GOTRID:

;Save the reelid
MOVEM
S1,VOLID
;Open up the tape
OPEN
IOCHN,OPNBLK
;Can't
SKIPA
JRST
TAPOPN
;Got it
$TEXT (,<Can't OPEN device 'WIDVNAM/:»
EXIT
1,
JRST
RETRY

TAPOPN:

$TEXT (,<Start of tape map for volume 'W/vOLIDI on drive 'WIDVNAM/: at 'H/[-1]/»

FILLOP:

FILCNT
;Clear the number of files seen
SETZM
$TEXT (,<File File Name Pro Gen Ver BlkSiz RecSiz Creation Expiration RecF Form»
$TEXT (, < ------ ---------------- ----- ------ ----- -------- --------- ------------ --------------- ------ ------"-> )
MOVE
S1,[LPLEN"LPBLK] ;Aim at the label parameter block
S1,
;Read the label params of the next file
TAPOP.
ALLDON
;Can't, see about EOF
JRST
AOS
FILCNT
;Wins, count this file
PUSHJ
;Print the info about this file
P,TYPFIL
S1,[2"T1]
;Aim at the argument block
MOVE
Tl,.TFFSF
;Code to skip a file
MOVX
;On this tape
T2,DVNAM
MOVE
S1,
;Skip it
TAPOP.
;Can't, so complain
SKIPA
JRST
FILLOP
;Wins, go get params on this file
$TEXT (,<Can't skip file on 'WIDVNAM/: file 'D/FILCNT/»
EXIT
1,
JRST
RETRY

;Here when we can't get params on some file.
ALLDON:

11-10

GETSTS
TXNN
JRST
TXC
TXCE
JRST
MOVE
MOVX

IOCHN,S1
S1,IO.ERR!IO.EOF
QUIT
S1,IO.ERR
S1,IO.ERR
EOFCHK
S1,[2"T1]
T1,.DFRES

The TAPOP. Monitor Call

;Read the standard device status
;Errors, or EOF?
;No errors, nor EOF, strange, so quit
;Complement the error bits
;All four error bits up?
;No, Could have been EOF
;Aim at the argument block
;Read extended device status

MOVE
T2,DVNAM
;On this tape
SI,
;Read 'em
DEVOP.
;Can't!?
SKIPA
JRST
G-OTERS
;Got 'em, see what the problem is
$TEXT (,<Can't Get label error status for 'W/DVNAM/: file 'D/FILCNT/»
1,
EXIT
JRST
RETRY
GOTERS: $TEXT (,<4abel error on 'W/DVNAM/: file 'D/FILCNTI
Error code:'D/S1I, 'T/@ERrTAB(Sl)/»
JRST
QUIT
EOFCHK:

TXNN
SI,IO.ERR
;Things other than EOF
JRST
QUIT
;No, just EOF, so that's alright
$TEXT (,<110 error on 'W/DVNAM/: file 'D/FILCNT/, 110 status = '0/S1I»
EXIT
1,
JRST
RETRY

QUIT:
$TEXT (,<End of tape map for volume 'W/VOLIDI on drive 'W/DVNAM/: at 'H/f-ll/
Total of 'D/FILCNTI files.»
EXIT
SUBTTL

TYPFIL - Print info about the current file

;This routine dumps the known info about the file
; that is currently open.
;It is assumed that the .TFLPR to get the label parameters has just
; been done.
;Most of the info typed by this routine comes from LPBLK.
;Other inputs:
; FILCNTI # of the current file
TYPFIL:

$SAVE

<Pl,P2,P3,P4>

;Do date-time conversion of creation date
SETZ
LOAD
PUSHJ
MOVE

SI,
;No time, just date
S2,LPBLK + .TPEXP,TPE.CR
;Get the creation date
P,CNVDT
;Convert to UDT
Pl,SI
;Save it

;Do date-time conversion of expiration date
SI,
;No time, just date
S2,LPBLK + .TPEXP,TPE.EX
;Get the expiration date
P,CNVDT
;Convert to UDT
PUSHJ
P2,SI
;Save expiration UDT
MOVE
SI,LPBLK + .TPREC,TR.RFM
LOAD
;Get the record format code
S2,LPBLK + .TPREC,TR.FCT
LOAD
;Get the form control code
$TEXT (, <'D4L1FILCNTI 'TI7L ILPBLK + .TPFNMI
'03RO/LPBLK + .TPPROI 'D4R/LPBLK + .TPGEN,TP.GENI

SETZ
LOAD

'D3R/LPBLK + .TPGEN,TP.VERI 'D6R/LPBLK + .TPBSZI
'D6R/LPBLK + .TPRSZI 'H9/P1I 'H9/P21 'T4L I@RECFMT(SI)/

'T4L
I@FORCON(S2)/> )

$RETT
SUBTTL

CNVDT

CONVERT DATE TO UDT

;THIS ROUTINE WILL MAKE A UDT OUT OF AN ARBITRARY DATE

,
;CALL

S21

TIME IN SECONDS
DATE IN SYSTEM FORMAT
(Y*12 + M)*31 + DAY SINCE 111164

S1I

UDT

S1I

,
;RETURN

,
;NOTE:

LEFT HALF DIVIDED BY 7 GIVES THE DAY OF THE WEEK

RADIX 10

;UNDER RADIX 10 **** NOTE WELL ****

The TAPOP. Monitor Call

11-11

CNVDT::

PUSHJ
PUSH
IDIVI
CAlLE
JRST
IDIVI
ADD
MOVE I
CAlL
MOVEI
MOVE
ADDI
IDIVI
CAIE
MOVE I
ADDI
MOVE
IMULI
ADD
HRREI
JUMPLE
IDIVI
SUB
CAIE
ADD
CAlLE
SETOM

GETNW1:
GETNW2:

POP
MOVEI
ASHC
DIV
CAMLE
ADDI
HRL
GETNWX: POPJ
MONTAB: EXP
RADIX
END

;PRESERVE T3
;SAVE TIME FOR LATER
;S2 = YEARS--1964
;SEE IF BEYOND 2217
;YES-RETURN -1
;T1=MONTHS-JAN, T2=DAYS--1
;T2= DAYS-JAN 1
;LEAP YEAR ADDITIVE IF JAN, FEB
;CHECK MONTH
;ADDITIVE IF MAR-DEC
;SAVE YEARS FOR REUSE
;OFFSET SINCE LEAP YEAR DOES NOT
; GET COUNTED
S2,4
;HANDLE REGULAR LEAP YEARS
T1,3
;SEE IF THIS IS LEAP YEAR
T3,0
;NO-WIPE OUT ADDITIVE
T2,<1964-1859>*365 + <1964-1859>/4+ <31-18> + 31(S2)
;T2 = DAYS BEFORE JAN 1,1964 + SINCE JAN 1
; + ALLOWANCE FOR ALL LEAP YEARS SINCE 64
S2,S1
;RESTORE YEARS SINCE 1964
;DAYS SINCE 1964
S2,365
T2,S2
;T2 = DAYS EXCEPT FOR 100 YR. FUDGE
S2,64-100-1(S1)
;S2 = YEARS SINCE 2001
S2,GETNW1
;ALL DONE IF NOT YET 2001
S2,100
;GET CENTURIES SINCE 2001
T2,S2
;ALLOW FOR LOST LEAP YEARS
T1,99
;SEE IF THIS IS A LOST L. Y.
T2,T3
;ALLOW FOR LEAP YEAR THIS YEAR
T2,A0377777
;SEE IF TOO BIG
T2
;YES-SET -1

P,.SAVET
P,S1
S2,12*31
S2,2217-1964
GETNW2
T1,31
T2,MONTAB(Tl)
T3,0
T1,2
T3,1
S1,S2
S2,3

P,S1
;GET MILLISEC TIME
S2,0
;CLEAR OTHER HALF
;POSITION
S1,-17
S1,[24*60*60*1000] ;CONVERT TO 112**18 DAYS
S2,[AD24 *AD60*AD60*AD 1000/2]
; OVER 112 TO NEXT?
S1,1
;YES, SHOULD ACTUALLY ROUND UP
S1,T2
;INCLUDE DATE
P,
;RETURN
0,31,59,90,120,151,UH,212,243,273,304,334,365
8
START

The program above produces the following results:
.MOU TAPE-EXAMPLE:STR/LABEL:EBCDIC/NOTIFY/NOWAIT/VOLID:HPPO02
[15:20:05]
[Mount re9uest TAPE-E}{AMPLE 9ueued, re9uest #1274]
FroITls}'steITI:
[Magtape HPP002 ITlounted on MTB283 I"rith logical nalTle STR ]
• RU TAPMAP
Start of tape ITlap for I.lOlUITle HPP002 on drive STR: at 10-Jul-80 15:24:34
File File NalTle Pro Gen

I.J e r

--------

3
4

11-12

IMPRESS
CLIMBER
CALENDAR
GEOMETRY

000
000
000
000

0
0
0
0

BlkSiz RecSiz Creation

Expiration RecF F 0 rlTI

------ ------ --------

----------

2741
2741
2741
2741

The TAPOP. Monitor Call

180
180
180
180

23-Feb-78
23-Feb-78
12-Mar-78
12-Mar-78

1-Jan-84
1-Jan-84
1-Jan-84
1-Jan-84

Undf
Undf
Undf
Undf

None
None
None
None

Chapter 12
Crash Procedures

GALAXY components may crash, that is, stop performing useful work, for
a number of reasons:
• A monitor call (UUO) that should not fail takes the error return and a
GALAXY stopcode occurs.
• One of the consistency checks, which are built into the various GALAXY
components, detects an error and a GALAXY stopcode occurs.
• An undetected error eventually causes an illegal memory reference and a
GALAXY stop code occurs.
• A component goes into event-wait (EW) state for an extended length of
time waiting for an event that is not likely to occur (perhaps due to some
hardware or software malfunction). In this case, the program hangs in
the UUO.
Crashes mayor may not be accompanied by GALAXY stopcode messages.
Those crashes without stopcode messages indicate that the system is hung
or has fallen into an unending loop. This chapter is concerned with crashes
that involve the PULSAR and QUASAR GALAXY components. Refer to
the TOPS-10 Crash Analysis Guide for an in-depth discussion of stopcodes
and crashes that involve the monitor.
To recover from PULSAR and QUASAR crashes, users should take the
steps outlined below.

12.1 Recovery Procedures
User procedures differ according to the GALAXY component involved in
the crash:

12-1

PULSAR
After a PULSAR crash, users accessing tape drives may find that their jobs
are unable to automatically continue when PULSAR is restarted. The user
who caused the crash will find that his/her job remains in EW state. Other
users will receive tape-labeling errors and will be unable to continue because PULSAR has lost record of any label-processing requests made before the crash. This is because PULSAR's data base has been reinitialized.
Thus users/jobs in the post restart state must type a control-C; give the
.FINISH command; then restart operations.

QUASAR
When QUASAR crashes, user requests for queue services are lost, even
though the failsoft master queue file is rebuilt when QUASAR is restarted.
For example, user requests for mounts/dismounts involving
GALAXY-controlled devices will not be serviced. Also, programs using the
QUEUE. UUO to interface with QUASAR will hang in EW state in the
UUO. To recover from these crashes, type a control-C and begin anew.
Tape drives allocated to a user will be set unavailable when QUASAR is
restarted. However, the user will continue to be able to access the drive,
both during the crash and after the restart, without problems.

12.2 Crash Analysis
If you are a system programmer, you might want to examine the ac registers as they existed at the time of the crash to determine the cause of the
crash. You can then continue the component's operations from the point
following the crash.
Whenever a GALAXY component generates a stopcode, it saves the contents of the ac registers and the text description of the crash. The ac registers are saved in an area called .SACS, from locations .SACS + 0 to
.SACS+ 17. The address of the ASCIZ text that describes the crash is saved
in WTOADR. A sample of the crash block information is shown below:
Figure 12-1: A Crash Block
?Stopcode - ILM - in module GLXINT on 16-Mar-81 on 15:49:43
Reason: Illegal memory reference at PC 631334
Program is ORION version 4(522) using GLXLlB version 1(1006)
Crash block starts at location 674000
Last GLXLlB error: 21 (No such pid)
Contents of the ACs:
0/
777777777777
4/
0
10/
16
14/
1000015

777627

Last 9 stack locations:
-2(P)/
-1 (P)/
200370
-4(P)/
400000522
-5(P)/
-8(P)/
-7(P)/
0

200370

o
o
o

o
o
777627

44622000000

120
77
400
777474000143
-3(P)/
-6(P)/
-9(P)/

312000000700
1

In the GALAXY code, the $STOP macro is used to process the fatal
condition.
12-2

Crash Procedures

GALAXY stopcodes are stored as S .. xxx, where xxx is the 3--character stopcode name. (Refer to Appendix A, Label-Processing Stopcodes.)
Stopcodes stop program execution, and execution cannot continue unless
DDT is loaded with the program. If DDT is loaded with the program, the
stopcode processor transfers control to DDT.
Stopcodes are conditionally sent to ORION to be recorded in the system log
and to be sent to any jobs running OPR. A flag in the program's initialization block (in IB.STP) enables this· conditional function.
To continue a program that has received a stopcode, first determine where
the stop code occurred. The address where the stop code was executed is located at S .. xxx, where xxx is the 3-character stop code name. Next, issue
the POP P,®QX instruction to restore the stack to its precrash value. You
can now continue the program from the point following the stopcode, provided that you have corrected the situation that caused the program to
receive a stopcode in the first place.
Other data areas that may help resolve a crash are:
.SPC

PC of the stop code

.SCODE

SIXBIT name of the stopcode

.SPTBL

Page table address

.SPRGM

SIXBIT program name

.SPVER

Program version number

.SPLIB

GLXLIB version number

.LGERR

Last GALAXY error code

.LGEPC

PC of the last $RETE

STPFLG

A stop code is in progress (-1)

PIDTAB

System PID table - indexed by SP.xxx, where xxx is the
processor name (QSR, OPR, and others)

RCVBLK

PDB of the last IPCF message received

PACKET

Contents of the last IPCF message received as a packet

PAGTBL

Memory page map (1 word per page)
Bits:

PG.USE = = 1B35 PG.WRK= = 1B34 PG.ADR = = 1B33 PG.INI = = 1B32 -

Page is in use
Page is in working set
Page is addressable
Page is part of initial core
image
PG.SWP= = 1B31 - Page is swapp able on a timer
trap

No bits lit indicates that the page does not exist.

Crash Procedures

12-3

Appendix A
Label-Processing Stopcodes

As discussed in Chapter 12, Crash Procedures, certain hardware and software malfunctions cause stopcodes to occur. These stopcodes result in system crashes. Appendix A lists the label-processing stopcode names and
their associated messages. These appear on the operator's terminal when
PULSAR experiences a stopcode crash.
ADF

AVR Label Destroy Failed
The .TFLDD TAPOP. failed. This function returns the label
DDB after volume recognition is performed.

BBF

Bad Backspace File
See SSR.

BBR

Bad Backspace Record
This stopcode is like SSR.

BCN

Bad Call to NXTFIL
PULSAR is confused about the position of the tape.

BCP

Bad Call to POST AP
POSTAP has been called with neither position by sequence nor
position by filename on in the status word.

CAS

Can't Append to SPT List
The system PUSHJs here if there are problems building the
SAT tables for a file structure.

CBP

Called on Bypass Label Processing
The system PUSHJ s here if any acti vi ty occurs on a bypass
label tape.

CCD

Can't Change Density
The .TFDEN TAPOP. failed.

A-I

A-2

CCF

Can't Create Failsoft File

CCP

Can't Checkpoint File

CCR

Can't Check Ring Status
The .TFWLK TAPOP. failed.

CCW

Can't Clear Watch Bits
The SETUUO to clear watch bits (so unload stats do not get
typed on a detached TTY) failed.

CDC

Can't Determine Density Capabilities
The .TFPDN TAPOP. failed.

CDF

Can't Delete Failsoft File

CDK

Can't Determine Kontroller Type
The .TFKTP TAPOP. failed.

CDS

Can't Get DSKCHR Status for (device name) error code = S2
DSKCHR for a file structure or a disk unit failed.

CDT

Can't Determine Track Status
The .TFTRK TAPOP. failed.

CEF

Can't Enter Failsoft File

CGC

Cannot Get a Failsoft Channel
There are no free (standard) channels for failsofting TCBs.

CGD

Can't Get Density

CGS

Can't GETTAB States Word

CMU

Can't Make TCB

CMV

Can't Make TCB

COF

Can't Open Failsoft Device

CPF

Clear Label Parameters Failed
The .TFLPR to set label parameters to 0 failed.

CRB

Can't Read Buffer Size
PULSAR cannot determine user's buffer size to set up a default
block count in HDRI.

CRM

Can't Read User's Mode

CRS

Can't RESDV. UFD Channel
The extended channel to read/write the UFD has gone a way or
been lost.

CRT

Can't Reopen Tape
The operation that causes dummy output to be written after the
label DDB channel is closed failed.

Label-Processing Stopcodes

CSB

Can't Set Blocksize
The .TFBSZ TAPOP. failed.

CSD

Can't Set Density

CSI

Can't Set Industry-Compatible Mode

CSM

Can't Set DIGITAL-Compatible Mode
The .TFMOD to set DEC-compatible I/O failed.

CSU

Can't Switch Units
The .TFLSU TAPOP. failed.

CUF

CHKACC UUO Failed

CWT

Can't Write Tape Mark
The .TFWTM TAPOP. failed.

DNI

DEVTYP UUO Not Implemented

DNZ

Device Name Is Zero in Failsoft File

DOF

Dummy OUT Failed
The operation that causes dummy output to be written whenever a tape is opened for I/O failed.

ELF

Error Looking Up Failsoft File

FTS

Failed Reading Track Status
The .TFTRK TAPOP. failed.

FUF

FILOP. UUO Failed
A FILOP for other than IN or OUT failed.

GNF

GETTAB for User's Name Failed

GSF

GETTAB for Serial Number Failed

GUF

GETTAB UUO Failed.

IDM

Invalid Date from Monitor
The date used 'in expiration checking is non-numeric.

IPF

Illegal Positioning Function
An illegal SIXBIT positioning code was passed to T$POS.

IRF

10 Error Reading Failsoft File

ISN

Illegal Structure Name
QUASAR has asked PULSAR to build a structure with a bad
FS name.

LDF

Label Destroy Failed
This stopcode is similar to ADF, but from a different context.

LGF

Label Get Failed
The .TFLBG TAPOP. failed. This condition was probably
caused by a bad device name in a recognize or monitor message.

Label-Processing Stopcodes

A-3

A-4

LRF

Label Release Failed
The .TFLRL TAPOP. failed.

MCF

MTAID. UUO Failed

NEB

No Error Bit
A call to translate DEC-IO I/O error bits to internal PULSAR
error bits was made with no DEC-IO I/O bits on.

NVD

No Valid Density
PULSAR cannot find any valid density in the possible density
mask.

PLM

Previous List TCB Has Been Meddled With
A return to the scheduler has been made on a process that was
different from the one the scheduler believed was running.

PNR

PULSAR Not Restartable

PRF

Positioning Request Failed
The TAPOP. function to move the tape failed.

RAT

Requesting Work for Active TCB
A disk TCB has requested that the HOM blocks for a unit be
read, but that unit's TCB is not idle.

RKT

Running a Killed TCB
The call to kill a TCB after all work has been finished has
returned, and it should never return!

RLT

Failed Reading Label Type
The .TFLBL TAPOP. failed.

RPF

Read Label Parameters Failed
The .TFLPR TAPOP. to get user's label parameters failed.

RSE

Reschedule from Exec Level
The process scheduler was called from non-TCB level.

RSF

TAPOP. to Read Statistics Failed
PULSAR cannot determine the number of bytes the user wrote
on the tape.

RTT

Releasing Tape Twice
A second call to T$RELE was done without an intervening call
to T$OPEN, or the TI.OPN bit was lost.

SIO

Switch Units with OPEN Label DDB
Attempt to switch units to a drive that is active, that is, open
for label I/O.

SLT

Set Label Type Failed
The .TFPLT TAPOP. failed.

SND

Switch Units with Nonexistent Device (device name)

Label-Processing Stopcodes

SPF

Set Label Parameters Failed
The .TFLPR TAPOP. to return label information to user failed.

SSR

Strange Skip Record
PULSAR is confused about the positioning of a tape on receipt
of a monitor message issued when a user tried to move a labeled
tape.

TRB

Can't Check Read-backwards

TUF

TAPOP. UUO failed
Some TAPOP. function failed. See the stack for details.

ULS

Unit Parameter List is Short
The list of unit parameter blocks that hang off a structure TCB
does not have as many entries as there are units in the FS.

WCF

Wait after Close Failed

WCL

Writing COBOL Labels
This stopcode is like CBP, but for the case of outputting COBOL
labels.

WNF

Waiting TCB not Found
When a TCB doing work finished, the TCB requesting the work
was not found.

WRF

Wait Request Failed
The .TFWAT TAPOP. (after every positioning except rewind
and unload) failed.

Label-Processing Stopcodes

A-5

Appendix B
Tape Label Implementation Levels

The ANSI standard allows for varying degrees of labeling capability among
systems involved in information interchange. It does this by establishing
three nested subsets of the full standard. These subsets, or levels, range
from the simplest processing at level 1 to the most complex at level 4
(representing the complete model). In meeting ANSI standards, a system
conforms to one of these levels.
There are three main identifiable constituents of a level: the file formats,
the labels used, and the record formats. The most familiar and useful options of these are as follows:
Constituent

Option

Constituent Configurations

Files

Single-file single-volume
+ single-file multivolume

Single-file single-volume
+ Single-file multivolume
+ multifile single-volume
+ multifile multivolume

VOLl HDRl EOFl EOVl

VOLl HDRl HDR2 EOVl EOV2 EOFl EOF2

Fixed

Fixed + variable

Fixed + variable + spanned

Labels

Record Format

B-1

Each level is made up of the option constituents as follows:
Files

Labels

Format

Levell

Level 2

Level 3

Level 4

Recall that TOPS-IO reads and writes level 4 labels and file formats. However, it does not ensure consistency between the file data and the label
specifications. Thus, TOPS-IO does not fully conform to the ANSI standard.

B-2

Tape Label Implementation Levels

Appendix C
Glossary

accessibility character

A character that indicates whether a tape volume is protected.
ANSI-ASCII mode

A data recording mode in which data is stored in 7-bit ASCII bytes (also called
7-bit mode). Data must be stored in this mode to meet ANSI standards.
Automatic Volume Recognition (AVR)

A TOPS-10 facility that automatically verities label information when a tape is
mounted, then assigns the tape to a waiting user.
beginning-of-tape marker

A physical marker on a magnetic tape used to indicate the beginning of the permissible recording area.
block

A collection of characters written or read as a unit. A physical record.
block size

The size in characters of a block. The default size is 128 words, 4 characters to a
word.
blocked record

A record contained in a file in which more than one record or segment can be
contained in a block.

C-l

bypass label processing

The processing mode wherein the system ignores tape labels.
density

The number of characters that can be recorded on one inch of magnetic tape.
EOT

The abbreviation for end-of-tape.
end-of-tape marker

A physical marker on a magnetic tape used to indicate the end of the permissible
recording area.
expiration date

The earliest date on which a tape may be overwritten.
file section

That part of a file recorded on anyone volume. The sections of a file do not have
sections of other files interspersed.
file set

A collection of one or more files recorded consecutively on a volume set.
fixed-length record

A record contained in a file in which all records are the same length. No indication
of the record length is contained in a fixed-length record.
frame

The smallest data aggregate on a magnetic tape. It consists of 8 bits (on a 9-track
tape) or 6 bits (on a 7-track tape), plus a parity bit.
interrecord gap (IRG)

Spaces or gaps between blocks on a magnetic tape.
label

A reserved record at the beginning of a volume, or at the beginning or end of a file
section, or at the end of a file, that identifies and delimits that volume or file
section. A label is not considered to be part of a file.
label group

A collection of one or more contiguous label sets.

C-2

Glossary

label number

A digit that indicates the sequence of labels within a label set. Label number one
is always present if the label set exists.
label set

A collection of one or more contiguous labels with the same three initial characters. These three characters are called the Label Identifier.
logical record

A record defined from the standpoint of its content, function, and use rather than
from its physical attributes.
multiple-volume file

A file that spans more than one magnetic tape reel.
physical record

A record identified from the standpoint of the manner or form in which it is stored
and retrieved; that is, one that is meaningful with respect to access. A block.
prestaging

The operator's mounting of volumes that a user has not yet requested.
protection code

A code indicating who can and cannot access the files on the magnetic tape.
record segment

That part of a spanned record that is contained in anyone block. The segments of a
record do not have segments of other records interspersed.
record size

The size in characters of a record. The default size is 128 characters.
scratch tape

A new tape that has been initialized and that contains only volume labels (if a
labeled tape), tape marks, and no user data, or an old tape whose data can be
overwritten.
Scratch tapes are assigned to users as needed and may be used for any purpose.
spanned record

A record contained in a file in which each record may begin in one block and end
in another. These records can span volumes.

Glossary

C-3

tape mark

On labeled tapes, a delimiter used to indicate the boundary between label groups
and between label groups and file data. On unlabeled tares, a tape mark separates
individual files. On both types of tapes, tape marks indicate the logical
end-of-tape.
unblocked record

A record contained in a single block with no other records or parts of records
contained in that block.
unspanned record

A record contained in a file in which each record by design ends in the same block
in which it begins.
variable-length record

A record contained in a file in which the records may have different lengths.

VID
The visual identification label; the physical label that is affixed to a magnetic tape
reel.

VOllD
The volume identifier; the string of characters that distinguishes a volume and is
written in the VOLl label during initialization.
volume

The equivalent term for a reel of magnetic tape. A volume may contain part of a
file, a complete file, or more than one file.
volume-set

A set of volumes that the user may logically treat as one large volume. A volume
set contains one file set only.
volume switch

The process of removing a volume from a tape drive and mounting the next volume of the set.

C-4

Glossary

Index
Access code field, 8-7
Accessibility character field, 3-5, 3-6, 8-4,
8-10
ALLOCATE command, 1-2,3-6
ANSI standard, 1-8, 3-3, 7-1, 8-1, 8-5,
Appendix B
ANSI-ASCII mode, 6-4
ASSIGN command, 3-8
Automatic Volume Recognition, 9-2 to 9-6
SET TAPE-DRIVE AVAILABLE
command and, 1-1
BACKSP ACE command, 3-8
BACKUP,
labeled tapes, 3-9
unlabeled tapes, 4-1
Batch control file, 9-1, 9-2
Batch log file, 9-3, 9-4
Block count field, 8-10, 8-15
Block length field, 8-13
Blocking, 7-1
Blocks, 1-2, 1-3, 5-1, 7-1
Buffer offset length field, 8-13
Bypass label processing, 1-8, 3-3, 9-5
CLOSE monitor call, 3-7, 4-2
Closing a file, 3-7
COBOL,
labeled tapes, 3-9
unlabeled tapes, 4-1
Commands,
ALLOCATE, 1-2,3-6
ASSIGN, 3-8
BACKSPACE, 3-8
COPY, 1-9
DEALLOCATE, 1-2
DIRECTORY, 3-9
DISABLE VOLUME-RECOGNITION, 9-4
DISMOUNT, 1-2,3-8

Commands (Cont.),
ENABLE VOLUME-RECOGNITION, 9-4
EOF, 3-8
MOUNT, 1-2,3-1 to 3-3, 3-6, 3-8, 4-1,
5-2
MOUNT, Chapter 9 (labeled tapes)
RECOGNIZE, 9-5
REWIND, 3-8
SET BLOCKSIZE, 5-1
SET TAPE-DRIVE AVAILABLE, 1-1
SET TAPE-DRIVE INITIALIZE,
Chapter 2
SKIP, 3-8
UNLOAD, 3-8
Conversion,
Unlabeled to labeled tape, 1-9
COpy command, 1-9
ICOUNT switch, 2-3
Crash procedures, Chapter 12
Creation date field, 8-10
Data modes, 5-2, Chapter 6
DEALLOCATE command, 1-2
DEC standard version field, 8-5
DEC-Compatible core dump mode, 6-1
DECSYSTEM-20, 8-4
Density, 5-2
IDENSITY switch, 2-3
DIRECTORY command,
labeled tapes, 3-9
DISABLE VOLUME-RECOGNITION
command, 9-4
DISMOUNT command, 1-2, 3-8
Eight-bit mode, 6-3
ENABLE VOLUME-RECOGNITION
command, 9-4
EOF command, 3-8
EOF labels, 8-2

Index-l

EOF1 label, 8-14
EOF2 label, 8-16
EOV labels, 3-7, 8-2, 9-4
EOV1 label, 8-17
EOV2 label, 8-18
Errors, 1-11
Expiration date field, 8-10

File access code field, 8-13
File identifier field, 8-9
File section identifier field, 8-9
File sequence identifier field, 8-9
File set identifier field, 8-9
Files, 5-4
multivolume, 3-3, 3-6
updating on labeled tapes, 3-7
updating on unlabeled tapes, 4-3
Fixed-length records, 7-2
Forms control character field, 8-13
FORTRAN,
labeled tapes, 3-9
Frames, 4-2, 6--1

GALGEN, 1-9
Generation number field, 8-10
Generation version field, 8-10
Glossary, Appendix C

HDR labels, 8-2
HDRllabel, 8-8 to 8-11
HDR2 label, 8-12

lAS, 8--11
IINCREMENT switch, 2-3
Industry-compatible core dump mode, 6-3
Initializing tapes, Chapter 2
labeled, 2-1
unlabeled, 2-2
Interrecord gaps, 5-2
Interrupts, 9-3, 9-5
IPCF, 1-2

Label fields,
access code, 8-7
accessibility character, 3-5, 3-6, 8-4, 8-10
block count, 8-10, 8-15
block length, 8-13
buffer offset length, 8-13
creation date, 8-10
DEC standard version, 8-5
Index-2

Label fields (Cont.),
expiration date, 8-10
file access code, 8-13
file identifier, 8--9
file section identifier, 8-9
file sequence identifier, 8-9
file set identifier, 8-9
forms control character, 8-13
generation number, 8-10
generation version, 8-10
label identifier, 1-6
label number, 1-6
label standard version, 8-5
machine code, 8-4, 8-5
owner identifier, 8-4
owner's directory, 8-13
owner's name, 8-7
project-programmer number, 8-7
record format, 8-13
record length, 8-13
system code, 8-11
system-dependent (HDR2), 8-13
volume identifier, 8-4
Label identifier field, 1-6
Label number field, 1-6
Label processing, 1-6, 1-7
SET TAPE-DRIVE AVAILABLE
command and, 1-1
Label processor, see PULSAR
Label standard version field, 8-5
ILABEL-TYPE switch, 2-3, 3-2, 3-3
Labeled tapes, 1-3 to 1-8, 2-1,
Chapter 3, Appendix B
Labels, 5-3, 5-4
contents, Chapter 8
EOF, 8-2
EOF1, 8-14
EOF2, 8-16
EOV, 3-7, 8-2, 9-4
EOV1, 8-17
EOV2, 8-18
formats, Chapter 8
groupings, 5-5
HDR, 8-2
HDR1, 8-8 to 8-11
HDR2, 8-12
processing, 1-6, 1-7
sequencing within volume sets, 5-6 to 5-9
system, 1-6
types, 5-5
user, 1-6
UVL1, 8-6
VOL1, 9-5
VOL1, 8-3 to 8-5
Logical EOT, 3-6, 3-7,4-2

Machine code, 8-4, 8-5
Marks, 5-2
MONGEN,8-5
MOUNT command, 1-2,3-1 to 3-3,3-6,3-8,
4-1
density, 5-2
labeled tapes, Chapter 9
unlabeled tapes, 4-1
MOUNT program, 9-2, 9-3

Records (Cont.),
fixed-length, 7-2
logical, 5-4, 7-1
spanned, 7-4
undefined, 7-6
variable-length, 7-2
REWIND command, 3-8
RMS, 8-11
RSTS/E, 8-11
RSX, 8-11
RT-11, 8-11

INOTIFY switch, 3-3, 4-1, 9-4
INOW AIT switch, 3-3

Operator, 1-10, 1-11, 5-2, 9-2 to 9-6
OPR, 1-1, 1-2 9-2 to 9-4
ORION, 1-1, 1-2,9-2 to 9-4
IOVERRIDE-EXPIRATION switch, 2-3, 9-4
Owner identifier field, 8-4
IOWNER switch, 2-4
Owner's directory field, 8-13
Owner's name field, 8-7
PDP-11, 8-4
PDP-15, 8-4
PDP-8, 8-4
Physical records, see blocks
Positioning,
labeled tapes, 3-8, Chapter 10
unlabeled tapes, 4-2
Premounting, 9-3, 9-6
Project-programmer number field, 8-7
Protection codes, 3-3 to 3-6, 8-10, 8-13
IPROTECTION switch, 2-4
PULSAR, 1-1, 1-2, 3-8, 9-1 to 9-6
crashes, 12-1, 12-2, Appendix A
file search algorithm, Chapter 10
QUASAR, 1-1, 1-2, 9-2 to 9-6
crashes, 12-1, 12-2
Reading,
labeled tapes, 3-7
unlabeled tapes, 4-2
RECOGNIZE command, 9-5
Record format field, 8-13
Record formats, Chapter 7
Record length field, 8-13
Records, 1-2, 5-4, 7-1
blocking, 7-1

SET BLOCKSIZE command, 5-1
SET TAPE-DRIVE AVAILABLE command,
1-1
SET TAPE-DRIVE INITIALIZE command,
Chapter 2
SIXBIT mode, 6-3
SKIP command, 3-8
Spanned records, 7-4
Stopcodes, Chapter 12
PULSAR, Appendix A
System administrator, 1-8, 1-9
System code field, 8-11
System-dependent field (HDR2), 8-13
Tape drives, 9-5
Tape structure, Chapter 5
ITAPE-DISPOSITION switch, 2-4
TAPOP. monitor call,
accessing records, 3-8, 4-2
blocks, 5-1, Chapter 11
clearing errors, 10-3
data modes, 6-1
EOV labels, 3-7
errors, 11-8
expiration date field, 8-10
file identifier field, 8-9
file protection code, 3-3
positioning, Chapter 10
program example, 11-9
reading label parameters, 3-8
record format, 5-4, 8-13
record size, 1-2
volume switching, 4-1
writing label parameters, 3-7
TOPS-20, 8-11
Tracks, 1-3, 6-2, 9-5
Undefined records, 7-6
Unlabeled tapes, 1-8,2-2, Chapter 4

Index--3

UNLOAD command, 3-8
USER--EOT mode, 4-2
UVL1 label, 8-6

Variable-length records, 7-2
VAX-II, 8-4
Visual identification label, 1-9, 1-11
VMS, 8-11
VaLl label, 8-3 to 8-5, 9-5
VaLID, see volume identifier
/VOLID switch, 3-3, 8-4
Volume identifier, 1-9, 3-2, 3-3

Index-4

Volume identifier field, 8-4
Volume sets, 1-3, 1-5
labeled tapes, 5-3
unlabeled tapes, 5-4
Volume switching, 9-3, 9-4
SET TAPE-DRIVE AVAILABLE
command and, 1-1
unlabeled tapes, 4-1
/VOLUME-ID switch, 2-4
Writing,
labeled tapes, 3-6
unlabeled tapes, 4-2

TOPS-10
Tape Processing Manual
AA-L412A-TB

READER'S COMMENTS

NOTE:

This form is for document comments only. DIGITAL will use comments submitted on
this form at the company's discretion. If you require a written reply and are eligible to
receive one under Software Performance Report (SPR) service, submit your comments on an SPR form.

Did you find this manual understandable, usable, and well-organized? Please make suggestions for improvement.

Did you find errors in this manual? If so, specify the error and the page number.

Please indicate the type of reader that you most nearly represent.

o Assembly language programmer
D Higher-level language programmer

D Occasional programmer (experienced)
D User with little programming experience

D Student programmer

o Other (please specify)~~~~~~~~~~~~~~~~~~~~~

Name __~~~~~~~~~~~~~~~~ Date _~~~~~~~~~_
Organization __~~~~~~~~~~~~~_ Telephone __~~~~~~~~
Street __~~~~~~~~~~~~___~~~______~______~~~~~~~~__
City _~~~~~___~~~~~~~~~~_

State __~~_ Zip Code _~~_
or Country

I
I
I
1
1

I
I
1

------~-.- -oDOmNotoTear -FOa He1reaod Tape ----------------------r~-II1--------~~~~;:;~:--A
id

if Mailed in the
United States

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

SOFTWARE PUBLICATIONS
200 FOREST STREET
MR01-2/L 12
MARLBOROUGH, MA
01752

I
I

- - - - - - - -

Do Not Tear - Fold Here and Tape

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ______ 1

I
I

I
1

I
I
~

.....

::I