Digital PDFs

AA-D340A-TC

July 1978

182 pages

Original

6.7MB

Document:	TRAX MACRO Reference Manual
Order Number:	AA-D340A-TC
Revision:	0
Pages:	182
Original Filename:	http://bitsavers.org/pdf/dec/pdp11/trax/AA-D340A-TC_TRAX_V1.0_MACRO_Reference_Manual_Jul78.pdf

OCR Text

July 1978

This manual describes tbe use of the MACRO assembler on
TRAX systems.

TRAX
MACRO Reference Manual
Order No. AA-D340A-TC

OPERATING SYSTEMS AND VERSIONS:

TRAX Version 1.0

To order additional copies of this document, contact the Software Distribution
Center, Digital Equipment Corporation, Maynard, Massachusetts 01754

digital equipment corporation · maunard, massachusetts

First Printing, April 1978

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 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.
The following are trademarks of Digital Equipment Corporation:

DIGITAL
DEC
PDP
DECUS
UNIBUS
COMPUTER LABS
COMTEX
DDT
DECCOMM
ASSIST-Il

DECsystem-10
DECtape
DIBOL
EDUSYSTEM
FLIP CHIP
FOCAL
INDAC
LAB-8
DECSYSTEM-20
RTS-8

MASSBUS
OMNIBUS
OS/8
PHA
RSTS
RSX
TYPESET-8
TYPESET-ll
TMS-ll
ITPS-lO

CONTENTS

PAGE
ix

PREFACE

ix
ix

0.1
0.2
0.3
0.4

MANUAL OBJECTIVES AND READER ASSUMPTIONS
STRUCTURE OF THE DOCUMENT
ASSOCIATED DOCUMENTS
DOCUMENT CONVENTIONS

PART

INTRODUCTION TO MACRO

CHAPTER

MACRO FEATURES

1-1

1.1
1.1.1
1.1.2

OVERVIEW OF MACRO
Assembly Pass 1
Assembly Pass 2

1-1
1-1
1-2

SOURCE PROGRAM FORMAT

2-1

2.1
2.2
2.2.1
2.2.2
2.2.3
2.2.4
2.3

PROGRAMMING STANDARDS AND CONVENTIONS
STATEMENT FORMAT
Label Field
Operator Field
Operand Field
Comment Field
FORMAT CONTROL

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

PART

PROGRAMMING IN MACRO ASSEMBLY LANGUAGE

CHAPTER

SYMBOLS AND EXPRESSIONS

3-1

3.1
3.1.1
3.1.2
3.1.3
3.2
3.2.1
3.2.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9

CHARACTER SET
Separating and Delimiting Characters
Illegal Characters
Unary and Binary Operators
MACRO SYMBOLS
Permanent Symbols
User-Defined and Macro Symbols
DIRECT ASSIGNMENT STATEMENTS
REGISTER SYMBOLS
LOCAL SYMBOLS
CURRENT LOCATION COUNTER
NUMBERS
TERMS
EXPRESSIONS

3-1
3-2
3-3
3-4
3-5
3-5
3-5
3-7
3-9
3-10
3-11
3-13
3-14
3-15

CHAPTER

RELOCATION AND LINKING

4-1

CHAPTER

ADDRESSING MODES

5-1

5.1
5.2
5.3

5-1
5-2
5-2

CHAPTER

iii

x
x

PAGE
5-3
5-3
5-3
5-4
5-4
5-4
5-5
5-6
5-6
5-7
5-8
5-8

5.4
5.5
5.6
5.7
5.8
5.9
5.10
5.11
5.12
5.13
5.14
5.15

AUTO INCREMENT DEFERRED MODE
AUTODECREMENT MODE
AUTODECREMENT DEFERRED MODE
INDEX MODE
INDEX DEFERRED MODE
IMMEDIATE MODE
ABSOLUTE MODE
RELATIVE MODE
RELATIVE DEFERRED MODE
SUMMARY OF ADDRESSING FORMS
BRANCH INSTRUCTION ADDRESSING
USING TRAP INSTRUCTIONS

PART

III

MACRO DIRECTIVES

CHAPTER

GENERAL ASSEMBLER DIRECTIVES

6-1

6.1
6.1.1
6.1.2
6.1.3
6.1.4
6.1.5
6.1.6
6.2
6.3
6.3.1
6.3.2
6.3.3
6.3.4
6.3.5
6.3.6
6.3.7
6.4
6.4.1
6.4.1.1
6.4.1.2

LISTING CONTROL DIRECTIVES
.LIST and .NLIST Directives
Page Headings
.TITLE Directive
.SBTTL Directive
.IDENT Directive
.PAGE Directive/Page Ejection
FUNCTION DIRECTIVES: .ENABL AND .DSABL
DATA STORAGE DIRECTIVES
.BYTE Directive
.WORD Directive
ASCII Conversion Characters
.ASCII Directive
.ASCIZ Directive
.RAD50 Directive
Temporary Radix-50 Control Operator: AR
RADIX AND NUMERIC CONTROL FACILITIES
Radix Control and Unary Control Operators
.RADIX Directive
Temporary Radix Control Operators: AD,
AO, and AB
Numeric Directives and Unary Control
Operators
.FLT2 and .FLT4 - Floating-Point Storage
Directives
Temporary Numeric Control Operators: AC
and AF
LOCATION COUNTER CONTROL DIRECTIVES
.EVEN Directive
.ODD Directive
.BLKB and .BLKW Directives
TERMINATING DIRECTIVES
.END Directive
.EOT Directive
PROGRAM BOUNDARIES DIRECTIVE: .LIMIT
PROGRAM SECTIONING DIRECTIVES
.PSECT Directive
Creating Program Sections
Code or Data Sharing
Memory Allocation Considerations
.ASECT and .CSECT Directives
SYMBOL CONTROL DIRECTIVE: .GLOBL
CONDITIONAL ASSEMBLY DIRECTIVES
Conditional Assembly Block Directives:
• IF, .ENDC
Subconditional Assembly Block Directives:
• IFF, • I FT, • I FTF

6-1
6-1
6-8
6-11
6-11
6-12
6-12
6-13
6-17
6-17
6-18
6-19
6-20
6-21
6-22
6-23
6-24
6-24
6-24

6.4.2
6.4.2.1
6.4.2.2
6.5
6.5.1
6.5.2
6.5.3
6.6
6.6.1
6.6.2
6.7
6.8
6.8.1
6.8.1.1
6.8.1.2
6.8.1.3
6.8.2
6.9
6.10
6.10.1
6.10.2

6-25
6-26
6-27
6-27
6-29
6-29
6-29
6-30
6-30
6-31
6-31
6-31
6-32
6-32
6-36
6-37
6-38
6-38
6-39
6-40
6-40
6-43

PAGE
6.10.3

Immediate Conditional Assembly Directive:
.IIF

6-45

MACRO DIRECTIVES

7-1

7.1
7.1.1
7.1.2
7.1.3
7.1.4
7.2
7.3

7-1
7-1
7-2
7-3
7-3
7-3

7.6.1
7.6.2
7.7
7.8

DEFINING MACROS
.MACRO Directive
.ENDM Directive
.MEXIT pirective
MACRO Definition Formatting
CALLING MACROS
ARGUMENTS IN MACRO DEFINITIONS AND MACRO
CALLS
Macro Nesting
Special Characters in Macro Arguments
Passing Numeric Arguments as Symbols
Number of Arguments in Macro Calls
Creating Local Symbols Automatically
Keyword Arguments
Concatenation of Macro Arguments
MACRO ATTRIBUTE_DIRECTIVES: .NARG, .NCHR,
AND .NTYPE
.NARG Directive
.NCHR Directive
.NTYPE Directive
.ERROR AND .PRINT DIRECTIVES
INDEFINITE REPEAT BLOCK DIRECTIVES: .IRP
AND .IRPC
.IRP Directive
.IRPC Directive
REPEAT BLOCK DIRECTIVE: .REPT, .ENDR
MACRO LIBRARY DIRECTIVE: .MCALL

PART

OPERATING PROCEDURES

CHAPTER

OPERATING PROCEDURES

8-1

8.1
8.1.1
8.1.2
8.1.3
8.1.4
8.1.5
8.1.6
8.1.7
8.2
8.3

TRAX OPERATING PROCEDURES
Invoking MACRO Under TRAX
TRAX Command String Format
TRAX Macro Qualifiers
TRAX File Specification Qualifiers
Cross-Reference Processor (CREF)
TRAX MACRO in Batch Mode
TRAX Indirect Command Files
TRAX FILE SPECIFICATION FORMAT
MACRO ERROR MESSAGES

8-1
8-1
8-1
8-2
8-3
8-4
8-6
8-6
8-7
8-8

MACRO CHARACTER SETS

A-I

ASCII CHARACTER SET
RADIX-50 CHARACTER SET

A-I

CHAPTER

7.3.1
7.3.2
7.3.3
7.3.4
7.3.5
7.3.6
7.3.7
7.4
7.4.1
7.4.2
7.4.3
7.5
7.6

APPENDIX A
A.l
A.2
APPENDIX B
B.l
B.2
B.3
APPENDIX C
C.l
C.2

MACRO ASSEMBLY LANGUAGE AND ASSEMBLER
DIRECTIVES

7-4
7-5
7-6
7-6
7-7
7-7
7-9
7-10
7-11
7-11
7-12
7-13
7-14
7-15
7-16
7-16
7-18
7-18

A-4
B-1

SPECIAL CHARACTERS
SUMMARY OF ADDRESS MODE SYNTAX
ASSEMBLER DIRECTIVES

B-1
B-1

PERMANENT SYMBOL TABLE (PST)

C-l

OP CODES
MACRO DIRECTIVES

C-l
C-4

B-2

PAGE
APPENDIX D
D.l
APPENDIX E
E.l
E.2
E.3
E.4
E.4.l
E.4.l.l
E.4.l.2
E.4.l.3
E.4.2
E.4.3
E.4.4
E.4.S
E.4.S.l
E.4.S.2
E.4.S.3
E.4.S.4
E.S
E.S.l
E.S.2
E.S.3
E.S.4
E.S.4.1
E.S.4.2
E.S.4.3
E.S.4.4
E.S.4.S
E.6
E.6.l
E.6.2
E.6.3
E.7
E.8
E.9
E.9.l
E.1O
E.1O.l
E.lO.2
APPENDIX F

DIAGNOSTIC ERROR MESSAGE SUMMARY

D-1

MACRO ERROR CODES

D-1

SAMPLE CODING STANDARD

E-1

INTRODUCTION
LINE FORMAT
COMMENTS
NAMING STANDARDS
Register Standards
General Purpose Registers
Hardware Registers
Device Registers
Processor Priority
Other Symbols
Using the Standard Symbolics
Symbols
Global Symbols
Symbol Examples
Program-Local Symbols
Macro Names
PROGRAM MODULES
General Comments on Programs
The Module Preface
Formatting the Module Preface
Modularity
Calling Conventions (Inter-Module)
Exiting
Intra-Module Calling Conventions
Success/Failure Indication
Module Checking Routines
FORMATTING STANDARDS
Program Flow
Common Exits
Code with Interrupts Inhibited
PROGRAM SOURCE FILES
FORBIDDEN INSTRUCTION USAGE
RECOMMENDED CODING PRACTICE
Conditional Branches
PDP-II VERSION NUMBER STANDARD
Displaying the Version Identifier
Use of the Version Number in the Program

E-1
E-1
E-2
E-2
E-2
E-2
E-2
E-2
E-3
E-3
E-3
E-3
E-3
E-4
E-4
E-5
E-5
E-5
E-5
E-7
E-8
E-8
E-9
E-9
E-9
E-9
E-9
E-9
E-11
E-12
E-12
E-12
E-13
E-13
E-13
E-14
E-15

SAMPLE ASSEMBLY AND CROSS REFERENCE LISTING

F-1

INDEX

Index-1

FIGURES
FIGURE

3-1
3-2

6-1
6-2
6-3
6-4
6-S
6-6

7-1
8-1

Assembly Listing Showing Local Symbol Block
Sample Assembly Results
Example of Line Printer Assembly Listing
Example of Terminal Assembly Listing
Listing Produced With Listing Control
Directives
Assembly Listing Table of Contents
Example of .ENABL and .DSABL Directives
Example of .BLKB and .BLKW Directives
Example of .IRP and .IRPC Directives
Sample CREF Listing

3-11
3-12

6-6
6-7
6-9
6-12
6-16
6-30
7-17
8-5

PAGE
TABLES
TABLE

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

6-1
6-2
6-3
6-4
6-5
6-6
8-1

Special Characters Used in MACRO
Legal Separating Characters
Legal Argument Delimiters
Legal Unary Operators
Legal Binary Operators
Symbolic Arguments of Listing Control
Directives
Symbolic Arguments of Function Control
Directives
Symbolic Arguments of .PSECT Directive
Non-TRAX Program Section Default Values
Legal Condition Tests for Conditional
Assembly Directives
Subconditional Assembly Block Directives
File Specification Default Values

vii

3-1
3-3
3-3
3-4
3-5
6-3
6-13
6-33
6-38
6-41
6-43
8-8

PREFACE

0.1

MANUAL OBJECTIVES AND READER ASSUMPTIONS

MACRO fs supported in TRAX to enable Application Programmers to write
transaction step tasks
(TSTs). In addition it may be used to write
Support Environment subroutines which do not perform.
This manual
provides a reference for the MACRO language. No prior knowledge of
the MACRO Relocatable Assembler is assumed.
Although the description of the assembly
language
is
wholly
self-contained within this manual, the reader is assumed to be
familiar with the PDP-II processors and related terminology, as
presented in the PDP-II Processor Handbooks. No attempt is made in
this document to describe the PDP-II hardware or the functions of the
various PDP-II instructions.
The development of transaction step tasks
(TSTs)
also requires
knowledge of the library of RMS-II macros as presented in the TRAX RMS
MACRO Reference Manual.
In presenting MACRO, a tutorial bias has been adopted to enlarge upon
the reference material. This posture s reflected in the examples and
the accompanying commentary describing MACRO language elements in
typical applications.
NOTE
Utilization of MACRO will
eliminate
compatibility of user applications with
future members of a planned family of
TRAX systems.

0.2

STRUCTURE OF THE DOCUMENT

This manual contains four parts. Part I, consisting of two chapters,
briefly introduces MACRO. Chapter 1 lists the key features of MACRO,
and Chapter 2 identifies the advantages of following programming
standards and conventions.
Also described is the format used in
coding MACRO source programs.
Part II, consisting of three chapters, presents general information
essential to programming with the MACRO assembly language. Chapter 3
describes the symbols, terms, and expressions that form the elements
of MACRO instructions. The character set is listed, and the types of
programming symbols that may be defined by the user are discussed.

Chapter 4 describes the output of MACRO and presents concepts
essential to the proper relocation and linking of object modules by
the Linker. Chapter 5 briefly describes how data stored in memory can
be accessed and manipulated using the addressing modes recognized by
the PDP-II hardware.
Part III, consisting of two chapters, describes the MACRO directives
that control the processing of source statements during assembly.
Chapter 6 discusses directives which accomplish generalized MACRO
functions,
while Chapter 7 deals with directives used in the
definition and expansion of macros.
Part IV, consisting only of Chapter 8, presents the operating
procedures essential to the assembly, linking, and initiating of MACRO
programs.
Finally, several appendixes are provided,
supplying
information of interest to the MACRO programmer.

additional

Appendix A lists the ASCII and Radix-50 character sets that may be
used ~n MACRO programs.
Appendix B lists the special characters
recognized by MACRO, summarizes the syntax of the various addressing
modes used in PDP-II processors, and briefly describes the MACRO
directives in alphabetical order. The permanent symbols that have
been defined for use with MACRO are listed alphabetically in Appendix
C.
The diagnostic error codes produced by MACRO to identify various types
of
errors
detected
during
the
assembly process are listed
alphabetically in Appendix D. Appendix E contains a sample coding
standard that is recommended practice in preparing MACRO programs.

0.3

ASSOCIATED DOCUMENTS

The reader should refer to the TRAX documentation
descriptions of documents associated with this manual.

0.4

directory

for

DOCUMENT CONVENTIONS

The symbols defined below are used throughout this manual.
Symbol

Definition

[]

Brackets indicate
optional.

that

the

enclosed

Vertical bars indicate that a single
made from a list of arguments.

argument
choice

must

is
be

Ellipsis indicates optional continuation of an argument
list in the form of the last specified argument.
UPPER-CASE
CHARACTERS

Upper-case characters indicate elements of the language
that must be used exactly as shown.

lower-case
characters

Lower-case characters indicate elements of the language
that are supplied by the programmer.

(n)

In some instances the symbol (n) is used following a
number to indicate the radix.
For example, 100(8)
indicates that 100 is an octal value, while 100(10)
indicates a decimal value.

PART I
INTRODUCTION TO MACRO

CHAPTER 1
MACRO FEATURES

The MACRO Assembler provides the following features:
1.

Program and command string control of assembly functions

Device and filename specifications for input and output files

Error listing on command output device

Alphabetized, formatted symbol table
cross-reference listing of symbols

Reloc'atable object modules

Global symbols for linking independent object modules

Conditional assembly directives

Program sectioning directives

User-defined macros and macro libraries

10.

Comprehensive system macro library

11.

Extensive program
functions

12.

An indirect command
assembly process.

1.1

and

command
file

string

facility

listing;

control
for

optional

listing

controlling

the

OVERVIEW OF MACRO

MACRO is a 2-pass assembler. The functions and operations relevant to
each assembly pass are described in the following sections.

1.1.1

Assembly Pass I

The main purpose of assembly pass 1 is to locate and read all required
macros from libraries;
to build symbol tables and program section
tables for the program; while also performing a rudimentary assembly
of each source statement.
The first stage of assembly pass 1 is the initialization of all impure
data areas that MACRO uses internally for the assembly process. These
areas include all dynamic storage areas and buffer areas used as file
storage regions.

1-1

MACRO FEATURES
After initializing memory areas, MACRO issues a call to a system
subroutine which transfers a command line into memory. This command
line contains the specifications of the files to be used during
assembly.
After scanning the command line for proper syntax, MACRO
initializes the specified output files. These files are opened to
determine if valid output file specifications have been passed in the
command line. They are then closed to minimize requirements for
active file space.
As the assembly process begins, MACRO initiates a routine which
retrieves source lines from the input file.
If no such file is
currently open, as is the case at the beginning of assembly, MACRO
opens the next input file specified in the command line previously
read and begins to assemble the source statements.
MACRO determines
the length of each instruction and assembles it accordingly as one
word, two words, or three words.
At the end of assembly pass 1, MACRO reopens the output files
described above and writes out information that is to be used later by
the Linker in linking the object modules.
Such information as the
object module name, the program version number, and the global symbol
directory (GSD) entries for each program section are output to the
object file.
After writing out the GSD entries for a given program
section, MACRO scans through the symbol tables to find all the global
symbols that are bound to that particular program section. MACRO then
writes out GSD records to the object file for these symbols.
This
process continues for each program section, bringing to a close
assembly pass 1.

1.1.2

Assembly Pass 2

As an integral part of pass 2, MACRO simultaneously writes the object
records to the output file and generates the assembly listing,
followed by the symbol table
listing
for
the
program.
A
cross-reference listing may also be generated. See Section 8.1.4.
Basically, assembly pass 2 consists of the same steps performed in
assembly
pass 1, except that all source statements containing
MACRO-detected errors are flagged with an error code as the assembly
listing file is created. The object file that is created as the final
consequence of pass 2 contains all the object records,
together with
relocation records containing information necessary for subsequent
Linker linking of the object file.
The information thus passed to the Linker enables the global symbols
in the object modules to be associated with absolute or virtual memory
addresses, thereby forming an executable body of code.
The user may wish to become familiar with the macro object file format
and description.
This information is presented in the TRAX Linker
Reference Manual (see Section 0.3 in the Preface) .

1-2

CHAPTER 2
SOURCE PROGRAM FORMAT

2.1

PROGRAMMING STANDARDS AND CONVENTIONS

Assembly level programming deals directly with the host hardware.
Hence, great care must be exercised in establishing programming
standards and conventions to enable code written by one group to be
interchanged easily with another group. Standards provide a number of
advantages.
When applied to the program
development
process,
standards make the programming effort easier to:
Plan
Comprehend
Test
Modify
Convert.
Even though standards must accommodate local requirements, many
aspects
of
the
program
development
process
have universal
applicability. The standards common to all of DIGITAL's PDP-II
software products are presented in Appendix E as a model for users.
Observance of these standards is beneficial to DIGITAL and its users,
by simplifying both communications and the continuing task of software
maintenance and enhancement.

2.2

STATEMENT FORMAT

A source program is composed of a sequence of source coding lines.
Each line contains a single assembly-language statement. MACRO will
accept a source line of 132 characters, but 80 characters is the
recommended length, because of constraints imposed by listing format
and terminal line size.
A MACRO statement may consist of as many as four fields.
These fields
are identified by their order of appearance within the statement
and/or by specified separating characters between fields. The general
format of a MACRO statement is:
Label:

Operator Operand

;Comment(s)

The label and comment fields are optional. The operator and operand
fields are interdependent,
i~e.,
when both fields are present in a
source statement, each field is evaluated by MACRO in the context of
the other.
A statement may contain an operator field and no operand field,
but
the reverse is not true. A statement containing an operand with no
operator does not conform to established MACRO coding conventions;
such a statement is currently interpreted by MACRO during assembly as
an implicit .WORD directive (see Section 6.3.2).
2-1

SOURCE PROGRAM FORMAT
MACRO interprets and processes source program statements one by one,
generating
one or more binary instructions or data words, or
performing a specified assembly process. Blank lines, although legal,
have no significance in the source program.
An assembly-language statement must be completed on one
no continuation lines are allowed in MACRO.

source

line;

The tab character can be used in the source statement to format the
fields into aligned columns in accordance with DIGITAL's standard
source program format, as shown below:
Label - begins in column 1
Operator - begins i.n column 9
Operand(s) - begin(s) in column 17
Comment(s) - begin(s) in column 33.
For example, the following statement should be formatted in the source
program into specific columns,
increasing its readability in the
assembly listing:
REGTST:BIT#MASK,VALUE;COMPARES BITS IN OPERANDS.
1

REGTST: BIT

#MASK,VALUE

;COMPARES BITS IN OPERANDS.

(columns)

The above formatting conventions are not mandatory in coding MACRO
programs
(free-field
coding
is permissible).
However,
it is
recommended that source programs be prepared in accordance with these
conventions for consistency and clarity.

2.2.1

Label Field

A label is a means of
program.

symbolically

referring

location

A label is a user-defined symbol which is assigned the value of the
current location counter and entered into the user-defined symbol
table. The current location counter is the means by which MACRO
assigns memory addresses to the source program statements as they are
encountered during the assembly process. The address value of the
label is absolute or relocatable, depending on whether the current
program section being assembled is absolute or relocatable.
(The
concept of program sections and the attributes that may be specified
for them are discussed in detail in Section 6.8.)
In the case of an absolute program section, the value of the current
location counter is likewise absolute, i.e., its value references an
absolute virtual memory address (such as location 100).
Similarly,
'the value of the current location counter in a relocatable program
section is also relocatable; however, a relocation bias calculated by
the Linker will be added to the apparent value of the current location
counter to establish its effective absolute virtual address at
execution time.

2-2

SOURCE PROGRAM FORMAT
If present, a label always appears as the first field
in a source
statement and must be terminated by a colon.
For example, if the
current location counter value is absolute 100(8), the statement:
ABCD:

MOV

A,B

assigns the value 100(8) to the label ABCD. Subsequent references to
this label would then yield a value of absolute 100(8). In this
example, if the location counter value were relocatable,
the final
value of ABCD would be 100(8)+K, where K represents the relocation
bias of the program section, as calculated by the Linker at link time.
More than one label may appear within a single label field.
Each
label so specified is assigned the same address value. For example,
if the current location counter value is 100(8), the multiple labels
in the following statement:
ABC:

$00:

A7.7:

MOV

A,B

are each assigned the value 100(8).
Multiple labels may also appear on successive lines.
statements
ABC:
$DD:
A7.7:

MOV

For example, the

A,B

likewise cause the same current location counter value to be
to all three labels.

assigned

Of the two methods of assigning multiple labels shown above,
the
second is preferred, because consistency of field positioning within
the source program improves readability.
A double colon (::) defines the label as a global symbol.
Such a
label can be referenced by independently-assembled object modules.
References to this label in other modules will be resolved by the
Linker when the modules are linked as a composite executable task.
For example, the statement
ABeD::

MOV

A,B

establishes the label ABCD as a global symbol.
The distinguishing
attribute of a global symbol is that it can be referenced from within
an object module other than the module in which the symbol is defined
(see Section 6.9).
The legal characters for defining labels are:
A through Z
o through 9
(Period)
$ (Dollar Sign)
NOTE
By convention, the dollar sign
($)
and
period
(.)
are reserved for
use in
defining
DIGITAL
system
software
symbols.
Therefore these characters
should not be used in defining labels in
MACRO source programs.

2-3

SOURCE PROGRAM FORMAT
A label may be any length;
however, only the first six characters are
significant and, therefore, must be unique among all the labels in the
source program. All labels are terminatd by a colon (:), which is not
considered part of the label.
It is a mandatory delimiter. An error
code (M) is generated in the assembly listing if the first six
characters in two or more labels are the same (see Appendix D).
A symbol used as a label must not be redefined within the source
program.
If the symbol is redefined,
a label with a multiple
definition results, causing MACRO to generate an error code (M) in the
assembly listing (see Appendix D).
Furthermore, any statement in the
source program which references a multi-defined label results
in an
additional diagnostic message;
in this case, an error code (D) is
generated in the assembly listing (see Appendix D).

2.2.2

Operator Field

The operator field specifies the action to be performed.
It may
consist of an instruction mnemonic (op code), an assembler directive,
or a macro call.
The operator field follows the label field in a source statement.
Chapters 6 and 7 describe these three types of operator field entries.
When the operator is an instruction mnemonic,
the mnemonic op code
specifies the machine instruction to be generated.
MACRO then
continues with the evaluation of the address (es)
of the operand{s)
which follow{s).
When the operator is a directive, the directive
causes MACRO to perform certain control actions or
processing
operations during the assembly of the source program. When the
operator is a macro call, MACRO inserts the code generated by the
macro expansion.
The operator field need not be preceded by a label;
but it may be
preceded by one or more labels and followed by one or more operands
and/or a comment. Furthermore, leading and trailing spaces or tabs in
the operator field have no significance;
such characters serve only
to separate the operator field from the preceding and following
fields.
An operator is terminated by a space, tab, or any non-RAD50 character,
as in the following examples:

iTHE SPACE TERMINATES THE OPERATOR

MOV A,B

iMOV.
MOV

A,B

MOV@A,B

iTHE TAB TERMINATES THE OPERATOR MOV.
iTHE @ CHARACTER TERMINATES THE
iOPERATOR MOV.

Although the statements above are all equivalent in function,
the
second statement is the recommended form because it conforms to MACRO
coding conventions.

2.2.3

Operand Field

When the operator field contains an instruction mnemonic
Cop code),
the operand field specifies those program variables that are to be
evaluated/manipulated by the operator. The operand field may also be

2-4

SOURCE PROGRAM FORMAT
used to supply arguments to MACRO directives and macro calls, as
described in Chapters 6 and 7, respectively.
Operands may be expressions or symbolic arguments (within the context
of the specified operation). Multiple expressions used in the operand
field of a MACRO statement must be separated by a commai
multiple
symbolic arguments similarly used may be delimited by any legal
separator, i.e., a comma, tab, and/or space.
An operand should be
preceded by an operator fieldi
if it is not, the statement is treated
by MACRO as an implicit .WORD directive (see Section 6.3.2).
When the operator field contains an op code, associated operands
always expressions, as shown in the following statement:
MOV

are

RO,A+2(RI)

On the other hand, when the operator field contains a MACRO directive
or a macro call, associated operands are normally symbolic arguments,
as shown in the following statement:
.MACRO

ALPHA

ARGI,ARG2

Refer to the description of each MACRO directive to determine the type
and number of operands required in issuing the directive.
The operand field is terminated by a semicolon when the field
followed by a comment.
For example, in the following statement:
LABEL:

MOV

A,B

iCOMMENT FIELD

the tab between MOV and A terminates the operator field and defines
the beginning of the operand fieldi
a comma separates the operands A
and Bi and a semicolon terminates the operand field and defines the
beginning of the comment field. When no comment field follows, the
operand field is terminated by the end of the source line.

2.2.4

Comment Field

The comment field normally begins in column 33 and extends through the
end of the line.
This field is optional and may contain any ASCII
characters
except
null,
RUBOUT,
carriage-return,
line-feed,
vertical-tab or form-feed.
All other characters appearing in the
comment field, even special characters reserved for use in MACRO,
are
checked only for ASCII legality and then included in the assembly
listing as they appear in the source text.
All comment fields must begin with the semicolon character(i).
When
lengthy comments extend beyond the end of the source line (column 80),
the comment may be resumed in a following line.
Such a line must
contain a leading semicolon, and it is suggested that the body of the
comment be continued in the same columnar position in which the
comment began.
A comment line can also be included as an entirely
separate line within the code body.
Comments do not affect assembly processing or program execution.
However, comments are useful in source listings for later analysis,
debugging, or documentation purposes.

2-5

SOURCE PROGRAM FORMAT
2.3

FORMAT CONTROL

Horizontal formatting of the source program is controlled by the space
and tab characters. These characters have no effect on the assembly
process unless they are embedded within a symbol, number, or ASCII
text string, or unless they are used as the operator field terminator.
Thus, the space and tab characters can be used to provide an orderly
and readable source program, as reflected by the following statements:
LABEL:MOV(SP)+,TAGiPOP VALUE OFF STACK.
No spaces or tabs have been used to separate the fields in this
statement.
Note the difficulty in recognizing where one field ends
and the next begins.
LABEL:

MOV

(SP)+,TAG

iPOP VALUE OFF STACK.

This statement conforms to the standard
horizontal
formatting
conventions,
i.e., the statement elements are separated into four
distinct fields and are therefore easily discernible.
Page formatting and assembly listing considerations are discussed in
Chapter 6 in the context of MACRO directives that may be specified to
accomplish desired formatting operations. Appendix E describes the
coding conventions used in all DIGITAL PDP-II operating system
software.

2-6

SOURCE PROGRAM FORMAT

PART II
PROGRAMMING
IN MACRO ASSEMBLY
LANGUAGE

CHAPTER 3
SYMBOLS AND EXPRESSIONS

This chapter describes the components of MACRO instructions.
The
character set, the conventions observed in constructing symbols, and
the use of numbers, operators, terms and expressions are discussed as
they relate to MACRO programming.

3.1

CHARACTER SET

The following characters are legal in MACRO source programs:
1.

The letters A through Z. Both upper- and lower-case letters
are acceptable, although, upon input, lower-case letters are
converted to upper-case (see Section 6.2, .ENABL LC).

The digits 0 through 9.

The characters
(period)
and $
characters
are
reserved for use
Corporation system program symbols.

The special characters listed in Table 3-1.

(dollar sign).
These
as Digital Equipment

Table 3-1
Special Characters Used in MACRO
Character

Designation

Colon

Label terminator.

Double colon

Label terminator; defines the
label as a global label.

Equal sign

Direct assignment
operator;
and macro keyword indicator.

Double equal
sign

Direct assignment operator;
defines the symbol as a global
symbol.

Percent sign

Tab

Item or field terminator.

Space

Item or field terminator.

Function

(continued on next page)
3-1

SYMBOLS AND EXPRESSIONS
Table 3-1 (Cont.)
Special Characters Used in MACRO
Character

Designation

Function

Number sign

Immediate expression
indicator.

At sign

Deferred addressing indicator.

(

Left parenthesis

Initial register indicator.

)

Right parenthesis

Terminal register indicator.

Period

Current location counter

Comma

Operand field separator.

Semicolon

Comment field indicator.

Left angle
bracket

Initial argument or expression
indicator.

Right angle
bracket

Terminal argument or expression indicator.

Plus sign

Arithmetic addition operator
or autoincrement indicator.

Minus sign

Arithmetic subtraction operator or autodecrement indicator.

Asterisk

Arithmetic mUltiplication operator.

Slash

Arithmetic division operator.

Ampersand

Logical AND operator.

Exclamation point

Logical inclusive OR operator.

Double quote

Double ASCII character indicator.

Single quote

Single ASCII character indicator;
or
concatenation
indicator.

Up arrow or
circumflex

Universal unary operator or
argument indicator.

Backslash

Macro call numeric argument
indicator.

3.1.1

Separating and Delimiting Characters

Legal separating characters and legal argument delimiters are
below in Tables 3-2 and 3-3 respectively.

3-2

defined

SYMBOLS AND EXPRESSIONS
Table 3-2
Legal Separating Characters
Character

Definition

Usage

Space

One or more spaces
and/or tabs

A space is a legal separator
between instruction fields and
between
symbolic
arguments
within the
operand
field.
Spaces within expressions are
ignored (see Section 3.9) •

Comma

A comma is a legal separator
between
symbolic
arguments
within the
operand
field.
Multiple expressions used in
the operand field must
be
separated by a comma.
Table 3-3
Legal Argument Delimiters

Character

Definition

Usage

<••• >

Paired angle brackets

angle brackets may be
used anywhere in a program to
enclose an
expression
for
treatment as a single term.
Paired angle brackets are also
used
to
enclose
a macro
argument, particularly
when
that argument contains separating characters (see Section
7.3) •

AX ••• x

Up-arrow (unary operator)
construction,
where the up-arrow is
followed
by an argument that is bracketed
by any paired printing
characters (x).

This construction is equivalent in function to the paired
angle brackets described above
and is generally used only
where the argument itself contains angle brackets.

3.1.2

Pai~ed

Illegal Characters

A character is determined to be illegal for one of two reasons:
1.

A character is not an element of the recognized MACRO
character set.
A character of this kind is replaced in the
listing by a question mark, and an error code (1) is printed
in the assembly listing (see Appendix D). The exception to
this is an embedded null which, when detected, terminates the
scan of the current line.

A legal MACRO character is illegal in the context of its
usage within the source statement,
i.e.,
its syntax is
illegal or questionable. Such a character causes an error
code (Q) to be printed in the assembly listing.
3-3

SYMBOLS AND EXPRESSIONS
3.1.3

Unary and Binary Operators

Legal MACRO unary operators are described in Table 3-4.
Unary
operators are used in connection with single terms (arguments or
operands) to indicate an action to be performed on that term during
assembly.
A term preceded by a unary operator is considered to
contain that operator. The term 'so specified thus becomes a value
which can be used alone or as an element of an expression.
Table 3-4
Legal Unary Operators
Unary
Operator

Explanation

Example

Plus sign

Produces
the
value of A.

Minus sign

-A

Produces
the
negative
(2' s complement) value of
A.

Up-arrow, universal unary operator.
(This usage is
described in detail
in Section 6.,; 4. )

"'C24

Produces the lis complement value of 24 (8) •

"'D127

Interprets 127 as a
decimal number.
3.0
Interprets
as
a
I-word,
floating-point
number.

....

"'F3.0

Effect
positive

"'034

Interprets 34 as an octal
number.

"'Bl1000lll

Interprets 11000111 as
binary number.

"'RABC

Evaluates ABC in Radix-SO
form.

Unary operators can be used adjacent to each other or in constructions
involving multiple terms, as shown below:
-"'DSO
"'C"'012

(Equivalent to -<"'DSO»
(Equivalent to ... C< .... 012»

Legal MACRO binary operators are described in Table 3-S. In contrast
to unary operators, binary operators specify actions to be performed
on multiple items or terms within an expression. Table 3-S shows the
relationships that can be established between expression terms through
the use of binary operators.

3-4

SYMBOLS AND EXPRESSIONS
Table 3-5
Legal Binary Operators
Binary
Operator

Explanation

Example

Addition

A+B

Subtraction

A-B

Multiplication

A*B (16-bit product returned)

Division

A/B (16-bit quotient returned)

Logical AND

A&B

Logical inclusive OR

A!B

All binary operators have equal priority.
Items or terms can be
grouped for evaluation within an expression by enclosing them within
angle brackets. Terms so enclosed are evaluated first, and remaining
operations are performed from left to right,.as shown in the examples
below:
• WORD
• WORD

3.2

1+2*3
1+<2*3>

iEQUALS 11(8).
iEQUALS 7(8) •

MACRO SYMBOLS

Three types of symbols may be defined for use within MACRO source
programs: permanent symbols, user-defined symbols, and macro symbols.
MACRO maintains three types of symbol tables:
the Permanent Symbol
Table
(PST), the User Symbol Table (UST), and the Macro Symbol Table
(MST). The PST contains all the permanent symbols defined within (and
thus automatically recognized by) MACRO and is part of the MACRO task
image. The UST and MST are constructed as the source program is
assembled.

3.2.1

Permanent Symbols

Permanent symbols consist of the instruction mnemonics
(see Appendix
C) and MACRO directives (see Chapters 6 and 7 and Appendix B). These
symbols are a permanent part of the MACRO task image and need not be
defined before being used in the operator field of a MACRO source
statement (see Section 2.2.2).

3.2.2

User-Defined and Macro Symbols

User-defined symbols are those symbols treated by the programmer as
labels
(see Section 2.2.1) or that are equated to a specific value
through a direct assignment statement (see Section 3.3) or appear as
macro names or dummy arguments. These symbols are added to the User
Symbol Table as they are encountered during assembly.
Macro symbols
are those symbols used as macro names (see Section 7.l)~ Similarly,
these symbols are added to the Macro Symbol Table ~ as they are
encountered during assembly.
. .
3-5

SYMBOLS AND EXPRESSIONS
User-defined and macro symbols can be composed of alphanumeric
(.) only;
any other
characters, dollar signs
($), and periods
character is illegal.
NOTE
The dollar sign
($)
and period
(.)
characters
are reserved for use in
defining Digital Equipment Corporation
system software symbols. For example,
READ$ is a file-processing system macro.
The user
is cautioned not to employ
these
characters
in
constructing
user-defined symbols or macro symbols in
order to avoid possible conflicts with
existing or future Digital Equipment
Corporation system software symbols.
The following rules govern the
symbols:

creation

user-defined

and

macro

The first character of a symbol must not be a number
in the case of local symbols;
see Section 3.5).

(except

The first six characters of a symbol must be unique.

A symbol can be written with more than six legal characters,
but the seventh and subsequent characters are checked only
for ASCII legality and are not otherwise evaluated or
recognized by MACRO.

Spaces, tabs, and illegal characters must not be embedded
within a symbol. The legal MACRO character set is defined in
Section 3.1.

The value of a symbol depends upon its use in the program.
When a
symbol appears in the operator field, it may be anyone of the three
symbol types described above i.e., permanent, user-defined, macro. To
determine the value of an operator-field symbol, MACRO searches the
symbol tables in the following order:
1.

Macro Symbol Table

Permanent Symbol Table

User-Defined Symbol Table

This search order allows redefinition of Permanent Symbol Table
entries as macro symbols. That is, permanent symbols may be used as
macro symbols. But the user must keep in mind the sequence in which
the search for symbols is performed in order to avoid incorrect
interpretation of the symbol's use.
When a symbol appears in the operand field,
the User-Defined Symbol
Table is searched first, then the Permanent Symbol Table is searched.
Depending on their use in the source program, user-defined symbols
have either a local
(internal)
attribute or a global (external)
attribute.
Normally, MACRO treats all user-defined symbols
their definition is limited to the module

3-6

as local,
that is,
in which they appear.

SYMBOLS AND EXPRESSIONS
However, symbols can be explicitly
through one of three methods:

declared

global

Use of the .GLOBL directive (see Section 6.9).

Use of the double colon (::) in defining a label
2.2.1).

Use of the double equal (==)
statement (see Section 3.3).

sign

All symbols within a module that remain undefined
assembly are treated as default global references.

(see Section

direct
at

symbols

assignment
the

end

NOTE
Undefined symbols at the end of assembly
are assigned a value of 0 and placed
into the user-defined symbol table as
undefined default global references.
If
the .DSABL GBL directive is in effect,
however,
(see
Section
6.2),
the
automatic
global
reference
default
function of MACRO is inhibited, causing
the statement containing the undefined
symbol to be flagged with an error code
(U)
in the
assembly
listing
(see
Appendix D).
Global symbols provide linkages between independently-assembled object
modules within the task image. A global symbol defined as a label,
for example, may serve as an entry-point address to another section of
code within the task image. Such symbols are referenced from other
source modules in order to transfer control throughout the task's
execution.
These global symbols are resolved by the Linker at link
time, ensuring that the resulting task image is a logically coherent
and complete body of code.

3.3

DIRECT ASSIGNMENT STATEMENTS

A direct assignment statement allows you to equate a symbol to a
specific value.
When a direct assignment statement is first used to
define a symbol, that symbol is entered into the User-Defined Symbol
Table.
A symbol defined in this manner may be redefined in a
subsequent direct assignment statement by assigning a new value to the
previously-defined symbol.
The general format for a direct assignment statement is:
symbol=expression
or
symbol==expression
where:

expression

- can have only one level of forward
(see 5. below).

reference

- cannot contajn an undefined global reference.

3-7

SYMBOLS AND EXPRESSIONS
A direct assignment statement embodying the double equal (==) sign, as
shown above, defines the symbol as global (see Section 6.9).
The following examples illustrate
statements:

the

coding

direct

assignment

A=l

i~HE SYMBOL A IS EQUATED TO THE
iVALUE 1.

B=A-l&MASKLOW

iTHE SYMBOL B IS EQUATED TO THE
iVALUE OF THE ENTIRE EXPRESSION
iWHICH FOLLOWS.

C:
D=.
E:

iTHE SYMBOL D IS EQUATED TO ., AND
MOV

#l,ABLE

iTHE LABELS C AND E ARE ASSIGNED A
iVALUE THAT IS EQUAL TO THE LOCATION
iOF THE MOV INSTRUCTION.

The last of the three examples above is provided only to illustrate
the performance of MACRO in such situations. See Section 3.6 for a
description of the period (.) as the current location counter symbol.
The following conventions apply to the
statements:

coding

direct

assignment

An equal sign (=) or double equal sign (==) must separate the
symbol from the expression defining the symbol's value.
Spaces preceding and/or following the direct assignment
operators,
although permissible, have no significance in the
resulting value.

The symbol being assigned in a direct assignment statement is
placed in the label field.

Only one symbol can be defined in a single direct
statement.

assignment

A direct assignment statement
comment field.

only

Only one level of forward referencing is allowed, as shown in
the following example:

may

X=Y

(Illegal forward reference)

y=z

(Legal forward reference)

followed

Z=l
The above example would result in the generation of an error code
(U)
in the assembly listing on the line containing the illegal forward
reference.
Although one level of forward referencing is allowed for
local
symbols, a global symbol defined in a direct assignment statement must
not contain a forward reference,
i.e.,
the
global
assignment
expression must not itself contain an undefined reference to another
symbol. Such a forward reference is illegal, causing an error code
(A) to be generated in the assembly listing.

3-8

SYMBOLS AND EXPRESSIONS
3.4

The eight general registers of the PDP-II processor are numbered 0
through 7 and can be expressed in the source program in the following
manner:
%0
%1

%7
where % indicates a reference to a register rather than a location.
The digit specifying the register can be replaced by any legal,
absolute term that can be evaluated during the first assembly pass.
Use standard symbolic names for all register references.
The register definitions listed below are automatically assigned by
MACRO, i.e., these definitions are the normal default values and
remain valid for all register references within the source program.
iREGISTER 0 DEFINITION.
iREGISTER 1 DEFINITION.
iREGISTER 2 DEFINITION.
iREGISTER 3 DEFINITION.
iREGISTER 4 DEFINITION.
iREGISTER 5 DEFINITION.
iSTACK POINTER DEFINITION.
iPROGRAM COUNTER DEFINITION.

RO=%O
Rl=%l
R2=%2
R3=%3
R4=%4
R5=%5
SP=%6
pc=%7

Note that registers 6 and 7 are given special names because
unique system functions.

their

A regi~ter ~ymbol may be defined in a direct assignment statement
appearIng In the program.
The defining expression of a register
symbol must be a legal, absolute value. Although you can reassign the
standard register symbols through the use of the .DSABL REG directive
(see Section 6.2), this practice is not recommended.
An attempt to
redefine a default register symbol without first specifying the .DSABL
REG directive to override the normal register definitions causes that
assignment statement to be flagged with an error code (R) in the
assembly listing.
The symbolic default names assigned to
the
registers, as listed above, are the conventional names used in all
DIGITAL-supplied PDP-II system programs. For this reason, you are
well advised to follow these conventions.
All non-standard register symbols must be defined before they are
referenced in the source program. A register expression less than 0
or greater than 7 is flagged with an error code (R)
in the assembly
listing.
The % character may be used with any legal term
specify a register. For example, the statement
CLR

%3+1

is equivalent in function to the statement
CLR

and clears the contents of register 4.

3-9

expression

SYMBOLS AND EXPRESSIONS
In contrast, the statement
CLR

clears the contents of virtual memory location 4.

3.5

LOCAL SYMBOLS

Local symbols are specially formatted symbols used as labels within a
block of coding that has been delimited as a local symbol block.
Local symbols are of the form n$, where n is a decimal integer from 1
to 65535, inclusive. Examples of local symbols are:
1$
27$
59$
104$
A local symbol block is delimited in one of three ways:
1.

The range of a local symbol block usually consists of those
statements between two normally-constructed symbolic labels
(see Figure 3-1). Note that a statement of the form:
ALPHA=expression
is a direct assignment statement (see Section 3.3), but does
not create a label and thus does not delimit the range of a
local symbol block.

The range of a local symbol block is normally terminated upon
encountering a
.PSECT,
.CSECT, or .ASECT directive in the
source program (see Figure 3-1).

The range of a local symbol block is delimited through
directives, as follows:
Starting delimiter:

.ENABL LSB

Ending delimiter:

.ENABL

MACRO

(see Section 6.2)

LSB

or
• DSABL

LSB

(see Section 6.2)

followed by one of:
.PSECT
.CSECT
.ASECT

Symbolic label

(see Section 6.8.1)
(see Section 6.8.2)
(see Section 6.8.2)

Local symbols provide a convenient means of generating labels for
branch instructions and other such references within a local symbol
block.
Using local symbols reduces the possibility of symbols with
multiple definitions appearing within a user program.
In addition,
the use of local symbols differentiates entry-point labels from local
labels, since local symbols cannot be referenced from outside their
respective local symbol block. Thus, local symbols of the same name
can appear in other local symbol blocks without conflict.
Local
symbols do not appear in cross-reference listings.

3-10

SYMBOLS AND EXPRESSIONS
Local symbols require less symbol table space than other types of
symbols.
Their use is recommended. When defining local symbols, use
the range from 1$ to 63$ first, then the range from 128$ to 65535$.
Local symbols within the range 64$ through 127$, inclusive, can be
generated automatically as a feature of MACRO. Such local symbols are
useful in the expansion of macros during assembly and are described in
detail in this context in Section 7.3.5.
Be sure to avoid multiple definitions of local symbols within the same
local symbol block. For example, if the local symbol 10$ is defined
two or more times within the same local symbol block, each symbol
represents a different address value. Such a multi-defined symbol
causes an error code (P) to be generated in the assembly listing.
For examples of local symbols and local symbol blocks as
in a source program, see Figure 3-1.
121
122
123
12'
125 000000
126 o ill 0 00
127 1!l~0"'0'
128 00"~06
129 01lJ00t2
IiIl

PROGRAM

I~ITIALIZATIO~

.PSECT
012700
005020

0Q1000~1

02270~

0rI1J00~'

lICTP~GIIMOV

iSl

10137'

CLR
CMP
8"'1

they

appear

CODE

XCTPRG,GBL
.IMPURt.,R0
(R0).

'IMPURE DATA INITIAL.IZATION

.IMPURT,~",

13~

131 000'" o III
132 00000W'l
133 0001110'
134 111"'0006
135 0"'0"'12
136
137 00""''''0
138 1/l00"1iIl'"
139 0~0"'04
140 01lJ01i106
1'1 U001?
1'2

"1270P
00502'"

PP01Hl0 '

Q!227~0

PiII"';tI0" ,

101374
11112700
01115020
022700
li11J74

11'00(1100'
011l11H""""

.PSECT
lCCTLINIC"OV
lSI
CLR
CMP
8111

Figure 3-1

3.6

.PSECT
lICTPASll MOV
CLR
lS.
CMP
8"'1

lICTPAS,G8L
·I~PPAS,~0

,PASS INITIALIZATION

(I~IIl) •

.I"'PPAT,~0

lIClL.IN,GBl
*IMPLIN,RQ!
(R0'.
.YMPL.IT,RQ!

'L.I~E

INITIAlIZATIO~

Assembly Listing Showing Local Symbol Block

CURRENT LOCATION COUNTER

The period e.) is the symbol for the current location counter.
When
used in the operand field of an instruction, it represents the address
of the first word of the instruction, as shown in the first example
below.
When used in the operand field of a MACRO directive, it
represents the address of the current byte or word, as shown in the
second example below.
A:

MOV

#. ,RO

iTHE PERIOD (.) REFERS TO THE ADDRESS
iOF THE MOV INSTRUCTION.

(The function of the # symbol is explained in Section S.9.)
SAL=O
• WORD

l77535,.+4,SAL

iTHE OPERAND .+4 IN THE .WORD

iDIRECTIVE REPRESENTS A VALUE
iTHAT IS STORED AS THE SECOND
iOF THREE WORDS DURING
iASSEMBLY.
Assume that the current value of the location counter is 500.
During
assembly, MACRO reserves storage in response to the .WORD directive
(see Section 6.3.2), eeginning with location 500.
The operands
accompanying the .WORD directive determine the values so stored. The

3-11

SYMBOLS AND EXPRESSIONS
value 177535 is thus stored in location 500. The value represented by
.+4 is stored in location 502; this value is derived as the current
value of the location counter (which is now 502), plus the absolute
value 4, thereby depositing the value 506 in location 502. Finally,
the value of SAL, previously equated to 0, is deposited in location
504.
Figure 3-2 illustrates the result of the example.
LOCATION

CONTENTS

500

177535

502

506

504

Figure 3-2

Sample Assembly Results

At the beginning of each assembly pass, MACRO resets the location
counter.
Normally, consecutive memory locations are assigned to each
byte of object data generated. However, the value of the location
counter can be changed through a direct assignment statement of the
following form:
.=expression
Similar to other MACRO symbols, the current location counter symbol
(.)
has an attribute of relocatability associated with it: it is
either absolute or relocatable, depending on the specific such
attribute of the current program section.
(A program section and its
attributes are defined through the use of the .PSECT directive
described in Section 6.8.1.) The existing attribute (or mode) of the
current location counter cannot be changed by specifying a defining
expression having a different attribute.
Furthermore, such a defining expression must not force the location
counter into another program section (.PSECT area), even though the
program sections so involved may both be absolute or relocatable. The
expression defining the location counter value must not contain a
forward reference, i.e., the expression must not contain a reference
to
a symbol that is not previously defined.
Such violations
constitute a general assembly error, resulting in an error code (A) in
the assembly listing.
Thus, the attribute (or mode) of the current location counter takes on
the
attribute of the current program section.
Therefore,
its
attribute from program section to program section can be changed only
through
the program sectioning directives
(.PSECT,
.ASECT, and
.CSECT), as described in Section 6.8.
The following coding illustrates
counter:

the

use

the

current

location

.ASECT
.=500
FIRST:

MOV

• +lO,COUNT

;SET LOCATION COUNTER TO
;ABSOLUTE 500(OCTAL) •
;THE LABEL "FIRST" HAS THE VALUE
;500 (OCTAL) •
;.+10 EQUALS 5l0(OCTAL). THE
;CONTENTS OF THE LOCATION

3-12

SYMBOLS AND EXPRESSIONS

.=520
SECOND: MOV

.,INDEX

;510 (OCTAL) WILL BE DEPOSITED
;IN THE LOCATION "COUNT."
;THE ASSEMBLY LOCATION COUNTER
;NOW HAS A VALUE OF
;ABSOLUTE 520{OCTAL).
;THE LABEL SECOND HAS THE
;VALUE 520{OCTAL).
;THE CONTENTS OF LOCATION
;520{OCTAL}, THAT IS, THE BINARY
;CODE FOR THE INSTRUCTION
iITSELF, WILL BE DEPOSITED IN THE
iLOCATION "INDEX."

.PSECT
.=.+20
THIRD:

• WORD

;SET LOCATION COUNTER TO
iRELOCATABLE 20 OF THE
iUNNAMED PROGRAM SECTION.
iTHE LABEL THIRD HAS THE
iVALUE OF RELOCATABLE 20.

Storage areas may be reserved in the program by advancing the location
counter. For example, if the current value of the location counter is
1000, each of the following statements:
.=.+40
or
.BLKB 40
or
.BLKW 20
reserves 40(8) bytes of storage space in the source program.
The
.BLKB and .BLKW directives, however, are recommended as the preferred
ways to reserve storage space (see Section 6.5.3).

3.7

NUMBERS

MACRO assumes that all numbers in the source program are to be
interpreted in octal radix, unless otherwise specified. An exception
to this is that operands associated with Floating Point Processor
instructions and Floating Point Data directives are treated as decimal
(see Section 6.4.2). This default radix can be altered with the
.RADIX directive {see Section 6.4.l.l}. Also, individual numbers can
be designated as decimal, binary, or octal numbers through temporary
radix control operators (see Section 6.4.1.2).
For every statement in the source program that contains a digit that
is not in the current radix, an error code {N} is generated in the
assembly listing. However, MACRO continues with the scan of the
statement and evaluates each such number encountered as a decimal
value.
Negative numbers must be preceded by a minus sign;
MACRO translates
such numbers into two's complement form. Positive numbers may (but
need not) be prec~ded by a plus sign.
A number containing more than 16 significant bits, i.e., greater than
l77777(8),
is truncated from the left and flagged with an error code
(T) in the assembly listing.

3-13

SYMBOLS AND EXPRESSIONS
Numbers are always considered to be absolute values,
not relocatable.

i.e.,

they

are

Single-word floating-point numbers may be generated with the AF
operator (see Section 6.4.2.2) and are stored in the following format:
14

7 6

8-bit
Exponent

7-bit
Mantissa

Refer to the appropriate PDP-II Processor Handbook for details of
floating-point number format.

3.8

the

TERMS

A term is a
following:

component

expression

and

may

A number, as defined in Section 3.7, whose
used.

A symbol, as defined in Section 3.2.
as follows:

one

l6-bit

Symbols

are

the

value

evaluated

A period (.) specified in an expression causes the
of the current location counter to be used.

value

A defined symbol is located in the
Table (UST) and its value is used.

Symbol

A permanent symbol's basic value is
substituted for the addressing modes.
all op codes and their values.)

An undefined symbol is assigned a value of zero and
inserted in the User-Defined Symbol Table as an undefined
default global reference. If the • DSABL GBL directive
(see Section 6.2)
is in effect, the automatic global
reference default function of MACRO is inhibited,
in
which case, the statement containing the undefined symbol
is flagged with an error code
(U)
in the assembly
listing.

User-Defined

used, with zero
(Appendix C lists

An ASCII conversion operation is performed, using either a
single quote followed by a single ASCII character, or a
double quote followed by two ASCII characters. This type of
expression construction is explained in detail in Section
6.3.3.

A term may also be an expression enclosed in angle brackets
«». Any expression so enclosed is evaluated and reduced to
a single term before the remainder of the expression in which
it appears is evaluated. Angle brackets, for example, may be
used to alter the left-to-right evaluation of expressions (as
in A*B+C versus A*<B+C», or to apply a unary operator to an
entire expression (as in -<A+B>).

3-14

SYMBOLS AND EXPRESSIONS
3.9

EXPRESSIONS

Expressions are combinations of terms joined together by bin~ry
operators
(see Table 3-5)
and which reduc~ to a 16-bit expresslon
value. The evaluation of an expression includes the determination of
its attributes.
A resultant expression value may be anyone of four
types (as described later in this section):
absolute, relocatable,
external, or complex relocatable.
Expressions are evaluated from left to right with no operator
hierarchy rules, except that unary operators take precedence over
binary operators. A term preceded by a unary operator is considered
to contain that operator.
(Terms are evaluated, where necessary,
before their use in expressions.) Multiple unary operators are valid
and are treated as follows:
-+-A
is equivalent to:
-(+(-A»
A missing te~m, expression, or external symbol is interpreted as a
zero.
A mlssing or illegal operator terminates the expression
analysis, causing an error code (A) or (Q), or both, to be generated
in the assembly listing, depending on the context of the expression
itself. For example, the expression:
TAG! LA 177777
is evaluated as
TAG ! LA
because the first non-blank character following the symbol LA is not a
legal binary operator, an expression separator (i.e., a comma), or an
operand field terminator (i.e., a semicolon or the end of the source
line). It should be noted that spaces within expressions are ignored.
The value of an external expression is equal to the value of the
absolute part of that expression.
For example, the expression
EXTERN+A, where "EXTERN" is an external symbol, has a value at
assembly-time that is equal to the value of the internal symbol A.
This expression, however, when evaluated by the Linker at link time
takes on the resolved value of the symbol EXTERN, plus the value of
symbol A.
Expressions, when evaluated by MACRO, are determined to be one of four
types:
absolute, relocatable, external
(or global), or complex
relocatable. The following distinctions are important:
1.

An exp~ession is absolute if its value is fixed.
An
expresslon whose terms are numbers and ASCII conversion
characters will reduce to an absolute value.
A relocatable
expression or term minus a relocatable term, where both
elements being evaluated belong to the same program section,
are also absolute, since such an expression is reduced to a
single term by MACRO upon completion of the expression scan.
For example, the expression TAG2-TAGl, where both TAGI and
TAG2 are defined in the same program section, is an absolute
expression.

3-15

SYMBOLS AND EXPRESSIONS
2.

An expression is relocatable if its value is fixed relative
to the base address of the program section in which it
appears, but it will have an offset value added at task-build
time.
Expressions whose terms contain labels defined in
relocatable program sections will have a relocatable value;
similarly, a period
,.)
in a relocatable program section,
representing the value of the current location counter, will
also have a relocatable value.

An expression is external (or global) if it contains a single
global reference (plus or minus an absolute expression value)
that is not defined within the current program.
Thus, an
external expression is only partially defined following
assembly and must be resolved by the Linker at link time.

An expression is complex relocatable if any of the
conditions applies:

following

- It contains a global reference and a relocatable symbol.
- It contains more than one global reference.
- It contains relocatable
program sections.

terms

belonging

different

The value resulting from the expression has more than one
level of relocation.
For example,
if the relocatable
symbols TAGI and TAG2 associated with the same program
section are specified in an expression construction in the
form TAGI+TAG2, two levels of
relocation
would
be
introduced, since each symbol is evaluated in terms of the
relocation bias in effect for the program section.
- An operation other than
undefined global symbol.

addition

specified

- An operation other than addition, subtraction, negation, or
complementation is specified for a relocatable value.
The evaluation of relocatable, external,
expressions is completed at link time.

3-16

and

complex

relocatable

CHAPTER 4

RELOCATION AND LINKING

The output of MACRO is an object module which must be processed by the
Linker before it can be loaded and executed. Essentially, the Linker
fixes (i.e., makes absolute) the values of external or relocatable
symbols in the object module, thus transforming the object module, or
several such object modules, into an executable task image.
This
process is called linking.
To enable the Linker to fix the value of an expression, MACRO issues
certain directives to the Linker,
together with other required
parameters. In the case of relocatable expressions in the object
module, the Linker adds the base of the associated relocatable program
section to the value of the relocatable expression provided by MACRO.
In the case of external expression values, the Linker determines the
value of the external term in the expression
(since the external
symbol must be defined in one of the other object modules being linked
together) and then adds it to the absolute portion of the external
expression, as provided by MACRO.
All instructions that require modification by the Linker are flagged
in the assembly listing, as illustrated in the example below. The
apostrophe (I)
following the octal expansion of the instruction
indicates that simple relocation is requiredi
the letter G indicates
that the value of an external symbol must be added to the absolute
portion of an expressioni
and the letter C indicates that complex
relocation analysis by the Linker is required in order to fix the
value of the expression.
EXAMPLE:
EXTERN (R5)

iTHE VALUE OF THE "EXTERN" SYMBOL IS
iASSEMBLED AS ZERO AND IS TO BE
iRESOLVED BY THE TASK BUILDER.

005065 CLR
000006G

EXTERN+6 (R5)

iTHE VALUE OF THE SYMBOL "EXTERN"
iIS TO BE "RESOLVED BY
iTHE TASK BUILDER AND ADDED TO
iTHE ABSOLUTE PORTION (+6) OF
iTHE EXPRESSION.

005065 CLR
000040 1

RELOC (R5)

iASSUMING THAT THE VALUE OF THE
iSYMBOL "RELOC" IS RELOCATABLE
i40, THE TASK BUILDER WILL ADD A
iRELOCATION BIAS TO THIS VALUE.

005065

-(EXTERN+RELOC> (R5) iTHIS EXPRESSION IS COMPLEX
RELOCATABLE BECAUSE IT REQUIRES
THE NEGATION OF AN EXPRESSION
THAT CONTAINS A GLOBAL (EXTERN)
REFERENCE AND A RELOCATABLE TERM.

005065

CLR

OOOOOOG

OOOOOOC

CLR

4-1

RELOCATION AND LINKING
For a complete description of object records output by MACRO, refer to
the TRAX Linker Reference Manual (see Section 0.3 in the Preface).

4-2

CHAPTER 5
ADDRESSING MODES

The program counter (PC) always contains the address of the next word
to be fetched,
i.e., the address of the next instruction to be
executed, or the second or third word of the current instruction.
In order to understand how the address modes operate and how they
assemble, the action of the program counter must be understood. The
key rule to remember is:
"whenever the processor implicitly uses the program counter
(PC)
to fetch a word from memory, the program counter is
automatically incremented by 2 after the fetch operation is
completed."
In the case of 2- or 3-word instructions, the processor uses the PC to
fetch the following words as well.
The following symbols are
throughout this chapter:

used

describing

modes

E is any expression, as defined in Chapter 3.

R is a register expression, i.e., any expression containing a
term preceded by a percent sign (%) or a symbol previously
equated to such a term, as shown in the examples below:
RO=%O
Rl=RO+l
R2=1+%1

jGENERAL REGISTER O.
jGENERAL REGISTER 1.
jGENERAL REGISTER 2.

The symbol R may also represent any of
register definitions (see Section 3.4).

5.1

addressing

the

normal

default

ER is a register expression or an absolute expression in
range 0 to 7, inclusive.

the

A is a general addressing specification which produces a
6-bit
mode address field,
as described in the PDP-II
Processor Handbooks. The addressing specification, A,
is
described in terms of E, R, and ER, as defined above. Each
addressing specification within this section is illustrated
using either the single operand instruction CLR or the double
operand instruction MOV.

The register itself (R) contains the operand to be manipulated by
instruction.

5-1

the

ADDRESSING MODES
Format for A:

Example:
CLR

5.2

iCLEARS REGISTER 3.

The register (R) contains the address of the operand to be manipulated
by the instruction.
Format for A:

@R or

(ER)

Examples:
CLR
CLR
CLR

5.3

@Rl
(Rl)
(1)

iALL THESE INSTRUCTIONS CLEAR
iTHE WORD AT THE ADDRESS
iCONTAINED IN REGISTER 1.

AUTOINCREMENT MODE

The contents of the register (ER) are incremented immediately
being used as the address of the operand (see Note below).
Format for A:

(ER)+

Examples:
CLR
CLR
CLR

(RO)+
(R4)+
(R2)+

iEACH INSTRUCTION CLEARS
iTHE WORD AT THE ADDRESS
iCONTAINED IN THE SPECIFIED
iREGISTER AND INCREMENTS
iTHAT REGISTER'S CONTENTS
iBY TWO.
NOTE

Certain special instruction/address mode
combinations, which are rarely or never
used, do not operate exactly the same on
all
PDP-II processors, as described
below.
In the autoincrement mode, both the JMP
and JSR instructions auto increment the
register
before
its
use
on
the
PDP-ll/40, but not on the PDP-ll/45 or
11/10.
In double operand instructions having
the
addressing
form
Rn,(Rn)+
or
Rn,-(Rn),
where
the
source
and
destination registers are the same, the
source operand is evaluated as
the
auto incremented
or
autodecremented
value, but the destination register, at
the time it is used, still contains the
originally-intended effective address.
In the following example, as executed on

5-2

after

ADDRESSING MODES
the PDP-ll/40, Register
contains 100(8):
MOV

RO, (RO) +

MOV

RO,-(RO)

originally·

iTHE QUANTITY 102 IS MOVED
iTO LOCATION 100.
iTHE QUANTITY 76 IS MOVED
iTO LOCATION 100.

should
be
The use of these forms
avoided,
since they are not compatible
family
of
PDP-II
with the entire
processors.
An error code
(Z)
is printed in the
assembly listing with each instruction
which ~s not compatible
among
all
members of the PDP-II family.

5.4

AUTOINCREMENT DEFERRED MODE

The register (ER) contains a pointer to the address of the operand.
The contents of the register are incremented after being used as a
pointer.
Format for A:

@(ER)+

Example:
CLR

5.5

@(R3)+

iTHE CONTENTS OF REGISTER 3 POINT
iTO THE ADDRESS OF A WORD TO BE
iCLEARED BEFORE THE CONTENTS OF THE
iREGISTER ARE INCREMENTED BY TWO.

AUTODECREMENT MODE

The contents of the register (ER) are decremented before being used as
the address of the operand (see Note above in Section 5.3).
Format for A:

-(ER)

Examples:

5.6

CLR

-(RO)

CLR
CLR

-(R3)
-(R2)

iDECREMENT THE CONTENTS OF THE SPECIiFIED REGISTER (0, 3, OR 2) BY TWO
iBEFORE USING ITS CONTENTS
iAS THE ADDRESS OF THE WORD TO BE
iCLEARED.

AUTODECREMENT DEFERRED MODE

The contents of the register (ER) are decremented before being used as
a pointer to the address of the operand.

5-3

ADDRESSING MODES
Format for A:

@-(ER)

Example:
CLR

5.7

@-(R2)

iDECREMENT THE CONTENTS OF
iREGISTER 2 BY TWO BEFORE
iUSING ITS CONTENTS AS A POINTER
iTO THE ADDRESS OF THE WORD TO BE
iCLEARED.

INDEX MODE

The value of an expression (E) is stored as the second or third word
of
the instruction.
The effective address of the operand is
calculated as the value of E, plus the contents of register ER.
The
value E is the offset of the instruction, and the contents of register
ER form the base.
Format for A:

E(ER)

Examples:

5.8

CLR

X+2(Rl}

MOV

RO,-2(R3}

iTHE EFFECTIVE ADDRESS OF THE WORD
iTO BE CLEARED IS X+2, PLUS THE
iCONTENTS OF REGISTER 1.
iTHE EFFECTIVE ADDRESS OF THE
iDESTINATION LOCATION IS -2, PLUS
iTHE CONTENTS OF REGISTER 3.

INDEX DEFERRED MODE

An expression (E), plus the contents of a register
(ER), yields a
pointer to the address of the operand. As in index mode above, the
value E is the offset of the instruction, and the contents of register
ER form the base.
Format for A:

@E(ER}

Example:
CLR

5.9

@114(R4}

iIF REGISTER 4 CONTAINS 100, THIS
iVALUE, PLUS THE OFFSET 114, YIELDS
iTHE POINTER 214. IF LOCATION 214
iCONTAINS THE ADDRESS 2000, LOCATION
i2000 WOULD BE CLEARED.

IMMEDIATE MODE

Immediate mode allows the operand itself
second or third word of the instruction.
autoincrement of the PC.
Format for A:

(E)
to be stored as the
This mode is assembled as an

Examples:
MOV
MOV

#lOO,RO
#X,RO

iMOVE THE VALUE 100 INTO REGISTER O.
iMOVE THE VALUE OF SYMBOL X INTO
iREGISTER O.
5-4

ADDRESSING MODES
The number sign
(i)
in the MACRO character set
has
special
significance as an addressing mode indicator. When this character
appears in the operand field,
as shown above,
it specifies the
immediate addressing mode, indicating to MACRO that the operand itself
immediately follows the instruction word.
The operation of this mode can be shown through
MOV ilOO,RO, which assembles as two words:
Location 20:

0 1 2 7 0 0

Location 22:

0 0 0 1 0 0

Location 24:

Next instruction

the

first

example,

Note that the source operand (the value 100) is assembled immediately
following the instruction word,
i.e., as the second word in the
instruction. Upon execution of the instruction, the processor fetches
the first word (MOV) and increments the PC by 2 so that it points to
location 22 (which contains the source operand).
After the next fetch and increment cycle, the source operand (100)
is
moved into register 0,
leaving the PC pointing to location 24 (the
next instruction).

5.10

ABSOLUTE MODE

Absolute mode is the equivalent of immediate mode deferred.
The
address expression @iE specifies an absolute address which is stored
as the second or third word of the instruction.
In other words,
the
value immediately following the instruction word is taken as the
absolute address of the operand. Absolute mode is assembled as an
autoincrement deferred of the PC.
Format for A:

@iE

Examples:
MOV

@#lOO,RO

CLR

@#X

iMOVE THE CONTENTS OF LOCATION 100
iINTO REGISTER RO.
iCLEAR THE CONTENTS OF THE LOCATION
iWHOSE ADDRESS IS SPECIFIED BY
iTHE SYMBOL X.

The operation of this mode can be shown through
MOV @ilOO,RO, which assembles as two words:
Location 20:

0 1 3 7 0 0

Location 22:

0 0 0 1 0 0

Location 24:

Next instruction

the

first

example,

Note that the absolute address 100 is assembled immediately following
the instruction word,
i.e., as the second word in the instruction.
Upon execution of the instruction, the processor fetches the first
word (MOV) and increments the PC by 2 so that it points to location 22
(which contains the absolute address of the source operand).
After
the next fetch and increment cycle, the contents of absolute address
100 (the source operand) are moved into register 0, leaving the PC
pointing to location 24 (the next instruction).

5-5

ADDRESSING MODES
5.11

RELATIVE MODE

Relative mode is the normal mode for memory references within your
program.
It is assembled as index mode, using the PC as the index
register.
Format for A:

Examples:
CLR

100

MOV

RO,Y

iCLEAR LOCATION 100, RELATIVE TO
iTHE CONTENTS OF THE PC.
iMOVE THE CONTENTS OF REGISTER a
iTO LOCATION Y, RELATIVE TO THE
iCONTENTS OF THE PC.

In relative mode, the offset for the address calculation is assembled
as the second or third word of the instruction. This value is added
to the contents of the PC (the base register) to yield the address of
the source operand.
~he operation of
relative mode can be
MOV 100,R3, which assembles as two words:

Location 20:

0 1 6 7 0 3

Location 22:

0 0 a 0 5 4

Location 24:

Next instruction

shown

with

the

statement

Note that the constant 54 is assembled immediately following the
instruction word, i.e., as the second word in the instruction. Upon
execution of the instruction, the processor fetches the first word
(MOV)
and increments the PC by 2 so that it points to location 22
(containing the value 54). After the next fetch and increment cycle,
the processor calculates the effective address of the source operand
by taking the contents of location 22 (the offset) and adding it to
the current value of the PC, which now points to location 24 (the next
instruction). Thus, the source operand address is the result of the
calculation
OFFSET+PC = 54+24 = 100(8),
causing the contents of
location 100 to be moved into register 3.
Since MACRO considers the contents of the current location counter (.)
as the address of the first word of the instruction, an equivalent
index mode statement is shown below:
MOV

100-.-4(PC),R3

This instruction has a relative addressing mode because the operand
address is calculated relative to the current value of the location
counter. The offset is the distance (in bytes) between the operand
and the current value of the location counter.

5.12 RELATIVE DEFERRED MODE
The relative deferred mode is similar in operation to the relative
mode above, except that the expression E is used as a pointer to the
address of the operand. In other words, the operand following the
instruction word is added to the contents of the PC to yield a pointer
to the address of the operand.
Format for A:

5-6

ADDRESSING MODES
Example:
MOV

5.13

@X,RO

iRELATIVE TO THE CURRENT VALUE OF
iTHE PC, MOVE THE CONTENTS OF THE
iLOCATION WHOSE ADDRESS IS POINTED
iTO BY LOCATION X INTO REGISTER O.

SUMMARY OF ADDRESSING FORMS

Each PDP-II instruction takes at least one word. Operands of the form
listed below do not increase the length of an instruction.
Meaning

Form

R
(ER)

@R or

(ER) +

Autoincrement mode

@(ER)+

Autoincrement deferred mode

- (ER)

Autodecrement mode

@- (ER)

Autodecrement deferred mode

Operands of the following forms add one word to the instruction length
for each occurrence of an operand of that form:
Form

Meaning

E (ER)

Index mode

@E (ER)

Index deferred mode

Immediate mode

@#E

Absolute mode

Relative mode

Relative deferred mode

(see Note below)

The syntax of the addressing modes is
Additional
discussion
of addressing
applicable PDP-II Processor Handbook.

summarized in Appendix B.
modes is provided in the

NOTE
An alternate form for
@R is
(ER).
However,
the
form
@(ER)
is only
logically, but not physically equivalent
to
the
expression
@O(ER).
The
addressing form @#E differs from form E
in that the second or third word of the
instruction
contains
the
absolute
address of the operand, rather than the
relative distance between the operand
and the PC. Thus, the instruction CLR
@#100 clears absolute location 100, even
if the instruction is moved from the

5-7

ADDRESSING MODES
point at which it was assembled.
See
the
description
of the .ENABL AMA
function in Section 6.2, which causes
all
relative
mode addresses to be
assembled as absolute mode addresses.

5.14

BRANCH INSTRUCTION ADDRESSING

The branch instructions are I-word instructions. The high-order byte
contains the operator, and the low-order byte contains an 8-bit signed
offset (seven bits, plus sign), which specifies the branch address
relative to the current value of the PC. The hardware calculates the
branch address as follows:
1.

Extends the sign of the offset through bits 8-15.

Multiplies the result by 2, creating
than a word offset.

Adds the result to the current value of the PC
effective branch address.

byte

offset
to

rather

form

the

MACRO performs the reverse operation to form the word offset from the
specified address.
Remember that when the offset is added to the
current value of the PC, the PC is pointing to the word following the
branch
instruction;
hence,
the
factor -2 in the following
calculation:
Word offset

= (E-PC)/2 truncated to eight bits.

Since the.value of the PC

= .+2, we have:

Word offset = (E-.-2)/2 truncated to eight bits.
In using branch instructions, you must
following error conditions:

exercise

avoid

the

Branching from one program section to another;

Branching to a location
(global) symbol; or

Specifying a branch address that is out of range,
i.e., the
branch offset is a value that does not lie within the range
-128(10) to +127(10).

that

defined

The above conditions cause an error code (A) to be
assembly listing for the statement in error.

5.15

care

generated

external

the

USING TRAP INSTRUCTIONS

The EMT and TRAP instructions do not use the low-order byte of the
instruction word, allowing information to be transferred to the trap
handlers in the low-order byte.
If the EMT or TRAP instruction is
followed by an expression, the value of the expression is stored in
the low-order byte of the word. However, if the expression is greater
than 377(8), it is truncated to eight bits and an error code (T) is
generated in the assembly listing.

5-8

ADDRESSING MODES

PART III
MACRO DIRECTIVES

Chapters 6 and 7 describe all the directives used with MACRO.
Directives
are statements that cause MACRO to perform certain
operations during assembly. Chapter 6 describes several types of
directives,
including those which control symbol interpretation,
listing header material, program sections, data storage formats,
and
assembly listings.
Chapter 7 describes those directives concerning
macros, macro arguments, and repetitive coding sequences.
MACRO directives can be preceded by a label
(subject to
any
restrictions associated with specific directives) and followed by a
comment. A MACRO directive occupies the operator field of a source
statement.
Only one directive can be included in any given source
line. The operand field may be occupied by one or more operands or
left blank; legal operands differ with each directive specified.

CHAPTER 6
GENERAL ASSEMBLER DIRECTIVES

This category of directives includes:
1.

Listing control

Function control

Data storage

Radix and numeric control

Location counter control

Terminators

Program boUndaries

Program sectioning

Symbol control

10.

Conditional assembly

11.

PAL-llR conditional assembly.

Each is described in its own section of this chapter.

6.1

LISTING CONTROL DIRECTIVES

Listing control directives control the content, format, and pagination
of all line printer and teleprinter listing output generated during
assembly. Facilities also exist for creating object module names and
other identification information in the listing output.

6.1.1

.LIST and .NLIST Directives

Listing control options can be specified in the text of a MACRO
program through the .LIST and .NLIST directives. These directives are
of the form:
.LIST
.LIST
.NLIST
.NLIST
where:

arg

arg
arg
represents one or more of the optional
arguments defined in Table 6-1.
6-1

symbolic

GENERAL ASSEMBLER DIRECTIVES
As indicated above, the listing control directives may be used without
arguments,
in which case the listing directives alter the listing
level count. The listing level count is initialized to zero. At each
occurrence
of
a
.LIST directive,
the listing level count is
incrementedi
at each occurrence of an .NLIST directive,
the listing
level count is decremented. When the listing level count is negative,
the listing is suppressed
(unless the line contains an error).
Conversely, when the listing level count is greater than zero, the
listing is always generated.
Finally, when the count is zero,
the
line is either listed or suppressed, contingent upon the other listing
controls currently in effect for
the program.
For example,
the
following macro definition employs the .LIST and .NLIST directives to
selectively list portions of the macro body when the macro
is
expanded:
• MACRO LTEST
A-THIS LINE SHOULD LIST
.NLIST
B-THIS LINE SHOULD NOT LIST
.NLIST
C-THIS LINE SHOULD NOT LIST
• LIST
D-THIS LINE SHOULD NOT LIST
.LIST
E-THIS LINE SHOULD LIST
.ENDM

• LIST
ME
LTEST
A-THIS LINE SHOULD LIST
E-THIS LINE SHOULD LIST

iLIST TEST
iLISTING LEVEL COUNT IS o.
iLISTING LEVEL COUNT IS -1.
iLISTING LEVEL COUNT IS -2.
iLISTING LEVEL COUNT IS -1 .
iLISTING LEVEL COUNT IS o.
iLISTING LEVEL COUNT IS BACK TO O.

iLIST MACRO EXPANSION •

iCALL THE MACRO
iLISTING LEVEL COUNT IS O.
iLISTING LEVEL COUNT IS BACK TO O.

An important purpose of the level count is to allow macro expansions
to be listed selectively and yet exit with the listing level count
restored to the value existing prior to the macro call.
When used with arguments, the listing directives do not alter the
listing level count;
however, the .LIST and .NLIST directives can be
used to override current listing control,
as shown in the example
below:
.MACRO

.LIST

iLIST NEXT LINE.

.NLIST

iDa NOT LIST REMAINDER OF MACRO

X=.
iEXPANSION •
• ENDM
• NLIST
XX

iDO NOT LIST MACRO EXPANSIONS •

X=.
The symbolic arguments allowed for use with the listing directives are
described in Table 6-1.
These arguments can be used singly or in
combination with each other.
If multiple arguments are specified in a
listing directive, each argument must be separated by a comma, tab, or
space. For any argument not specifically included in a listing

6-2

GENERAL ASSEMBLER DIRECTIVES

control statement, the associated default assumption (List or No list)
is applicable throughout the source program. The default assumptions
for the listing control directives also appear in Table 6-1.
Table 6-1
Symbolic Arguments of Listing Control Directives
Argument

Default

Function

SEQ*

List

Controls the listing of source line
sequence
numbers.
MACRO
assigns
sequence number 1 to the first source
line in a file, and increments the
sequence number for each additional line
in
the
file.
If
this field is
suppressed
through
an
.NLIST
SEQ
directive,
MACRO
generates
a tab,
effectively allocating space for the
field, but fills the field with blanks.
Thus, the inter-positional relationships
of subsequent fields in the listing
remain undisturbed. During the assembly
process, MACRO examines each source line
for possible error conditions. For any
line in error, an appropriate error flag
is printed preceding the line sequence
number field
(see Appendix D). MACRO
does not assign sequence numbers for
files that have had sequence numbers
assigned by other programs, such as an
editor.

LOC*

List

Controls the listing of the current
location counter field. Normally, this
field is not suppressed. However, if it
is suppressed through the .NLIST LOC
directive, MACRO does not generate a
tab, nor does it allocate space for the
field, as is the case with the source
line
sequence
number
field
(SEQ)
described above. Thus, the suppression
of the current location counter (LOC)
field effectively left-justifies
all
subsequent
fields
(while preserving
inter-positional relationships) to that
position otherwise normally occupied by
this field.

BIN*

List

Controls the listing of generated binary
code.
If this field is suppressed
through
an
.NLIST
BIN
directive,
left-justification of the source code
field occurs
in
the
same
manner
described above for the current location
counter (LOC) field.
(continued on next page)

* If the .NLIST arguments SEQ, LOC, BIN, and SRC are in effect at the
same time, i.e., if all four significant fields in the listing are to
be suppressed, the printing of the resulting blank line is inhibited.

6-3

GENERAL ASSEMBLER DIRECTIVES
Table 6-1 (Cont.)
Symbolic Arguments of Listing Control Directives
Argument

Default

Function

BEX

List

Contr'ols
the
listing
of
binary
extensions,
i.e., the locations and
binary contents beyond those that will
fit on the source statement line. This
is a subset of the BIN argument.

SRC*

List

Controls the listing of source lines.

COM

List

Controls the listing of comments.
This
is a subset of the SRC argument. The
.NLIST COM directive reduces listing
time and space when comments are not
desired.

List

Controls
the
listing
of
macro
definitions and repeat range expansions.

List

Controls the listing of macro calls
repeat range expansions.

No list

Controls
the
expansions.

MEB

No list

Controls the listing of macro expansion
binary code.
A .LIST MEB directive
causes only
those
macro
expansion
statements that generate binary code to
be listed. This is a subset of the ME
argument.

CND

List

Controls the listing of
unsatisfied
conditional coding and associated .IF
and .ENDC directives in the
source
program.
This
argument
permits
conditional assemblies to be
listed
without
including
unsatisfied
conditional coding.

No list

Controls the listing of all listing
directives having no arguments, i.e.,
those listing directives that alter the
listing level count.

TOC

List

Controls the listing of the table of
contents during assembly pass 1 (see
Section 6.1.4 describing the
.SBTTL
directive).
This argument does not
affect the printing of the full assembly
listing during assembly pass 2.

listing

and
macro

(continued on next page)
* If the .NLIST arguments SEQ, LOC, BIN, and SRC are in effect at the
same time, i.e., if all four significant fields in the listing are to
be suppressed, the printing of the resulting blank line is inhibited.

6-4

GENERAL ASSEMBLER DIRECTIVES
Table 6-1 (Cont.)
Symbolic Arguments of Listing Control Directives
Argument

Default

Function

SYM

List

Controls the listing of the symbol table
resulting
from the assembly of the
source program.

TTM

List

Controls the listing output format.
The
default
be set by the system
can
manager. If the system manager does not
set a default, it is set to line printer
format.
Figure 6-1 illustrates the line
output
printer
format.
Figure 6-2
illustrates
the
teleprinter
output
format.

An example of an assembly listing, as sent to a l32-column line
printer, is shown in Figure 6-1. Note that binary extensions for
statements generating more than one word are formatted horizontally on
the source line.
An example of an assembly listing, as sent to a teleprinter
(in the
same format as for an 80-column line printer), is shown in Figure 6-2.
Notice that binary extensions for statements generating more than one
word are printed on subsequent lines. There is no explicit truncation
of output to 80 characters by the assembler.
Any argument specified in a .LIST/.NLIST directive other than those
listed in Table 6-1 causes the directive to be flagged with an error
code CA) in the assembly listing.
The listing control options can also be specified at assembly time
through switches included in the command string to MACRO (see Chapter
8). The use of these switches overrides all corresponding listing
control (.LIST or .NLIST) directives specified in the source program.

6-5

eSITST _. TEST OF CSI1 AND CSI2 MACRO M0707
READ AND PARSE COMMA~O LINES

I
~

209
210
211 001230
212 01211244
213 001246
214 001254
:Z15 01211300
216 001324
217 001326
218 001332
219 001336
220 001360
221 001404
222 01111412
223 001416
224 001424
225 001450
226 01211474
227
228 001416
229 001!)02
230 001504
231 0~1512
232 O01520
233 001522
234 001526
23!5 001552
236 0~1!)76
237 O01602
238 001620
239 001622
240 O01626
241 001634
242 001636
243 001662
244 001666
245 001704
246 01111706
247 0/i!1712
248 001120
249 001722

09-JUL-74 15147

GETLN,
103003
lSI
103064
016046
166A16
066060
162660
166060

000020
000004
000016
000002
00001 b

000020
000002

00065!)
005760
001652
112767
132760
001402
105267

251

000~02

000060
Pl00040

176432
000001

176416

103441
132760
00136O

,,"'0020

000001

000020

000001

103407
132760
001360
000412

Figure 6-1

PAGE 5

,saTTL

READ AND PARSE

GCMLS
BCC
EXITSS
TYPE
CSISI
BCC
MOV
SUB
TYPE
TYPE
ADD
SUB
SUB
TVPE
TYPEM
BR

'GCLBLI<

TST
BEQ
MOvB
BITB
BEQ
INCS
TYPEM
1051
TYPE'"
OPARSEI CALL
CSIS2
Bes
CALL
BITS
BNE
TYPEM
IPARSEI CALL
CS152
BCS
CALL
SITB
BNE
SR

COMMA~D

LINES

'GET LINE VIA GCML
'SKIP IF NO ERROR
HLSt, UIT
G,CMLD.2(R0),G,CMLD(R0)"!0
'SEND OUT TME INPUT LINE
'CSIBLK,GCLBLK.G.CMLO+2,GCLBLK.G.CMLO
25
'SRANCH I~ NO ERROR DETECTED
C,FILD.2(R0),~(SP)
'PUT STRING ERROR ADDR IN STK
C.CMLD+2(~0),(SP)
'CALCULATE LENGTM OF FIRST PART
C.CMLD+2CR0),(SP),,'S
'SEND OUT FIRST PART O~ STRING
C.FILD+2(R0),C,FILD(H0),.'5
,SEND OUT SECOND PART
C.FILO(R0),C,FILO.2(H0) 'CALC AOOR OF LAST PART OF STRING
(SP)+,C,CMI.O(R0)
'DEDUCT LENGTM OF FIRST PART
C.FILD(A0),C,C~LD(R0)
'CALC LENGTH OF LAST PAAT
C.FILO+2CR0),C,CMLDCH0), •• 0
'SEND OUT LAST PART
STX,40
'SEND SYNTAX ERROR MESSAGE
GETLN
,un- FOR MORE
11

C,CMI.O(R0)
'CHECK LENGTM OF LINE
GETLN
,IF NULL, SKIP S.CK FOR ~ExT LINE
.'0,EQUBIT
'ASSUME EQUAL SIGN NOT FOUNO
'CS,EQU,C,STAT(R0) 'C~ECK STATUS
US
'SKIP If EQUAL SIGN NOT S~EN
[QUBIl
'ELSE, INDICATE EQUAL SIGN FOUND
EQU,40
,SEND EQUAL SIGN STATUS MESSAGE
OPT,40
'SEND OUTPUT SCAN MESSAGE
INIT2
'INIT LOCNS FOR CSI2 CALL/TEST
,OUTPUT,*SwTBL 'PARSe OUTPUT SPEC
CS2ERR
'SKIP ON ERROR
EVALU8
'EVALUATE RESULTS OF SEMANTIC PARSE
'CS.MD~,C,STAT(R0)
'ADDITIONAL OUTPUT SPECS?
OPARSE
'YES, CONTINUE wITH OUTPUT SCAN
IPT,40
'SEND INPUT SCAN MESSAGE
INIT2
IINIT LOC~S FOR CSI2 CALL/TEST
,INPUT"swTSL
'PARSE INPUT SPEC
CS2EHR
'SKIP ON ERROR
EVALU8
'EVALUATE RESULTS OF SEMANTIC PARSE
'CS,MOR,C,STAT(R0) 'ADOITIO~AL INPUT SPECS?
IPARSE
'YES, CONTINUE wITH INPUT SCAN
JMPGET
'GET ANOTHER COMMAND LINE

Example of Line Printer Assembly Listing

(j)

tr:l

1:'"1

)II

til
til
tr:l

3:
Il1

1:'"1
tr:l

!:tI
tj
t-I
~

tr:l

(')

t-3

t-I

t;j
til

GENERAL ASSEMBLER DIRECTIVES

CSIT5T •• TEST OF CSIl AND C5I2
READ AND PARSE COMMAND LINES
209
210
211 001230
212 001244
213 001246
214 001254
215 001300
216 001324
217 001326
218 001332
219 001336
220 001360
221 001404
222 001412
223 001416
224 001424
225 0014M'J
226 1/101474
221
228 01111476
229 0015P12
230 001504
231 001512
232 001520
233 001!522

GETLIIII
103003
lSI

103064
016046
1/100020
166016
000004

M0101

09~JUL.74

15159

PAGE 5

• SBTTL

READ AND PARSE COMMAND LIIIIES

GCMLS
Bec
Exnss
TyPE
CSISl
ace
MO¥

'GCLBLK
15

,GET LINE VIA GCML
,SKIP IF NO ERROR
,ELSE, EXIT
G.CMLD.2(R0),G.CMLD(R0),"0
'SEND OUT THE INPUT LINf
.CSIBLK,GCLBLK.G.CMLD.2,GCLBLK.G.CMLD
2S
'BRANCH IF NO ERROR DETECTED
C.FILD.2(R0),-(SP)
,PUT STRING ERROR ADDR IN STK

SUB
TYPE
TYPE
ADD

066060
000016
000020
162660
000002
166061/1
000016
000002

'CALCULATE LENGTH OF

251

1~5261

FI~ST

'DEDUCT LENGTH OF FIRST PART

SUB

C.FILD(R0),C.C~LD(R0)

ICALC LENGTH OF LAST PART

TYPE
TYP[M
BH

C.FILO.2(R0),C.CMLD(R0),'40
'SE~D OUT LAST PART
STX,40
,SEND SYNTAX ERROR MESSAGE
GETLN
,TRY FOW MORE

TST

C.CMLD(R0)

,CHECK LENGTM OF lINE

BEQ
MOVB

GETL"~

.'0,EQUBIT

,IF NUL~, SKIP BACK FOR NExT
,ASSUME EQUAL SIGN NOT FOU~D

BIT8

.CS,EQU,C.STAT(R0)

BEQ
INC8

105
EQUe IT

'SKIP IF EQUAL SIGN NOT SEEN
'ELSE, I~DICATE EQUAL SIG~ FOUND

EQU,40
OPT,40
INIT2
,OUTPUT,.SwTBL
eS2E"RR
EVALU8

,SEND EQUAL SIGN STATUS MESSAGE
,SEND OUTPUT SCAN MESSAGE
,INIT LOCNS FOR CsI2 CALL/TEST
,PARSE OUTPUT SPEC
'SKIP ON ERROR
,EVALUATE RESULTS OF SEMANTIC PARSE

~INE

'CHECK STATuS

176416
234 0A1526
235 001552
236 001576
237 0~1602
238 001620
239 01211622
240 01/11626
241 0\H6:H
242 001636
243 001662
244 001666
245 (IHH704
246 001106
241 001712
248 001120

1051

103441

TYPEM
TYPEM
OPARSEt CALL
CSU2
BCS
CALL

132160
01110020
000001
001360

BITB

'CS.MOR,C.STAT(H0)

BNE
TYPEM

OPARSE

'ADDITIONA~

OUTPUT SPECS?

132760
000020
000001
001360

BITB

,YES, CONTINUE wIT~ OUTPUT SCAN
,SEND INPUT SCAN MESSAGE
INIT2
,INIT LOCNS FOR CSI2 CAL~/TEST
,INPUT,.SwTBL
'PARSE INPUT SPEC
CS2ERR
'S~IP ON ERROR
EVALU8
'EVALUAT~ RESULTS OF SEMANTIC PARSE
.CS,"OR,C.STAT(R0) 'ADDITIONAL INPUT SPECS1

BNE

IPARSE

Figure 6-2

Example of Terminal Assembly Listing

IPARSEI
103401

CAL~

CSIS2
BCS
CALL

PART

C.CMLD.2(R0),(SP),.'S
'SEND OUT FIRST PART OF STRING
C.FILD.2(R0),C.FILD(RS),.'S
,SEND OUT SECO~D PART
C.FILD(R0),C.FILD.2(H0) 'CALC ADDR OF ~AST PART OF STRI~G

SUB

000655
005761/1
000002
001652
112167
000060
176432
132760
000040
000001
001402

MAC~O

IPT,4~

IYES, CONTINUE wITM INPUT SCAN

6-7

GENERAL ASSEMBLER DIRECTIVES
Figure 6-3 shows a listing, produced in line printer
format,
reflecting the use of the .LIST and .NLIST directives in the source
program and the effects such directives have on the assembly listing
output.

6.1.2

Page Headings

MACRO prints each assembly page in the format shown in either Figure
6-1 or Figure 6-2, depending on the listing mode (see TTM, Table 6-1).
On the first line of each page, MACRO prints the following (from left
to right):
1.

Title of the object module, as established through the .TITLE
directive (see next section).

Assembler version identification.

Date.

Time-of-day.

Page number.

The second line of each assembly listing page contains the subtitle
text specified in the last-encountered .SBTTL directive (see Section
6.1.4).

6-8

27
28
29 000062
1/)1/)0062
000070

COM
IS T COM
,1II0RD
1,2,3,4,5

~STMAC
,N~

011J111001
00011J04

001/) 0f/.I 2
130001/15

1/)00003

30
31
32 000074

011J1B074

1/)00001

000002

000003

,~IST

COM

L!TMAC
, NLlST
,!IIORD
.I.IST

cCOM,8EX>
COM,BEX

'COMMENT LINES TEST

'COMMENT LINES AND EXTENDED BINARY TEST

1,2,3,4,5

COM,8EX

en
ttl

2:
trJ

t"4

,MAIN.

MACRO 1040707

e9 .. JUI. .. 74 16129

)II
til
til

PAGE 1-1

ttl

3:
0'\

I
~

tEl
t"4

33
34

trJ

36
3'
38 000106
100106
000110
000112
000114
I1JCI)IIJ 116

GJ0000t
000002
000003
000004
000005

,NARROW LISTING MODE IS IN EFFECT

L!TMAC
• NL 1ST
,wORD

SEQ

,SEQUENCE NUM8ERS T~ST

,THIS IS A COMMENT

SEQ

1,2,3,4,5

000001

SEQ

~STMAC

SEx
BEX

,NLIST
,WORD
,~IST

1,2,3,4,5

BEX

JEXTENDED 8INARY TEST
,THIS IS A COMMENT

000001 '

t-3

til
.~IST

<:
trJ

000120

t:1

41 000120

,~IST

TTM

,END

Figure 6-3 Listing Produced With Listing Control Directives

,MAIt\I.

MACRO M0'0'

09-JUL-14 16.29

PAGE 1
,NLIST
.LIST

1
2

'WIDE LISTING MODE IS IN EFFECT
.LIST MACHO EXPANSIONS

TT'"
ME

4
LISTI~G

,MACRO
.NLIST
.wORD
• LI ST
.[NOM

8
9
10

11
12
13
14
15
16
17 0U012
000001
1/100004
0'\

I-'
0

CONTROL TEST MACRO
LSTMAC ARG
ARG
1,2,3,4,5
AHG

'TMIS IS A COMMENT
Ci)
~

000002
000005

• NLI S T
.wORD

000003

LSTMAC LOC
Loe
1,2,3,4,5
.L.IST

.LOCATION COUNTER TEST

~
I:"'t

.TMIS IS A COMMENT

til
til
~

L.OC

I:"'t

20 000024
,NLIST
,!IIORD

0f1l0024

BIN
1,2,3,4,5

LSTMAC

BIN

'GE~ERATED

BINARY TEST

,TMIS IS A COMMENT
BIt\I
.UST

22
23 000036
000001

000002

000003

24
25
26 000050
0AB0S0
000056

t-f
~

000036

000001
000004

0"'0002
000005

LSTMAC
,NU S T
.wORO
.LIST

BEX

L.STMAC

SRC

.LIST

SRC

SEX

BEX
1,2,3,4,5

,EXTENDED BINARY TEST

t-f

,TMIS IS A COMMENT

til

'SOURCE L.INES TEST

0000O3

Figure 6-3 (Cont.) Listing Produced With Listing Control Directives

GENERAL ASSEMBLER DIRECTIVES
6.1.3

.TITLE Directive

The .TITLE directive is used to assign a name to the object module as
the first entry in the header of each page in the assembly listing.
The name so assigned is the first six non-blank characters following
the .TITLE directive. This name should be six Radix-50 characters or
less in length; any characters beyond the first six are checked for
ASCII legality, but they are not used as part of the object module
name. For example, the directive:
.TITLE

PROGRAM TO PERFORM DAILY ACCOUNTING

causes the assembled object module to be named PROGRA. Note that this
6-character name bears no relationship to the filename of the object
module, as specified in the command string to MACRO. The name of an
object module
(specified in the .TITLE directive) appears in the
Linker load map. This is also the module name which the Librarian
will recognize.
If the .TITLE directive is not specified, MACRO assigns the default
name .MAIN. to the object module. If more than one .TITLE directive
is specified in the source program, the last .TITLE directive
encountered establishes the name for the entire object module.
All spaces and/or tabs up to the first non-space/non-tab character
following the .TITLE directive are ignored by MACRO when evaluating
the text string.
If the .TITLE directive is specified without an object module name, or
if the first non-space/non-tab character in the object module name is
not a Radix-50 character, the directive is flagged with an error code
(A) in the assembly listing.
Section A.2 of Appendix A contains a table of Radix-50 characters.

6.1.4

.SBTTL Directive

The .SBTTL directive is used to produce a table of
contents
immediately preceding the· assembly listing and to further identify
each page in the listing. In the latter case, the text following the
.SBTTL directive is printed as the second line of the header of each
page in the listing, continuing until altered by a subsequent .SBTTL
directive in the program. For example, the directive:
.SBTTL

CONDITIONAL ASSEMBLIES

causes the text
CONDITIONAL ASSEMBLIES
to be printed as the
listing.

second

line

the

header

the

assembly

During assembly pass 1, a table of contents is printed for the
assembly listing, containing the line sequence number, the page
number, and the text accompanying each .SBTTL directive. The listing
of the table of contents is suppressed whenever an .NLIST TOC
directive is encountered in the source program (see Table 6-1).
An
example of a table of contents listing is shown in Figure 6-4.

6-11

GENERAL ASSEMBLER DIRECTIVES

CSITST •• TEST OF CSl1 AND CSI2 MACRO M0707
TABLE OF CONTENTS

MACRO DEFINITIONS
MESSAGE STRINGS
MISCELLANEOUS DATA
READ AND P'~SE COMMAND LINES
EVALUATE THE SEMANTIC ANALYSIS
SUBROUTINES

2- 55
3- 74

4-153
5·209
6-255

'-345

Figure 6-4

6.1.5

09.JUL.74 15,47

Assembly Listing Table of Contents

.IDENT Directive

The .IDENT directive provide~ an additional means of labeling the
object module produced by MACRO.
In addition to the name assigned to
the object module with the .TITLE directive
(see Section 6.1.3), a
character string up to six Radix-50 characters can be specified
between paired printing delimiters to label the object module with the
program version number. This directive takes the following form:
.IDENT
where:

/string/

string

represents six legal Radix-50 characters or less
which establish the program identification or
version number. This number is included in the
global symbol directory of the object module.

represent delimiting characters. These delimiters
may be any paired printing characters, other than
the equal sign (=), the left angle bracket «), or
the semicolon
(i),
as long as the delimiting
character is not contained within the text string
itself.
If the delimiting characters do not
match, or if an illegal delimiting character is
used,
the
.IDENT directive is flagged with an
error code (A) in the assembly listing.

An example of the .IDENT directive is shown below:
.IDENT

/V05A/

The character string V05A is converted to Radix-50 representation and
included in the global symbol directory of the object module. This
character string also appears in the Linker load map and the Librarian
directory listings.
When more than one .IDENT directive is encountered in a given program,
the last such directive encountered establishes the character string
which forms part of the object module identification.

6.1.6

.PAGE Directive/Page Ejection

Page ejection is accomplished in one of four ways:

After reaching a count of 58 lines in
automatically performs a page eject
perforations on line printer paper
teleprinter output into pages.
6-12

the listing, MACRO
to skip over page
and
to
formulate

GENERAL ASSEMBLER DIRECTIVES
2.

In addition, the .PAGE directive is used within the source
program to perform a page eject at desired points in the
listing. The format of this directive is:
• PAGE
This directive takes no arguments and causes a skip to the
top of the next page when encountered. It also causes the
page number to be incremented. The .PAGE directive does not
appear in the listing.
When used within a macro definition, the .PAGE directive is
ignored during the assembly 'of the macro definition. Rather,
the page eject operation is performed as the macro itself is
expanded. In this case, the page number is also incremented.

6.2

A page eject is performed when a form-feed character is
encountered.
If the form-feed character appears within a
macro definition, a page eject occurs during the assembly of
the macro definition, but not during the expansion of the
macro itself. A page eject resulting from the use of the
form-feed character likewise causes the page number to be
incremented.

Encountering a new source file causes the page number
incremented and the line sequence count to be reset.

FUNCTION DIRECTIVES:

.ENABL AND .DSABL

Several function control options are provided by MACRO through the
.ENABL and . DSABL directives.
These directives are included in a
source program to invoke or inhibit certain MACRO functions and
operations
incidental
to
the assembly process itself.
These
directives take the following form:
.ENABL
• DSABL
where:

arg

arg
arg
represents one or more of the optional
arguments defined in Table 6-2.

symbolic

Table 6-2
Symbolic Arguments of Function Control Directives
Argument

Default

Function

AMA

Disable

Enabling this
causes
all
function
relative addresses (address mode 67) to
be assembled as
absolute
addresses
(address mode 37) •
This function is
useful during the debugging phase of
program development.

"._

...•...

(continued on next page)

6-13

GENERAL ASSEMBLER DIRECTIVES
Table 6-2 (Cont.)
Symbolic Arguments of Function Control Directives
Argument

Default

Function

CDR

Disable

Enabling this function causes source
columns 73 and greater, i.e., to the end
of the line, to be treated as a comment.
The most common use of this feature is
to permit sequence numbers in
card
columns 73-80.

CRF

Enable

Disabling this function inhibits the
generation of cross-reference output.
This function only has
meaning
if
cross-reference
output generation is
specified in the command string.

FPT

Disable

Enabling this function causes floatingpoint
truncation;
disabling
this
function causes floating-point rounding.

Disable

Enabling this function causes MACRO to
accept lower-case ASCII input instead of
converting it to upper-case.
If this
function ~s not enabled, all text is
converted to upper-case.

LSB

Disable

This argument permits the enabling or
disabling
of a local symbol block.
Although a local
symbol
block
is
normally established by encountering a
new symbolic label or a .PSECT directive
in the source program, an .ENABL LSB
directive establishes a new local symbol
block which is not terminated until (1)
another .ENABL LSB is encountered, or
(2)
another symbolic label or .PSECT
directive is encountered following a
paired .DSABL LSB directive.
Although the
.ENABL
LSB
directive
permits a local symbol block to cross
.PSECT boundaries, local symbols cannot
be defined in a program section other
than the one that was in effect when the
block was entered. The basic function
of this directive
with
regard
to
.PSECT's is limited to those instances
where it is desirable to leave a program
section
temporarily
to store data,
followed by a return to the original
program section.
Attempts to define
local symbols in an alternate program
section are flagged with an error code
(P) in the assembly listing.
An example of the .ENABL LSB and
.DSABL
LSB directives, as typically used in a
source program, is shown in Figure 6-5.
(continued on next page)

6-14

GENERAL ASSEMBLER DIRECTIVES

Table 6-2 (Cont.)
Symbolic Arguments of Function Control Directives
Arugment

Default

Function

PNC

Enable

Disabling this function inhibits binary
output until an .ENABL PNC statement is
encountered within the same module.

REG

Enable

When specified, the .DSABL REG directive
inhibits
the
normal
MACRO default
register definitions;
if not disabled,
the default definitions .listed below
remain in effect.
RO=%O
Rl=%l
R2=%2
R3=%3
R4=%4
R5=%5
SP=%6
PC=%7
The .ENABL REG statement may be used as
the logical complement of the .DSABL REG
directive. The use of these directives,
however,
is
not
recommended.
For
logical consistency, use the
normal
default
register
definitions listed
above.

GBL

Enable

When the .ENABL
GBL
directive
is
specified,
MACRO
treats all symbol
references that are undefined at the end
of assembly pass 1 as default global
references;
when
the
• DSABL
GBL
directive is specified, MACRO treats all
such references as undefined symbols.
In assembly pass 2, if the .DSABL GBL
function is still in effect,
these
undefined symbols are flagged with an
error code (U) in the assembly listing;
otherwise, they continue to be regarded
by MACRO as global references.

Specifying any argument in an .ENABL/.DSABL directive other than those
listed in Table 6-2 causes that directive to be flagged with an error
code (A) in the assembly listing.

6-15

SQUEEZE MACRO M0707

m
I
.....,

272
273
274
275
276
277
278 003142
279 003144
280 ",,3146
281 003Hi0
282 003 HI2
283 003156
284 0QJ3160
285 0A3162
286
287 003164
288 003166
289 003170
290 003172
291 003176
292 003200
293 003204
294 003206
295 003210
296 003212
297 003214
298 0A3216
299 003220
300
301
302

09-JUL-74 15113

PAGE 4

,ENABI.
010103
060203
020301
001422
124327
001373
010302
000412
060102
020201
001412
124227
001773
121227
001770
005202
1&0102
000241
000401
000261

000073

FNOSMI I MOV
ADD
CMP
lSI
BEQ
CMPB
Bt.JE
MOV
BR

Rt,R3
R2,R3
HJ,H1
305
-(R3),*SEMIC

SKPBL.I< I ADO
CMP
105 I

R1,R2
R2,R1
305
.. (R2),UAB
U5

SEQ

000011
0~0040

Figure 6-5

L.SB

2051

CMPB
SEQ
CMP8
BEQ
INC
SUR
CLoC
SR

30$1
40$1

SEC
RETURN
,OSABL

R3,R2
205

(R2),*BLAN~

H2
R1, R2

'PUT AOOR OF LINE IN ~3
,POINT R3 PAST LAST CHAR IN LINE
'DOES RJ POINT TO START OF LINE?
IIF SO, LEAVE INDICATING FAILURE
'15 THE LAST CHARACTER SEMICO~ON?
INO, CONTINUE LOO~ING
IYES, POINT R2 PAST NEW ENO-OF-LINE
'LEAVE VIA COMMON SUCCESS CODE
IPOINT R2 PAST ENO-Of-LINE
100ES R2 POINT' TO SURT OF LINE?
'IF SO, LEAVE w!T~ FAILURE
lIS THE LAST CHARACTER A TAB?
I IF SO, IGNORE' IT
lIS IT A BLANK?
'IF SO, IGNORE IT
,NON-BLANK C~ARACTER--POINT PAST IT
,RE-COMPUTE LINE LENGTH
,INOICATE SUCCESS
'BRANC~ TO LEAV~
.INOICATE FAILURE

LSB

Example of .ENABL and .DSABL Directives

Cj')

tzl

2:
tzl

t"1

>
til

til
tzl

3:
c:g
t"1

tzl

!:t'

t-t

!:t'
tzl

t-3

t-t

tzl
til

GENERAL ASSEMBLER DIRECTIVES
6.3

DATA STORAGE DIRECTIVES

A wide range of data and data types can be generated with the
following directives, ASCII conversion characters, and radix-control
operators:
.BYTE
. WORD

.ASCII
.ASCIZ
.FLT2
.FLT4
.RAD50
AB
AC
AD
AF
AO
AR

These MACRO facilities are described in the following sections.

6.3.1

.BYTE Directive

The .BYTE directive is used to generate successive bytes
data in the object module. The directive is of the form:

binary

• BYTE

exp

iSTORES THE BINARY VALUE OF THE
iEXPRESSION "EXP" IN THE NEXT BYTE.

.BYTE

expl,exp2,expn

iSTORES THE BINARY VALUES OF THE LIST
iOF EXPRESSIONS IN SUCCESSIVE BYTES.

A legal expression must reduce to eight bits of data or less.
The
operands of a .BYTE directive are evaluated as word expressions before
b~ing truncated to the low-order eight bits.
The l6-bit value of the
specified expression must have a high-order byte (which is truncated)
that is either all zeros (0) or all ones (1). Each expression value
is
stored
in the next byte of the object module.
Multiple
expressions, which must be separated by commas, are stored in
successive bytes, as described below:
SAM=5
.=410
.BYTE

AD48,SAM

iTHE VALUE 060 (OCTAL EQUIVALENT OF 48
iDECIMAL) IS STORED IN LOCATION 410.
iTHE VALUE 005 IS STORED IN LOCATION
i4ll.

If the high-order byte of the expression reduces to a value other than
o or -1, the value is truncated to the low-order eight bits and
flagged with an error code (T) in the assembly listing.
The construction AD in the first operand of the .BYTE directive above
illustrates the use of a temporary radix-control operator. The
function of such special unary operators is described in Section
6.4.1.2.
At link time, it is likely that a relocatable expression will result
in a value having more than eight bits, in which case the Linker

6-17

GENERAL ASSEMBLER DIRECTIVES
issues a truncation diagnostic for the object module in question.
example, the following statements create such a possibility:
.BYTE

• BYTE

For

iSTORES OCTAL 23 IN NEXT BYTE.

A:
iRELOCATABLE VALUE A WILL PROBABLY

iCAUSE LINKER TRUNCATION
iDIAGNOSTIC.
If an expression following the
.BYTE directive
interpreted as a zero, as described below:
.=420
.BYTE

,,,

null,

iZEROS ARE STORED IN BYTES 420, 421,
i422, AND 423.

Note that in the above example, four bytes of storage result from the
.BYTE directive.
The three commas in the operand field represent an
implicit declaration of four null values, each separated from the
other by a comma. Hence, four bytes, each containing a value of zero
(0), are reserved in the object module.

6.3.2

.WORD Directive

The .WORD directive is used to generate successive words
the object module. The directive is of the form:

data

• WORD

exp

iSTORES THE BINARY EQUIVALENT OF THE
iEXPRESSION EXP IN THE NEXT WORD.

• WORD

expl,exp2,expn

iSTORES THE BINARY EQUIVALENTS OF THE
iLIST OF EXPRESSIONS IN SUCCESSIVE
iWORDS.

A legal expression must result in 16 bits of data or less.
Each
expression is stored in the next word of the object program. Multiple
expressions must be separated by commas and stored in successive
words, as shown in the following example:
SAL=O
.=500
• WORD

177535,.+4,SAL

iSTORES THE VALUES 177535, 506, AND

i 0 IN WORDS 500, 502, AND 504,
iRESPECTIVELY.

If an expression following the .WORD directive contains a null value,
it is interpreted as a zero, as shown in the following example:
.=500
• WORD

,5,

iSTORES THE VALUES 0, 5, AND 0 IN
iLOCATION 500, 502, AND 504,
iRESPECTIVELY.

A statement containing a blank operator field, i.e., a symbol that is
not recognized by MACRO as a macro call, an instruction nmemonic, a
MACRO directive, or a semicolon is interpreted during assembly as an
implicit .WORD directive, as shown in the example below:
.=440
LABEL:

100,LABEL

iSTORES THE VALUE 100 IN LOCATION 440
iAND THE VALUE 440 IN LOCATION 442.

6-18

GENERAL ASSEMBLER DIRECTIVES
CAUTION
You should not use this technique to
generate .WORD directives because it may
not be included
in
future
PDP-II
assemblers.

6.3.3

ASCII Conversion Characters

The single quote (I) and the double quote
(")
characters are unary
operators that can appear in any MACRO expression. When so used,
these characters cause a l6-bit expression value to be generated.
When the single quote is used, MACRO takes the next character in the
expression and converts it from its 7-bit ASCII value to a l6-bit
expression value. The 16-bit value is then used as an absolute term
within the expression. For example, the statement:
MOV

#'A,RO

results in the following 16-bit
register 0:

expression

value

being

moved

into

Binary Value of ASCII A
Thus, i~ the example above, the expression 'A results in a value of
101(8).
Note that the high-order byte is always zero (0) in the
resulting expression value when the single quote unary operator is
used.
The 1 character must not be followed by a
RUBOUT, line-feed, or form-feed character;
(A) is generated in the assembly listing.

carriage-return, null,
if it is, an error code

When the double quote is used, MACRO takes the next two characters in
the expression and converts them to a l6-bit binary expression value
from their 7-bit ASCII values. This 16-bit value is then used as an
absolute term within the expression. For example, the statement:
MOV

#"AB,RO

results in the following l6-bit
register 0:

expression

value

being

moved

into

Binary Value of ASCII A
Binary Value of ASCII B
Thus, in the example above, the expression "AB results in a
041101 (8) •

6-19

value

GENERAL ASSEMBLER DIRECTIVES
The" character also must not be followed by a carriage-return, null,
RUBOUT, line-feed, or form-feed character;
if it is, an error code
(A) is likewise generated in the assembly listing.
The ASCII character set is listed in Section A.l, Appendix A.

6.3.4

.ASCII Directive

The .ASCII directive translates character strings into their 7-bit
ASCII equivalents and stores them in the object module. The format of
the .ASCII directive is as follows:
.ASCII
where:

/string l/ •.. /string n/

string

is a string of printable ASCII characters. All
printable ASCII characters
are
legal.
The
vertical-tab, null, line-feed, RUBOUT,
and all
other non-printable ASCII
characters,
except
carriage-return
and
form~feed,
are
illegal
characters.
Such
an
illegal
non-printing
character is flagged with an error code (I) in the
assembly
listing.
The
carriage-return
and
form-feed characters terminate the scan of the
source line. This premature termination of the
.ASCII statement results in the generation of an
error code (A) in the assembly listing, because
MACRO is unable to complete the scan of the
matching delimiter at the end of the character
string.

represent delimiting characters. These delimiters
may be any paired printing characters, other than
the equal sign (=), the left angle bracket «), or
the semicolon
(;), as long as the delimiting
character is not contained within the text string
itself.
If the delimiting characters do not
match, or if an illegal delimiting character is
used, the .ASCII directive is flagged with an
error code (A) in the assembly listing.

A non-printing character can be expressed in an .ASCII statement
by enclosing its equivalent octal value within angle brackets.
set of angle brackets so used represents a single character.
example, in the following statement:
.ASCII

only
Each
For

the expressions <15>, <A+2>, <S>, and <4> represent the values of
non-printing characters. Furthermore, the expressions must reduce to
eight bits of absolute data or less, subject to the same rules for
generating data as with the .BYTE directive (see Section 6.3.1).
Angle brackets can be embedded between delimiting characters in the
character string, but angle brackets so used do not take on their
usual significance as delimiters for non-printing characters.
For
example, the statement:
.ASCII

/ABC<expression>DEF/

6-20

GENERAL ASSEMBLER DIRECTIVES

contains a single ASCII character string, and performs no evaluation
of the embedded, bracketed expression. This use of the angle brackets
is shown in the third example of the .ASCII directive below:

iSTORES THE BINARY REPRESENTATION
iOF THE LETTERS HELLO IN FIVE
iCONSECUTIVE BYTES •

.ASCII

/HELLO/

• ASCII

/ABC/<lS><12>/DEF/ iSTORES THE BINARY REPRESENTATION
iOF THE CHARACTERS A,B,C,CARRIAGE
iRETURN,LINE FEED,D,E,F IN EIGHT
iCONSECUTIVE BYTES .

• ASCII

/A<lS>B/

iSTORES THE BINARY REPRESENTATION
iOF THE CHARACTERS A, <, 1, S, >,
iAND B IN SIX CONSECUTIVE BYTES.

The semicolon (i) and equal sign
(=)
can be used as delimiting
characters in an ASCII string, but care must be exercised in so doing
because of their significance as a comment indicator and assignment
operator, respectively, as illustrated in the examples below:
.ASCII

iABCi/DEF/

iSTORES THE BINARY REPRESENTATION OF
iTHE CHARACTERS A, B, C, D, E, AND F
iIN SIX CONSECUTIVE BYTESi NOT
iRECOMMENDED PRACTICE •

• ASCII

/ABC/iDEFi

iSTORES THE BINARY REPRES~NTATIONS OF
iTHE CHARACTERS A, B, AND C IN THREE
iCONSECUTIVE BYTESi THE CHARACTERS D,
iE, F, AND: ARE TREATED AS A COMMENT •

• ASCII

/ABC/=DEF=

iSTORES THE BINARY REPRESENTATION
:OF THE CHARACTERS A, B, C, D, E, AND
:F IN SIX CONSECUTIVE BYTES: NOT
:RECOMMENDED PRACTICE.

An equal sign is treated as an assignment operator when it appears as
the first character in the ASCII string, as illustrated by the
following example:
.ASCII

6.3.S

=DEF=

:THE DIRECT ASSIGNMENT OPERATION
:.ASCII=DEF IS PERFORMED, AND A Q
: (SYNTAX) ERROR IS GENERATED UPON
:ENCOUNTERING THE SECOND = SIGN.

.ASCIZ Directive

The .ASCIZ directive is equivalent to the .ASCII directive described
above, except that a zero byte is automatically inserted as the final
character of the string. Thus, when a list or text string has been
created with an .ASCIZ directive, a search for the null character in
the last byte can effectively determine the end of the string, as
reflected by the coding below:
CR=lS
LF=12
HELLO:

.ASCIZ
.EVEN

<CR><LF>/MACRO VOIA/<CR><LF>

6-21

:INTRODUCTORY MESSAGE

GENERAL ASSEMBLER DIRECTIVES

10$:

MOV
MOV
MOVB
BNE

fHELLO,Rl
fLINBUF,R2
(Rl) +, (R2) +
10$

iGET ADDRESS OF MESSAGE.
iGET ADDRESS OF OUTPUT BUFFER.
iMOVE A BYTE TO OUTPUT BUFFER.
iIF NOT NULL, MOVE ANOTHER BYTE.

The .ASCIZ directive is subject to the same checks for character
legality and proper character string construction as described above
for the .ASCII directive.

6.3.6

.RADSO Directive

The .RADSO directive allows the user to generate data in Radix-50
packed format.
Radix-50 form allows three characters to be packed
into sixteen bits (one word)i therefore, any 6-character symbol can
be stored in two consecutive words. The form of the directive is:
.RADSO
where:

Istring l/ ••• /string

string

represents a series of characters to be packed
(three characters per word).
The string must
consist of the characters A through Z, 0 through
9, dollar sign ($), period (.) and space ( ). An
illegal printing character causes an error flag
(Q) to be printed in the assembly listing.
If fewer than three characters are to be packed,
the string is packed left-justified within the
word, and trailing spaces are assumed.
As with the .ASCII directive described in Section
6.3.4, the vertical-tab, null, line-feed, RUBOUT,
and all other non-printing characters, except
carriage-return
and
form-feed,
are
illegal
characters, resulting in an error code (I) in the
assembly listing. Similarly, the carriage-return
and form-feed characters result in an error code
(A)
because these characters end the scan of the
line, preventing
MACRO
from
detecting
the
terminating matching delimiter.

represent delimiting characters. These delimiters
may be any paired printing characters, other than
the equal sign (=), the left angle bracket «), or
the semicolon
(i), provided that the delimiting
character is not contained within the text string
itself.
If the delimiting characters do not
match, or if an illegal delimiting character is
used, the .RADSO directive is flagged with an
error code (A) in the assembly listing.

Examples of .RADSO directives are shown below:
.RADSO
.RADSO
.RADSO

IABCI
IAB/
IABCD/

.RADSO

IABCDEF/

iPACKS ABC INTO ONE WORD.
iPACKS AB (SPACE) INTO ONE WORD.
iPACKS ABC INTO FIRST WORD AND
iD (SPACE) (SPACE) INTO SECOND WORD.
iPACKS ABC INTO FIRST WORD, DEF INTO
iSECOND WORD.

6-22

GENERAL ASSEMBLER DIRECTIVES
Each character is translated into
indicated in the following table:
Character

its

Radix-50

equivalent,

(Cl,C2,C3)

are

Radix-50 Octal Equivalent

(space)
A-Z
$

1-32
33
34
35
36-47

(undefined)
0-9

The Radix-50 equivalents for characters 1
combined as follows:
Radix-50 Value

through

«Cl*50)+C2)*50+C3

For example:
Radix-50 Value of ABC = «1*50)+2)*50+3 = 3223
Refer to Section
equivalents.

A.2

Appendix

for

table

Radix-50

Angle brackets «» must be used in the .RAD50 directive whenever
special codes are to be inserted in the text string, as shown in the
example below:
• RAD50

IAB/<35>

.RAD50

<CHRl><CHR2><CHR3> iEQUIVALENT TO .RAD50 IABC/.

iSTORES 3255 IN ONE WORD .

CHRl=l
CHR2=2
CHR3=3

6.3.7

Temporary Radix-50 Control Operator:

The AR operator specifies that an argument is to be converted to
Radix-50 format.
This allows up to three characters to be stored in
one word. The AR operator is coded as follows:
A
RccC
where ccc represents a maximum of three characters to be converted to
a 16-bit Radix-50 value. If more than three characters are specified,
any following the third character are ignored. If fewer than 3 are
specified, it is assumed that the trailing characters are blanks. The
following example shows how the AR operator might be used to pack a
3-character file type specifier (MAC) into a single 16-bit word.
MOV

#ARMAC,FILEXT

iSTORE RAD50 MAC AS FILE EXTENSION

The number sign (#) is used to indicate immediate data, i.e., data to
be assembled directly into object code.
AR specifies that the
characters MAC are to be converted to Radix-50. This value is then
stored in location FILEXT.

6-23

GENERAL ASSEMBLER DIRECTIVES
6.4

RADIX AND NUMERIC CONTROL FACILITIES

6.4.1

Radix Control and Unary Control Operators

The normal default assumption for numeric values or expression values
appearing in a MACRO source program is octal. However, numerous
instances may occur where an alternate radix is useful for portions of
a program or for variables within a given statement.
It may be
useful, for example, to declare a given radix for applicability
throughout a program or to specify a numeric value or expression value
in a manner that causes it to be interpreted as a binary, octal, or
decimal value during assembly.
In other such instances, it may be
useful to complement numeric values or expression values. These MACRO
facilities are described in the following sections.
NOTE
When two or more unary operators appear
together, modifying the same term, the
operators are applied, from right to
left, to the term.

6.4.1.1 .RADIX Directive - Numbers used in a MACRO source program are
initially considered to be octal valuesi however, you can declare any
one of the following radices for applicability throughout the source
program or within specific portions of the program:
2, 8, 10

This is accomplished via a .RADIX directive of the form:
.RADIX
where:

n
represents one of the three acceptable radices
listed above.
If the argument n is not specified,
the octal default radix is assumed.

The argument in the .RADIX directive is always interpreted as a
decimal value.
Any alternate radix declared in the source program
through the .RADIX directive remains in effect until altered by the
occurrence of another such directive, i.e., a given radix declaration
is valid throughout a program until changed.
For example, the
statement:
.RADIX

. RADIX

iBEGINS A SECTION OF CODE HAVING A
iDECIMAL RADIX.

iREVERTS TO OCTAL RADIX •

Any value other than null, 2, 8, or 10 specified as an argument in the
.RADIX directive causes an error code
CA) to be generated in the
assembly listing.
In general, macro definitions should not contain or rely on radix
settings established with the
.RADIX directive. Rather, temporary
radix control operators should be used within a macro definition.
Where a possible radix conflict exists within a macro definition or in

6-24

GENERAL ASSEMBLER DIRECTIVES
possible future uses of that code, it is recommended that the user
specify numeric or expression values using the temporary radix control
operators described below.

6.4.1.2 Temporary Radix Control Operators:
AD, AO, and AB - Once the
user has specified a given radix for a section of code or has decided
to use the default octal radix, he may discover a number of cases
where an alternate radix is more convenient or desirable (particularly
within macro definitions). The creation of a mask word, for example,
might best be accomplished through the use of a binary radix.
MACRO has three unary operators that allow the user
alternate radix, as shown below:
AD"number"
AO"number"
AS"number"

establish

("number" is evaluated as a decimal number)
("number" is evaluated as an octal number)
("number" is evaluated as a binary number)

Thus, an alternate radix can be declared temporarily to meet a
localized requirement in the source program. Such a declaration can
be made at any time, regardless of the existence of the default octal
radix or another specific radix declaration elsewhere in the program.
In other words, the effect of a temporary radix control operator is
limited to the term or expression immediately following the operator.
Any value specified in connection with a temporary radix control
operator is evaluated during assembly as a 16-bit entity. Temporary
radix control declarations can be included in the source program
anywhere a numeric value is legal.
The expressions below are representative of the methods of
temporary radix control operators:
AD123
AO 47
AS 00001101
AO<A+13>

specifying

Decimal radix
Octal Radix
Binary Radix
Octal Radix

Note that the up-arrow and the radix control operator may not be
separated, but the radix control operator and the following term or
expression can be physically separated by spaces or tabs
for
legibility or formatting purposes. A multi-element term or expression
that is to be interpreted in an alternate radix should be enclosed
within angle brackets, as shown in the last of the four temporary
radix control expressions above.
The following example also illustrates the use of angle brackets to
delimit an expression that is to be interpreted in an alternate radix:
.RADIX

• WORD

AO<A+lO>*lO

A=lO
When the temporary radix expression in the .WORD directive above is
evaluated, it effectively yields the following equivalent statement:
• WORD

180.

MACRO also allows a temporary radix change to decimal by specifying a
number, immediately followed by a decimal point (.), as shown below:
100.
1376.
128.

Equivalent to 144(8)
Equivalent to 2540(8)
Equivalent to 200(8)
6-25

GENERAL ASSEMBLER DIRECTIVES
The above expression forms are equivalent in function to those
below:

listed

~DIOO

~D1376
~D128

6.4.2

Numeric Directives and Unary Control Operators

Two storage directives and two numeric control operators are available
to simplify the use of the floating-point hardware on the PDP-II.
These facilities allow floating-point data to be created in the
program,
and numeric values to be complemented or treated as
floating-point numbers.
A floating-point number is represented by a string of decimal digits.
]xe string
(which can be a single digit in length) may optionally
contain a decimal point, and may be followed by an optional exponent
indicator in the form of the letter E and a signed decimal integer
exponent. The number may not contain embedded blanks, tabs or angle
brackets and may not be an expression. Such a string will result in
one or more errors (A or Q) in the assembly listing.
The list of numeric representations below contains seven
valid representations of the same floating-point number:

distinct,

3
3.
3.0
3.0EO
3EO

.3EI
300E-2
As can be inferred, the list could be extended indefinitely
(e.g.,
3000E-3,
.03E2, etc.). A leading plus sign is optional (e.g., 3.0 is
considered to be +3.0). A leading minus sign complements the sign
bit. No other operators are allowed (e.g., 3.0+N is illegal).
All floating-point numbers are evaluated as 64 bits in
format:
64

EEEEEEEE

the

following

MMM ••••• MMMI
Mantissa
Exponent
Sign

(55 bits)
(8 bits)
(1 bit)

MACRO returns a value of the appropriate size and precision via one of
the floating-point directives. The values returned may be truncated
or rounded (see Section 6.2).
Floating-point numbers are normally rounded.
That is, when
a
floating-point number exceeds the limits of the field in which it is
to be stored, the high-order bit of the unretained word is added to
the low-order bit of the retained word, as shown below. For example,
if the number is to be stored in a 2-word field, but more than 32 bits
are needed to express its exact value, the highest bit (32) of the
unretained field is added to the least significant bit
(0)
of the
6-26

GENERAL ASSEMBLER DIRECTIVES
retained field (see illustration below). The .ENABL FPT directive is
used to enable floating-point truncation;
.DSABL FPT is used to
return to floating-point rounding (see Table 6-2).
Bit
32

Bit

o 32 31

Retained
field

I I

Unretained
field

Note that all numeric operands associated with Floating
Point
Processor instructions are automatically evaluated as single-word,
decimal, floating-point values unless a temporary radix control
operator is specified.
For example, to add (floating) the octal
constant 41040 to the contents of floating accumulator zero,
the
following instruction must be used:
#~041040,FO

ADDF
where:

FO is assumed to represent floating accumulator zero.

Floating-point numbers are described
applicable PDP-II Processor Handbook.

greater

detail

the

6.4.2.1 .FLT2 and .FLT4 - Floating-Point Storage Directives - MACRO
supports
two directives that evaluate successive floating-point
numbers and store the results in the object module. These directives
are similar to the .WORD directive and are of the form:
• FLT2
• FLT4
where:

argl,arg2, •••
argl,arg2, •••

argl,arg2, •••

represent one or more floating point numbers
as described in Section 6.4.2.
Multiple
arguments must be separated by commas •

. FLT2 causes two words of storage to be generated for each argument,
while ".FLT4 generates four words of storage for each argument.

6.4.2.2 Temporary Numeric Control Operators:
AC and AF - The ~C
unary operator allows you to specify an argument that is to be
complemented as it is evaluated during assembly.
The ~F unary
operator allows you to specify an argument consisting of a I-word
floating-point number.
As with the radix control operators described above, the numeric
control operator (~C) can be used anywhere in the source program that
an expression value is legal. Such a construction is evaluated by
MACRO as a l6-bit binary value before being complemented. For
example, the following statement:
TAG4:

• WORD

causes the lis complement of the value 151 (octal) to be stored as a
l6-bit value in the program. The resulting value expressed in octal
form is 177626(8).

6-27

GENERAL ASSEMBLER DIRECTIVES

Because the ~C construction is a unary operator, the operator and its
argument are regarded as a term. Thus, more than one unary operator
may be applied to a single term.
For example, the following
construction:

causes the decimal value 25 to be complemented during assembly.
The
resulting binary value, when expressed in octal form, reduces to
177746(octal).
The term created through the use of the temporary numeric control
operator thus becomes an entity that can be used alone or in
combination with other expression elements.
For
example,
the
following construction:

is equivalent in function to:

This expression is evaluated during assembly as the lis complement of
2, plus the absolute value of 6. When these terms are combined, the
resulting expression value generates a carry beyond
the
most
significant bit, leaving 000003(8) as the reduced value.
As shown above, when the temporary numeric control operator and its
argument are coded as a term within an expression, angle brackets
should be used as delimiters to ensure precise evaluation and
readability.
MACRO also supports a unary operator for numeric control which allows
you to specify an argument consisting of a I-word floating-point
number. For example, the following statement:
A:

MOV

creates a I-word floating-point number at location A+2 containing
value 3.7 formatted as shown below.
BIT IS

EEEEEEEE

MMMMMMMI

Sign (bit 15)

Exponent (bits 14-7)

Mantissa (bits 6-0)

The importance of ordering with respect to unary
below.
~Fl.O

~F-l.O
-~FI.O

-~F-I.O

the

operators

shown

020400
120400
157400
057400

The value created by the ~F unary operator and its argument is then
term that can be used)b~QOitself or in an expression. For example:

is equivalent to:

6-28

GENERAL ASSEMBLER DIRECTIVES
For this reason, the use of angle brackets is advised.
Expressions
used as terms or arguments of a unary operator must be explicitly
grouped.

6.5

LOCATION COUNTER CONTROL DIRECTIVES

The directives used in controlling the value of the current location
counter and in reserving storage space in the object program are
described in the following sections.
In this connection, it should be noted that several MACRO statements
may cause an odd number of bytes to be allocated, as listed below:
1.

.BYTE directive

.BLKB directive

.ASCII or .ASCIZ directive

.000 directive

A direct assignment statement of the form .=.+expression,
which results in the assignment of an odd address value.

In cases that yield an odd address value, the next word-boundaried
instruction automatically forces the location counter to an even
value, but that instruction is flagged with an error code (B)
in the
assembly listing.

6.5.1

.EVEN Directive

The .EVEN directive ensures that the current location counter contains
an even value by adding I if the current value is odd. If the current
location counter is already even, no action is taken.
Any operands
following an .EVEN directive are flagged with an error code (Q) in the
assembly listing.
The .EVEN directive is used as follows:

6.5.2

.ASCIZ
• EVEN

/THIS IS A TEST/

• WORD

XYZ

;ENSURES THAT THE NEXT STATEMENT WILL
;BEGIN ON A WORD BOUNDARY •

.ODD Directive

The .000 directive ensures that the current location counter contains
an odd value by adding I if the current value is even. If the current
location counter is already odd, no action is taken.
Any operands
following an .000 directive are also flagged with an error code (Q) in
the assembly listing.

6-29

GENERAL ASSEMBLER DIRECTIVES
6.5.3

.BLKB and .BLKW Directives

Blocks of storage can be reserved in the object program by means of
the .BLKB and .BLKW directives.
The .BLKB directive is used to
reserve byte blocks; similarly, the .BLKW directive reserves word
blocks. The two directives are of the form:
.BLKB
.BLKW

exp
exp

exp

where:

represents the specified number of bytes or words
to be reserved in the object program. If no
argument is present, a default value of I
is
assumed.
These directives should not be used
without arguments.
Any
expression
that
is
completely
defined at assembly-time and that
reduces to an absolute value is legal.
If the
expression specified in either of these directives
is not an absolute value, the statement is flagged
with an error code CA) in the assembly listing.

Figure 6-6 illustrates the use of the .BLKB and .BLKW directives.
166 100000
161 0all10
168
169 001111
171 111011
171 111114
172 11111014
173 001115
174 001016
175 001010
176
177
178
179
180 001020
181 001024
182 110025
183 001026
184 101031
185 000032
186 000034
181 101136
188 001111

PASS ••

1001U)3

.PSECT
.8LKW

.PSECT
SYMBOL.I.8LKW
MODE I.
FLAGS II • BLK8
SECTOR II. BLK8
VALUEII .BLI<W
RELLVLI •• BLKW
.REPT
.BLKW
.ENOR
CLCNAM.I.BLKW
CL.CFGSII.BL.KB
CLCSECII.BLKB
CLCLOCI •• BLKW
CLCMAX.I.8L.l<w
CHRPNT II. BLKlrf
SYMBEG.,.BLKW
ENDH,' II. BL.KW
.PSECT

Figure 6-6

IMPUHE,D
1

IMPPAS,D,GBL

,PASS FLAG
'NEXT GROUP MUST STAY TOGETHER
,SYMBOL ACCUMULATOR
,MODE/FLAGS BYTE

,SYMBOL/EXPRESSION TYPE
'EXPRESSION VALUE
,RELOCATION LEVEL
MAXXMT.« •• SYMBOL~/2>
1

,'CURRENT LOCATION COUNTER NAME
,

2
1
1
1
1
1
1

,END OF GROUPED DATA
,CHARACTER POINTER
,POINTER TO START OF SYMBOL.

Example of .BLKB and .BLKW Directives

The .BLKB directive in a source program has the
following statement:

same

effect

the

.=.+expression
which causes the value of the expression to be added to the current
value of the location counter.
The .BLKB directive, however, is
easier to interpret in the context of the source code in which it
appears and is therefore recommended.

6.6

TERMINATING DIRECTIVES

6-30

GENERAL ASSEMBLER DIRECTIVES
6.6.1

.END Directive

The .END directive indicates the logical end of the source input,
takes the following form:
.END
where:

and

exp
represents an optional expression value which,
if
present, indicates the program-entry point, i.e.,
the transfer address at which program execution is
to begin.

exp

When MACRO encounters a valid occurrence of the .END directive,
it
terminates the current assembly pass. Any additional text beyond this
point in the current source file, as well as in additional source
files identified in the command line, will be ignored.
When creating a task image consisting of several object modules, only
one object module may be terminated with an .END exp statement
specifying the starting address. All other object modules must be
terminated with an .END statement without an address argument;
otherwise, the Linker will issue a diagnostic message. If no starting
address is specified in any of the object modules, task execution will
begin at location 1 of the task and immediately fault because of an
odd addressing error.
The .END statement must not be used within a macro expansion or a
conditional assembly block;
if it is so used, it is flagged with an
error code (0) in the assembly listing. The .END statement may be
used, however, in an immediate conditional statement (see Section
6.10.2).
If the source program input is not terminated with an .END
an error code (E) results in the assembly listing.

6.6.2

directive,

.EOT Directive

Under the TRAX operating system, the MACRO .EOT directive is ignored
and simply treated as a directive without effect, i.e., as a no-ope

6.7

PROGRAM BOUNDARIES DIRECTIVE:

.LIMIT

It is often desirable to know the upper and lower address boundaries
of the task image.
When the .LIMIT directive is specified in the
source program, MACRO effectively generates the following instruction:
.BLKW

causing two storage words to be reserved in the object module. Later,
at link time, the address of the bottom of the task's stack is
inserted into the first reserved word, and the address of the first
free word following the task image is inserted into the second
reserved word.
During linking, the size of the task image is rounded
nearest 2-word boundary.

6-31

upward

the

GENERAL ASSEMBLER DIRECTIVES
For a discussion of task memory allocation and mapping, refer to the
applicable Linker reference manual (see Section 0.3 in the Preface).

6.8

PROGRAM SECTIONING DIRECTIV£S

The MACRO program sectioning directives are used to declare names for
program sections and to establish certain program section attributes
essential to Linker proc~ssing.

6.8.1

.PSECT Directive

The .PSECT directive allows absolute control over
the
memory
allocation of a program at link time, because any program attributes
established through this directive are passed to the Linker.
For example, if you are writing programs for a multi-user environment,
a program section containing pure code
(instructions only) or a
program section containing impure code (data only) may be explicitly
declared through the .PSECT directive. Furthermore, these program
sections may be explicitly declared as read-only code, qualifying them
for use as protected, reentrant programs.
In addition, program sections exhibiting the global
(GBL)
attribute
can be explicitly allocated in a task's overlay structure at link
time.
The advantages gained through sectioning programs in this manner
therefore relate primarily to control of memory allocation, program
modularity, and more effective partitioning of memory. Refer to the
applicable
Linker reference manual for a discussion of memory
allocation (see Section 0.3 in the Preface).
The .PSECT directive is formatted as follows:
.PSECT
where:

name,argl,arg2, ••• argn

name

represents the symbolic name of
section, as described in Table 6-3.

the

represents any legal separator (comma, tab
space).
argl,
arg2, ...
argn

program
and/or

represent one or more of the legal symbolic
arguments defined for use with the .PSECT
directive, as described in Table 6-3.
The slash
separating each pair of symbolic arguments listed
in the table indicates that
these
optional
arguments are mutually exclusive, i.e., one or the
other, but not both, may be specified.
Multiple
arguments must be separated by a legal separating
character. Any symbolic argument specified in the
.PSECT directive other than those listed in Table
6-3 will cause that statement to be flagged with
an error code (A) in the assembly listing.

6-32

GENERAL ASSEMBLER DIRECTIVES

Table 6-3
Symbolic Arguments of .PSECT Directive
Argument

Default

Meaning

NAME

Blank

Establishes the program section name,
which
is
specified as one to six
Radix-50 characters. If this argument
is omitted, a comma must appear in place
of the name parameter.
The Radix-50
character set is listed in Section A.2
of Appendix A.

RO/RW

Defines which
type
of
access
permitted to the program section:

RO=Read-Only Access
RW=Read/Write Access
NOTE
lAS and RSX-llD set hardware
protection
for
RO
program
sections. TRAX does not provide
such protection.
I/O

Defines
the
program
section
as
containing either instructions
(I) or
data (D). These attributes allow the
Linker to differentiate global symbols
that
are
program
entry-point
instructions
(1)
from those that are
data values (D).

GBL/LCL

LCL

Defines the scope
of
the
program
section, as subsequently interpreted by
the Linker.
In building single-segment programs, the
GBL/LCL
arguments
have no meaning,
because the total memory allocation for
the
program will go into the root
segment of the task.
The
GBL/LCL
arguments apply only in the case of
overlays.
If an object module contains a local
program
section,
then
the storage
allocation for that module will occur
within the segment in which the module
resides.
Many modules can reference
this
same program section, and the
memory allocation for each module is
either concatenated or overlaid within
the segment, depending on the argument
of the program section (.PSECT) defining
(continued on next page)

6-33

GENERAL ASSEMBLER DIRECTIVES

Table 6-3 (Cont.)
Symbolic Arguments of .PSECT Directive
Argument

Default

Meaning

GBL/LCL
(cont'd)

LCL

its allocation requirements (see CON/OVR
below). If an object module contains a
global
program
section,
the
contributions to this program section
are collected across segment boundaries,
and the allocation of memory for that
section will go into the segment nearest
the root in which the first contribution
to this program section appeared.
(The
term contribution implies an allocation
of memory to the program section.)

ABS/REL

REL

Defines the relocatability attribute
the program section:

ABS=Absolute
(non-relocatable).
When
the ABS argument is specified, the
program section is regarded by the
Linker as an absolute module, thus
requiring
no
relocation.
The
program section is assembled and
loaded, starting at absolute virtual
address O.
The location of data in absolute
program sections must fall within
the virtual memory limits of the
segment
containing
the
program
section;
otherwise,
an
error
results at link time. For example,
the following code, although valid
at during assembly, may generate a
Linker error message if
virtual
location
100000
is outside the
segment's virtual address space:
.PSECT
.=.+100000
. WORD

ALPHA,ABS
X

The above coding assembles properly,
but the resulting load address may
be outside the respective segment's
boundaries.
In such cases, the
Linker recognizes this as an attempt
to load data outside the task image
and rwith an error message.
REL=Relocatable. When the REL argument
is specified, the Linker calculates
a relocation bias and adds it to all
references to locations within the
program
section,
i.e.,
all
references to the program section
must have a relocation bias added to
them to make them absolute.
(continued on next page)
6-34

GENERAL ASSEMBLER DIRECTIVES
Table 6-3 (Cont.)
Symbolic Arguments of .PSECT Directive
Argument

Default

CONjOVR

CON

Meaning
Defines the allocation
the program section:

requirements

CON=Concatenated. All program section
contributions are to be concatenated
with other references to this same
program
section
in
order
to
determine
the
total
memory
allocation
requirement
for this
program section.
OVR=Overlaid.
All
program
section
contributions are to be overlaid.
Thus,
the
total
allocation
requirement for the program section
is equal to the largest allocation
request
made
by any individual
contribution
to
this
program
section.
The only argument in the .PSECT directive that is position-dependent
is NAME.
If it is omitted, a comma must be used in its place.
For
example, the directive:
.PSECT

,GBL

shows a .PSECT directive with a blank name argument and the GBL
argument.
Default values
(see Table 6-3) are assumed for all other
unspecified arguments.
Once the attributes of a program section are declared through a .PSECT
directive, MACRO assumes that these attributes remain in effect for
all subsequent .PSECT directives of the same name that are encountered
within the module.
MACRO provides for 256(10) program sections, as listed below:
1.

One default absolute program section (. ABS.)

One default unnamed relocatable program section

Two-hundred-fifty-four named program sections.

The .PSECT directive enables the user to:
1.

Create program sections (see Section 6.8.1.1)

Share code and
6.8.1.2).

data

among

For each program section specified or
following information:

program

sections

implied,

Program section name

Contents of the current location counter

6-35

MACRO

(see

Section

maintains

the

GENERAL ASSEMBLER DIRECTIVES
3.

Maximum location counter value encountered

Program section attributes,
described in Table 6-3 above.

i.e.,

the

.PSECT

arguments

6.8.1.1 Creating Program
Sections - MACRO
automatically
begins
assembling source statements at relocatable zero of the unnamed
program section, i.e., the first statement of a source program is
always an implied .PSECT directive.
The first occurrence of a .PSECT directive with a given name assumes
that the current location counter is set at relocatable zero. The
scope of this directive then extends until a directive declaring a
different program section is specified.
Further occurrences of a
program section name
in subsequent .PSECT statements cause the
resumption of assembly where that section previously ended.
For
example:
A:

B:
C:
X:

Y:
D:

.PSECT
• WORD
• WORD
• WORD
.PSECT
• WORD
• WORD
.PSECT
• WORD

o
o
a
ALPHA

jDECLARES UNNAMED RELOCATABLE PROGRAM
jSECTION ASSEMBLED AT RELOCATABLE
jADDRESSES 0, 2, AND 4.
jDECLARES RELOCATABLE PROGRAM SECTION
jNAMED ALPHA ASSEMBLED AT RELOCATABLE
jADDRESSES 0 AND 2.
jRETURNS TO UNNAMED RELOCATABLE
iPROGRAM SECTION AND CONTINUES ASSEMjBLY AT RELOCATABLE ADDRESS 6.

A given program section may be defined completely upon encountering
its first .PSECT directive. Thereafter, the section can be referenced
by specifying its name only, or by completely respecifying its
attributes.
For example,
a program section can be declared through
the directive:
.PSECT

ALPHA,ABS,OVR

and later referenced through the equivalent directive:
.PSECT

ALPHA

which requires no arguments.
By maintaining separate location counters for each program section,
MACRO allows the user to write statements that are not physically
contiguous within the program, but that can be loaded contiguously
following assembly, as shown in the following example.

A:
B:

ST:

.PSECT
• WORD
. WORD
• WORD
CLR
CLR
CLR
.PSECT
• WORD
.PSECT
INC
BR

SEC1,REL,RO

a
o
o

jSTART A RELOCATABLE PROGRAM SECTION
jNAMED SECl ASSEMBLED AT RELOCATABLE
iADDRESSES 0, 2, AND 4.

jASSEMBLE CODE AT RELOCATABLE

iADDRESSES 6 THROUGH 12.

SECA,ABS

.+2,A
SECl

A
ST

iSTART AN ABSOLUTE PROGRAM SECTION
iNAMED SECA. ASSEMBLE CODE AT
jABSOLUTE ADDRESSES a AND 2 •
jRESUME RELOCATABLE PROGRAM SECTION
jSEC1. ASSEMBLE CODE AT RELOCATABLE
jADDRESSES 14 AND 16.

6-36

GENERAL ASSEMBLER DIRECTIVES
All labels in an absolute program section are absolute; likewise, all
labels in a relocatable section are relocatable. The current location
counter symbol (.) is also relocatable or absolute when referenced in
a relocatable or absolute program section, respectively.
Any labels appearing on a line containing a .PSECT
(or .ASECT or
.CSECT) directive are assigned the value of the current location
counter before the .PSECT (or other) directive takes effect. Thus, if
the first statement of a program is:
A:

.PSECT

ALT,REL

the label A is assigned to relocatable address zero of the unnamed (or
blank) program section.
It is not known during assembly where relocatable program sections
will be loaded, therefore all references between relocatable sections
in a single assembly are translated by MACRO to references relative to
the base of the referenced section. Thus, MACRO provides the Linker
with the necessary information to resolve the linkages between various
program sections.
Such information is not necessary, however, when
referencing an absolute program section, because all instructions in
an absolute program section are associated with an absolute virtual
address.
In the following example, references to the symbols X and Yare
translated into references relative to the base of the relocatable
program section named SEN.
.PSECT
.=.+1000
CLR
A:

ENT,ABS

JMP

.PSECT
MOV
JMP
HALT
• WORD

SEN,REL
RO,Rl
A

iASSEMBLED AS CLR BASE OF
iRELOCATABLE SECTION + 10.
iASSEMBLED AS JMP BASE OF
iRELOCATABLE SECTION + 6.
iASSEMBLED AS JMP 1000.

0
NOTE
In the preceding example,
using
a
constant in conjunction with the current
location counter symbol (.) in the form
.=1000 would result in an error, because
constants are always absolute and are
always associated with the program's
.ASECT (. ABS.). If the form .=1000
were
used,
a
program
section
incompatibility would be detected.
See
Section 3.6 for a discussion of the
current location counter.

6.8.1.2 Code or Data Sharing - Named relocatable program sections
with the arguments GBL and OVR operate in the same manner as FORTRAN
COMMON, i.e., program sections of the same name with the arguments GBL
and OVR from different assemblies are all loaded at the same location
by the Linker. All other program sections,
i.e., those with the
argument CON, are concatenated.

6-37

GENERAL ASSEMBLER DIRECTIVES
Note that no conflict exists between internal symbolic names and
program section names, i.e., it is legal to use the same symbolic name
for both purposes. Considering FORTRAN again, using the same symbolic
name is necessary to accommodate the following statement:
COMMON

jXj A,B,C,X

where the symbol X represents the base of the program section and also
the fourth element of that section.

6.8.1.3 Memory Allocation Considerations - The assembler does not
generate an error when a module ends at an odd location. This allows
you to place odd length data at the end of a module.
However, when
several modules contain object code contributions to the same program
section having the concatenate attribute (see Table 6-3), odd length
modules
(except the last) may cause the Linker to link succeeding
modules starting at odd locations, thereby making the linked program
unexecutable.
To avoid this problem, code and data should be
separated from each other and be placed in separately named program
sections. This permits the Linker to automatically begin each program
section on an even address. Refer to the applicable Linker reference
manual for further information on memory allocation of tasks (see
Section 0.3 in the Preface).

6.8.2

.ASECT and .CSECT Directives

TRAX assembly-language programs use the .PSECT and .ASECT directives
exclusively, since the .PSECT directive provides all the capabilities
of the .CSECT directive defined for other PDP-II assemblers.
MACRO
will accept both .ASECT and .CSECT directives, but assembles them as
though they were .PSECT directives with the default attributes listed
in Table 6-4. Also, compatibility exists between other MACRO programs
and the TRAX Linkers, since the respective Linkers recognize the
.ASECT and .CSECT directives that appear in such programs and likewise
assign the default values listed in Table 6-4.
Table 6-4
Non-TRAX Program Section Default Values
Default Value
Attribute
.ASECT

.CSECT (named)

.CSECT (unnamed)

Name

. ABS.

name

Blank

Access

Type

Scope

GBL

LCL

Relocation

ABS

REL

Allocation

OVR

CON

6-38

GENERAL ASSEMBLER DIRECTIVES
The allowable syntactical forms of the .ASECT
are:
.ASECT
.CSECT
.CSECT

and

.CSECT

directives

are

assumed

symbol

Note that the statement:
.CSECT

JIM

is identical to the statement:
.PSECT

JIM,GBL,OVR

because the .CSECT default values GBL and
named program section.

6.9

SYMBOL CONTROL DIRECTIVE:

OVR

for

the

.GLOBL

MACRO produces a relocatable object module and a listing file
containing the assembly listing and symbol table. The Linker joins
separately-assembled object modules into a single executable task
image.
During linking, object modules are relocated as a function of
the specified base of the module. The object modules are then linked
via global symbols, such that a global symbol in one module, defined
either by a global assignment operator (==), a global label operator
(::), or the .GLOBL directive can be referenced from another module.
Thus, all symbols which will be referenced by other program modules
must be singled out as global symbols in the defining modules.
The .GLOBL directive is provided to define (and thus provide linkage
to)
symbols not otherwise defined as global symbols within a module.
For example, if the .DSABL GBL directive is in effect
(see Section
6.2),
.GLOBL directives might be included in a source program to
effect linkage to library routines. For a global symbol definition,
the directive .GLOBL A,B,C is equivalent to:
A==expression (or A::)
B==expression (or B::)
C==expression (or C::)
Thus, the general form of the .GLOBL directive is:
.GLOBL
where:

syml,sym2, •.• symn

syml,
sym2, •••
symn

represent legal symbolic names. When multiple
symbols are specified, they are separated by any
legal separator (comma, space, and/or tab).

A .GLOBL directive may also embody a
field.

label

field

and/or

comment

At the end of assembly pass 1, MACRO determines whether a given global
symbol is defined within the current program module or whether it is
to be treated as an external symbol. All internal symbols appearing
within a given program must be defined at the end of assembly pass I
or they will be assumed to be default global references.
Refer to
Section
6.2 for a description of enabling/disabling of global
references.

6-39

GENERAL ASSEMBLER DIRECTIVES
In the example below, A and B are entry-point symbols. The symbol A
has been explicitly defined as a global symbol by means of the .GLOBL
directive, and the symbol B has been explicitly defined as a global
label by means of the double colon (::). Since the symbol C is not
defined as a label within the current assembly,
it is an external
(global) reference.
DEFINE A SUBROUTINE WITH 2 ENTRY POINTS WHICH CALLS AN
EXTERNAL SUBROUTINE

A:
X:
B··

• PSECT
.GLOBL
MOV
MOV
JSR
RTS
MOV
CLR
BR

A
@(RS)+,RO
#X,Rl
PC,C
RS
(RS)+,Rl
R2

iDECLARE THE UNNAMED PROGRAM SECTION •
iDEFINE A AS A GLOBAL SYMBOL.
iDEFINE ENTRY POINT A.
iCALL EXTERNAL SUBROUTINE C.
iEXIT.
iDEFINE ENTRY POINT B.

External symbols can appear in the operand field of an instruction or
MACRO directive as a direct reference,. as shown in the examples below:
CLR
• WORD
CLR

EXT
EXT
@EXT

External symbols may also appear as a term within
shown below:
CLR
• WORD
CLR

expression,

EXT+A
EXT-2
@EXT+A(Rl)

It should be noted that an undefined external symbol cannot be used in
the evaluation of a direct assignment statement or as an argument in a
conditional assembly directive (see Sections 6.10.1 and 6.10.3).

6.10

CONDITIONAL ASSEMBLY DIRECTIVES

Conditional assembly directives allow you to include or exclude blocks
of source code during the assembly process, based on the evaluation of
stated condition tests within the body of the program.
This
capability allows several variations of a program to be generated from
the same source code.

6.10.1

Conditional Assembly Block Directives:

.IF, .ENDC

The general form of a conditional assembly block is as follows:
• IF

cond,argument(s)

iSTART CONDITIONAL ASSEMBLY BLOCK •

range

iRANGE OF CONDITIONAL ASSEMBLY BLOCK.

• ENDC

iEND OF CONDITIONAL ASSEMBLY BLOCK •

6-40

GENERAL ASSEMBLER DIRECTIVES
cond

where:

represents a specified condition that must be met
if the block is to be included in the assembly.
The conditions that may be
tested
by
the
conditional assembly directives are defined in
Table 6-5.
represents any
and/or tab).

legal

separator

(comma,

space,

argument(s)

represent(s)
the
symbolic
argument(s)
or
expression(s)
of the specified conditional test.
These arguments are thus a function of
the
specified condition to be tested (see Table 6-5).

range

represents the body of code that is
either
included in the assembly or excluded, depending
upon whether the specified condition is met.

.ENDC

terminates the conditional assembly block.
This
directive must be present to end the conditional
assembly block.

A condition test other than those listed in Table 6-5, an illegal
argument, or a null argument specified in an .IF directive causes that
line to be flagged with an error code (A) in the assembly listing.
Table 6-5
Legal Condition Tests for Conditional Assembly Directives
Conditions
Arguments

Assemble Block If:

Expression

Expression is equal to 0
(or not equal to 0).

Expression

Expression
is
greater
than 0 (or less than or
equal to 0).

Expression

Expression is less than 0
(or greater than or equal
to 0).

NDF

Symbolic
argument

Symbol is defined (or not
defined).

Macro-type
argument

Argument is blank (or
non-blank).

ION

OIF

Two macro-type
arguments

Arguments are identical
(or different).

Expression

Same as EQ/NE.

Expression

Same as GT/LT.

Positive

Complement

6-41

GENERAL ASSEMBLER DIRECTIVES
NOTE
A macro-type argument (which is a
form
of symbolic argument), as shown below,
is enclosed within angle brackets or
denoted with an up-arrow construction
(as described in Section 7.3.1).
(A,B,C)
~/124/

An example of a conditional assembly directive follows:
• IF EQ

ALPHA+l

iASSEMBLE BLOCK IF ALPHA+l=O •

.ENDC
The two operators & and! have special meaning within DF and NDF
conditions, in that they are allowed in grouping symbolic arguments.
Logical AND operator

Logical inclusive OR operator
For example, the conditional assembly statement:
.IF DF

SYMl & SYM2

.ENDC
results in the assembly of the conditional block if the
and SYM2 are both defined.

symbols

SYMI

Nested conditional directives take the form:
Conditional Assembly Directive
Conditional Assembly Directive

.ENDC
.ENDC
For example, the following conditional directives:
.IF DF
.IF DF

SYMI
SYM2

.ENDC
.ENDC
can govern whether assembly is to occur.
In the example above, if the
outermost condition is unsatisfied, no deeper level of evaluation of
nested conditional statements within the program occurs.

6-42

GENERAL ASSEMBLER DIRECTIVES
Each conditional assembly block must be terminated with an .ENDC
directive.
An .ENDC directive encountered outside a conditional
assembly block is flagged with an error code
(0)
in the assembly
listing.
MACRO permits a nesting depth of 16(10) conditional assembly levels.
Any statement that attempts to exceed this nesting level depth is
flagged with an error code (0) in the assembly listing.

6.10.2

Subconditional Assembly Block Directives:

Subconditional directives may be placed
blocks to indicate:

conditional

assembly

The assembly of an alternate body of code when the
of the block tests false.

condition

The assembly of a non-contiguous body of code within the
conditional assembly block, depending upon the result of the
conditional test in entering the block.

The unconditional assembly
conditional assembly block.

within

.IFF, .IFT, .IFTF

body

code

within

The subconditional directives are described in detail
in Table 6-6.
If a subconditional directive appears outside a conditional assembly
block, an error code (0) is generated in the assembly listing.
Table 6-6
Subconditional Assembly Block Directives
Subconditional
Directive

Function

.IFF

If the condition tested
u~on
entering
the
conditional assembly block 1S false,
the code
following this directive, and continuing up to the
next occurrence of a subconditional directive or
to the end of the conditional assembly block,
is
to be included in the progr~m.

.IFT

If the condition tested
u~on
entering
the
conditional assembly block 1S true,
the code
following this directive, and continuing up to the
next occurrence of a subconditional directive or
to the end of the conditional assembly block,
is
to be included in the program.

.IFTF

The code following this directive, and continuing
up to the next occurrence of a subconditional
directive or to the end of the
conditional
assembly block, is to be included in the program,
regardless of the result of the condition tested
upon entering the conditional assembly block.

The implied argument of a subconditional directive is the condition
test specified upon entering the conditional assembly block, as
reflected by the initial directive in the conditional coding examples

6-43

GENERAL ASSEMBLER DIRECTIVES
below.
Conditional or subconditional directives in nested conditional
assembly blocks are not evaluated if the previous (or outer) condition
in the block
is not satisfied.
Examples 3 and 4 below illustrate
nested directives that are not evaluated because
of
previous
unsatisfied conditional coding.
EXAMPLE 1: Assume that symbol SYM is defined.
.IF OF

SYM

iTESTS TRUE, SYM IS DEFINED. ASSEMBLE
iTHE FOLLOWING CODE.
iTESTS FALSE. SYM IS DEFINED. DO NOT
iASSEMBLE THE FOLLOWING CODE.

.IFF

.IFT

iTESTS TRUE. SYM IS DEFINED. ASSEM-

iBLE THE FOLLOWING CODE.
iASSEMBLE FOLLOWING CODE UNCONDITION-

.IFTF

iALLY.
.IFT

iTESTS TRUE. SYM IS DEFINED. ASSEMiBLE REMAINDER OF CONDITIONAL ASSEMiBLY BLOCK •

• ENDC
EXAMPLE 2:

Assume that symbol X is defined and
defined.

. IF DF
. IF DF
.IFF

X
Y

• IFT

that symbol Y is not

iTESTS TRUE, SYMBOL X IS DEFINED •
iTESTS FALSE, SYMBOL Y IS NOT DEFINED •
iTESTS TRUE, SYMBOL Y IS NOT DEFINED,
iASSEMBLE THE FOLLOWING CODE.
iTESTS FALSE, SYMBOL Y IS NOT DEFINED •

iDO NOT ASSEMBLE THE FOLLOWING CODE •
. ENDC
.ENDC
EXAMPLE 3:

Assume that symbol A is defined and that symbol B is not
defined.

. IF DF

iTESTS TRUE.

A IS DEFINED •

iASSEMBLE THE FOLLOWING CODE.
MOV

A,Rl

.IFF
MOV

iTESTS FALSE. A IS DEFINED.
DO NOT
iASSEMBLE THE FOLLOWING CODE.
Rl,RO

.IF NDF B

iNESTED CONDITIONAL DIRECTIVE IS NOT

6-44

GENERAL ASSEMBLER DIRECTIVES
iEVALUATED •
• ENDC
.ENDC
EXAMPLE 4:

Assume that symbol X is not defined and that symbol Y is
defined.

. IF DF

iTESTS FALSE. SYMBOL X IS NOT DEFINED •
iDO NOT ASSEMBLE THE FOLLOWING CODE.

.IF DF

iNESTED CONDITIONAL DIRECTIVE IS NOT
iEVALUATED.
iNESTED SUBCONDITIONAL DIRECTIVE IS
;NOT EVALUATED.

.IFF

.IFT

;NESTED SUBCONDITIONAL DIRECTIVE IS
;NOT EVALUATED •

• EN DC
.ENDC

6.10.3

Immediate Conditional Assembly Directive:

.IIF

An immediate conditional assembly directive provides a means for
writing a I-line conditional assembly block.
In using this directive,
no terminating .ENDC statement is required, and the condition to be
tested
is completely expressed within the line containing the
directive.
Immediate conditional assembly directives are of the form:
.IIF
where:

cond

cond,arg,statement
represents one of the legal condition
tests
defined for conditional assembly blocks in Table
6-5.
represents any
and/or tab).

arg

legal

separator

(comma,

represents the argument associated
immediate
conditional
directive,
expression, symbolic argument,
or
argument, as described in Table 6-5.

space,

with
the
i.e.,
an
macro-type

represents the separator between the conditional
argument
and
the
statement field.
If the
preceding argument is an expression, then a comma
must be used; otherwise, a comma, space, and/or
tab may be used.
statement

represents the specified statement to be assembled
if the condition is satisfied.

For example, the immediate conditional statement:
.IIF

DF FOO,BEQ ALPHA

6-45

GENERAL ASSEMBLER DIRECTIVES

generates the code
SEQ
if the symbol

ALPHA

Foa is defined within the source program.

As with the .IF directive, a condition test other than those listed in
Table 6-5, an illegal argument, or a null argument specified in an
.IIF directive results in an error code CA) in the assembly listing.

6-46

CHAPTER 7
MACRO DIRECTIVES

7.1

DEFINING MACROS

In assembly-language programming, it is often convenient and desirable
to generate a recurring coding sequence by invoking a single statement
within the program. In order to do this, the desired coding sequence
is first established with dummy arguments as a macro definition. Once
a macro has been defined, a single statement calling the macro by name
with a list of real arguments
(replacing the corresponding dummy
arguments in the macro definition) generates the desired coding
sequence. This sequence is called the macro expansion.

7.1.1

.MACRO Directive

The first statement of a macro definition must be a .MACRO
This directive takes the form:
.MACRO

label:
where:

directive.

name, dummy argument list

label

represents an optional statement label.

name

represents the programmer-assigned symbolic name
of the macro. This name may be any legal symbol
and may be used as a label elsewhere in the
program.
represents any
and/or tab).

dummy
argument
list

legal

separator

(comma,

space,

represents a number of legal symbols (see 3.2.2)
that may appear anywhere in the body of the macro
definition, even as a label. These dummy symbols
can be used elsewhere in the program with no
conflict of definition. Multiple dummy arguments
specified in this directive may be separated by
any legal separator. The detection of a duplicate
or an illegal symbol in a dummy argument list
terminates the scan and causes an error code to be
generated.

A comment may follow the dummy argument list in a .MACRO directive, as
shown below:
• MACRO

ABS

A,B

iDEFINES MACRO ABS WITH TWO ARGUMENTS •

7-1

MACRO DIRECTIVES
NOTE
Although it is legal for a label to
appear
on a
. MACRO directive, this
practice is discouraged, especially in
the case of nested macro definitions,
because
invalid.
labels
or
labels
constructed
with
the
concatenation
character will cause the macro directive
to be ignored.
This may result in
improper termination
of
the
macro
definition.
This NOTE also applied to
.IRP, .IRPC, and .REPT.

7.1.2

.ENDM Directive

The final statement of
directive of the form:
.ENDM
where:

every

definition

must

.ENDM

name
represents an optional argument specifying the
symbolic name of the macro being terminated by the
directive, as shown in the following example:

name

.ENDM
.ENDM

macro

iTERMINATES THE CURRENT
iMACRO DEFINITION.
ABS

iTERMINATES THE CURRENT
iMACRO DEFINITION NAMED ABS.

If specified, the symbolic name in the .ENDM statement must match the
name specified in the corresponding .MACRO directive. Otherwise, the
statement is flagged with an error code (A) in the assembly listing
(see Appendix D).
In either case, the current macro definition is
terminated. Specifying the macro name in the .ENDM statement thus
permits MACRO to detect missing .ENDM statements or improperly-nested
macro definitions.
The .ENDM directive may be followed by a comment field, but
contain a label, as shown below:
. MACRO
JSR
• WORD
.ENDM

TYPMSG MESSGE
RS ,TYP~ISG
MESSGE

must

not

iTYPE A MESSAGE •

iEND OF TYPMSG MACRO.

An .ENDM statement encountered by MACRO outside a macro definition is
flagged with an error code (O) in the assembly listing (see Appendix
D) •

NOTES
1.

Labels on .ENDM directives are ignored.

Illegal labels will cause the
to be bypassed.

7-2

directive

MACRO DIRECTIVES

7.1.3

.MEXIT Directive

The .MEXIT directive may be used to terminate a macro expansion before
the end of the macro is encountered. This directive is also legal
within repeat blocks (see Sections 7.6 and 7.7). It is most useful in
the context of nested macros. The .MEXIT directive terminates the
current macro as though an .ENDM directive had been encountered.
Using the .MEXIT directive bypasses the complexities of nested
conditional directives and alternate assembly paths, as shown in the
following example:
.MACRO

ALTR N,A,B

• IF EQ

iSTART CONDITIONAL ASSEMBLY BLOCK •

• MEXIT
• ENDC

iTERMINATE MACRO EXPANSION •
iEND CONDITIONAL ASSEMBLY BLOCK •

• ENDM

iNORMAL END OF MACRO •

Considering the above macro, in an assembly where the real argument
for the dummy symbol N is equal to zero
(see Table 6-5), the
conditional block would be assembled, and the macro expansion would be
terminated by the .MEXIT directive. When macros are nested, a .MEXIT
directive causes an exit to the next higher level of macro expansion.
A .MEXIT directive encountered outside a macro definition
with an error code (0) in the assembly listing.

7.1.4

flagged

MACRO Definition Formatting

A form-feed character used within a macro definition causes a page
eject during the assembly of the macro definition. A page eject,
however, is not performed when the macro is expanded.
Conversely, when the .PAGE directive is specified within a macro
definition, it is ignored during the assembly of the macro definition,
but a page eject is performed when that macro is expanded.

7.2

CALLING MACROS

A macro definition must be established by means of the
• MACRO
directive
(see Section 7.1.1) before the macro can be expanded within
the source program. Macro calls are of the general form:
label:
where:

name

real arguments

label

represents an optional statement label.

name

represents the name of the macro, as specified
the .MACRO directive (see Section 7.1.1).

7-3

MACRO DIRECTIVES
real
arguments

represent
symbolic
arguments
which
replace
the dummy arguments specified in the
• MACRO
directive. When multiple arguments are specified,
they are separated by any
legal
separator.
Arguments
to the macro call are treated as
character strings whose usage is determined by the
macro definition.
Note that MACRO accepts the
ASCII value of lower-case alphabetic characters
when .ENABL LC has been specified.

When a macro name is the same as a user label, the appearance of the
symbol in the operator field designates the symbol as a macro calli
the appearance of the symbol in the operand field designates it as a
label, as shown below:
ABS:

7.3

MOV

(RO) ,Rl

iABS IS DEFINED AS A LABEL.

ABS

iABS IS CONSIDERED TO BE A LABEL.

ABS

#4,ENT,LAR

iABS IS A MACRO CALL.

ARGUMENTS IN MACRO DEFINITIONS AND MACRO CALLS

Arguments within a macro definition or macro call are separated from
other arguments by any of the legal separating characters described in
Section 3.1.1.
Macro definition arguments (dummy) and macro call arguments
(real)
normally maintain a strict positional relationship.
That is, the
first real argument in a macro call corresponds with the first dummy
argument in a macro definition. Only the use of keyword arguments in
a macro call can override this correspondence (see Section 7.3.6).
For example, the following macro definition and its
expansion contain multiple arguments:
.MACRO

REN A,B,C

REN

ALPHA,BETA,<CI,C2>

associated

macro

Arguments which themselves contain separating characters must be
enclosed in paired angle brackets, as shown above.
For example, the
macro call:
REN

<MOV

X,Y>,#44,WEV

causes the entire expression
MOV

X,Y

to replace all occurrences of the symbol A in the macro definition.
Real arguments within a macro call are considered to be character
strings and are treated as a single entity during the macro expansion.

7-4

MACRO DIRECTIVES
The up-arrow (A) construction is provided to allow angle brackets to
be passed as part of the argument. This construction, for example,
could have been used in the above macro call, as follows:
REN

A/<MOV X,Y>/,i44,WEV

causing the entire character string
argument.

passed

The following macro call:
A
REN
i44,WEV /MOV X,Y/
however, contains only two arguments (i44 and WEVA/MOV X,Y/), because
the up-arrow is a unary operator (see Section 3.1.3) and it is not
preceded by an argument separator.
As shown in the examples above, spaces can be used within bracketed
argument constructions to increase the legibility of such expressions.

7.3.1

Macro Nesting

The nesting of macros, where the expansion of one macro includes a
call to another, causes one set of angle brackets in the macro
definition to be removed from an argument with each nested call.
The
depth of nesting allowed is dependent - upon the amount of dynamic
memory used by the source program being assembled.
To pass an argument containing legal argument delimiters to nested
macros, the argument in the macro definition should be enclosed within
one set of angle brackets for each level of nesting, as shown in the
coding sequence below.
It should be noted that this extra set of
angle brackets for each level of nesting is required in the macro
definition, not in the macro call.
• MACRO
LEVEL2
LEVEL2
.ENDM

LEVELl DUMI,DUM2
<DUMI>
<DUM2>

• MACRO
DUM3
ADD
"MOV
.ENDM

LEVEL2 DUM3
iIO,RO
RO, (Rl) +

A call to the LEVELl macro, as shown below, for example:
LEVELl

<MOV

X,RO>,<MOV

R2,RO>

causes the following macro expansion to occur:
MOV
ADD
MOV
MOV
ADD
MOV

X,RO
ilO,RO
RO,(Rl)+
R2,RO
ilO,RO
RO, (Rl)+

7-5

MACRO DIRECTIVES
When macro definitions are nested, i.e., when a macro definition is
contained entirely within the definition of another macro, the inner
definition is not a callable macro until the outer macro has been
called and expanded.
For example, in the following coding:
• MACRO

LVI A,B

• MACRO

LV2 C

.ENDM
.ENDM
the LV2 macro cannot be called and expanded until the LVI macro has
been so invoked.
Likewise, any macro defined within the LV2 macro
definition cannot be called and expanded until LV2 has also been
invoked.

7.3.2

Special Characters in Macro Arguments

An argument may include special characters without enclosing them in a
bracketed construction if that argument does not contain spaces, tabs,
semicolons, or commas.
For example, the macro definition:
• MACRO
MOV
.ENDM

PUSH ARG
ARG,-(SP)

PUSH

X+3(%2)

causes the following code to be generated:
MOV

7.3.3

X+3(%2),-(SP)

Passing Numeric Arguments as Symbols

When macro arguments are passed, an absolute symbol value can be
passed which is treated by the macro as a numeric string. An argument
preceded by the unary operator backslash (\) is treated as a numer1C
value in the current program radix. The ASCII characters representing
this value are inserted in the macro expansion, and their function is
defined in the context of the resulting code, as shown in the
following example:
. MACRO
CON

INC A,B
A,\B

B=B+l
A'B:

C=o

. ENDM
. MACRO
• WORD
.ENDM

INC

;B IS TREATED AS A NUMBER IN CURRENT
;PROGRAM RADIX •

CON A,B
4

;A'B IS DESCRIBED IN SECTION 7.3.6 •

X,C

7-6

MACRO DIRECTIVES
The above macro call (INC) would thus expand to:
xO:

• WORD

Note in this expanded code that the label XO:
is the result of the
concatenation of two real arguments.
The single quote (I) character
'in the label AlB:
causes the real arguments X and a to be
concatenated as they are passed during the expansion of the macro.
This type of argument construction is described in further detail
in
Section 7.3.6.
A subsequent call to the same macro would generate the following code:
Xl:

• WORD

and so on, for later calls. The two macro definitions are necessary
because the symbol associated with dummy argument B (i.e., C) cannot
be updated in the CON macro definition, because its numeric value has
already been substituted for its symbolic name, i.e., the character a
has replaced C in the argument string.
In the CON macro definition,
the number passed is treated as a string argument.
(Where the value
of the real argument is 0, only a single a character is passed to the
macro expansion.)
Passing numeric values in this manner is useful in identifying source
listings.
For
example,
versions of programs created through
conditional assemblies of a single source program can be identified
through such coding as that shown below. Assume, for example, that
the symbol ID in the macro call (IDT) has been equated elsewhere in
the source program to the value 6 •
• MACRO
• IDENT
.ENDM

lOT SYM
/V05A 1 SYM/

IDT

\ID

iASSUME THAT THE SYMBOL ID TAKES
iON A UNIQUE 2-DIGIT VALUE •
iWHERE VaSA IS THE UPDATE
iVERSION OF THE PROGRAM.

The above macro call would then expand to:
.IDENT

/V05A6/

where 6 is the numeric value of the symbol 10.

7.3.4

Number of Arguments in Macro Calls

If more arguments appear
in the macro call than in the macro
definition,
an error code (Q) is generated in the assembly listing.
If fewer arguments appear
in the macro call than in the macro
definition, missing arguments are assumed to be null values.
The
conditional directives .IF Band .IF NB (see Table 6-5)
can be used
within the macro to detect missing arguments. The number of arguments
can also be specified using the .NARG directive (Section 7.4.1). Note
that a macro can be defined with no arguments.

7.3.5

Creating Local Symbols Automatically

A label is often required in an expanded macro.

macro

facilities

In· the conventional
thus far described, such a label must be explicitly

7-7

MACRO DIRECTIVES
specified as an argument with each macro call. Be careful in issuing
subsequent calls to the same macro, to avoid specifying a duplicate
label as a real argument. This concern can be eliminated through a
feature of MACRO which creates a unique symbol where a label is
required in an expanded macro.
As noted in Section 3.5, MACRO can automatically create local symbols
of the form n$, where n is a decimal integer within the range 64
through 127, inclusive. Such local symbols are created by MACRO in
numerical order, as shown below:
64$
65$

126$
127$
This automatic facility is invoked on each call of a macro whose
definition contains a dummy argument preceded by the question mark (?)
character, as shown in the macro definition below:
.MACRO

ALPHA, A,?B

TST
BEQ
ADD

A
B
i5,A

;CONTAINS DUMMY ARGUMENT B PRECEDED BY
;QUESTION MARK.

B:
.ENDM
A local symbol is generated automatically by MACRO only when a real
argument of the macro call is either null or missing, as shown in
Example 1 below, which reflects the expansion of the ALPHA macro
defined above.
If the real argument is specified in the macro call, however, MACRO
inhibits the generation of a local symbol and normal argument
replacement occurs, as shown in Example 2 below.
EXAMPLE 1:

Generate a Local Symbol for the Missing Argument:

ALPHA
TST
BEQ
ADD

Rl
Rl
64$
i5,Rl

; SECOND ARGUMENT IS MISSING.
iLOCAL SYMBOL IS GENERATED.

64$:
EXAMPLE 2:

Do Not Generate a Local Symbol:

ALPHA
TST
BEQ
ADD

R2,XYZ
R2
XYZ
i5,R2

;SECOND ARGUMENT XYZ IS SPECIFIED.
iNORMAL ARGUMENT REPLACEMENT OCCURS.

XYZ:
Automatically-generated local symbols are
16(10) arguments of a macro definition.

restricted

the

first

Note that automatically-created local symbols resulting local symbols
from the expansion of a macro, as described above, do not in any way
influence local symbol block boundaries.
In other words,
such
automatically-created local symbols do not establish a local symbol
block in their own right.

7-8

MACRO DIRECTIVES
However, when a macro has several arguments earmarked for
automatic
local symbol generation,
substituting a specific label for one such
argument introduces a risk that assembly errors will result.
This
is
because MACRO constructs its argument substitution list at the point
of macro invocation. Therefore, the appearance of any label,
the
.ENABL LSB directive, or the .PSECT directive, in the macro expansion
will create a new local symbol block. This could leave local symbol
references in the previous block and the symbol definitions in the new
one, resulting in error codes in the assembly listing
(see Appendix
D).
Furthermore,
a subsequent macro expansion that generates local
symbols in the new block may duplicate one of the symbols in question,
resulting in an additional error code (P) in the assembly listing.

7.3.6

Keyword Arguments

Macros may be defined with and/or invoked with keyword
keyword argument has the following form:

arguments.

name=string
where
name

represents the dummy argument,

string

represents the real symbolic argument.

The keyword argument may not contain embedded argument
unless properly delimited as described in section 7.3.

separators

When a keyword argument appears in the dummy argument list of a macro
definit~on,
the specified string becomes the default real argument at
macro call.
When a keyword argument appears in the real argument list of a macro
call,
the specified string becomes the real argument for the dummy
argument that exactly matches the specified name, whether or not the
dummy argument was defined with a keyword.
If a match fails, the
entire argument specification is treated as the next positional real
argument.
A keyword argument may be specified anywhere in the dummy
argument list of a macro definition and is part of the positional
ordering of argument.
On the other hand, a keyword argument may be
specified anywhere in the real argument list of a macro call but does
not affect the positional correspondence of the remaining arguments.
.LIST

1
2
3
4
5
6

DEFINE A MACRO HAVING KEYWORDS IN DUMMY ARGUMENT LIST
;

• MACRO
• WORD
• WORD
• WORD
.ENDM

7
8
9
10
11

12
13
14
15
16
17 000000
000000

TEST CONTRL=l,BLOCK,ADDRES=TEMP
CONTRL
BLOCK
ADDRES

;

NOW INVOKE SEVERAL TIMES

OOOOOOG

TEST
• WORD

A,B,C

7-9

MACRO DIRECTIVES
000002
000004
18
19 000006
000006
000010
000012
20
21 000014
000014
000016
000020
22
23 000022
000022
000024
000026
24
25 000030
000030
000032
000034
26
27 000036
000036
000040
000042
28
29
30

7.3.7

OOOOOOG
OOOOOOG

• WORD
• WORD

B
C

000040
000030
000020

TEST
• WORD
• WORD
• WORD

ADDRES=20,BLOCK=30,CONTRL=40
40
30
20

000001
000005

TEST
• WORD
• WORD
• WORD

BLOCK=5
1
5
TEMP

TEST
• WORD
• WORD
• WORD

CONTRL=5,ADDRES=VARIAB
5

OOOOOOG
000005
000000

OOOOOOG

TEST
• WORD
. WORD
• WORD

000001
000000
ooooooe

TEST
• WORD
• WORD
• WORD

000001

.END

000001
000000

VARIAB
1
TEMP
ADDRES=JACK!JILL
1
JACK!JILL

Concatenation of Macro Arguments

The apostrophe or single quote character
(') operates as a legal
delimiting character in macro definitions.
A single quote that
precedes and/or follows a dummy argument in a macro definition is
removed, and the substitution of the real argument occurs at that
point. For example, in the following statements:
A'B:

• MACRO
.ASCIZ
.BYTE
.ENDM

DEF A,B,e
/e/
' 'A,' 'B

when the macro DEF is called through the statement:
DEF

X,Y,<MACRO>

it is expanded, as follows:
XY:

.ASCIZ
.BYTE

/MACRO/
'X, 'y

In expanding the first line, the scan for the first argument
terminates upon finding the first 'character. Since A is a dummy
argument, the ' is removed. The scan then resu~es with Bi B is also
noted as another dummy argument. The two real arguments X and Yare
then concatenated to form the label XY:. The third dummy argument is
noted in the operand field of the .ASCIZ directive, causing the real
argument MACRO to be substituted in this field.

7-10

MACRO DIRECTIVES
When evaluating the arguments to the .BYTE directive during expansion
of the second line, the scan begins with the first' character. Since
it is neither preceded nor followed by a dummy argument, this
character remains in the macro expansion. The scan then encounters
the second' character, which is followed by a dummy argument and is
therefore discarded.
The scan of argument A is terminated upon
encountering the comma (,). The third' character is neither preceded
nor followed by a dummy argument and again remains in the macro
expansion. The fourth (and last) , character is followed by another
dummy
argument and is likewise discarded.
(Note that four
characters were necessary in the macro definition to generate two
characters in the macro expansion.)

7.4

MACRO ATTRIBUTE DIRECTIVES:

.NARG, .NCBR, AND .NTYPE

Three directives are available in MACRO which allow the user to
determine certain attributes of macro arguments. The use of these
directives permits selective modifications of a macro expansion,
depending on the nature of the arguments being passed.
These
directives are described separately below.

7.4.1

.NARG Directive

The .NARG directive is used to determine the number of arguments in
the macro call currently being expanded. Hence, the .NARG directive
can appear only within a macro definition;
if it does not, an error
code
(0)
is generated in the assembly listing. This directive takes
the form:
label:
where:

.NARG

symbol

label

represents an optional statement label.

symbol

represents any legal symbol.
This symbol is
equated to the number of arguments in the macro
call currently being expanded. If a symbol is not
specified, the .NARG directive is flagged with an
error code (A) in the assembly listing.

An example of the .NARG directive follows:

7-11

MACRO DIRECTIVES
t
2
3

.TITLF

.MARt;

Sy~

.IF

r.t;I,~y,..

6
1

.MEXIT

.MAc~n

NA~G

"OPP,t.JI)M

.T~F

.~EPT

NU~

N('\I'
.ENDM
.ENOC
.EN/)M

10
\t
\2
13
\4

1'5 ot9l000P1

~opp

.NARG

SVM

.IF

EQ,SY~

.MExtT

.IFF
.REPT
NOf)
.EN['I~

.~~OC

16
11
18 0~A'H"A

N(')PP

• ~'ARG

SYM
EO,SYt.1

.J~
.~FxtT

.TFF
.~EPT

Nnp

.E'N"M
00000~

0~"'''''2

00000a

00(l10~6

000P110
01;11l21t;'1\2

0V1e24~

0101"2401

0I0IClI2aOl
POl0l2al2l
00,02aCll
0I0I02a'"

N(')P
Nt'lP
N~P

NOP
N("IP

N"P
.E~DC

000001
7.4.2

.EN"

.NCBR Directive

The .NCHR directive, which can appear anywhere in a MACRO program,
is
used to determine the number of characters in a specified character
string. This directive, which is useful in calculating the length of
macro arguments, takes the following form:
label:
where:

.NCHR

syrnbol,(string>

label

represents an optional statement label.

symbol

represents any legal symbol.
This symbol is
equated
to the number of characters in the
specified character string. If a symbol is not
specified, the .NCHR directive is flagged with an
error code
(A)
in the assembly listing
(see
Appendix 0).

7-12

MACRO DIRECTIVES
represents any
and/or tab).
<string>

legal

separator

(comma,

space,

represents a string of printable characters.
The
character string need be enclosed within angle
brackets
or up-arrows
(~)
only if
the
specified
character
string contains a legal
separator (comma, space, and/or tab).
If the
delimiting characters do not match or if the
ending delimiter cannot be detected because of a
syntactical error in the character string (thus
prematurely terminating its evaluation), the .NCHR
directive is flagged with an error code (A) in the
assembly listing.

«»

An example of the .NCHR directive follows:

1
2

.TtTLF. NC-H~

3
IJ

.lo1ACPO
• ~;CHQ

CHA'l,MESS
SYM,Mf:SS

.WOR!,)

SV~

.ASCII
.EVEN

IMESSI

.EN,,'"

tl
Q

1('1
11 PI"'~OI"''''

MSGlt
"''''~~~c;

000L'1t'10

",01"0105

00~e~?

t 1~

0~0I0P.3

1'''5

0"'~t'I~"

'1a

0Q'~~~5

"''''0010&

C~A~

....!C~p

<~F.LLO>

SVM,tofE'LLC'

.wn~D

Sv~

.!SCYr

IHFLLOI

t,a

1\7
.eVE~

12
t3
llJ

7.4.3

"'''''''~0!1

.EN'"

.NTYPE Directive

The .NTYPE directive is used to determine the addressing mode of a
specified macro argument. Hence, the .NTYPE directive can appear only
within a macro definition; if it appears elsewhere, it is flagged
with an error code (0) in the assembly listing. This directive takes
the form:
label:

.NTYPE

symbol,aexp

where:

label

represents an optional statement label.

symbol

represents any legal symbol.
This symbol is
equated to the 6-bit addressing mode of the
following argument. If a symbol is not specified,
the .NTYPE directive is flagged with an error code
(A) in the assembly listing.
represents any
and/or tab).

7-13

legal

separator

(comma,

space,

MACRO DIRECTIVES
represents any legal address expression, as used
with an opcode.
If no argument is specified, the
result will be zero.

aexp

An example of the use of an .NTYPE directive in a macro definition
shown below:
1
2

.TYTLE NTVPE

.~Ar.R(l

SAVE',A~G

u
5
6
7
8
q
10
t1
12

.~TVPE

SVM,ARG

.IF

F~,SVMlt7~

~Ov

.~G,.(SP'

,~EGISTE~

.TFF
M"V
.ENDC

*ARG,-(SP)

,NON-REGISTER

MODE
"~O"E

.E~"M

P10Q11?1~0

",,,,,,,01,,,01

TF.~Pt

• IrJ

OR I')

1"
15
16 QI"''''~0I2

0"''''''''''2

'"
Rt

0100010'

SAVE
.t.JTVPF:'

"'lp1a~

.TF
Mnv

~l,-(SP)

,~EGISTr:R

.IFF
a.10V

.Rt,-eSp)

,t..iON-REGISTFR M('IOE

SAVE'
• t.:TvPE

SVM,TE~P

.eNnc

Sv~,~t
F:Q,SV~~71l'

'40["lE'

18
tQ "'''''0010a
"''''0I~~7

0"'0p(1Ia

t'l127U~

TE~P

.IF

EQ,SV~1t.7~

~t"V

TFMP,-~~P)

J~E'C:ISTER

.TFF
Mnv

*TEMP,_(SP)

, t.J(H·J- RFG I ST FR

MnOF
M(,OE

~t'lQ!"OIP'

.EN"'C
20

21
22

PI"'0~01

.FNO

For additional information concerning
Chapter 5 and Appendix B, Section B.2.

7.5

addressing

modes,

refer

.ERROR AND .PRINT DIRECTIVES

The .ERROR directive is used to output messages to the listing file
during assembly pass 2. A common use of this directive is to provide
a diagnostic announcement of a rejected or erroneous macro call or to
alert the user to the existence of an illegal set of conditions
specified in a conditional assembly.
If the listing file
is not
specified, the
.ERROR messages are output to the command output
device. The .ERROR directive takes the form:
label:

.ERROR

where:

label

expr

;text
represents an optional statement label.

7-14

MACRO DIRECTIVES
expr

represents an optional expression whose value is
output when the .ERROR directive is encountered
during assembly.
denotes the beginning of the text string.

text

represents the specified message
the .ERROR directive.

Upon encountering an .ERROR directive anywhere in
MACRO outputs a single line containing:

associated
source

An error code (P)

The sequence number of the .ERROR directive statement

The value of the current location counter

The value of the expression, if one is specified

The source line containing the .ERROR directive.

with

program,

For example, the following directive:
.ERROR

iINVALID MACRO ARGUMENT

causes a line in the following form to be output to the listing file:

Seq. Loc.
No. No.

Exp.
Value

512 005642

000076

Text
.ERROR

iINVALID MACRO ARGUMENT

The .PRINT directive is identical in function to the .ERROR directive,
except that it is not flagged with the P error code.

7.6

INDEFINITE REPEAT BLOCK DIRECTIVES:

.IRP AND .IRPC

An indefinite repeat block is a structure that is similar to a macro
definitioni
essentially a macro definition that has only one dummy
argument. At each expansion of the indefinite repeat range,
this
dummy argument is replaced with successive elements of a specified
real argument list. An indefinite repeat block directive and its
associated repeat range are coded in-line within the source program.
This type of -macro definition and expansion does not require calling
the macro by name, as required in the expansion of conventional macros
previously described in this section.
An indefinite repeat block can appear either within or outside another
macro definition,
indefinite repeat block, or repeat block (see
Section 7.7).
The rules for specifying indefinite repeat block
arguments are the same as for specifying macro arguments (see Section
7.3).

7-15

MACRO DIRECTIVES

7.6.1

.IRP Directive

The .IRP directive is used to replace a dummy argument with successive
real arguments specified in an argument string. This replacement
process occurs during the expansion of an indefinite repeat block
range. This directive takes the following form:
label:

.IRP

sym,<argument list>

(range of indefinite repeat block)

.ENDM
where:

label

represents an optional statement label.

sym

represents a dummy argument that is successively
replaced
with
the
specified real arguments
enclosed within the angle brackets. If no dummy
argument is specified, the .IRP directive is
flagged with an error code
(A)
in the assembly
listing.
represents any
and/or tab).

legal

separator

(comma,

space,

represents a list of real arguments enclosed
within angle brackets that is to be used in the
expansion of the indefinite repeat range. A real
argument may consist of one or more characters;
multiple arguments must be separated by any legal
separator
(comma, space, and/or tab).
If no real
arguments are specified, no action is taken.

range

represents the block of code to be repeated once
for each occurrence of a real argument in the
list.
The range
may
contain
other
macro
definitions
and
repeat
ranges.
The .MEXIT
directive (see Section 7.1.3) is legal within the
range of an indefinite repeat block.

.ENDM

indicates the end of the indefinite
range.

repeat

block

An example of the use of the .IRP directive is shown in Figure 7-1.

7.6.2

.IRPC Directive

The .IRPC directive is available to
permit
single
character
substitution, rather than argument substitution. On each iteration of
the indefinite repeat range, the dummy argument is replaced with each
successive character in the specified string. The .IRPC directive is
specified as follows:
label:

.IRPC

sym,<string>

(range of indefinite repeat block)

7-16

MACRO DIRECTIVES

.ENDM
where:

label

represents an optional statement label.

sym

represents a dummy argument that is successively
replaced
with
the
specified real arguments
enclosed within the angle brackets. If no dummy
argument is specified, the .IRPC directive is
flagged with an error code
(A)
in the assembly
listing.
represents any
and/or tab).

legal

separator

(comma,

space,

represents a list of characters enclosed within
angle brackets to be used in the expansion of the
indefinite repeat range.
Although the
angle
brackets
are
required only when the string
contains separating characters, their use
is
recommended for legibility.

range

represents the block of code to be repeated once
for each occurrence of a character in the list.
The range may contain macro definitions and repeat
ranges.
The .MEXIT directive (see Section 7.1.3)
is legal within the range of an indefinite repeat
block.

.ENDM

indicates the end of the indefinite
range.

repeat

block

An example of the use of the .IRPC directive is shown in Figure 7-1.
1

• TITLE

2
J

.LIST

IRPTST

4
5

ClJ00GlIGUlG
01110P100G
001000G
0P10000G
1Il11l00t1110G
01110000G

, IMP
MOV
• END'"
MOV
MOV
MOV
MOV
MOV
MOV

111I11000G
000000G
100100G
11I00000G
00111011.1G
10000111G

,IMPC
Mova
.ENDM
MOVe
Mova
Mova
Mova
Mova
... ova

6
7
8

00111000
0~0004
PI;J)0~10

000014
1/100020
0I1l0024

016720
016720
01672111
016720
01672111
016720

X,c,.,B8,CC,DD,EE,FF~

X,(R0) •
&A, eR0).
BR,(M0).
ee,(NI).
DD,(R0).
EE,(NI).
FF,(R0).

1IIl

11
12
13
o IlH/JII1 30
IUI/l034
11.10111040
000044
00P11!1~0

000054

112741
112741
112741
112741
112741
112741

x,cAaCDEF~

,IX,-(Rn
u,-o~t)

.B,-CRt)
.e, - CR1)
'D,-CRt)
'E,-(Rt)
IF,-(Nl)

14
15

,END

000001'

Figure 7-1

Example of .IRP and .IRPC Directives

7-17

MACRO DIRECTIVES
7.7

REPEAT BLOCK DIRECTIVE:

.REPT, .ENDR

It is sometimes useful to duplicate a block of code a number of times
in-line
with other source code.
This duplication of code is
accomplished by creating a repeat block, using a directive in the
form:
label:

.REPT

exp

(range of repeat block)

.ENDM
where:

7.8

label

represents an optional statement label.

exp

represents any legal expression
whose
value
controls the number of times the block of code is
to be assembled within the program.
When the
expression value is less than or equal to zero
(0), the repeat block is not assembled.
If this
expression is not an absolute value, the .REPT
statement is flagged with an error code (A) in the
assembly listing.

range

represents the block of code to be repeated the
number
of times determined by the specified
expression value. The repeat block may contain
macro definitions,
indefinite repeat blocks, or
other repeat blocks.
The .MEXIT directive is
legal within the range of a repeat block.

.ENDM
or
• ENDR

indicates the end of the repeat block range. The
terminating statement in a repeat block can be
either an .ENDM directive or an .ENDR directive •

MACRO LIBRARY DIRECTIVE:

• MCALL

The .MCALL directive allows you to indicate in advance those system
and/or user-defined macro definitions that are required in the
assembly of the source program. The .MCALL directive allows you to
specify the names of all system or user macro definitions not defined
within the source program but which are required to assemble the
program. The .MCALL directive must appear before the first occurrence
of a call to any externally-defined macro. The .MCALL directive is of
the form:
.MCALL
where:

argl,arg2, ••• argn

argl,
arg2, •.•
argn

represent the symbolic names of the macro
definitions required in the assembly of the source
program.
The symbolic macro
names
may
be
separated by any legal separator (comma, space,
and/or tab).

The • MCALL directive thus provides the means to
access
user-defined and system macro libraries during assembly.

7-18

both

MACRO DIRECTIVES

The /LIBRARY qualifier is specified in connection with an input file
specification, to indicate to MACRO that the file is a macro library.
When a macro call is encountered in the source program, MACRO first
searches the user macro library for the named macro definitions, and,
if necessary, continues the search with the system macro library.
Any number of such user-supplied macro files may be designated.
In
cases of multiple library files,
the search for the named macros
begins with the last such file specified.
The search continues in
reverse
order until the required macro definitions are found,
terminating again, if necessary, with a search of the system macro
library.
If any named macro is not found upon completion of the search,
i.e.,
if the macro is not defined, the .MCALL statement is flagged with an
error code (U) in the assembly listing.
Furthermore, a statement
elsewhere in the source program which attempts to expand such an
undefined macro is flagged with an error code
(0)
in the assembly
listing.
The TRAX command strings to MACRO, through which a file specification
is supplied, are described in detail in Sections 8.1.2 and 8.2.2,
respectively.

7-19

MACRO DIRECTIVES

PART IV
OPERATING PROCEDURES

CHAPTER 8
OPERATING PROCEDURES

MACRO assembles one or more ASCII source files containing MACRO
statements into a single relocatable binary object file. The output
of MACRO consists of a binary object file and a file containing the
table of contents listing, the assembly listing, and the symbol table
listing. An optional cross-reference listing of symbols and macros is
available. A sample assembly listing is provided in Appendix I.

8.1

TRAX OPERATING PROCEDURES

The following sections describe the use of MACRO under TRAX.

8.1.1

Invoking MACRO Under TRAX

The MACRO command is used under TRAX to begin MACRO assembler
operations.
The command causes MACRO to assemble one or more ASCII
source files containing MACRO statements into a relocatable binary
object file.
The assembler will also produce an assembly listing,
followed by a symbol table listing.
A cross-reference listing can
also be produced, by means of the /CROSSREFERENCE qualifier (see
8.2.2, below).
The command can be issued whenever the TRAX Support Environment is at
command level in interactive mode. The MACRO command can be input
either directly from the terminal (interactive mode) or from a batch
file
(batch mode). When the specified assembly has completed, MACRO
terminates operations and returns control to the Support Environment.
(Refer to the TRAX Support Environment Userls Guide for further
information about interactive and batch mode operations.)

8.1.2

TRAX Command String Format

A MACRO command string can be specified using either an interactive
mode or a batch mode.
A MACRO command string under TRAX has the
following format:
MACRO[/QUALIFIER[S]]

FILESPEC[/QUALIFIER[S]]

where:
MACRO - is the invoking command.
[/QUALIFIER[S]] - are the optional qualifiers.

8-1

[+FILESPEC+ ••• ]

OPERATING PROCEDURES
FILESPEC -

is the MACRO source file to be assembled.

[/QUALIFIER[S]] - is the optional qualifiers for the source file.
[+FILESPEC+] - is the additional source files to be assembled.
If the FILESPEC parameter is not given the system
the prompt:

will

respond

with

qualifiers

are

entered

and

The TRAX Macro Qualifiers direct the MACRO assembler
source file with the following options:

process

the

)FILES?
The desired file names
processing continues.

8.1.3

and

optional

TRAX Macro Qualifiers

QUALIFIERS

SPECIFIES ONE OR MORE OF THE FOLLOWING:

output
/OBJECT[:filespec]

Produce an object file as specified by
filespec (see Section 8.3). The default
is a file with the same filename as the
last named source file and an .OBJ
extension.
/OBJECT
is
always the
default condition.

/NOOBJECT

Do not produce an object file.

output
/LIST[:filespec]

Produce
an
assembly
listing
file
according to filespec (see Section 8.3).
If
filespec
is not specified,
the
listing is printed on the line printer.
The default is /NOLIST.

/NOLIST

Do not produce a listing file.
The
default in interactive mode is /NOLIST
and in batch mode is /LIST.
NOTE
When no listing file is specified, any
errors encountered in the source program
are displayed at the terminal from which
MACRO was initiated.

/CROSSREFERENCE[:argl ••• arg4]
Produce a cross-reference listing. Argl
through arg4 are as described in Section
8.1.5.
FUNCTION

/SWITCHES
/LI:arg
/NL:arg

Listing control switches;
these options
accept ASCII switch values
(arg) which
are equivalent in function and name to
and override the arguments of the .LIST

8-2

OPERATING PROCEDURES
and .NLIST directives specified in the
source program (see Section
6.1.1).
This switch overrides the arguments and
remains in effect
for
the
entire
assembly process.
Function control switches; these options
accept ASCII switch values
(arg) which
are equivalent in function and name to
and override the arguments of the .ENABL
and .DSABL directives specified in the
source program (see Section 6.2).
This
switch
overrides
the arguments and
remains in effect
for
the
entire
assembly process.

/EN:arg
/DS:arg

TRAX accepts the MACRO or $MACRO command as input and initializes the
MACRO assembler, which in turn processes the specified files according
to the options indicated in the command string. When the operation is
complete, MACRO returns control to PDS to obtain the next command line
either from the terminal or from the batch stream.

8.1.4

TRAX File Specification Oualifiers

The following
specification:
/PASS:N

/LIBRARY

optional

qualifiers

may

appended

file

If N=l, assemble the associated file
the first pass.

If N=2, assemble the associated file
the second pass.

specifies that an input file is a macro
library file. As noted in Section 7.8,
any macro that is defined externally
must
be
identified
in the • MCALL
directive before it can be retrieved
from a macro library file and assembled
with the user program.
In locating
macro definitions, MACRO initiates a
fixed search algorithm, beginning with
the last user macro file specified,
continuing in reverse order with each
such file specified, and terminating, if
necessary, with a search of the system
macro library file.
If a required macro
definition is not found upon completion
of the search, an error code (U) results
in the assembly listing
(see Appendix
D).
This means that a user macro
library file must be specified in the
command line prior to the source file(s)
that uses any macros defined in the
library file. If more than one library
file is specified, the libraries will be
searched in right-to-left order.

8-3

OPERATING PROCEDURES
8.1.5

Cross-Reference Processor (CREF)

The CREF processor is used to produce a listing that includes
cross-references to symbols that appear in the source program. The
cross-reference listing is appended to the assembly listing.
Such
cross-references
are helpful .in debugging and in reading long
programs.
A cross-reference listing can include up to four sections:
1.

User-defined symbols

Macro symbols

Permanent symbols

To generate a cross-reference listing, specify the /CR switch in
MACRO command string. Optional arguments can also be specified.
form of the switch is:

JCR

the
The

SYM
MAC
REG
PST

where:
SYM

specifies user-defined symbols (default)

MAC

specifies macro symbols (default)

REG

specifies register symbols

PST

specifies permanent symbols.

If you wish to generate listings for user-defined
only, simply use JCR. No argument is necessary.
However,
if an argument
is
specified,
only
cross-reference listing is generated. For example:

and

macro
that

symbols
type

/CR:SYM
produces a cross-reference listing of user-defined symbols only.
No
listing of macro symbols is generated. Thus, to produce all four
types of cross-reference listings, you must specify all four arguments
(the order in which they are specified is not significant). Use a
colon to separate arguments.
For example:
/CR:REG:SYM:MAC:PST
The CREF processor is more fully described in the Utilities
Manual supplied with your system.
Figure 8-1 illustrates a complete cross-reference listing.

8-4

Reference

OPERATING PROCEDURES
P!RMAN!NT SYMBOL TABLE C~OSS REFERENCE
SYMBOL

REFERENCES

ADD
Bec
Bes
BEQ;BISB
BIT
BIT8

2-227
2-241
2-271
2-203
2-237
2-235
2-250
2-232
2-225
2-282
2-207
2-304
2-2"'2
2-252
2-285
2-205
2-28\
2-204
2-245
2-268
2-31'"

810

BN!
BR
CLR
CLRB
C~P

INC
JMP
JSR

MOV

Move
RTS
SEC
TST
TSTB
.8lKB
.BlI<W
.BYTE
.END
.ENOe
.IDENT
.IF

.IFF
.lIST
.r.t'CRO
.NARG
.NlIST
.PSEtT
.RAD50
.TITLE
.WORD

2-23~

2-2t?
2-2C,2
2-246
2-29q
2-151
2-17
2-122
2-152
2-tbq
2-333
1-17q
1-3
1-174
2-16q
2-208
2-220
2-240
2-25e
2-281
2-104
1-t9q
1-46
2-40
2-205
2-311

I-I

2-15
2-79
1-2
2-135

C~E'

VIII

2-256

2-313

2-236

2-275

2-2C,5

2-251

2-273

2-302

2-264

2-267

2-280

2-2q8

2-224
2-253
2-331
2-208
2-311
2-206
2-247
2-276
2-312
2-239
2-320

2-231

2-272

2-3211

2-314

2-Z1Gl1

2-21 9

2-220

2-222

2-20q
2-248
2-277
2-31e
2-307

2-221
2-24q
2-278
2-32q

2-223
2-255
2-27q
2-330

2-226
2-260
2-284

2-~0

2-124
2-164
2-t10

2-81
2-125
2-t65
2-170

2-82
2-12b
2-t6'5

2-83
2-127
2-166

2-166

2-167

2-168

2-10 4

2-U~4

2-2G1'8
2-222
2-243
2-258
2-281
2-165

2-165
2-170
2-210
2-222
2-243
2-258
2-281
2-106

2-105
2-205
2-210
2-222
2-243
2-258
2-311
2-108

2-166
2-205
2-210
2-222
2-243
2-211
2-]11
2-110

1-75

1-83

t-fle

1-123

l-t45

2-208

2-210

2-22~

2-222

2-241

2-88
2-149
2-1
2-138

2-118
2-164

2-t2q
2-t65

2-U~6

2-161

2-13e,

2-140

2-141

2-142

2-300

2-S16

2-305
2-274

2-318
2-78
2-123
2-164
2-tb9
2-165
2-2
2-164
2-t69
2-208
2-220
2-243

Z-25e

2-281

Figure 8-1

2-11~

Sample CREF Listing

8-5

OPERATING PROCEDURES
8.1.6

TRAX MACRO In Batch Mode

MACRO command strings can be processed via the TRAX batch mode
facility.
The batch mode facility imposes no restrictions on MACRO.
Batch mode processing has the following format:
SUBMIT

FILESPEC

where:
SUBMIT

- is the command to invoke the batch facility.

FILESPEC

- is the file containing the MACRO command strings.

The file must have the following structure:
$JOB
MACRO
$DATA
SOURCE DATA
$EOD
$EOJ
where:
$JOB

- defines

MACRO

- invokes the MACRO assembler.

$ DATA

- specifies the data to be processed.

SOURCE DATA

- is the MACRO command strings.

$EOD

- specifies the end of data.

$EOJ

- specifies the end of the job.

For a more complete description of the TRAX batch processing facility
please see the TRAX System command Language Reference Manual.

8.1.7

TRAX Indirect Command Files

The indirect command file facility can be used with MACRO command
strings. This is accomplished by creating an ASCII file that contains
the desired command strings (or portions thereof) in the forms shown
in Section 8.1.2. When an indirect command file reference is used in
a MACRO command string, the contents of the specified file are taken
as all or part of the command string. An indirect command file
reference is specified in the form:
@filespec
where:

specifies that the name that
file.

filespec

is the file specification of a file (see Section 8.2)
that contains a command string. The default extension
for the file name is .CMD.

follows

indirect

An indirect command file reference must always be the rightmost
in the command.
8-6

entry

OPERATING PROCEDURES
8.2

TRAX FILE SPECIFICATION FORMAT

The general form for a file specification in TRAX systems is shown
below.
Detailed information is provided in the applicable system
user's guide or operating procedures manual (see Section 0.3 in the
Preface).
dev: [g,m]name.extiver
where:
dev:

is the name of the physical device where the desired
file resides. A device name consists of two characters
followed by a 1- or 2-digit device unit number
(octal)
and a colon
(e.g., DPl:, DKO:,OT3:). The default
device under TRAX is established initially by the
system manager for each user and can be changed through
the SET command, or is given a default value as
specified in Table 8-1 if none is specified.

[g,m]

is the User File Directory
(UFO) code.
This code
consists of a group number (octal), a comma (,), and an
owner (member) number (octal) all enclosed in brackets
([]). An example of a UFO code is:
[200,30].
The
default
Identification
lAS, this can
command.

UFO
is
equivalent
to
the
User
Code (UIC) given at log-in time. Under
be changed through the SET DEFAULT

name

is the filename and consists of one through nine
alphanumeric characters.
There is no default for a
filename •

• ext

is a 1- to 3-alphanumeric character filename extension
or type that is preceded by a period (.). An extension
is normally used to identify the nature of the file.
Default values depend on the context of the file
specification and are as follows:
.CMD
Indirect command (input) file
.LST
A listing (print format) file
.MAC = MACRO source module (input file)
.OBJ = MACRO object module (output file)
.CRF = Intermediate CREF input file created
by MACRO.

iver

is an octal number between 1 and 77777 that is used to
differentiate between versions of the same file. This
number must be prefixed by a semicolon (i).
For input files,
the default value is
version number of the file that exists.

the

highest

For output files, the default value is the highest
verSion number of the file that exists increased by 1.
If no version number exists, the value 1 is used.

8-7

OPERATING PROCEDURES
Table 8-1
File Specification Default Values
Default Value
File

8.3

Device
Directory

Filename

Type

Object
File

System
device.

Current.

None

.OBJ

Listing
File

Device used
for object
file.

Directory
used in
Object file.

None

.LST

Source 1
File

System
device.

Current.

None

.MAC

Source 2
to
Sourc% n
File

Device used
for source 1 or
last source file
specified.

Directory
used for
source 1 or
last source
file specified.

None

.MAC

User
Macro
Library

System device,
if macro file
is specified
first; if not,
device used
by last source
file is used.

None
Current, if
macro file
is specified
first; if not,
directory of
last source
file is used.

.MLB

System
Macro
Library

System
device.

[1,1]

TRAXMAC

.SML

Indirect
Command
File

System
device.

Current.

None

.CMD

MACRO ERROR MESSAGES

MACRO outputs an appropriate error message to the command output
device when one of the error conditions described below is detected.
These error messages reflect operational problems and should .not be
confused with the diagnostic error messages (see Appendix D) produced
by MACRO during assembly.
All the error messages listed below, with the exception of the "MAC -COMMAND I/O ERROR" message, result in the termination of the current
assembly; MACRO then attempts to restart by reading another command
line. In the case of a command I/O error, however, MACRO exits, since
it is unable to obtain additional command line input.

8-8

OPERATING PROCEDURES
Meaning

Error Message
MAC -- COMMAND FILE OPEN FAILURE

Either the file from which
MACRO is reading a command
could not be opened initially
or between assemblies;
or,
the indirect command
file
specified as "@filename" in
the MACRO command line could
not be opened.
See "OPEN
FAILURE ON INPUT FILE" for
meaning.

MAC -- COMMAND I/O ERROR

An error was returned by the
file system during MACRO's
attempt to read a command
line.
This
is
an
unconditionally fatal error,
causing MACRO to exit. No
MACRO restart is attempted
when this message appears.

MAC -- COMMAND SYNTAX ERROR

An error was detected in the
syntax of the MACRO command
line.

MAC -- ILLEGAL FILENAME

Neither the device name nor
the filename was present in
the input file specification
(i.e.,
the
input
file
specification is null), or a
"wild
card"
convention
(asterisk) was employed in an
input
or
output
file
specification.
"Wild card"
options (*) are not permitted
in MACRO file specifications.

MAC -- ILLEGAL SWITCH

An
illegal
switch
was
specified
for a file,
an
illegal value was specified
with a switch, or an invalid
use of a switch was detected
by MACRO.

MAC -- INDIRECT COMMAND SYNTAX ERROR

The name of the
indirect
command
file
(@filename)
specified
in
the
MACRO
command line is syntactically
incorrect.

MAC -- INDIRECT FILE DEPTH EXCEEDED

An attempt to exceed
the
maximum allowable number of
nested indirect command files
has occurred.
(Only three
levels of indirect command
files
are
permitted
in
MACRO.)

8-9

OPERATING PROCEDURES
Meaning

Error Message
MAC -- INSUFFICIENT DYNAMIC MEMORY

There is not enough physical
memory available for MACRO to
page
its
symbol
table.
Reinstall MACRO in a larger
partition;
or see Section
F.3.

MAC -- INVALID FORMAT IN MACRO LIBRARY

The library file has been
corrupted
or
it was not
produced by the
Librarian
Utility Program (LBR).

MAC -- I/O ERROR ON INPUT FILE

In reading a record from a
source input file or macro
library file,
an error was
detected by the file system,
e.g., a line containing more
than 132(10) characters is
encountered.
This
message
may
also indicate that a
device problem exists or that
either a source file or a
macro library file has been
corrupted
with
incorrect
data.

MAC -- I/O ERROR ON MACRO LIBRARY FILE

Same meaning as I/O ERROR ON
INPUT FILE, except that the
file is a macro library file
and not a source input file.

MAC -- I/O ERROR ON OUTPUT FILE

In writing a record to the
object output file or the
listing output file, an error
was
detected by the file
system.
This message
may
also indicate that a device
problem exists or that the
storage space on a device has
been exhausted
(i.e.,
the
device is full).

MAC -- I/O ERROR ON WORK FILE

A
read
or
write
error
occurred on the work file
used to store the
symbol
table.
This error is most
likely caused by a problem on
this device, or by attempting
to write to a device that is
full.

8-10

OPERATING PROCEDURES
Error Message

Meaning

MAC -- OPEN FAILURE ON INPUT FILE

1.
2.
3.
4.
S.
6.

MAC -- OPEN FAILURE ON OUTPUT FILE

1.
2.
3.
4.
S.
6.

MAC -- 64K STORAGE LIMIT EXCEEDED

The MACRO Assembler uses

it~

Specified device does not
exist.
The volume is not
mounted.
A problem exists with the
device.
Specified directory file
does not exist.
Specified file does not
exist.
User does not have access
to the file directory or
the file itself.
Specified device does not
exist.
The volume is not
mounted.
A problem exists with the
device.
Specified directory file
does not exist.
User does not have access
to the file directory
The volume is full or the
device is write
protected.

64K words of work file memory
are available to MACRO. This
message indicates that the
assembler has generated so
many symbols (on the order of
13,000 to 14,000), it has run
out of space.
This means
either the source program is
too large to start with, or
it contains a condition that
leads to excessive size, such
as a macro expansion that
recursively
calls
itself
without
a
terminating
condition.

stack for the following purposes:

Symbol Table. Four words for every symbol, including macro
names and local symbols.
(Local symbol space, however, is
reused.)

Control Section Information.
.ASECT, or .CSECT.

Storage of macro-definition text. Each and every character,
including comments, between a .MACRO and the corresponding
.ENDM is stored on the stack.

Work space (for code conversion, etc.).

8-11

Five

words

for

each

PSECT,

OPERATING PROCEDURES

If, during its execution, the MACRO Assembler reports a stack
overflow,
then
either
the
demands
of one or more of the
above-described categories must be reduced, or the program must be
broken up into smaller modules and linked together at Task-Build time.
The default stack size for MACRQ is 4K.
This allows
1000 symbols with the trade-offs mentioned above.

8-12

approximately

APPENDIX A
MACRO CHARACTER SETS

A.l

ASCII CHARACTER SET
EVEN
PARITY
BIT

7-BIT
OCTAL
CODE

CHARACTER

o
I

000
001

NUL
SOH

002

STX

003

ETX

004

EOT

005

ENQ

006
007
010

ACK
BEL
BS

011
012

HT
LF

013
014

VT
FF

015

016

017

020
021

DLE
DCl

023

DC3

024

DC4

REMARKS
Null, tape feed, CONTROL/SHIFT/P.
Start of heading; also SOM, start
of message, CONTROL/A.
Start of text; also EOA, end of
address, CONTROL/B.
End of text;
also EOM, end of
message, CONTROL/C.
End of transmission
(END);
shuts
off TWX machines, CONTROL/D.
Enquiry
(ENQRY);
also
WRU,
CONTROL/E.
Acknowledge;
also RU, CONTROL/F.
Rings the bell. CONTROL/G.
Backspace;
also
FEO,
format
effector.
backspaces
some
machines, CONTROL/H.
Horizontal tab. CONTROL/I.
Line feed or Line space (new line);
advances
paper
to
next line,
duplicated by CONTROL/J.
Vertical tab (VTAB). CONTROL/K.
Form Feed to top of next page
(PAGE). CONTROL/L.
Carriage return to beginning of
line; duplicated by CONTROL/M.
Shift out; changes ribbon color to
red. CONTROL/N.
Shift in; changes ribbon color to
black. CONTROL/O.
Data link escape. CONTROL/P (DCO).
Device
control
1;
turns
transmitter
(READER) on, CONTROL/Q
(X ON).
0 022 DC2 Device control
2;
turns punch or auxiliary on.
CONTROL/R (TAPE, AUX ON).
Device
control
3;
turns
transmitter (READER)
off, CONTROL/S (X OFF).
Device control 4;
turns punch or
auxiliary
off.
CONTROL/T
(AUX
OFF) •

A-I

MACRO CHARACTER SETS
EVEN
PARITY
BIT

7-BIT
OCTAL
CODE

CHARACTER

025

NAK

Negative acknowledge;
ERROR. CONTROL/U.

026

SYN

Synchronous
CONTROL/V.

027

ETB

030
031
032
033
034
035
036
037
040
041
042
043
044
045
046
047
050
051
052
053
054
055
056
057
060
061
062
063
064
065
066
067
070
071
072
073
074
075
076
077
100
101
102
103
104
105
106
107
110

CAN
EM
SUB
ESC
FS
GS
RS
US
SP

End of transmission block;
also
LEM,
logical
end
of
medium.
CONTROL/W.
Cancel (CANCL). CONTROL/X.
End of medium. CONTROL/Y.
Substitute. CONTROL/Z.
Escape. CONTROL/SHIFT/K.
File separator. CONTROL/SHIFT/L.
Group separator. CONTROL/SHIFT/M.
Record separator. CONTROL/SHIFT/N.
Unit separator. CONTROL/SHIFT/O.
Space.

1
1

o
o
o

o
o

1
1

o
o

1
1

o
1
o
1
1

o
1
o
o

1
1

o
o
1

1
1

o
o
o

III

REMARKS
also

file

!
II

i
$
%
&

•

Accent acute or apostrophe.

(
)

/
o
1
2

3
4

5
6
7
8
9

>
?
@

A
B

E
F
G
H
I

A-2

ERR,

(SYNC) •

MACRO CHARACTER SETS
EVEN
PARITY
BIT

7-BIT
OCTAL
CODE

1
0
1
0
0
1
0
1
1
0
1
0
0
1
1
0
0
1
0
1
1
0
0

112
113
114
115
116
117
120
121
122
123
124
125
126
127
130
131
132
133
134
135
136
137
140

1
1
0
1
0
0
1
1
0
0
1
0
1
1
0
1
0
0
1
0
1
1
0
0
1
1
0
0

141
142
143
144
145
146
147
150
151
152
153
154
155
156
157
160
161
162
163
164
165
166
167
170
171
172
173
174
175
176

177

* ...

CHARACTER

REMARKS

J
K

L
M
N
0

P
Q

R
S
T
U
V
W

Y
Z
[

]

shift/k.
shift/I.
shift/me

*
**
Accent grave.
a
b
c
d
e
f
g
h
i
j

k
1

m
n
0

P
q

r
s
t
u
v

w
x
Y

This code generated by ALTMODE.
This code generated by prefix key
(if present).
Delete, Rubout.

Appears as i or t on some machines.

** _ Appears as

on some machines.

A-3

MACRO CHARACTER SETS

A.2

RADIX-50 CHARACTER SET

Character

ASCII Octal Equivalent

Radix-50 Equivalent

space

A-Z

101-132

1-32

34
35

unused
0-9

60-71

36-47

The maximum Radix-50 value is, thus,
47*50**2+47*50+47=174777
The following table provides a convenient means of translating between
the ASCII character set and its Radix-50 equivalents. For example,
given the ASCII string X2B, the Radix-50 equivalent is (arithmetic is
performed in octal):
X=113000
2=002400
B=000002
X2B=115402

Single Char.
or
First Char.
Space
A
B
C
D

E
F
G
H

I
J
K
L
M

0
p
Q

R
S
T
U
V

X
y
Z

000000
003100
006200
011300
014400
017500
022600
025700
031000
034100
037200
042300
045400
050500
053600
056700
062000
065100
070200
073300
076400
101500
104600
107700
113000
116100
121200
124300

Second
Character
Space
A
B
C
D

E
F
G
H

I
J
K
L
M

0
P
Q

R
S

T
U
V

X
y
Z

000000
000050
000120
000170
000240
000310
000360
000430
000500
000550
000620
000670
000740
001010
001060
001130
001200
001250
001320
001370
001440
001510
001560
001630
001700
001750
002020
002070

A-4

Third
Character
Space
A
B
C
D

E
F
G
H

I
J
K
L
M

0
p
Q

R
S
T
U
V

X
y
Z

000000
000001
000002
000003
000004
000005
000006
000007
000010
000011
000012
000013
000014
000015
000016
000017
000020
000021
000022
000023
000024
000025
000026
000027
000030
000031
000032
000033

MACRO CHARACTER SETS

SINGLE CHAR.
OR
FIRST CHAR.

SECOND
CHARACTER

127400

Unused 132500
0
1
2
3
4
5
6
7
8
9

135600
140700
144000
147100
152200
155300
160400
163500
166600
171700

Unused
0
1
2
3
4
5
6
7
8
9

002140
002210
002260
002330
002400
002450
002520
002570
002640
002710
002760
003030

A-5

THIRD
CHARACTER

Unused
0
1
2
3
4
5
6
7
8
9

000034
000035
000036
000037
000040
000041
000042
000043
000044
000045
000046
000047

APPENDIX B
MACRO ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

B.l

SPECIAL CHARACTERS
Character
=
%

tab
space
i
@
(
)

, (comma)
+

• (apostrophe)

>
,.

\
vertical tab

B.2

Function
Label terminator
Direct assignment operator
Register term indicator
Item terminator or field terminator
Item terminator or field terminator
Immediate expression indicator
Deferred addressing indicator
Initial register indicator
Terminal register indicator
Operand field separator
Comment field indicator
Arithmetic addition operator or auto
increment indicator
Arithmetic subtraction operator or auto
decrement indicator
Arithmetic multiplication operator
Arithmetic division operator
Logical AND operator
Logical OR operator
Double ASCII character indicator
Single ASCII character indicator or
concatenation indicator
Assembly location counter
Initial argument indicator
Terminal argument indicator
Universal unary operator or argument
indicator
Macro call numeric argument indicator
Source line terminator

SUMMARY OF ADDRESS MODE SYNTAX

Address mode syntax is e~pressed in the summary below using the
following symbols:
n IS an integer between 0 and 7 representing a
register number: R is a register expression:
E is an expression:
and ER is either a register expression or an expression in the range 0
to 7.

B-1

MACRO ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

Format

Address
Mode
Name

Address
Mode
Number

@R or
(ER)

(ER) +

Autoincrement

The contents of the register
spec ified
as
(ER)
are
incremented after being used
as the address of the operand.

@(ER)+

Autoincrement
Deferred

The register specified as (ER)
contains the pointer to the
address of the operand;
the
register (ER)
is incremented
after use.

-(ER)

Autodecrement

The contents of the register
specified
as
(ER)
are
decremented before being used
as the address of the operand.

@- (ER)

Autodecrement
Deferred

The contents of the register
specified
as
(ER)
are
decremented before being used
as the pointer to the address
of the operand.

E(ER)

Index

The expression E, plus the
contents
of
the
register
specified as
(ER), form the
address of the operand.

@E(ER)

Index Deferred

The expression E, plus the
contents
of
the
register
specified as
(ER), yield a
pointer to the address of the
operand.

Immediate

The expression
operand itself.

the

@#E

Absolute

The expression
E
is
address of the operand.

the

Relative

The address of the operand E,
relative to the instruction,
follows the instruction.

Relative
Deferred

The address of the operand is
pointed to by E whose address,
relative to the instruction,
follows the instruction.

B.3

ASSEMBLER DIRECTIVES

Meaning
R

contains

the
the ad-

The MACRO assembler directives are summarized in the following table.
For a detailed description of each directive, the table contains

B-2

MACRO ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES
references to the appropriate sections in the body of tne manual.

Form

Section
Reference

Operation

6.3.3
7.3.6

A
single quote
(apostrophe)
followed by one ASCII character
generates a word which contains the
7-bit ASCII representation of the
character ·in the low-order byte and
zero in the high-order byte.
This
character
is
also
used as a
concatenation indicator
in
the
expansion of macro arguments (see
Section 7.3.6).

6.3.3

A double quote followed by two
ASCII characters generates a word
which contains the 7-bit
ASCII
representation
of
the
two
characters. The first character is
stored in the low-order byte;
the
second character is stored in the
high-order byte.

"'Bn

6.4.1.2

Temporary radix control;
causes
the value n to be treated as a
binary number.

"'Cexpr

6.4.2.2

Temporary numeric control;
causes
the expression's value to be onescomplemented.

6.4.1.2

Temporary radix control;
causes
the value n to be treated as a
decimal number.

6.4.2.2

Temporary numeric control;
causes
the value n to be treated as a
sixteen-bit floating-point number.

6.4.1.2

Temporary radix control;
causes
the value n to be treated as an
octal number.

"'Rccc

6.3.7

Convert ccc to Radix-50 form •

• ASCII /string/

6.3.4

Generates
a
block
of
data
containing the ASCII equivalent of
the character string
(enclosed in
delimiting
characters),
one
character per byte.

B-3

MACRO ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

Form

Section
Reference

Operation

.ASCIZ /string/

6.3.5

Generates
a
block
of
data
containing the ASCII equivalent of
the character string
(enclosed in
delimiting
characters),
one
character per byte, with a zero
byte
terminating
the specified
string.

.ASECT

6.8.2

Begin or
resume
program section.

.BLKB exp

6.5.3

Reserves a block of storage space
whose length in bytes is determined
by the specified expression.

.BLKW exp

6.5.3

Reserves a block of storage space
whose length in words is determined
by the specified expression.

.BYTE expl,exp2, .•

6.3.1

Generates successive bytes of data;
each byte contains the value of the
corresponding specified expression •

• CSECT [name]

6.8.2

Begin or resume named or unnamed
relocatable program section. This
directive
is
provided
for
compatibility
with other PDP-II
assemblers.

.DSABL arg

6.2

Disables the function specified
the argument.

.ENABL arg

6.2

Enables
(invokes)
the
function
specified by the argument.

.END [exp]

6.6.1

Indicates the logical end of the
source
program.
The
optional
argument specifies the
transfer
address where program execution is
to begin.

.ENDC

6.10.1

Indicates the end of a
assembly block.

.ENDM [name]

7.1.2

Indicates the end of the current
repeat
block, indefinite repeat
block, or macro definition.
The
optional name,
if used, must be
identical to the name specified in
the macro definition.

.ENDR

7.7

Indicates the end of the current
repeat· block.
This directive is
provided for compatibility
with
other PDP-II assemblers.

.EOT

6.6.2

Ignored;
indicates
end-of-tape
(which is detected automatically by
the hardware). It is included for
compatibility
with
earlier
assemblers.
B-4

the

absolute

conditional

MACRO ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

Form

Section
Reference

Operation

.ERROR exp;text

7.5

User-invoked
error
directive;
causes output to the listing file
or the
command
output
device
containing the optional expression
and the statement containing the
directive.

• EVEN

6.5.1

Ensures that the current location
counter contains an even address by
adding 1 if it is odd.

.FLT2 argl,arg2, •••

6.4.2.1

Generates
successive
2-word
floating-point equivalents for the
floating-point numbers specified as
arguments.

.FLT4 argl,arg2, •••

6.4.2.1

Generates
successive
4-word
floating-point equivalents for the
floating-point numbers specified as
arguments.

.GLOBL syml,sym2, •••

6.9

Defines the symbol(s) specified
global symbol(s).

.IDENT /string/

6.1.5

Provides a means of labeling the
object
module with the program
version number. The version number
is the Radix-50 string appearing
between
the
paired
delimiting
characters.

.IF cond,argl

6.10.1

Begins a conditional assembly block
of source code which is included in
the assembly only if the stated
condition is met with respect to
the argument(s) specified.

.IFF

6.10.2

Appears only within a conditional
assembly
block,
indicating the
beginning of a section of code to
be assembled if the condition upon
entering the block tests false.

.IFT

6.10.2

Appears only within a conditional
assembly
block,
indicating the
beginning of a section of code to
be assembled if the condition upon
entering the block tests true.

.IFTF

6.10.2

Appears only within a conditional
assembly
block,
indicating the
beginning of a section of code to
be assembled unconditionally.

.IIF cond,arg,
statement

6.10.3

Acts
as
a I-line
conditional
assembly block where the condition
is
tested
for
the
argument
specified.
The
statement
is
assembled only if the condition
tests true.

B-5

MACRO ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

Form

Section
Reference

Operation

.IRP sym,
(argl,arg2, ••• >

7.6.1

Indicates
the
beginning of an
indefinite repeat block
in which
the symbol specified is replaced
with successive elements of the
real argument list enclosed within
angle brackets.

.IRPC sym,(string>

7.6.2

Indicates the beginning
of
an
indefinite repeat block in which
the specified symbol takes on the
value
of successive characters,
optionally enclosed within angle
brackets.

.LIMIT

6.7

Reserves two words into which the
Task Builder inserts the low and
high addresses of the task image.

.LIST [arg]

6.1.1

Without an argument,
the
.LIST
directive increments the listing
level count
by
1.
With
an
argument, this directive does not
alter the listing level count,
but
formats
the
assembly
listing
according
to
the
argument
specified •

• MACRO name,argl,
ar 9 2, .•.

7.1.1

Indicates
definition
name
and
arguments •

• MCALL argl,arg2, .••

7.8

Specifies the symbolic names of the
user or system macro definitions
required in the assembly of the
current user program, but which are
not defined within the program.

.MEXIT

7.1.3

Causes an exit from the current
macro
expansion
or
indefinite
repeat block.

.NARG symbol

7.4.1

Can appear only within a macro
definition:
equates the specified
symbol to the number of arguments
in the macro call currently being
expanded .

• NCHR symbol,(string>

7.4.2

Can appear anywhere in a source
program:
equates
the
symbol
specified
to
the
number
of
characters in the specified string.

.NLIST [arg]

6.1.1

Without an argument,
the
.NLIST
directive decrements the listing
level count
by
1.
With
an
argument, this directive suppresses
that
portion
of
the
listing
specified by the argument.

B-6

the start of a macro
having the
specified
the
following dummy

MACRO ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

Form

Section
Reference

Operation

.NTYPE symbol,aexp

7.4.3

Can appear only within a macro
definition;
equates the symbol to
the 6-bit addressing mode of the
specified address expression.

.000

6.5.2

Ensures that the current location
counter contains an odd address by
adding 1 if it is even.

• PAGE

6.1.6

Causes the assembly listing to skip
to the top of the next page, and to
increment the page count.

.PRINT exp;text

7.5

User-invoked
message
directive;
causes output to the listing file
or the
command
output
device
containing the optional expression
and the statement containing the
directive.

.PSECT name,attl, •..
attn

6.8~1

Begin or resume a named or unnamed
program
section
having
the
specified attributes.

.RADIX n

6.4.1.1

Alters the current program radix to
n, where n is 2, 8, or 10 •

• RAD50 /string/

6.3.6

Generates
a
block
of
data
containing the Radix-50 equivalent
of the character string enclosed
within delimiting characters.

.REPT exp

7.7

Begins a repeat block; causes the
section of code up to the next
.ENDM or .ENDR directive to be
repeated
the
number
of times
specified as exp.

.SBTTL string

6.1.4

Causes the specified string to be
printed as part of the assembly
listing page header.
The string
component of each .SBTTL directive
is collected into a
table
of
contents at the beginning of the
assembly listing.

.TITLE string

6.1.3

Assigns the first six
Radix-50
characters in the string as an
object module name and causes the
string to appear on each page of
the assembly listing.

.WORD expl,exp2, ••

6.3.2

Generates successive words of data;
each word contains the value of the
corresponding specified expression.

B-7

APPENDIX C
PERMANENT SYMBOL TABLE (PST)

The permanent symbol table (PST)
contains those symbols which are
automatically recognized by MACRO. These symbols consist of both op
codes and assembler directives. The op codes (i.e., the instruction
set) are listed first, followed by the directives which cause specific
actions during assembly.
For a detailed description of the instruction set, see the appropriate
PDP-II Processor Handbook.

C.l

OP CODES

MNEMONIC

OCTAL
VALUE

ADC
ADCB
ADD
ASH
ASHC
ASL
ASLB
ASR
ASRB
BCC
BCS
BEQ
BGE
BGT
BHI
BHIS
BIC
BICB
BIS
BISB
BIT
BITB
BLE
BLO
BLOS
BLT

005500
105500
060000
072000
073000
006300
106300
006200
106200
103000
103400
001400
002000
003000
101000
103000
040000
140000
050000
150000
030000
130000
003400
103400
101400
002400

FUNCTIONAL NAME
Add Carry
Add Carry (Byte)
Add Source To Destination
Shift Arithmetically
Arithmetic Shift Combined
Arithmetic Shift Left
Arithmetic Shift Left (Byte)
Arithmetic Shift Right
Arithmetic Shift Right (Byte)
Branch If Carry Is Clear
Branch If Carry Is Set
Branch If Equal
Branch If Greater Than Or Equal
Branch If Greater Than
Branch If Higher
Branch If Higher Or Same
Bit Clear
Bit Clear (Byte)
Bit Set
Bit Set (Byte)
Bit Test
Bit Test (Byte)
Branch If Less Than Or Equal
Branch If Lower
Branch If Lower Or Same
Branch If Less Than

C-l

PERMANENT SYMBOL TABLE (PST)

MNEMONIC

OCTAL
VALUE

BMI
BNE
BPL
BPT
BR
BVC
BVS
CALL
CCC
CLC
CLN
CLR
CLRB
CLV
CLZ
CMP

100400
001000
100000
000003
000400
102000
102400
004700
000257
000241
000250
005000
105000
000242
000244
020000

CMPB

120000

COM
COMB

005100
105100

DEC
DECB

005300
105300

DIV
EMT
FADD
FDIV
FMUL
FSUB
HALT
INC
INCB

071000
104000
075000
075030
075020
075010
000000
005200
105200

lOT
JMP
JSR
MARK
MFPI

000004
000100
004000
006400
006500

MFPS

106700

MOV
MOVB

010000
110000

MTPI

006600

MTPS

106400

MUL
NEG
NEGB
NOP
RESET
RETURN
ROL
ROLB
ROR

070000
005400
105400
000240
000005
000207
006100
106100
006000

FUNCTIONAL NAME
Branch If Minus
Branch If Not Equal
Branch If Plus
Breakpoint Trap
. Branch Unconditional
Branch If Overflow Is Clear
Branch If Overflow Is Set
Jump To Subroutine (JSR PC,xxx)
Clear All Condition Codes
Clear C Condition Code Bit
Clear N Condition Code Bit
Clear Destination
Clear Destination (Byte)
Clear V Condition Code Bit
Clear Z Condition Code Bit
Compare Source To
Destination
Compare Source To
Destination (Byte)
Complement Destination
Complement Destination
(Byte)
Decrement Destination
Decrement Destination
(Byte)
Divide
Emulator Trap
Floating Add
Floating Divide
Floating Multiply
Floating Subtract
Halt
Increment Destination
Increment Destination
(Byte)
Input/Output Trap
Jump
Jump To Subroutine
Mark
Move From Previous
Instruction Space
Move from PS
(LSI-II)
Move Source To Destination
Move Source To Destination
(Byte)
Move To Previous
Instruction Space
Move to PS
(LSI-II)
Multiply
Negate Destination
Negate Destination (Byte)
No Operation
Reset External Bus
Return From Subroutine (RTS PC)
Rotate Left
Rotate Left (Byte)
Rotate Right

C-2

PERMANENT SYMBOL TABLE (PST)

MNEMONIC

OCTAL
VALUE

RORB
RTI

106000
000002

RTS
RTT

000200
000006

SBC
SBCB
SCC
SEC
SEN
SEV
SEZ
SOB
SUB

005600
105600
000277
000261
000270
000262
000264
077000
160000

SWAB
SXT
TRAP
TST
TSTB
WAIT
XOR

000300
006700
104400
005700
105700
000001
074000

FUNCTIONAL NAME
Rotate Right (Byte)
Return From Interrupt
(Permi ts a tr ace
trap)
Return From Subroutine
Return From Interrupt
(inhibits trace trap)
Subtract Carry
Subtract Carry (Byte)
Set All Condition Code Bits
Set C Condition Code Bit
Set N Condition Code Bit
Set V Condition Code Bit
Set Z Condition Code Bit
Subtract One And Branch
Subtract Source From
Destination
Swap Bytes
Sign Extend
Trap
Test Destination
Test Destination (Byte)
Wait For Interrupt
Exclusive OR

OP CODES FLOATING POINT PROCESSOR ONLY
MNEMONIC

OCTAL
VALUE

ABSD
ABSF
AnDD
ADDF
CFCC

170600
170600
172000
172000
170000

CLRD
CLRF
CMPD
CMPF
DIVD
DIVF
LDCDF

170400
170400
173400
173400
174400
174400
177400

LDCFD

177400

LDCID

177000

LDCIF

177000

LDCLD

177000

LDCLF

177000

LDD
LDEXP

172400
176400

FUNCTIONAL NAME
Make Absolute Double
Make Absolute Floating
Add Double
Add Floating
Copy Floating Condition
Codes
Clear Double
Clear Floating
Compare Double
Compare Floating
Divide Double
Divide Floating
Load And Convert From
Double To Floating
Load And Convert From
Floating To Double
Load And Convert Integer To
Double
Load And Convert Integer To
Floating
Load And Convert Long
integer To Double
Load And Convert Long
Integer To Floating
Load Double
Load Exponent

C-3

PERMANENT SYMBOL TABLE (PST)

C.2

MNEMONIC

OCTAL
VALUE

LDF
LDFPS
MFPD

172400
170100
106500

MODO

171400

MODF

171400

MTPD
MULD
MULF
NEGD
NEGF
SETD
SETF
SETI
SETL
SPL
STCDF

106600
171000
171000
170700
170700
170011
170001
170002
170012
000230
176000

STCDI

175400

STCDL

175400

STCFD

176000

STCFI

175400

STCFL

175400

STD
STEXP
STF
STFPS
STST
SUBD
SUBF
TSTD
TSTF

174000
175000
174000
170200
170300
173000
173000
170500
170500

FUNCTIONAL NAME
Load Floating
Load FPPs Program Status
Move From Previous Data
Space
Multiply And Integerize
Double
Multiply And Integerize
Floating
Move To Previous Data Space
Multiply Double
Multiply Floating
Negate Double
Negate Floating
Set Double Mode
Set Floating Mode
Set Integer Mode
Set Long Integer Mode
Set Priority Level
Store And Convert From
Double To Floating
Store And Convert From
Double To Integer
Store And Convert From
Double To Long Integer
Store And Convert From
Floating To Double
Store And Convert From
Floating To Integer
Store And Convert From
Floating To Long Integer
Store Double
Store Exponent
Store Floating
Store FPPs Program Status
Store FPPs Status
Subtract Double
Subtract Floating
Test Double
Test Floating

MACRO DIRECTIVES
DIRECTIVE
• ASCII
.ASCIZ
.ASECT
.BLKB
.BLKW
.BYTE
.CSECT

FUNCTIONAL SIGNIFICANCE
Translates character string to ASCII equivalents •
Translates character string to ASCII equivalents;
inserts zero byte as last character.
Begins absolute program section
(provided for
compatibility with other PDP-II assembliers).
Reserves byte block in accordance with value of
specified argument.
Reserves word block in accordance with value of
specified argument.
Generates successive byte data in accordance with
specified arguments.
Begins relocatable program section
(provided for
compatibility with other PDP-II assemblers).

C-4

PERMANENT SYMBOL TABLE (PST)

DIRECTIVE
• DSABL
• ENABL
• END
• ENDC
.ENDM
.ENDR
• EOT
• ERROR
• EVEN
.FLT2
• FLT4
• GLOBL
.IDENT
• IF
.IFF
.IFT
.IFTF
.IIF
.IRP
.IRPC
.LIMIT
.LIST
.MCALL
.MEXIT
.NARG
.NCHR
.NLIST
.NTYPE
• ODD
• PAGE
• PRINT
.PSECT
• RADIX
.RADSO
.REPT

FUNCTIONAL SIGNIFICANCE
Disables specified function •
Enables specified function •
Defines logical end of source program •
Defines end of conditional assembly block •
Defines end of macro definition, repeat block, or
indefinite repeat block.
Defines end of current repeat block (provided for
compatibility with other PDP-II assemblers).
Define End of Tape condition (ignored) •
Outputs diagnostic message to listing file or
command output device.
Word-aligns the current location counter •
Causes two words of storage to be generated for
each floating-point argument •
Causes four words of storage to be generated for
each floating-point argument.
Declares global attribute for specified symbol(s) •
Labels object module with
specified
program
version number.
Begins conditional assembly block •
Begins
subconditional
assembly
block
(if
conditional assembly block test is false).
Begins
subconditional
assembly
block
(if
conditional assembly block test is true).
Begins subconditional assembly block
(whether
conditional assembly block test is true or false).
Assembles immediate conditional assembly statement
(if specified condition is satisfied).
Begins
indefinite
repeat
block;
replaces
specified symbol with specified successive real
arguments.
Begins
indefinite
repeat
block;
replaces
specified
symbol
with
value
of successive
characters in specified string.
Reserves two words of storage for high and low
addresses of task image.
Controls listing level count and
format
of
assembly listing.
.MACRO Denotes start of macro
definition.
Identifies required
macro
definition(s)
for
assembly.
Exit from current macro definition or indefinite
repeat block.
Equates specified symbol to
the
number
of
arguments in the macro expansion.
Equates specified symbol to
the
number
of
characters in the specified character string.
Controls listing level count
and
suppresses
specified portions of the assembly listing.
Equates specified symbols to the addressing mode
of the specified argument.
Byte-aligns the current location counter •
Advances form to top of next page •
Prints specified message on command output device •
Begins specified program section having specified
attributes.
Changes current program radix to specified radix •
Generates data block having Radix-50 equivalents
of specified character string.
Begins repeat block and replicates it according to
the value of the specified expression.

C-S

PERMANENT SYMBOL TABLE (PST)
DIRECTIVE
.SBTTL
.TITLE
• WORD

FUNCTIONAL SIGNIFICANCE
Prints specified subtitle text as the second line
of the assembly listing page header.
Prints specified title text as object module name
in the first line of the assembly listing page
header.
Generates successive word data in accordance with
specified arguments.

The MACRO directives listed above are summarized in greater detail
Appendix B.

C-6

APPENDIX D
DIAGNOSTIC ERROR MESSAGE SUMMARY

0.1

MACRO ERROR CODES

A diagnostic error code is printed as the first character in a source
line which contains an error detected by MACRO. This error code
identifies a syntactical problem or some other type of error condition
detected during the processing of a source line. An example of such a
source line is shown below:

26 000236

010102

MOV Rl,R2,A

The extraneous argument A in the MOV instruction above causes the line
to be flagged with a Q (syntax) error.
Error Code

Meaning
Assembly error. Because many different types of
error conditions produce this diagnostic message,
all the possible directives which may yield a
general assembly error have been categorized below
to reflect specific classes of error conditions:
CATEGORY 1:

ILLEGAL ARGUMENT SPECIFIED •

• RADIX -- A value other than
specified as a new radix .

2, 8,

• LIST/.NLIST -- Other than a legally defined
argument
(see Table 6-1) is specified with the
directive •
• ENABL/.DSABL -- Other than a legally defined
argument
(see Table 6-2) is specified with the
directive •
• PSECT -- Other than a legally-defined argument
(see
Table
6-3)
is
specified with the
directive •
• IF/.IIF -- Other than
a
legally
defined
conditional test (see Table 6-5) or an illegal
argument expression value is specified with the
directive •
• MACRO -- An illegal or duplicate symbol
in dummy argument list.

found

DIAGNOSTIC ERROR MESSAGE SUMMARY

Meaning

Error Code

A
(Cont'd)

CATEGORY 2:

NULL ARGUMENT OR SYMBOL SPECIFIED.

.TITLE -- Program name is not specified in the
directive,
or
first
non-blank
character
following the directive is a
non-Radix-SO
character •
• IRP/.IRPC -- No dummy argument is specified in
the directive •
• NARG/.NCHAR/.NTYPE -- No symbol
in the directive •
• IF/.IIF -- No
conditional
specified in the directive.

specified

argument

CATEGORY 3: UNMATCHED DELIMITER/ILLEGAL
CONSTRUCTION •

ARGUMENT

• ASCII/.ASCIZ/.RAD50/.IDENT -- Character string
or argument string delimiters do not match, or
an illegal character is used as a delimiter, or
an illegal argument construction is used in the
directive •
• NCHAR -- Character string delimiters do not
match, or an illegal character is used as a
delimiter in the directive.
CATEGORY 4:

GENERAL ADDRESSING ERRORS.

This type of error results from one of
possible conditions:

several

Permissible range of a branch instruction,
i.e., from -128(lO} to +l27(lO} words, has
been exceeded.

A statement makes invalid
use of
the
current
location
counter,
e.g.,
a
".=expression" statement attempts to force
the
current location counter to cross
program section (.PSECT) boundaries.

A statement contains an invalid address
expression.
In cases where an absolute
address expression is required, specifying
a global symbol, a relocatable value, or a
complex relocatable value (see Section 3.9)
results in an invalid address expression.
Similarly, in cases where a relocatable
address expression is required, either a
relocatable
or
absolute
value
is
permissible, but a global symbol or a
complex relocatable value in the statement
likewise results in an invalid address
expression. Specific cases of this type of
error are those which follow:

D-2

DIAGNOSTIC ERROR MESSAGE SUMMARY
Error Code

Meaning
.BLKB/.BLKW/.REPT -- Other than an absolute
value or an expression which reduces to an
absolute value has been specified with the
directive.
4.

Multiple expressions are not separated by a
comma.
This condition causes the next
symbol to be evaluated as part of the
current expression.

CATEGORY 5:

ILLEGAL FORWARD REFERENCE.

This type of error results from either
possible conditions:
1.

A
global
assignment
(symbol==expression)
contains
reference to another symbol.

two

statement
forward

An expression defining the value of the
current location counter contains a forward
reference.

Bounding error. Instructions or word data are
being assembled at an odd address. The location
counter is incremented by 1.

Doubly-defined symbol referenced.
Reference was
made to a symbol which is defined more than once.

End directive not found. When the end-of-file is
reached during source input and the .END directive
has not yet been encountered, MACRO generates this
error code, ends assembly pass 1, and proceeds
with assembly pass 2.

Illegal character detected.
Illegal characters
which are also non-printable are replaced by a
question mark (?) on the listing.
The character
is then ignored.

Input line is greater than 132(10) characters in
length. Currently, this error condition is caused
only through excessive substitution
of
real
arguments for dummy arguments during the expansion
ofa macro.

Multiple definition of a label.
A label was
encountered which was equivalent (in the first six
characters) to a label previously encountered.

A number contains a digit that is not in the
current program radix. The number is evaluated as
a decimal value.

Opcode
error.
Directive
out
of
context.
Permissible nesting level depth for conditional
assemblies has been exceeded. Attempt to expand a
macro which was unidentified after .MCALL search.

D-3

DIAGNOSTIC ERROR MESSAGE SUMMARY
Error Code

Meaning

Phase error. A label's definition of value varies
from one assembly pass to another or a multiple
definition of a local symbol has occurred within a
local symbol block. Also, when in a local symbol
block defined by the
.ENABL LSB directive,
an
attempt has occurred to define a local symbol in a
program section other than that which was
in
effect when the block was entered. A P error code
also appears if an .ERROR directive is assembled.

Questionable syntax. Arguments are missing,
too
many arguments are specified, or the instruction
scan was not completed.

Register-type error.
An invalid use
of
or
reference to a register has been made, or an
attempt has been made to redefine a standard
register symbol without first issuing the .DSABL
REG directive.

Truncation error. A number generated more than 16
bits in a word, or an expression generated more
than 8 significant bits during the use of the
.BYTE directive or trap (EMT or TRAP) instruction.

Undefined symbol.
An
undefined
symbol
was
encountered
during
the
evaluation
of
an
expression;
such an undefined symbol is assigned
a value of zero. Other possible conditions which
result in this error code include unsatisfied
macro names in the list of .MCALL arguments and a
direct assigment
(symbol=expression)
statement
which contains a forward reference to a symbol
whose
definition
also
contains
a
forward
reference;
also, a local symbol may have been
referenced that does not exist in the current
local symbol block.

Instruction error. The instruction so flagged
is
not compatible among all members of the PDP-II
family.
See Section 5.3 for details.

0-4

APPENDIX E

SAMPLE CODING STANDARD

E.l

INTRODUCTION

Standards eliminate variability and the requirement to make
a
decision. Much of the difficulty in establishing standards stems from
the notion that they should be optimal. However, to be successfully
applied,
standards must represent an agreement on certain aspects of
the programming process.
This Appendix contains DIGITALis PDP-II Program Coding Standard.
It
is suggested that this be used as a model to assist users in preparing
standards for their own installations.

E.2

LINE FORMAT

All source lines shall co·nsist of from one to a maximum of eight
characters.
This program is described in the DEC Editor Reference
Manual (see Section 0.3 in the Preface) •
Assembly language code lines shall have the following format:
1.

Label Field - if present, the label shall start at tab stop 0
(column 1).

Operation field - the operation field shall start at tab stop
1 (column 9).

Operand field - the operand field shall start at tab
(column 17).

Comm~nts

stop

field - the comments field shall st~rt at tab stop 4
(column 33) and may continue to column 80.

Comment lines that are included in the code body shall be delimited by
a line containing only a leading semicolon.
The comment itself
contains a leading semicolon and starts in column 3.
Indents shall be
1 tab.
If the operand field extends beyond tab stop 4
(column 33)
simply
leave a space and start the comment. Comments which apply to an
instruction but require continuation should always line up with the
character position which started the comment.

E-l

SAMPLE CODING STANDARD
E.3

COMMENTS

Comment all coding to convey the global role of an instruction, rather
than simply a literal translation of the instruction into English.
In
general this will consist of a comment per line of code.
If a
particularly difficult, obscure, or elegant instruction sequence is
used, a paragraph of comments must immediately precede that section of
code.
Preface text, which describes formats,
algorithms, program-local
variables, etc., will be delimited by the character sequence ;+ at the
start of the text and ;- at the end;
these delimiters facilitate
automated extraction of narrative commentary. The comment itself will
start in column 3.
For example:
i+
i
i
i-

THE INVERT ROUTINE ACCEPTS
A LIST OF RANDOM NUMBERS AND
APPLIES THE KOLMOGOROV ALGORITHM
TO ALPHABETIZE THEM.

E.4

E.4.1

NAMING STANDARDS

E.4.1.1 General Purpose Registers - Only
and may
permitted as register names;
purpose:
RO=%O
RI=%1
R2=%2
R3=%3
R4=%4
R5=%5
SP=%6
PC=%7

the
not

following names are
be used for any other

;REG 0
;REG I
;REG 2
;REG 3
;REG 4
;REG 5
;STACK POINTER (REG 6)
iPROGRAM COUNTER (REG 7)

registers
must
be
named
E.4.1.2 Hardware
Registers - These
For example, PS and SWR.
identically to the hardware definition.

E.4.1.3 Device Registers - These are symbolically named identically
to the hardware notation. For example, the control status register
for the RK disk is RKCS. Only this symbolic name may be used to refer
to this register.

E-2

SAMPLE CODING STANDARD
E.4.2

Processor Priority

Testing or altering the processor priority is done using the symbols
PRO, PRl, PR2, •••••• PR7
which are equated to their corresponding priority bit pattern.

E.4.3

Other Symbols

Frequently-used bit patterns such as CR and
conventional symbolics on an as-needed basis.

E.4.4

will

made

Using the Standard Symbolics

The register standards will be defined within the assembler.
All
other standard symbols will appear in a file and will be linked prior
to program execution.

E.4.S

Symbols*

E.4.5.l Global Symbols - Global symbols should be easily recognized
by their format. The following standards apply and completely define
symbol standards for PDP-II Medium/Large software products.

symbol

pos-1

pos-2

pes-3

pes-4

pes-5

pes-6

length

nen-glbl-sym

letter

a-numl
null

>=1

glbl-sym

$1.

a-num
null

a-numl
null

>=1

***

letter

$1.

a-num

a-numl
null

>=3

$1.

a-numl

>=4

glbl-offset

***

gl bl-pit-ptrn

letter

a-num

***

lecal-sym

number

>=2

* Symbols that are branch targets are also called labels, but we will
always use the term "symbol".

** Number is in the range O<number<65535.
*** The use of $ or • for global names is reserved for DEC-supplied
software.

E-3

SAMPLE CODING STANDARD
where:
a-num
non-glbl-sym
local-sym
glbl-sym
glbl-offset
glbl-bit-ptrn

is an alphanumeric cha~acter.
are non-global symbols.
local symbols, as defined by
MACRO.
are global symbols (addresses).
are global offsets (absolute
quantities).
are global bit patterns.

A program never contains a .GLOBL statement without showing cause.

E.4.S.2

Symbol Examples

Non-Global Symbols
AlB
ZXCJl
INSRT
Global Address Symbols
$JIM
.VECTR
$SEC
Global Absolute Offset Symbols
A$JIM
A$XT
A.ENT
Global Bit Pattern Symbols
Al$20
B3.6

JI.M
Local Symbols
37$
271$
6$

E.4.S.3 Program-Local Symbols - Self-relative
address
arithmetic
(.+n) is absolutely forbidden in branch instructions; its use in
other contexts must be avoided if at all possible and practical.

E-4

SAMPLE CODING STANDARD
Target symbols for branches that exist solely for positional reference
will use local symbols of the form
<num>$:
Use of non-local symbols is restricted, within reason, to those cases
where
reference
to
the
code
occurs external to the code.
Local-symbols are formatted such that the numbers proceed sequentially
down the page and from page to page.

E.4.5.4 Macro Names - The last two characters
(with
the
last
character possibly being null) have special significance. The next to
last character is a $, the last, a character specifying the mode of
the macro.
For example, in the three macro forms in-line, stack, and p-section,
the in-line form has no suffix, the stack has an <S>, and the
p-section a <C>. Thus the Queue I/O macro can be written as any of
QIO$
QIO$S
QIO$C
depending on the form required. These are not reserved letters.
the form of the name is standard.

E.5

E.5.l

Only

PROGRAM MODULES

General Comments on Programs

In our software, a program provides a single distinct function.
No
limits exist on size, but the single function limitation should make
modules larger than lK a rarity. Since any software may eventually
exploit the virtual memory capacity of the 11/40 and 11/45, programs
should make every attempt to maintain a dense reference locus
(don't
promiscuously branch over page boundaries or ·over a large absolute
address distance).
All code is read-only. Code and data areas are distinct and each
contains explanatory text. Read-only data should be segregated from
read-write data.

E.5.2

The Module Preface

Each program module in the system shall exist as a separate file. The
filename will reflect the name of the module and the file extension
shall be of the form 'NNN'. The 'NNN' signifies the edit number or
the version number. The version number shall be changed only when a
new base level is:·creat~d. Furthermore, if no corrections are made to
a file from one~base level to the next, the version number will not be
changed. The availability of File Control Services and File Control
Primitives will greatly simplify version number maintenance. Program
modules adhere to a strict. format.
This format adds to
the
readability and understandability of the module.
The following
sections are included in each module:
E-5

SAMPLE CODING STANDARD
For the Code Section:
1.

A .TITLE statement that specifies the name of the module.
If
a module contains more than one routine, subtitles may be
used.

An .IDENT statement specifying the version number.
PDP-II version number standard appears in section E.IO.

The

A .PSECT statement that defines the program section in
the module resides.

which

A copyright statement, and the disclaimer.
COPYRIGHT (C) 1976
DIGITAL EQUIPMENT CORPORATION, ,MAYNARD, MASS.
THIS SOFTWARE IS FURNISHED UNDER A LICENSE FOR USE
ONLY ON A SINGLE COMPUTER SYSTEM AND MAY BE COPIED
ONLY WITH THE INCLUSION OF THE ABOVE COPYRIGHT
NOTICE.
THIS SOFTWARE, OR ANY OTHER COPIES THEREOF,
MAY NOT BE PROVIDED OR OTHERWISE MADE AVAILABLE TO
ANY OTHER PERSON EXCEPT FOR USE ON SUCH SYSTEM AND TO
ONE WHO AGREES TO THESE LICENSE TERMS. TITLE TO AND
OWNERSHIP OF THE SOFTWARE SHALL AT ALL TIMES REMAIN
IN DEC.
THE INFORMATION IN THIS DOCUMENT IS SUBJECT TO CHANGE
WITHOUT NOTICE AND SHOULD NOT BE CONSTRUED AS A
COMMITMENT BY DIGITAL EQUIPMENT CORPORATION.
DEC ASSUMES NO RESPONSIBILITY FOR THE
USE
OR
RELIABLILITY OF ITS SOFTWARE ON EQUIPMENT WHICH IS
NOT SUPPLIED BY DEC.

The version number of the file.
The PDP-II version number standard is
E.IO.

described
on

section

The name of the principal author and the date
module was first created.

which

the

The name of each modifying author and
the
date
of
modification.
Names and modification dates appear one per
line and in chronological order.

A brief statement of the function of the module.
Note: Items 1-8 should appear on the same page.

A list of the definitions of all equated local symbols used
in the module. These definitions appear one per line and in
alphabetical order.

10.

All local macro definitions, preferably in alphabetical order
by name.

11.

All local data.
a.
b.
c.

The data should indicate

Description of each element (type, size, etc.)
Organization (functional, alpha, adjacent, etc.)
Adjacency requirements

E-6

SAMPLE CODING STANDARD
12.

A more detailed definition of the function of the module.

13.

A list of the inputs expected by the module.
the
calling
sequence
if non-standard,
settings, and global data settings.

14.

A list of the outputs produced as a result of entering this
module.
These include delivered results, condition code
settings, but not side effects.
(All these outputs are
visible to the caller.)

15.

A list of all effects (including side effects) produced as a
result of entering this module. Effects include alterations
in the state of the system not explicitly expected in the
calling sequence, or those not visible to the caller.

16.

The module code.

E.5.3

This includes
condition code

Formatting the Module Preface

Rules:
1.

The first eight items appear on the same page and will not
have explicit headings.
Item 3 may be omitted if the blank
p-section is being used.

Headings start at the left
indented 1 tab position.

Items 7-14 will have headings which start at the left margin,
preceded and followed by lines containing only a leading <i>.
Items which do not apply may be omitted.

margin*i

descriptive

text

A template for the module preface follows.
FILE-EXAMPL.SOI
.TITLE
.IDENT
.PSECT

EXAMPLE

/01/
KERNEL

COPYRIGHT (C) 1976
DIGITAL EQUIPMENT COPORATION, MAYNARD, MASS.
THIS SOFTWARE IS FURNISHED UNDER A LICENSE FOR USE ONLY ON A
SINGLE COMPUTER SYSTEM AND MAY BE COPIED ONLY WITH THE
INCLUSION OF THE ABOVE COPYRIGHT NOTICE. THIS SOFTWARE, OR
ANY OTHER COPIES THEREOF, MAY NOT BE PROVIDED OR OTHERWISE
MADE AVAILABLE TO ANY OTHER PERSON EXCEPT FOR USE ON SUCH
SYSTEM AND TO ONE WHO AGREES TO THESE LICENSE TERMS. TITLE
TO AND OWNERSHIP OF THE SOFTWARE SHALL AT ALL TIMES REMAIN
IN DEC.
THE INFORMATION IN THIS DOCUMENT IS SUBJECT TO CHANGE WITHOUT
NOTICE AND SHOULD NOT BE CONSTRUED AS A COMMITMENT BY DIGITAL
EQUIPMENT CORPORATION.

*The left margin consists of a <i> a <space> then the heading, so the
text of the heading begins in column 3.

E-7

SAMPLE CODING STANDARD
DEC ASSUMES NO RESPONSIBILITY FOR THE USE OR RELIABILITY OF
ITS SOFTWARE ON EQUIPMENT WHICH IS NOT SUPPLIED BY DEC.
VERSION 01
JOE PASCUSNIK I-JAN-72
MODIFIED BY:
RICHARD DOE 21-JAN-73
SPENCER THOMAS 12-JUN-73
Brief statement of the module's function
EQUATED SYMBOLS
List equated symbols
LOCAL MACROS
Local Macros
;

LOCAL DATA
Local data
;+
Module function-details
INPUTS:
Description of inputs
OUTPUTS:
Description of outputs
EFFECTS:
Description of effects
i-

Begin Module Code

E.S.4

Modularity

No other characteristic has more impact on the ultimate engineering
success of a system than does modularity. Modularity for PDP-II
Software Engineering's products consists of the application of the
single-function philosophy described in section E.S.l, and adherence
to a set of calling and return conventions.

E.S.4.1 Calling Conventions
conventions must be observed.

(Inter-Module) - The

E-8

following

calling

SAMPLE CODING STANDARD
Transfer of Control
Macros will exist for call and return. The actual transfer will
be via a JSR PC instruction.
For register save routines, a
JSR Rn,SAVE will be permitted.
The CALL macro is:
CALL

subr-name

The RETURN macro is:
RETURN
Register Conventions
On entry, a subroutine minimally saves all registers it intends
to alter except result registers.
On exit it restores these
registers.
(State preservation is assumed across calls.)
Argument Passing
Any registers may be used, but their use should follow a coherent
pattern.
For example, if passing three arguments, pass them in
RO, RI and R2 rather than RO, R2, RS.
Saving and restoring
occurs in one place.

E.S.4.2
macro.

Exiting - All subroutine exits occur through a single

RETURN

E.S.4.3 Intra-Module Calling Conventions - Designer optional,
but
consistency favors a calling sequence identical to that of the
inter-module sequence.

E.S.4.4 Success/Failure Indication - The C bit will be used to return
the success/failure indicator, where success equals 0, and failure
equals 1. The argument registers can be used to return values or
additional success/failure data.

E.S.4.S Module Checking Routines - Modules are
responsible
for
verifying the validity of arguments passed to them. The design of a
module's calling sequence should aim at minimizing the validity checks
by minimizing invalid combinations.
Programmers may add test code to
perform additional checks during checkout. All code should aim at
discovering an error as close (in terms of instruction executions) to
its occurrence as possible.

E.6

E.6.l

FORMATTING STANDARDS

Program Flow

Programs will be organized on the listing such that they flow down the
page, even at the cost of an extra branch or jump.

E-9

SAMPLE CODING STANDARD
For example:

PROCESS

BBB

AAA

COMMON

shall appear on the listing as:
TST
BNE

BBB

CMN

AAA:

BBB:
CMN:

Rather than:
TST
BNE

BBB

CMN

AAA:

CMN:

BBB:

E-IO

SAMPLE CODING STANDARD
E.6.2

Common Exits

A common exit appears as the last code sequence on the listing.
the flow chart:

EXIT

will appear on the listing as:
PR1:
BR

EXIT

PR2:

....

PR3:

PR4-:

EXIT:
And not as:
PRl:

EXIT:

PR2:

EXIT

IiR

EXIT

PR3:

PR4:

E-ll

Thus

SAMPLE CODING STANDARD
E.6.3

Code with Interrupts Inhibited

Code that is executed with interrupts inhibited, shall be flagged by a
three semicolon (iii) comment delimiter.
For example:
;ENABLE BY RETURNING
iBY SYSTEM SUBROUTINES,

•• ERTZ:
BIS
BIT
BEQ
RTT

#PR7,PS
#PR7,+2(SP)
10$

10$:

E.7

; ; ;
; ; ;
; ; ;
; ; ;
; ; ;

INHIBIT INTERRUPTS
C
0

i ; ;
; ; ;
; ; ;
; ; ;

M
M
E
N
T
S

PROGRAM SOURCE FILES

Source creation and maintenance shall be done in base levels. A base
level is defined as a point at which the program source files have
been frozen.
From the freeze point to the next base
level,
corrections will not be made directly to the base level itself.
Rather a file of corrections shall be accumulated for each file in the
base
level.
Whenever an updated source file is desired,
the
correction file will be applied to the base file.
The accumulation of corrections shall proceed until a logical breaking
point has occurred
(i.e.
a milestone or significant implementation
point has been reached). At this time all accumulated corrections
shall be applied to the previous base level to create a new base
level. Correction files will then be started for the new base level.

E.8

FORBIDDEN INSTRUCTION USAGE
1.

The use of instructions or index words
previous instruction. For example:
MOV

@PC,Register

BIC

Src,Dst

literals

the

uses the bit clear instruction as a literal. This may seem
to be a very "neat" way to save a word but what about
maintaining a program using this trick?
To compound the
problem, it will not execute properly if I/D space is enabled
on the 11/45. In this case @PC is a D bank reference.
2.

The use of the MOV instruction instead of a JMP instruction
to
transfer program control to another location.
For
example:
MOV

#ALPHA,PC

transfers control to location ALPHA. Besides taking longer
to execute
(2.3 microseconds for MOV vs. 1.2 for JMP) the
use of MOV instead of JMP makes it nearly impossible to pick
up someone else's program and tell where transfers of control

E-12

SAMPLE CODING STANDARD
take place. What if one would like to get a jump trace of
the execution of a program (a move trace is unheard of)? As
a more general issue, perhaps even other operations such as
ADD and SUB from PC should be discouraged. Possibly one or
tWb words can be saved by using these operations but how many
such occurrences are there?
3.

The seemingly "neat" use of all single word instructions
where one double-word instruction could be used and would
execute faster and would not consume additional memory.
Consider the following instruction sequence:
CMP

-(Rl) ,(-Rl)

CMP

-{Rl} ,-{Rl}

The intent of this instruction sequence is to subtract 8 from
register Rl
(not to set condition codes).
This can be
accomplished in approximately 1/3 the time via a
SUB
instruction (9.4 vs. 3.8 microseconds) at no additional cost
in memory space. Another question here is also, what if RI
is odd?
SUB always wins since it will always execute
properly and is always faster!

E.9

E.9.1

RECOMMENDED CODING PRACTICE

Conditional Branches

When using the PDP-II conditional branch instructions,
it
is
imperative that the correct choice be made between the signed and the
unsigned branches.
SIGNED

UNSIGNED

BGE
BLT
BGT
BLE

BHIS
BLO
BHI
BLOS

(BCC)
(BCS)

A common pitfall is to use a signed branch (e.g. BGT) when comparing
two memory addresses.
All goes well until the two addresses have
opposite signs; that is, one of them goes across the 16K
(lOOOOO{8»
bound.
This type of coding error usually shows itself as a result of
re-linking at different addresses and/or a change in size of the
program.

E.IO

PDP-II VERSION NUMBER STANDARD

The PDP-II Version Number Standard applies to all modules, parameter
files, complete programs, and libraries which are written or caused to
be written, as part of the PDP-II Software Development effort.
It is
used to provide unique identification of all released, pre-released,
and in-house software.
It is limited in that, as currently specified, only six characters of
identification
are
used.
Future implementations of the Macro
Assembler, Task Builder, and Librarian should provide for at least

E-13

SAMPLE CODING STANDARD
nine characters, and possibly twelve.
It
standard will be enhanced as the need arises.

expected that this

Version Identifier = <form> <version> <edit> <patch>
<form>

Used to identify a particular form of a module or
program, where applicable, as in the case of
LINK-II. One alphabetic character, if used, and
null (i.e., a binary 0) if not used.

Used to identify the release, or generation, of a
program.
Two decimal digits, starting at 00, and
incremented at the discretion of the project in
order to reflect what, in their opinion, is a
major change.

<edit>

Used to identify the level to which a particular
release, or generation, of a program or module has
been edited.
An edit is defined to be
an
alteration
to the source form.
Two decimal
digits, beginning at 01, and incremented with each
edit; null if no edits.

<patch>

Used to identify the level to which a particular
release, or generation, of a program or module has
been patched. A patch is defined as an alteration
to a binary form.
One alphabetic character,
starting at S, and running sequentially toward Z,
each time a set of patches is released; null if
no patches.

These fields are interrelated.
When <version> is changed, then
<patch> and <edit> must be reset to nulls. It is intended that when
<edit> is incremented, then <patch> will be re-set to null, because
the various bugs have been fixed.

E.lO.l

Displaying the Version Identifier

The visible output of the version identifier should appear as:
Key <letter> <form> <version> - <edit> <patch>,
where the following Key Letters have been identified:
V
X
y

released or frozen version
in-house experimental version
field test, pre-release, or in-house release version

Note that 'X' corresponds roughly to individual support, 'Y' to
support, and 'V' to company support.

group

The dash which separates <version> from <edit> is used only if <edit>
and/or <patch> is not nUll. When a version identifier is displayed as
part of program identification, then the format is:
Program

Name

E-14

SAMPLE CODING STANDARD

Examples:
PIP X03
LINK VB04-C
MACRO YOS-Ol

E.lO.2

Use of the Version Number in the Program

All sources must contain the version number in an .IDENT directive.
For programs
(or libraries) which consist of more than one module,
each individual module will follow this version number standard.
The
version number of the program or library is not necessarily related to
the version numbers of the constituent modulesi
it is perfectly
reasonable, for example, that the first version of a new FORTRAN
library, VOO, contain an existing SIN routine, say VOS-Ol.
Parameter files are also required to contain the version number in an
.IDENT directive. Because the assembler records the last .IDENT seen,
parameter files must precede the program.
Entities which consist of a collection of modules or programs, will
have
an
identification
module
in
the
first position.
An
identification module exists solely to provide identification, and
normally consists of something like:
iOTS IDENTIFICATION
.TITLE FTNLIB
.IDENT /003010/
.END

E-lS

APPENDIX F

SAMPLE ASSEMBLY AND CROSS REFERENCE LISTING

COPYRIGHT (C) 1977 BY
DIGITAL EQUIPMENT CORPORATION, MAYNARD, MASS.
THIS SOFTWARE IS FURNISHED UNDER A LICENSE AND MAY BE USED AND COPIED
ONLY IN ACCORDANCE WITH THE TERMS OF SUCH LICENSE AND WITH THE
INCLUSION OF THE ABOVE COPYRIGHT NOTICE. THIS SOFTWARE OR ANY OTHER
COPIES THEREOF MAY NOT BE PROVIDED OR OTHERWISE MADE AVAILABLE TO ANY
OTHER PERSON. NO TITLE TO AND OWNERSHIP OF THE SOFTWARE IS HEREBY
TRANSFERRED.
THE INFORMATION IN THIS SOFTWARE IS SUBJECT TO CHANGE WITHOUT NOTICE
AND SHOULD NOT BE CONSTRUED AS A COMMITMENT BY DIGITAL EQUIPMENT
CORPORATION.
DIGITAL ASSUMES NO RESPONSIBILITY FOR THE USE OR RELIABILITY OF ITS
SOFTWARE ON EQUIPMENT WHICH IS NOT SUPPLIED BY DIGITAL.

F-1

~
I\J

1
2
3
4
5 000000
6 000000
7 000002
8 000004
9 000006
10 000010
11 000012
12 000012
13 000014
14 000014
15 000016
16 000016
17 000022
18 000022
19 000024
20 000026
21 000030
22 000032
23 000034

.TITLE
.SBTTL

TST205
LOCAL DATA STRUCTURES

:
FPTRPY:
000004
000000
000012'
000014'
000016'
000043
000000
000000

000000

030060
030461
031062
031463
025052
027056

• WORD
MSGRPY: • WORD
• WORD
• WORD
. WORD
SIZE:
• WORD
REPLYN:
• WORD
STATUS:
. WORD
FUNC:
• WORD
• WORD
• WORD
• WORD
• WORD
. WORD

4
0
SIZE
REPLYN
STATUS

REPLY MESSAGE FPT
ADDRESS OF MESSAGE
ADDRESS OF MSG SIZE
ADDRESS OF REPLY NUM
ADDRESS OF STATUS RETURN

~
to

35 •

t"I

til

):II

0,0

til
til
til

:I:
~

"00
"11
"22
"33

t"I

t;
0

l:tI

0
til

TST205 MAIN
25
26
27
28
29
30
31
32
33
34
35 000036
36 000036
37
38
39 000042
40 000046
41 000052
42 000052
43 000054
44 000056

til

en
.SBTTL

l:tI

TST205 MAIN

TST205 IS USED IN TRANSACTION TIM08N TO TEST THE TIM
ACTION CODE C.KEY. THE TST RECEIVES THE TEXT ASSOCIATED
WITH THE FUNCTION KEY DEPRESSED. THE TEXT IS 35 CHARACTERS
IN LENGTH. EACH KEY HAS A SPECIFIED TEXT. DEPENDING
ON THE TEXT, THEN, TST206 SENDS ONE OF 7 REPLIES TO DISPLAY
THE TEXT. THE MULTIPLE REPLIES ENABLE THE TEXT OF EACH FUNCTION
KEY TO BE DISPLAYED ON A DIFFERENT LINE

l:tI
til
2:

til

t"1

1-1

til
to.;
1-1

;

TSTEP: :
016504

000002

MOV

2(R5),R4

ADDRESS OF EXCHANGE MSG

012702
012701

000006
000022'

MOV
MOV

#6,R2
#FUNC,Rl

INITIALIZE
ADDRESS OF POSSIBLE REPLY TEXT

CMP
BEQ
SOB

(Rl)+,(R4)
40$
R2,20$

LOOP UNTIL FUNC TEXT FOUND

20$:
022114
001401
077203

til
"I:J
til

END LOOP

2:
G'l

45 000060
46 000060
47
48
49 000064
50
51
52 000070
53 000074
54 000100
55

40$:
010267

177730

MOV

R2,REPLYN

DEFINE REPLY NUMBER

010467

177712

MOV

R4,MSGRPY

SAVE ADDRESS OF EM FOR FPT

012705
004767
000207
000001

000000'
OOOOOOG

MOV
CALL
RETURN
.END

iFPTRPY,R5
REPLY

SEND REPLY

til

=
~

til

til
til
til

SYMBOL TABLE
FPTRPY
FUNC

OOOOOOR
000022R

• ABS.

1
w

000000
000102
ERRORS DETECTED:

MSGRPY 000002R
REPLY = ****** GX

REPLYN
SIZE

000014R
000012R

STATUS

000016R

TSTEP

000036RG

:.:
tX'

000
001

VIRTUAL MEMORY USED: 84 WORDS
DYNAMIC MEMORY: 16142 WORDS
ELAPSED TIME: 00:00:03
TST205,TST205=TST205

( 1 PAGES)
62 PAGES)

til
til

=
"III
til

l:O
til

SYMBOL TABLE
FPTRPY OOOOOOR
FUNC = ****** GX
• ABS. 000000
000
000100
001
ERRORS DETECTED: 0

til

MSGRPY 000002R
REPLY = ****** GX

VIRTUAL MEMORY USED: 84 WORDS ( 1 PAGES)
DYNAMIC MEMORY: 3532 WORDS
13 PAGES)
ELAPSED TIME: 00:00:02
LI,OBJ/CRF=TST20S.MAC

REPLYN
SIZE

000014R
000012R

STATUS

000016R

TSTEP

000034RG

l:'1

I-f

til

~
I-f

t:l

til

ttl
t:"I
t!3

>'
til
CREF

SYMBOL CROSS REFERENCE
SYMBOL

I
~

FPTRPY
FUNC
MSGRPY
REPLY
REPLYN
SIZE
,STATUS
TSTEP

000000
******
000002
******
000014
000012
000016
000034

R
GX
R
GX
R
R
R
RG

#1-5
1-38
#1-7
1-51
1-9
1-8
1-10
#1-33

:B:

t"'I

REFERENCES

VALUE

VOl

til
t!3

1-50

(')
~

*1-47
#1-13
#1-11
#1-15

*1-44

otil
til

t!3
t'IiI
t!3
~

tzl

lZ:

(')

tzl

tot

....til

toi
....
lZ:

(j)

INDEX

Absolute addresses, 6-13
Absolute binary output, 6-13
Absolute expression, 3-16
Absolute mode, 5-5, 5-7
~solute module, 6-34
Absolute program section, 6-37
Address boundaries, 6-31
Address mode syntax, B-1
Address modes, 5-1
Addressing forms, summary, 5-7
Allocating byte data, 6-17
Allocating word data, 6-18
Allocation requirements, 6-35
Alternate radix, 6-25
Ampersand, 3-2
Angle brackets, 3-3, 3-15, 6-4,
6-25, 6-28, 7-4 to 7-5,
7-16 to 7-17
Apostrophe, 7-10
Argument substitution, 7-16
Arithmetic addition operator
or autoincrement indicator,
3-2
Arithmetic division operator,
3-2
Arithmetic multiplication
operator, 3-2
Arithmetic subtraction
operator or autodecrement
indicator, 3-2
ASCII character set, A-I
ASCII conversion, 3-14
ASCII conversion characters,
6-19
.ASCII directive, 6-20
.ASCIZ directive, 6-21
.ASECT directive, 6-38
Assembler directives, 6-1,
B-1, B-2
Assembler version, 6-8
Assembly language, B-1
Assembly listing, 2-6
Assembly pass 1, 1-1
Asterisk, 3-2
At sign (@), 3-2
Attribute of the current
location counter, 3-12
Autodecrement deferred
mode, 5-3, 5-7
Autodecrement mode, 5-7
Autoincrement deferred
mode, 5-3, 5-7
Autoincrement mode, 5-2, 5-7

AB operator, 6-25
Backslash, 3-2
Binary operators, 3-15
Blank lines, 2-2
.BLKB directive, 6-30
.BLKW directive, 6-30
Blocks of storage, reserving,
6-30
Branch instruction addressing,
5-8
.BYTE directive, 6-17

AC operator, 6-27
Calling conventions, E-8
Calling macros, 7-3
Changing default radix, 3-13
Changing value of location
counter, 3-12
Character set, 3-1
Character substitution, 7-16
Code and data separation, 6-38
Code or data sharing, 6-38
Coding standard, E-l
Colon, 3-1
Comma, 3-2
Command string format, 8-1
Comment, 6-14, E-2
Comment field, 2-5
Comment field indicator, 3-2
Complementing an argument, 6-27
Complex relocatable expression,
3-16
Complex relocation, 4-1
Concatenated, 6-35
Concatenation of macro arguments,
7-10
Conditional assembly block, 6-41
Conditional assembly directive,
6-41, 6-42
Conditional branches, E-13
Continuation lines, 2-2
Creating local symbols
automatically, 7-7
Creating program sections, 6-36
Cross-reference listing (CREF) ,
8-5
Cross-reference processor, 8-4
.CSECT directive, 6-17, 6-38
Current location counter, 2-2,
3-11, 3-14, 5-6, 6-29

Index-l

INDEX (Cont.)

AD operator, 6-25
Data storage directives, 6-17
Date, 6-8
Default object module name,
6-11
Default register definitions,
6-15
Deferred addressing indicator,
3-2
Defining macros, 7-1
Device registers, E-2
Diagnostic, 7-14
Diagnostic error message
summary, D-1
Direct assignment operator, 3-1
Direct assignment statements,
3-7
Directives, 2-5, 5-9, 6-1
Double ASCII character indicator,
3-2
Double colon, 3-1, 3-7
Double equal sign, 3-1, 3-7
Double quote, 3-2, 3-14, 6-19
.DSABL directive, 3-7, 3-9,
6-13 to 6-15, 6-27
Duplication of code, 7-17

EMT, 5-8
.ENABL directive, 5-8, 6-13 to
6-15, 6-27
.END directive, 6-31
.ENDC directive, 6-41
.ENDM directive, 7-2
End of the source input, 6-31
.ENDR directive, 7-18
Entry-point instructions, 6-33
.EOT directive, 6-31
Equal sign, 3-1
Error codes, D-1
.ERROR directive, 7-14
Error messages, 8-8 to 8-11
Evaluation of expressions, 3-15
.EVEN, 6-29
Exclamation point, 3-2
Exiting, E-9
Expressions, 3-14, 3-15
External expression, 3-15, 3-16
External symbols, 6-40
Externally-defined macro, 7-18

AF operator, 3-14, 6-27
File specification format, 8-7
File Specification Qualifiers
TRAX, 8-3

Finding address mode of macro
arguments, 7-13
Finding number of characters
in strings, 7-12
Floating point,
data, 6-26
number, 6-28
number specification, 6-27
rounding, 6-14, 6-26
storage directives, 6-27
truncation, 6-14,- 6-27
.FLT2 directive, 6-27
.FLT4 directive, 6-27
Forbidden instruction usage,
E-12
Form-feed, 6-13, 7-3
Format control, 2-6
Formatting standards, E-9
Forward referencing, 3-8
Function directives, 6-13

General purpose registers, E-2
General registers, 3-9
Global,
label, 6-40
references, 6-15
symbol, 2-3, 6-40
symbol directory, 1-2
.GLOBAL directive, 3-7, 6-39
GSD, 1-2

Hardware registers, E-2
Horizontal formatting, 2-6

.IF directive, 6-40
.IFF directive, 6-43, 6-44
.IFT directive, 6-43
.IFTF directive, 6-43
.IIF directive, 6-45
Illegal characters, 3-3
Immediate conditional assembly,
6-45
Immediate expression indicator,
3-2
Immediate mode, 5-4, 5-7
Immediate mode deferred, 5-5
Implicit .WORD directive, 2-5,
6-18
Indefinite repeat block
directives, 7-15
Index deferred mode, 5-4, 5-7
Index mode, 5-4, 5-7

Index-2

INDEX (Cont.)

Indirect command files, 8-6
Initial argument or expression
indicator, 3-2
Initial register indicator, 3-2
Instruction set, C-l
Invoking MACRO under TRAX, 8-1
.IRP directive, 7-15, 7-16
.IRPC directive, 7-15, 7-16
Item or field terminator, 3-1

Keyword arguments, 7-4, 7-9

Label field, 2-2
Label terminator, 3-1
Left angle bracket, 3-2
Left parenthesis, 3-2
.LIMIT directive, 6-31
Line format, E-l
Linking, 4-1
.LIST directive, 6-1
Listing conditional assemblies,
6-4
Listing control directives, 6-1
Listing control switches, 8-2
Listing level count, 6-2
Listing of binary extensions, 6-4
Listing of comments, 6-4
Listing of generated binary code,
6-3
Listing of macro calls, 6-4
Listing of macro definitions, 6-4
Listing "of macro expansion binary
code, 6-4
Listing of repeat range
expansions, 6-4
Listing of source line sequence
numbers, 6-3
Listing of source lines, 6-4
Listing of the current location
counter, 6-3
Listing of the symbol table, 6-5
Local symbol block, 6-14
Local symbol block delimiters,
3-10
Local symbols, 3-6, 3-10, 3-11
Location counter, 6-36
Location counter control
directives, 6-29
Logical AND operator, 3-2, 6-42
Logical inclusive OR operator,
3-2, 6-42
Lower-case ASCII, 6-14

Macro arguments, 7-6
Macro attribute directives, 7-11
Macro call, 2-5, 7-3, 7-5
Macro call arguments, 7-4
Macro call numeric argument
indicator, 3-2
MACRO character sets, A-l
Macro definition, 7-1, 7-15
Macro definition arguments, 7-4
Macro definition formatting, 7-3
Macro definition termination, 7-2
MACRO directives, 5-9, C-4
Macro directives, 7-1
Macro expansion termination, 7-3
Macro library directive, 7-18
Macro name, 7-1, 7-4
Macro names, E-5
Macro nesting, 7-5
Macro qualifiers, 8-2
Macro symbol table, 3-6
MACRO symbols, 3-5
.MCALL directive, 7-18
Memory allocation, 6-32, 6-33,
6-38
Memory allocation and mapping,
6-32
.MEXIT directive, 7-3
Minus sign, 3-2
Modularity, E-8
Module checking routines, E-9
Module preface, E-5
Multi-defined label, 2-4
Multiple definitions of local
symbols, 3-11
Multiple labels, 2-4

Naming standards, E-2
.NARG directive, 7-11
.NCHR directive, 7-11, 7-12
Negative numbers, 3-13
Nested conditional directives,
6-43
Nested macros, 7-3, 7-5
.NLIST directive, 6-1, 6-11
.NTYPE directive, 7-11, 7-13
Number of arguments in macro
calls, 7-7, 7-11
Number sign, 3-2
Numbers, 3-13
Numeric control, 6-24
Numeric control operators, 6-26,
6-27
Numeric directives, 6-26

Index-3

INDEX (Cont.)

~O operator, 6-25
Object module, 4-1
Object module name, 6-11
Octal radix, 3-13
.000 directive, 6-29
Op codes, 2-4, C-l
Operand field, 2-4
Operand field separator, 3-2
Operating procedures, 8-1
Operator field, 2-4
Order of symbol table search, 3-6
Other symbols, E-3
Overlaid, 6-35
Overlays, 6-33

.PAGE directive, 6-12
Page eject, 7-3
Page ejection, 6-13
Page formatting, 2-6
Page headings, 6-8
Page number, 6-8
Passing numeric arguments as
symbols, 6-45
Percent sign, 3-2
Permanent symbol table, 3-5, C-l
Plus sign, 3-2
.PRINT directive, 7-14
Processor priority, E-3
Program boundaries directive,
6-31
Program counter, 3-9, 5-1
Program modules, E-5
Program section access, 6-33
Program section name, 6-33
Program sections, 3-12, 6-32
Program source files, E-12
Program-local symbols, E-4
Programming standards and
conventions, 2-1
.PSECT directive, 3-12, 6-32,
6-35

AR operator, 6-23
.RAD50 directive, 3-13, 6-22
Radix control, 6-24
Radix control operators, 6-25
.RADIX directive, 3-13, 6-24
Radix-50 character set, A-4
Radix-50 control operator, 6-23
Radix-50 data, 6-22
Read-only access, 6-33
Read/write access, 6-33
Register deferred mode, 5-2
Register expression, 5-1
Register, mode, 5-1, 5-7

Register standards, E-2
Register symbols, 3-9
Register term indicator, 3-1
Relative addresses, 6-13
Relative addressing mode, 5-6
Relative deferred mode, 5-6, 5-7
Relative mode, 5-6, 5-7
Relocatability, 6-34
Relocatable expressions, 3-16, 4-1
Relocatable module, 6-34
Relocatable program sections, 6-37
Relocation, 4-1
Relocation bias, 2-2, 6-34
Repeat block directive, 7-18
.REPT directive, 7-18
Reserving storage, 6-30
Reserving storage space, 3-13,
6-30
Right parenthesis, 3-2

.SBTTL directive, 6-8, 6-11
Scope of the program section, 6-33
Semicolon, 3-2
Sending messages to listing file,
7-14
Separating and delimiting
characters, 3-2
Single ASCII character indicator,
3-2
Single quote, 3-2, 3-14, 6-19,
7-10
Slash, 3-2
Source line sequence numbers, 6-3
Space, 3-1
Special characters, B-1
Special characters in macro
arguments, 7-6
Stack pointer, 3-9
Statement format, 2-1
Storing Radix-50 data, 6-23
Subconditional assembly, 6-43
Subtitle, 6-8
Success/failure indication, E-9
Symbol control directive, 6-39
Symbol examples, E-4
Symbol table listing, 1-2
Symbolic arguments of listing
control directives, 6-3, 6-4
Symbols, E-3
Symbols and expressions, 3-1
System macro libraries, 7-18

Tab, 3-1
Tab character, 2-2
Table of contents, 6-4, 6-11

Index-4

INDEX (Cont.)

Teleprinter mode, 6-5
Terminal argument or expression
indicator, 3-2
Terminal register indicator, 3-2
Terminating directives, 6-31
Terms, 3-14
Time-of-day, 6-8
.TITLE directive, 6-11
Title of the object module, 6-8
Translating to ASCII, 6-20, 6-21
Translating to Radix-50, 6-22
Trap instructions, 5-8
TRAX Command String Format, 8-1
TRAX File Specification Format,
8-7
TRAX File Specification Qualifiers,
8-3
TRAX Indirect Command Files, 8-6
TRAX MACRO in Batch Mode, 8-6
TRAX MACRO Qualifiers, 8-2
TRAX Operating Procedures, 8-1

Version number, 6-12
Version number standard, E-13

Unary and binary operators, 3-5
Unary control, 6-24
Unary operator ordering, 6-27

.WORD directive, 3-11, 6-18

Unary operators, 3-15
Unconditional assembly, 6-43
Undefined symbols, 3-7, 3-14
Universal unary operator or
argument indicator, 3-2
Up arrow or circumflex, 3-2
Up-arrow, 3-3
Up-arrow (A) construction, 7-5
User symbol table, 3-5
User-defined and macro symbols,
3-5
User-defined macro libraries,
7-18
Using the standard symbolics, E-3

Index-5

TRAX MACRO
Reference Manual
AA-D340A-TC

READER'S COMMENTS

NOTE:

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

"Did you find errors in this manual?

If so, specify by page.

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

Is there sufficient documentation on associated system programs
required for use of the software described in this manual? If not,
what material is missing and where should it be placed?

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

o Assembly language programmer
o Higher-level language programmer
o Occasional programmer (experienced)
o User with little programming experience
o Student programmer
o Non-programmer interested in computer concepts and capabilities
Name

Date __________________________

Organization _______________________________________________________________
Street ______________________________________________________________________
City ___________________________ State _____________ Zip Code ______________

,.._.. ............

-------------------------------------------------------------F'old lIere-----------------------------------------------------------.

.----------------------------------------------- Do Not Tear - Fold lIere and Staple ----------------------------------------------.

FIRST CLASS
PERMIT NO. 33
MAYNARD, MASS.
BUSINESS REPLY MAIL
NO POSTAGE STAMP NECESSARY IF MAILED IN THE UNITED STATES

Postage will be paid by:

~DmDDmD
Software Documentation
146 Main Street ML5-5/E39
Maynard, Massachusetts 01754