Digital PDFs

AA-V027A-TC

March 1983

225 pages

Original

8.3MB

Document:	PDP-11 MACRO-11 Language Reference Manual Version 5
Order Number:	AA-V027A-TC
Revision:	0
Pages:	225
Original Filename:	http://bitsavers.org/pdf/dec/pdp11/rsx11m_s/RSX11M_V4.1_Apr83/4_ProgramDevelopment/AA-V027A-TC_macro11_198303.pdf

OCR Text

PDP-11 MACR0-11
Language Reference Manual
AA-V027A-TC

March 1983

This document describes how to use the MACR0-11 relocatable assembler to develop PDP-11 assembly language programs. Although no
prior knowledge of MACR0-11 is required, the user should be familiar
with the PDP-11 pmcessor addressing modes and instruction set. This
manual presents detailed descriptions of MACR0-11 's features, including source and command string control of assembly and listing functions, directives for conditional assembly and program sectioning, and
user-defined and system macro libraries. The chapters on operating
procedures previously were found in two separate manuals (the
PDP-11 MACR0-11 Language Reference Manual and the /AS/RSX
MACR0-11 Reference Manual). This manual should be used with a
system-specific user's guide as well as a Linker or a Task Builder manual.
This manual supersedes previous editions, Order Numbers
AA-50758-TC, published 1980, AA-5075A-TC, published 1977, and
DEC-11-0IMRA-B-D, published 1976.

Operating System:

Software:

VAX/VMS Version 3
RSTS/E Version 8
RSX-11 M Version 4
RSX-11 M-PLUS Version 2

MACR0-11 Version 5

To order additional documents from within DIGITAL, contact the Software Distribution
Center, Northboro. Massachusetts 01532.
To order additional documents from outside DIGITAL, refer to the instructions at the back
of this document.

digital equipment corporation . maynard, massachusetts

First Printing, August 1977
Revised, January 1980
Updated, December 1981
Revised, March 1983

The information in this document is subject to change without notice and should not
be construed as a commitment by Digital Equipment Corporation. Digital Equipment
Corporation assumes no responsibility for any errors that may appear in this document.
The software described in this document is furnished under'a license and may be used
or copied only in accordance with the terms of such license.
No responsibility is assumed for the use or reliability of software on equipment that is
not supplied by DIGITAL or its affiliated companies.

©Digital Equipment Corporation 1977, 1980, 1981, 1983.
All Rights Reserved.
Printed in U.S.A.
A postage-paid READER'S COMMENTS form is included on the last page of this
document. Your comments will assist us in preparing future documentation.
The following are trademarks of Digital Equipment Corporation:

~nmnomn™
DEC
DEC mate
DECsystem-10
DECSYSTEM-20
DEC US
DECwriter
DIBOL

MASSBUS
PDP
P/OS
Professional
Rainbow
RSTS
RSX

M19400

UNIBUS
VAX
VMS
VT
Work Processor

CONTENTS

Page
PREFACE

PART I

MACR0-11: ASSEMBLY AND FORMATTING

CHAPTER

THE MACR0-11 ASSEMBLER

1-1

1.1
1.2

ASSEMBLY PASS 1
ASSEMBLY PASS 2

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

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

CHAPTER

Label Field
Operator Field
Operand Field
Comment Field
FORMAT CONTROL

1-1

PART II

PROGRAMMING IN MACR0-11 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

DIRECT ASSIGNMENT STATEMENTS
REGISTER SYMBOLS
LOCAL SYMBOLS
CURRENT LOCATION COUNTER
NUMBERS
TERMS
EXPRESSIONS

3-1
3-3
3-3
3-4
3-6
3-6
3-6
3-8
3-10
3-11
3-13
3-15
3-15
3-16

CHAPTER

RELOCATION AND LINKING

4-1

CHAPTER

ADDRESSING MODES

5-1

5.1
5.?.
5.3
5.4
5.5
5.6
5.7
5.8
5.9

REGISTER MODE
REGISTER DEFERRED MODE
AUTOINCREMENT MODE
AUTOINCREMENT DEFERRED MODE
AUTODECREMENT MODE
AUTODECREMENT DEFERRED MODE
INDEX MODE
INDEX DEFERRED MODE
IMMEDIATE MODE

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

Separating and Delimiting Characters
Illegal Characters
Unary and Binary Operators
MACR0-11 SYMBOLS

Permanent Symbols
User-Defined and Macro Symbols

iii

5-5
5-5
5-6

5 .10

5.11
5.12
5.13
5.14
PART III

ABSOLUTE MODE
RELATIVE MODE
RELATIVE DEFERRED MODE
BRANCH INSTRUCTION ADDRESSING
USING TRAP INSTRUCT!ONS

5-6
5-7
5-8

5-9
5-9

MACR0-11 DIRECTIVES

CHAPTER
6.1
6. l . l
6 .1. 2
6. l . 3
6 .1. 4
6 .1. 5
fi.1.6

6.2
6.2.1
6.2.2

6.3
6.3.1

6.3.2
6.3.3
6.3.4
6.3.5
6. 3. ()
6.3.7

6.3.8
') • 4
6.4.1
5.4.1.1
6.4.1.2
6.4.2

6.4.2.1
6.4.2.2
5.5
11.5.1
6.5.2
fi.5.3
6.5.4
6.6
f, • 7

6.7.1
6.7.1.1
6.7.1.2
fi.7.1.3
6.7.2

6.7.3
11.7.4

6.8
6.8.1
6.8.2
e:;. 9
fJ.9.l
h. 9.?.

G.9.3
6 .10
5.10.1
fi.10.2

GENERAL ASSEMBLER DIRECTIVES

6-1

LISTING CONTROL DIRECTIVES
.LIST and .NLIST Directives
.TITLE Directive
.SBTTL Directive
. !DENT Di re ct i ve
.PAGE Directive/Page Ejection
.REM Directive/Begin Remark Lines
FUNCTION DIRECTIVES
.ENABL and .DSABL Directives
Cross-Reference Directives: .CROSS
and .NOCROSS
DATA STORAGE DIRECTIVES
.BYTE Directive
.WORD Directive
ASCII Conversion Characters
.ASCII Directive
.ASCIZ Directive
.RAD50 Directive
Temporary Radix-50 Control Operator
.PACKED Directive
RADIX AND NUMERIC CONTROL FACILITIES
Radix Control and Unary Control Operators
.RADIX Directive
Temporary Radix Control Operators
Numeric Directives and Unary Control
Operators
Floating-Point Storage Directives
Temporary Numeric Control Operators:
"'c and "'F
LOCATION COUNTER CONTROL DIRECTIVES
.EVEN Directive
.ODD Directive
.BLKB and .BLKW Directives
.LIMIT Directive
TERMINATING DIRECTIVE: .END DIRECTIVE
PROGRAM SECTIONING DIRECTIVES
.PSECT Directive
Creating Program Sections
Code or Data Sharing
Memory Allocation Considerations
.ASECT and .CSECT Directives
.SAVE Directive
.RESTORE Directive
SYMBOL CONTROL DIRECTIVES
.GLOBL Directive
.WEAK Directive
CONDITIONAL ASSEMBLY DIRECTIVES
Conditional Assembly Block Directives
Subconditional Assembly Block Directives
Immediate Conditional Assembly Directive
FILE CONTROL DIRECTIVES
.LIBRARY Directive
.INCLUDE Directive

6-4
6-9

6-15
6-15

6-16
6-17

IS-18
6-18
6-19
6-22
o-23
6-23
'1-24
6-25
6-26
1)-28

n-29
'S-31
6-31
6-32
6-32

6-32
n-33
6-34
6-35

6-36
6-37
6-38
6-38
6-38
6-39

6-40
6-40
5-41
6-45
6-47
11-47
6-48
5-49
6-49
6-51
6-51
6-52
S-53

fi-53
6-56
6 -59
6-6~

6-6~

6-61

CHAPTER

7-1

MACRO DIRECTIVES

7.1
7 .1.1
7 .1. 2
7 .1. 3
7 .1. 4

DEFINING MACROS
7-1
.MACRO Directive
7-1
7-2
.ENDM Directive
.MEXIT Directive
7-3
MACRO Definition Formatting
7-4
7-4
CALLING MACROS
ARGUMENTS IN MACRO DEFINITIONS AND MACRO
CALLS
7-4
7-6
Macro Nesting
Special Characters in Macro Arguments
7-7
Passing Numeric Arguments as Symbols
7-7
Number of Arguments in Macro Calls
7-8
Creating Local Symbols Automatically
7-8
7-10
Keyword Arguments
7-11
Concatenation of Macro Arguments
MACRO ATTRIBUTE DIRECTIVES: .NARG, .NCHR, AND
7-12
.NTYPE
7-12
.NARG Directive
7-13
.NCHR Directive
7-14
.NTYPE Directive
7-16
.ERROR AND .PRINT DIRECTIVES
INDEFINITE REPEAT BLOCK DIRECTIVES: .IRP AND
7-17
. IRPC
7-17
.IRP Directive
7-18
.IRPC Directive
7-20
REPEAT BLOCK DIRECTIVE: .REPT, .ENDR
7-20
MACRO LIBRARY DIRECTIVE: .MCALL
7-21
MACRO DELETION DIRECTIVE: .MDELETE

7.2
7.3
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
7.6.1

7.6.2
7.7
7.8
7.9
PART IV

OPERATING PROCEDURES

CHAPTER

IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES 8-1

8.1

RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES
Initiating MACR0-11 Under RSX-llM/
RSX-llM-PLUS
Method 1 - Direct MACR0-11 Call
Method 2 - Single Assembly
Method 3 - Install, Run Immediately, and
Remove On Exit
Method 4 - Using the Indirect Command
Processor
Default File Specifications
MCR Command String Format
DCL Operating Procedures
MACR0-11 Command String Examples
IAS MACR0-11 OPERATING PROCEDURES
Initiating MACR0-11 Under IAS
IAS Command String
IAS Indirect Command Files
IAS Command String Examples
CROSS-REFERENCE PROCESSOR (CREF)
IAS/RSX-llM/RSX-llM-PLUS FILE SPECIFICATION
MACR0-11 ERROR MESSAGES UNDER IAS/RSX-llM/
RSX-llM-PLUS

8 .1.1

8.1.1.1
8.1.1.2
8.1.1.3
8.1.1.4
8 .1. 2

8 .1. 3
8 .1. 4
8 .1. 5

8.2
8.2.1
8.2.2
8.2.3

8.2.4
8.3

8.4
8.5
CHAPTER

8-1

8-2
8-2
8-2
8-2
8-3
8-3

8-4
8-8
8-13
8-14
8-14
8-14
8-16

8-16
8-17
8-19
8-20

RSTS/RT-11 OPERATING PROCEDURES

9-1

9.1

MACR0-11 UNDER RSTS
RT-11 Through RSTS

9-1
9-1

9 .1.1

9-1

9.5.1
9.5.2
9.5.3

RSX Through RSTS
INITIATING MACR0-11 UNDER RT-11
RT-11 COMMAND STRING
FILE SPECIFICATION OPTIONS
CROSS-REFERENCE (CREF) TABLE GENERATION
OPTION
Obtaining a Cross-Reference Table
Handling Cross-Reference Table Files
MACR0-11 Error Messages Under RT-11

MACR0-11 CHARACTER SETS

A-1

A .1

ASCII CHARACTER SET
RADIX-50 CHARACTER SET

A-1
A-4

MACR0-11 ASSEMBLY LANGUAGE AND ASSEMBLER
DIRECTIVES

B-1

B.l
B.2
B.3

SPECIAL CHARACTERS
SUMMARY OF ADDRESS MODE SYNTAX
ASSEMBLER DIRECTIVES

B-1
B-1
B-3

PERMANENT SYMBOL TABLE

C-1

C.l
C.2

OP CODES
MACR0-11 DIRECTIVES

C-1
C-6

APPENDIX

ERROR MESSAGES

D-1

APPENDIX

SAMPLE CODING STANDARD

E-1

E.l
E.2
E.3
E.3.1
E.3.1.1
E.3.1.2
E.3.1.3
E.3.2
E.3.3
E.3.3.1
E.3.3.2
E.3.3.3
E.3.3.4
E.3.3.5
E.4
E.4.1
E.4.2
E.4.3
E.4.4
E.4.4.1

LINE FORMAT
COMMENTS
NAMING STANDARDS
Registers
General Purpose Registers
Hardware Registers

E-1
E-1
E-2
E-2
E-2
E-2

Oonic:t-orc:

E-2

9 .1. 2

9.2
9.3
9.4
9.5

APPENDIX

A.2
APPENDIX

APPENDIX

E.4.4.2
E.4.4.3
E.4.4.4
E.5
E.5.1
E.5.2
E.5.3
E.5.4
E.6
E.6.1
E.6.2

E.7

nc:iuir-c:i
~-·~--

---J~~--~~

(PST)

Processor Priority
Symbols
Symbol Examples
Local Symbols
Global Symbols
Macro Names
General Symbols
PROGRAM MODULES
The Module Pref ace
The Module
Module Example
Modularity
Calling Conventions (Inter-Module/
Intra-Module)
Exiting
Success/Failure Indication
Module Checking Routines
CODE FORMAT
Program Flow
Common Exits
Code with Interrupts Inhibited
Code in System State
INSTRUCTION USAGE
Forbidden Instructions
Conditional Branches
PROGRAM SOURCE FILES
vi

9-2
9-2
9-Ll

9-5
9-5

9-7
9-7

E-2
E-3
E-3
E-4
E-4
E-4
E-4
E-5
E-5
E-5
E-7
E-8
E-9
E-9
E-9
E-9
E-10
E-10
E-11

E-12
E-12
E-13
E-13
E-14
E-14

E.8
E.8.1
E.8.2

PDP-11 VERSION NUMBER STANDARD
Displaying the Version Identifier
Use of the Version Number in the Program

E-14
E-15

ALLOCATING VIRTUAL MEMORY

F-1

F.l

GENERAL HINTS AND SPACE-SAVING GUIDELINES
MACRO DEFINITIONS AND EXPANSIONS
OPERATIONAL TECHNIQUES

F-1

F.2
F.3

WRITING POSITION-INDEPENDENT CODE

G-1

G.l
G.2

INTRODUCTION TO POSITION-INDEPENDENT CODE

EXAMPLES

G-1
G-2

APPENDIX

SAMPLE ASSEMBLY AND CROSS-REFERENCE LISTING

H-1

APPENDIX

OBSOLETE MACR0-11 DIRECTIVES, SYNTAX, AND
COMMAND LINE OPTIONS

I-1

I. l
I. 2

OBSOLETE DIRECTIVES AND SYNTAX
OBSOLETE COMMAND LINE OPTION

I-1
I-1

RELEASE NOTES

J-1

J.l

CHANGES
CHANGES
CHANGES

APPENDIX

J.2
J.3

ALL VERSIONS OF MACR0-11
MACR0-11/RSX VERSION ONLY
MACR0-11/RT VERSION ONLY

E-15

F-2
F-3

J-1
J-3
J-3

FIGURES
FIGURE

3-1
3-2
6-1
6-2
6-3
6-4
6-5
6-6
6-7
7-1
7-2
7-3

7-4
8-1

G-1
G-2

Assembly Listing Showing Local Symbol Block
Sample Assembly Results
Example of Line Printer Assembly Listing
Example of Teleprinter 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 .SAVE and .RESTORE Directives
Example of .NARG Directive
Example of .NCHR Directive
Example of .NTYPE Directive in Macro
Definition
Example of .IRP and .IRPC Directives
Sample CREF Listing
Example of Position-Dependent Code
Example of Position-Independent Code

3-12
3-13
6-5
F,-7

6-13
6-16
6-21
6-39
6-50
7-13

7-14
7-15

7-19
8-19
G-3
G-3

TABLES
TABLE

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

5-1

Special Characters Used in MACR0-11
Legal Separating Characters
Legal Argument Delimiters
Legal Unary Operators
Legal Binary Operators
Addressing Modes
vii

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

5-2
6-1
6-2

n-3
6-4
6-5
6-6
6-7
8-1
8-2
8-3
8-4
9-1
9-2
9-3
I-1

Symbols Used in Chapter 5
Directives in Chapter ry
Symbolic Arguments of Listing Control
Directives
Symbolic Arguments of Function Control
Directives
Symbolic Arguments of .PSECT Directive
Program Section Default Values
Legal Condition Tests for Conditional
Assembly Directives
Subconditional Assembly Block Directives
File Specification Default Values
MACR0-11 File Specification Switches
DCL Command Qualifiers
DCL Parameter Qualifiers
Default File Specification Values
File Specification Options
/C Option Arguments
Old and New Directives and Syntax

viii

5-2
6-1
6-10

6-19
6-41
6-48
6-54
6-57
8-4
8-6
8-8
8-13
9-3
9-4
9-6
I-1

PREFACE

0.1

MANUAL OBJECTIVES AND READER ASSUMPTIONS

This manual is intended to enable users to develop programs
the MACR0-11 assembly language.

coded

No prior knowledge of the MACR0-11 Relocatable Assembler is assumed,
but the reader should be familiar with the PDP-11 processors and
related terminology, as presented in the PDP-11 Processor Handbooks.
The reader is also encouraged to become familiar with the linking
process, as presented in the applicable system manual
(see Section
0.3), because linking is necessary for the development of executable
programs.
If a terminal is available to the reader, he/she is advised to try
some of the examples in the manual or to write a few simple programs
that illustrate the concepts covered.
Even experienced programmers
find that working with a simple program helps them to understand a
confusing feature of a new language.
The examples in this manual were done on an RT-11 system.
MACR0-11
may also be used on IAS/RSX-llM, RSX-llM-PLUS and RSTS systems (see
Part IV for information about operating procedures).
It can be assumed that all references to RSX-llM also apply to
RSX-llM-PLUS with the exception of those in Chapter 8, which deals
with each system individually.

0.2

STRUCTURE OF THE DOCUMENT

This manual has four parts and eight appendices.
Part I introduces MACR0-11.
Chapter 1 lists the key features of MACR0-11.
Chapter 2 identifies the advantages of following programming
standards and conventions and describes the format used in coding
MACR0-11 source programs.
Part II presents general information essential to programming with the
MACR0-11 assembly language.
Chapter 3 lists the character set and describes the
terms, and expressions that form the elements of
instructions.

symbols,
MACR0-11

Chapter 4 describes the output of MACR0-11 and presents concepts
essential to the proper relocation and linking of object modules.
Chapter 5 describes how data stored in memory can be accessed and
manipulated
using
the addressing modes recognized by the PDP-11
hardware.
Part III describes the MACR0-11 directives that control the processing
of source statements during assembly.
Chapter 6 discusses
functions.

directives

Chapter 7 discusses
expansion of macros.
Part IV presents the
programs.

used

directives

operating

for

generalized

used

the

procedures

for

assembling

MACR0-11

definition

and

MACR0-11

Chapter 8 covers the IAS, RSX-llM, and RSX-llM-PLUS systems.
Chapter 9 covers the RSTS/RT-11 systems.
Appendix A lists the
MACR0-11 programs.

ASCII

and

Radix-50

character

sets

used

Appendix B lisls
the !:SI:Jec1a.L character::; recognized by MACR0-11,
summarizes the syntax of the various addressing modes used in PDP-11
processors, and briefly describes
the
MACR0-11
directives
in
alphabetical order.
Appendix C lists alphabetically the permanent symbols that
defined for use with MACR0-11.

have

been

Appendix D lists alphabetically the error codes produced by MACR0-11
to
identify various types of errors detected during the assembly
process.
Appendix E contains a coding standard that is recommended practice
preparing MACR0-11 programs.

Appendix F discusses several methods of conserving dynamic memory
systems who may experience difficulty in
space
for
users of small
assembling MACR0-11 programs.
Appendix G is a discussion of oosition-independent code <PTC).
Appendix H contains an assembly and cross-reference listing.
Appendix I contains obsolete MACR0-11 directives, syntax, and
line options.
Appendix J
MACR0-11.

describes

the

differences

from

the

last

command

release

0.3

ASSOCIATED DOCUMENTS

For descriptions of documents associated with this manual,
the applicable documentation directory listed below:

refer

IAS Documentation Directory
RSX-llM-PLUS Information Directory and Index
RSX-llM/RSX-llS Information Directory and Index
Guide to RT-11 Documentation
RSTS/E Documentation Directory

0.4

DOCUMENT CONVENTIONS

The color red is used in command
type-in.

string

examples

indicate

user

The symbols defined below are used throughout this manual.
Definition

Symbol
[]

Brackets indicate
optional.

that

the

enclosed

argument

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.

CHAPTER 1
THE MACR0-11 ASSEMBLER

MACR0-11 provides the following features:
1.

Source 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

Relocatable object modules

Global symbols for linking object modules

Conditional assembly directives

Program sectioning directives

User-defined macros and macro libraries

10.

Comprehensive system macro library

11.

Extensive source
functions.

and

command

string

listing;

control

optional

listing

MACR0-11 assembles one or more ASCII source files containing MACR0-11
statements into a single relocatable binary object file.
The output
of MACR0-11 consists of a binary object file and a file containing the
table of contents, the assembly listing, and the symbol table. An
optional cross-reference listing of symbols and macros is available.
A sample assembly listing is provided in Appendix H.

1.1

ASSEMBLY PASS 1

During pass 1, MACR0-11 locates and reads all required macros from
libraries, builds symbol tables and program section tables for the
program, and performs a rudimentary assembly of each source statement.
In the first step of assembly pass 1, MACR0-11 initializes all the
impure data. areas (areas containing both code and data) that will be
used internally for the assembly process.
These areas include all
dynamic storage and buffer areas used as file storage regions.

1-1

THE MACR0-11 ASSEMBLER
MACR0-11 then calls a system subroutine which transfers a command line
into memory.
This command line contains the specifications of all
files to be used during assembly.
After scanning the command line for
proper syntax, MACR0-11 initializes the specified output files.
These
files are opened to determine if valid output file specifications have
been passed in the command line. ·
MACR0-11 now initiates a routine which retrieves source lines from the
input file.
If no input file is open, as is the case at the beginning
of assembly, MACR0-11 opens the next
input file
specified
in the
command line and starts assembling the source statements.
MACR0-11
first determines the length of the instructions, then assembles them
according to length as one word, two words, or three words.
At the end of assembly pass 1,
MACR0-11
reopens the output files
described
above.
Such
information as the object module name, the
program version number, and the global symbol directory (GSD) for each
program section are output to
the object file to be used later in
linking the object modules.
After writing out the GSD for
a given
program section, MACR0-11 scans through the symbol tables to find all
the global symbols that are bound to that particular program section.
MACR0-11 then writes out GSD records to the object file for these
symbols.
This process is done for each program section.

1.2

ASSEMBLY PASS 2

On pass 2 MACR0-11 writes the object records to the output file while
generac1ng
both the assembly listing and the symbol table listing for
the program.
A cross-reference listing may also be generated.
Basically, assembly pass 2 consists of the some steps performed
in
assembly
pass 1,
except that all
source statements containing
MACR0-11-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 a i i
cne object records,
together with
relocation records that hold the information necessary
for linking the object file.
The information in the object file, when passed to the Task Builder or
Linkeri 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 applicable
system manual (see Section 0.3 in the Preface).

1-2

CHAPTER 2
SOURCE PROGRAM FORMAT

2.1

PROGRAMMING STANDARDS AND CONVENTIONS

Programming standards and conventions allow code written by a person
(or group)
to be easily understood by other people. These standards
also make the program easier to:
Plan
Comprehend
Test
Modify
Convert
The actual standard used must meet local user requirements. A sample
coding standard is provided in Appendix E. Used by DIGITAL and its
users, this coding example simplifies both communications and the
continuing task of software maintenance and improvement.

2.2

STATEMENT FORMAT

A source program is composed of assembly-language statements.
Each
statement must be completed on one line. Although a line may contain
132 characters (a longer line causes an error
(L)
in the assembly
listing),
a
line of 80 characters is recommended because of
constraints imposed by listing format and terminal line size.
~Jank
~.~~es, although legal, have no s_ignificance in the source program.
A MACR0-11 statement may have as many as four
fields.
These fields
are identified by their order within the statement and/or by the
separating characters between the fields.
The general format of a
MACR0-11 statement is:
[Label:]

Operator Operand

[;Comment(s)]

The label and comment fields are optional. The operator and operand
fields are interdependent;
in other words! when both fields are
present in a source statement, each field is evaluated by MACR0-11 in
the context of the other.
A statement may contain an operator and no operand, but the reverse is
not true.
A statement containing an operand with no operator is
illegal and is interpreted by ~ACR0-11 during assembly as an implicit
.WORD directive (see Section 11.3.2).
MACR0-11 interprets and processes source program statements one by
one.
Each statement causes MACR0-11 either to perform a specified
assembly process or to generate one or more binary instructions or
data words.
2-1

SOURCE PROGRAM FORMAT
2.2.1

Label Field

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 used by MACR0-11 to assign
memory addresses to the source program statements as they are
encountered during the assembly process. Thus, a label is a means of
symbolically referring to a specific statement.
When a program section is absolute, the value of the current location
counter is absolute;
its value references an absolute virtual memory
address (such as location 100). Similarly, when a program section is
relocatable, the value of the current location counter is relocatable;
a relocation bias calculated at link time is added to the apparent
value of the current location counter to establish its effective
absolute virtual address at execution time.
(For a discussion of
program sections and their attributes, see Section 6.7.)
If present, a label must be the first field in a source statement and
must be terminated by a colon (:~~ For example, if the value of the
current location counter is absolute 100(8), the statement:
ABCD:

MOV

A,B

assigns the value 100(8) to the label ABCD.
If the location counter
value were relocat~ble,
the final value of ABCD would be 100(8)+K,
where K represents the relocation bias of the program section, as
calculated by the Task Builder or 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 value of the current location counter is 100(8), the multiple
labels in the following statement are each assigned the value 100(8):
ABC:

$DD:

A7.7:

A,B

MOV

Multiple labels may also appear on successive lines.
statements
ABC:
$DD:
A7.7:

MOV

For example, the

A,B

likewise cause the same value to be assigned to all three labels.
This second method of assigning multiple labels is preferred because
positioning the fields consistently within the source program makes
the program easier to read (see Section 2.3).
A double colon (: :) defines
example, the statement
ABCD::

MOV

the

label

global

symbol.

For

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 6.8)
or by independently assembled object modules.
References to this label in other modules are resolved when the
modules are linked as a composite executable image.

2-2

SOURCE PROGRAM FORMAT
The legal characters for defining labels are:
A through z
0 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
MACR0-11 source programs.
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. An error code (M)
is generated in the assembly
listing if the first six characters in two or more labels are the
same.
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 MACR0-11 to generate an error code (M)
in
the assembly listing.
Furthermore, any statement in the source
program which references a multi-defined label generates an error code
(D) in the assembly listing.

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. Chapters 6 and 7 describe these three types of
operators.
When the operator is an instruction mnemonic, a machine instruction is
generated and MACR0-11 evaluates the addresses of the operands which
follow. When the operator is a directive MACR0-11 performs certain
control actions or processing operations during the assembly of the
source program. When the operator is a macro call, MACR0-11 inserts
the code generated by the macro expansion.
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,
character*, as in the following examples:
MOV A,B
MOV
MOV@A,B

tab!

any

non-RAD50

;The space terminates the operator MOV.
A,B

;The tab terminates the operator MOV.
;The @ character terminates the operator MOV.

Appendix A.2 contains a table of Radix-50 characters.
2-3

SOURCE PROGRAM FORMAT
Although the statements above are all equivalent in
second statement is the recommended form be ca use
MACR0-11 coding conventions.

2.2.3

function, the
it conforms to

Operand Field

When the operator is an instruction mnemonic (op code) , the operand
field contains program variables that are to be evaluated/manipulated
by the operator. The operand field may also supply arguments to
MACR0-11 directives and macro calls, as described in Chapters 6 and 7,
respectively.
Operands may be expressions or symbols, depending on the operator.
Multiple expressions used in the operand field of a MACR0-11 statement
must be separated by a comma; multiple symbols similarly used may be
delimited by any legal separator (a comma, tab, and/or space). An
operand should be preceded by an operator field;
if it is not, the
statement is treated by MACR0-11 as an impli~it .WORD directive (~ee
Section 6.3.2).
When the operator field contains an op code, associated operands
always expressions, as shown in the following statement:
MOV

are

R0,A+2(Rl)

On the other hand, when the operator field contains a MACR0-11
directive or a macro call, associated operands are normally symbols,
as shown in the following statement:
.MACRO

ALPHA

SYM1,SYM2

Refer to the description of each MACR0-11 directive (Chapter 7)
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

;Comment field

the tab between MOV and A terminates the operator field and defines
the beginning of the operand field;
a comma separates the operands A
and B; 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 MACR0-11,
are checked only for ASCII legality and then included in the assembly
listing ?S they appenr in the source text~

2-4

SOURCE PROGRAM FORMAT
All comment fields must begin with a
semicolon
(;).
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 necessary in source listings for later analysis,
debugging, or documentation purposes.

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.
DIGITAL's standard source line format is shown below:
Label - begins in column 1
Operator - begins in column 9
Operands - begin in column 17
Comments - begin in column 33.
These formatting conventions are not mandatory;
free-field coding
is
permissible.
However, note the increase readability after formatting
in the example below:
REGTST:BIT#MASK,VALUE;COMPARES BITS IN OPERANDS.
1

REGTST: BIT

(columns)

#MASK,VALUE

;Compares bits in operands.

Page formatting and assembly listing considerations are discussed
in
Chapter 6 in the context of MACR0-11 directives that may be specified
to accomplish desired formatting operations.
Appendix E contains a
sample coding standard.

2-5

CHAPTER 3
SYMBOLS AND EXPRESSIONS

This chapter describes the components of MACR0-11 instructions:
the
character set, the conventions observed in constructing symbols, and
the use of numbers, operators, terms and expressions.

3.1

CHARACTER SET

The following characters are legal in MACR0-11 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 6.2.1, .ENABL LC).

The digits 0 through 9.

The characters
{period)
and $
{dollar sign) •
These
characters
are
reserved for use as Digital Equipment
Corporation system program symbols.

The special characters listed in Table 3-1.
Table 3-1
Special Characters Used in MACR0-11

Character

Designation

Function

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.

Equal sign colon

Direct assignment
operator;
macro
keyword
indicator;
causes error (M) in listing if
an attempt is made to change
the value of the symbol.
(continued on next page}
3-1

8YMBULS AND EXPRESSIONS

Table 3-1
(Cont.)
Special Characters Used in MACR0-11
Character

Designation

Function

==:

Double equal
sign colon

Direct assignment
operator;
defines the symbol as a global
s ym bo 1 ;
ca uses e r r or ( M) i n
listing if an attempt is made
to change the value of the
symbol.

Percent sign

Tab

Item or field terminator.

Space

Item or field terminator.

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
npPrrlrnr

rl11tnni:>rri:>mi:>nt

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.

(continued on next page)

3-2

SYMBOLS AND EXPRESSIONS
Table 3-1
(Cont.)
Special Characters Used in MACR0-11
Character

3.1.1

Designation

Function

Single quote

Single ASCII character
indicator; or concatenation
indicator.

Up arrow or
circumflex

Universal unary operator
argument indicator.

Backslash

Macro call numeric argument
indicator.

Separating and Delimiting Characters

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

defined

Table 3-2
Legal Separating Characters
Usage

Character

Definition

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.

3.1.2

Illegal Characters

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

recognized MACR0-11
I f a character is not an element of the
character set, it is replaced in the listing by a question
mark, and an error code (I)
is printed in the assembly
1 i sting.
The exception to this is an embedded null which,
when detected, terminates the scan of the current line.

If a legal MACR0-11 character is used in a source statement
with illegal or questionable syntax, an error code {Q) is
printed in the assembly listing.

3-3

SYMBOLS AND EXPRESSIONS
Table 3-3
Legal Argument Delimiters
Character

Definition

Usage

<••• >

Paired angle
brackets

Paired 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).

x ••• 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.

3.1.3

Unary and Binary Operators

Legal MACR0-11 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.
Because a term preceded by a unary operator is considered
to contain that operator, a term so specified can be used alone or as
~......
.....

, c111c111...
_._ - - ~
...

- &
Vl.

- CUI

- ... - - - - - : - CAJ::1Lc;::,;::,,1.v11.

Table 3-4
Legal Unary Operators
Unary
Operator

Explanation

Example

Plus sign

Produces the positive
value of A.

Minus sign

-A

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

Effect

(continued on next page)

3-4

SYMBOLS AND EXPRESSIONS
Tab 1 e 3 - 4 ( Cont. )
Legal Unary Operators
Unary
Operator

Explanation

Example

Up-arrow, universal
unary operator.
(This usage is
described in
detail in
Section 6.4.)

"'C24

Produces
the
l's
complement value of
24(8).

"'Dl27

int:erprets 127
decimal number.

as a

Interprets 3.0
1-word,
floating-po int
number.

"034

Interprets 34
octal number.

"'BlHHHHll

Interprets
11000111
as a binary number.

"RABC

Evaluates
ABC
Radix-50 form.

Effect

Unary operators can be used adjacent to each other or in constructions
involving multiple terms, as shown below:
-"'D50
"'c"'o12

(Equivalent to -<"D50>)
(Equivalent to "c<"Ol2>)

Legal MACR0-11 binary operators are described in Table 3-5.
In
contrast to unary operators, binary operators specify actions to be
performed on multiple items or terms within an expression.
Table 3-5
Legal Binary Operators
Binary
Operator

Explanation

Example

Addition

A+B

Subtraction

A-B

Multiplication

A*B

(signed 16-bit
product returned)

Division

A/B

(signed 16-bit
quotient returned)

Logical AND

A&B

Logical inclusive OR

A!B

3-5

SYMBOLS AND EXPRESSIONS
All binary operators have equal priority.
Terms enclosed by angle
brackets 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>

; Equals 11 (8) .
;Equals 7 (8).

MACR0-11 SYMBOLS

MACR0-11 maintains a symbol table for each of the three symbol types
that may be defined
in a MACR0-11 source program:
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) MACR0-11 and is part of
the MACR0-11
image.
The UST (for user-defined symbols) and MST (for
macro symbols) are constructed as the source program is assembled.

3.2.1

Permanent Symbols

Permanent symbols consist of the instruction mnemonics
(see Appendix
C)
and MACR0-11 directives
(see Chapters 6 and 7 and Appendix B).
These symbols are a permanent part of the MACR0-11 image and need not
be defined betore being used
in the operator field of a MACR0-11
source statement (see Section 2.2.2).

3.2.2

User-Defined and Macro Symbols

User-defined symbols are those symbols that are equated to a specific
value through a direct assignment statement (see Section 3.3), appear
as labels (see Section 2.2.1), or act as dummy arguments (see Section
7.1.1).
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 7.1).
They are added
to the Macro Symbol Table as they are encountered
during assembly.
The following rules govern the
symbols:

creation

and

macro

Symbols can be composed of alphanumeric characters,
signs ($), and periods ( .) only (see Note below).

dollar

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 MACR0-11.

Spaces, tabs, and illegal characters must not be embedded
within a symbol.
The legal MACR0-11 character set is defined
in Section 3.1.

3-6

user-defined

SYMBOLS AND EXPRESSIONS
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 value of a symbol depends upon its use in the program.
A symbol
in the operator field may be any one of the three symbol types
described above; permanent, user-defined, or macro. To determine the
value of an operator-field symbol, MACR0-11 searches the symbol tables
in the following order:
1.

Macro Symbol Table

Permanent Symbol Table

User-Defined Symbol Table

This search order allows permanent symbols to be used as macro
symbols.
But the user must keep in mind the sequence in \·Jh i ch 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 search order is:
1.

User-Defined Symbol Table

Permanent Symbol Table

Depending on their use in the source program, user-defined symbols
have either a local
(internal)
attribute or a global (external)
attribute.
Normally, MACR0-11 treats all user-defined symbols as local, that is,
their definition is limited to the module in which they appear.
However, symbols can be explicitly declared to be global symbols
through one of three methods:
1.

Use of the .GLOBL directive (see Section 6.8.1).

Use of the double colon

(: : )

in defining a label (see Section

2.2.1).
3.

Use of the double equal sign {==) or double equal colon sign
(==:) in a direct assignment statement (see Section 3. 3) •

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

3-7

the

end

SYMBOLS AND EXPRESSIONS
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
6.2.1)
the
statement
containing
the
undefined
symbol is flagged with an error code (U)
in the assembly listing.
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 image. Such symbols are referenced from other source
modules in order to transfer control throughout execution.
These
global symbols are resolved at link time, ensuring that the resulting
image is a logically coherent and complete body of code.

3.3

DIRECT ASSIGNMENT STATEMENTS

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 contain an undefined global reference.
The colon 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 contain an undefined global reference.
All the direct assignment statements above allow the user to equate a
symbol with a specific value. After the symbol has been defined it is
entered into the User-Defined Symbol Table. If the general format is
used (= or ==} the value of the symbol may be changed in subsequent
direct assignment statements. If, however, the colon format is used
(=:
or ==:) any attempt to change the value of the symbol will
generate an error (M) in the assembly listing~
A direct assignment statement embodying either the double equal (==)
sign or the double equal colon {==:) sign, as shown above, defines the
symbol as global (see Section 6.8.1).

3-8

SYMBOLS AND EXPRESSIONS

The following examples illustrate
statements.

the

coding

direct

assignment

Example 1:
A=l0

;Direct assignment

B==30

;Global assignment

A=l5

;Legal reassignment

L=:5

;Equal colon assignment

M==:A+2

;Double equal colon assignment
;M becomes equal to 17

L=4

;Illegal reassignment
;M error is generated

Example 2:
C:

D=.

U ,ABLE

MOV

;The symbol Dis equated to., and
;the labels C and E are assigned a
;value that is equal to the location
;of the MOV instruction.

The code in the second example above would not usually be used and is
shown
only to illustrate the performance of MACR0-11 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 (=),double equal sign (==), equal colon sign
(=:) or double equal colon 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
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.
The
following example would cause an error code (U) in the
assembly listing on the line containing the illegal forward
reference:

may

X=Y

(Illegal forward reference)

Y=Z

(Legal forward reference)

Z=l

3-9

followed

SYMBOLS AND EXPRESSIONS

Although one level of forward referencing is allowed for local
symbols, no forward referencing is allowed for global symbols. In
other words, the expression being assigned to a global symbol can
contain only previously defined symbols. A forward reference in a
direct assignment statement defining a global symbol will cause an
error code (A) to be generated in the assembly listing.

3.4

The eight general registers of the PDP-11 processor are numbered 0
through 7 and can be expressed in the source program in the following
manner:
%0
%1

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.
The register definitions listed below are the normal default values
and remain valid for all register references within the source
program.
R0=%0
Rl=%1
R2=%2
R3=%3
R4=%4
R5=%5
SP=%6
r.,,_o. ..,.

r1.,,--o'

;Register 0 definition.
;Register 1 definition.
;Register 2 definition.
;Register 3 definition.
;Register 4 definition.
;Register 5 definition.
;Stack pointer definition.
__
;Program ................ ._ .... ,...V"
,:i,...-1=;_;~.;

'-VUJll...C.L

UC'J....LJl.Ll....LVll•

Registers 6 and 7 are given special names because of their unique
system
functions.
The symbolic default names assigned to the
registers, as listed above, are the conventional names used in all
DIGITAL-supplied PDP-11 system programs.
For this reason, you are
advised to follow these conventions.
A register symbol 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 between 0 and 7, inclusive, or
an error code (R) will appear in the assembly listing. Although you
can reassign the standard register symbols through the use of the
.DSABL REG directive (see Section 6.2.1), 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. All non-standard
register symbols must be defined before they are referenced in the
source program.

3-10

SYMBOLS AND EXPRESSIONS
The % character may be used with any legal
term
specify a register.
For example, the statement
CLR

expression

%3+1

is equivalent in function to the statement
CLR

and clears the contents of register 4.
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, .ASECT, or .RESTORE directive
in the source program (see Figure 3-1).

The range of a
local symbol block
MACR0-11 directives, as follows:
Starting delimiter:

.ENABL LSB

3-11

delimited

(see Section 6.2.l)

through

GYMDCLG AND EXFREGGIONS
Ending delimiter:

one of the following:
Symbolic label (see Section
.PSECT (see Section 6.7.1)
.CSECT (see Section 6.7.2)
.ASECT (see Section 6.7.2)
.RESTORE (see Section 6.7.4)

j
}::.--

,.,-~··

encountered after
Section 6.2.1).

:-·DSABL.

LSB

(see

·., _,,/

Local symbols provide a convenient means of generating labels for
branch instructions and other such references within local symbol
blocks. 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 blocks. 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 and require less
symbol table space than other types of symbols.
Their use is
recommended.
When defining local symbols, use the range from 1$ to 29999$ first.
Local symbols within the range 30000$ through 65535$, inclusive, can
be generated automatically as a feature of MACR0-11.
Such local
symbols are useful in the expansion of macros during assembly (see
Section 7.3.5).
Be sure to avoid multiple definitions of local symbols within the same
local symbol block. For example, if the local symbol 10$ 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 ossembly listing.
For examples of local symbols and local symbol blocks as
in a source program, see Figure 3-1.

they

appear

1
2
3
4

; SiaPle illustration of local
; b~ the label XCrPAS.

s~abolsi

the second block is deliaited

6 000000

0.12700

XCTPRG: HOV

IIHPURE, RO

f Point

CRO>t

;clear a word
;rest if at toP of area

OOOOOOG
7 000004
8 000006

005020
020027

1$:

OOOOOOG
9 000012
10

CLR
CHP

RO, tIHPURT

001374

BNE

012700

XCTPAS: HOV

IIHPPAS, RO

;Point to Pass storase area

1$:

CLR
CMP

<RO>+
RO,tIMPPAr

iClear the area
;rest if at toP of area

BNE
Rrs

iiterate of not
;Return if so

;Iterate if not
iFall in to Perfora Pass initialization

12 000014

OOOOOOG
13 000020
14 000022

to iaPure area

005020
020027

OOOOOOG
15 000026
16 000030

001374
000207

Figure 3-1

Assembly Listing Showing Local Symbol Block

3-12

SYMBOLS AND EXPRESSIONS
3.6

CURRENT LOCATION COUNTER

The period (.) is the symbol for the current location counter.
When
used in the operand field of an instruction, the period 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 MACR0-11
directive, it represents the address of the current byte or word,
as
shown in the second example below.
A:

MOV

if., R0

(The function of the

;The period (.) refers to the address
;of the MOV instruction.
n~~ber

sign (#)

is explained in Section 5.9.)

SAL=0
.WORD

177535,.+4,SAL

;The operand .+4 in the .WORD
;directive represents a value
;that is stored as the second
;of three words during
;assembly.

Assume that the current value of the location counter is 500.
During
assembly, MACR0-11 reserves storage in response to the .WORD directive
(see Section 6.3.2),
beginning with location 500.
The operands
accompanying
the .WORD directive determine the values so stored.
The
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 50n 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, MACR0-11 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
The current
relocatable,
section.

location counter symbol
(.)
is either absolute or
depending on the attribute of the current program

3-13

SYMBOLS AND EXPRESSIONS
The attribute of the current location counter can be changed only
through the program sectioning directives (.PSECT, .ASECT, .CSECT and
.RESTORE), as described in Section 6.7.
Therefore, assigning
to the
counter an expression having an attribute different than that of the
current program section will generate an error code
(A)
in the
assembly listing.
Furthermore, an expression assigned to the counter may not contain a
forward
reference
(a
reference to a symbol that is not previously
defined) • The user must also be sure that the expression assigned
will not force the counter into another program section, even if both
sections involved have the same relocatability.
Either of these
conditions causes MACR0-11 to generate incorrect object file code, and
may cause statements following the error to be flagged with an error
code (P) in the assembly listing.
The following coding illustrates
counter:

the

use

the

current

location

.ASECT
.=500
FIRST:

MOV

.+10,COUNT

.=520
SECOND: MOV

.,INDEX

;Set location counter to
; absolute 500 (octal).
;The label "FIRST" has the value
;500(octal).
; • + 1 0 e qua 1 s 5 10 ( o ct a 1 ) • The
;contents of the location
;5i0(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
;itself, will be deposited in the
;location "INDEX" •

• PSECT
.=.+20
THIRD:

. WORD

;Set location counter to
;relocatable 20 of the
;unnamed program section.
;The label "THIRD" has the
;value 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 foll~wing 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 the preferred ways to reserve
storage space (see Section 6.5.3).

3-14

SYMBOLS AND EXPRESSIONS
3.7

NUMBERS

MACR0-11 assumes that all numbers in the source program are to be
interpreted
in octal radix, unless otherwise specified. An exception
to this assumption 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 6.4.1.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, MACR0-11 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;
MACR0-11
translates such numbers into two's complement form.
Positive numbers
may (but need not) be preceded by a plus sign.
A number containing more than 16 significant bits
(greater than
177777(8)),
is truncated from the left and flagged with an error code
{T) in the assembly listing.
Numbers are always considered to be absolute values;
are never relocatable.

therefore,

they

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:

Sign
Bit

8-bit
Exponent

7-bit
Mantissa

Refer to the appropriate PDP-11 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 3e2e
as follows:

one

16-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.)

3-15

User-Defined

used, with zero
(Appendix C lists

SYMBOLS AND EXPRESSIONS
D.

3.9

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 6.2.1)
is in effect, the automatic global
reference default function of MACR0-11 is inhibited,
and
the statement containing the undefined symbol is flagged
with an error code (U) in the assembly listing.

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.

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>).*

A unary operator followed by a symbol or number.

EXPRESSIONS

Expressions are combinations of terms
joined together by binary
operators (see Table 3-5).
Expressions reduce to a 16-bit value.
The
evaluation of an expression includes the determination of
its
attributes.
A resultant expression value may be any one of four types
(as described
later
in this section):
relocatable,
absolute,
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.
t~erms
are evaluated, where necessary,
before their use in expressions.)
Multiple unary operators are valid
and are treated as follows:*
-+-A
is equivalent to:
-<+<-A>>

* The maximum depth of an expression
is governed by the MACR0-11
assembler's expression stack space.
If an expression exceeds the
assembler's maximum expression depth, the statement is marked with an
(E) error, and processing continues.
3-16

SYMBOLS AND EXPRESSIONS
A missing term, expression, or external symbol is interpreted as a
zero.
A missing or illegal operator terminates the expression
analysis, causing error codes (A) and/or (Q), to be generated in the
assembly listing, depending on the context of the expression itself.
For example, the expression:
A + B 1 77777
is evaluated as

A + B
because the first non-blank character following the symbol B is not a
legal binary operator, an expression separator
(a
comma), or an
operand field terminator (a semicolon or the end of the source line).
NOTE
Spaces within exoressions can s~rve
delimiters only~ between symbols.
other words, the expressions

as
In

A + B

and
A+B
are the same, but the symbols
Bl7
and
B

are not

17
(B

17 is not a single symbol).

At assembly time the value of an external (global) 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
(local) symbol A. This expression, however, when evaluated at link
time takes on the resolved value of the symbol EXTERN, plus the value
of symbol A.
Expressions, when evaluated by MACR0-11, are one of four types:
relocatable,
absolute,
external,
or complex relocatable.
The
following distinctions are important:
1.

~n expression is relocatable if its value is
fixed relative
to the base address of the program section in which it
appears; it will have an offset value added at link time.
Terms that contain labels defined in relocatable program
sections will have a relocatable val~e; similarly, a period
(.)
in a relocatable program section, representing the value
of the current location counter, will also have a relocatable
value.

3-17

SYMBOLS AND

~XPH~SS1UN8

An expression is absolute
if
its value
is
fixed.
An
expression 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,
is an absolute expression.
This is because every term in a
program section has the same relocation bias. When one term
is subtracted
from another,
the resulting bias is zero.
MACR0-11 can then treat the expression as absolute and reduce
it to a single term upon completion of the expression scan.
Terms that contain labels defined in an absolute section will
also have an absolute 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 at link time.

An expression is complex relocatable
following conditions applies:

any

one

It contains a global reference and a relocatable symbol.

It contains more than one global reference.

It contains relocatable
program sections.

terms

belonging

the

different

- The value resulting from the expression has more than one
level of relocation.
For example,
if the relocatable
symbols TAGl and TAG2, associated with
the same program
section,
are specified
in the expression TAGl+TAG2, two
levels of relocation will be introduced, since each symbol
is evaluated in terms of the relocation bias in effect for
the program section.
-

operation o~ner ~nan
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,
and complex relocatable
expressions
is completed at link time.
The maximum number of terms
that can be specified in a complex expression is limited by the
maximum size of the object record.
The maximum number of terms is 20
{decimal).

3-18

CHAPTER 4
RELOCATION AND LINKING

The output of MACR0-11 is an object module that must be processed or
linked before it can be loaded and executed. Essentially, linking
fixes (makes absolute) the values of relocatable or external symbols
in the object module, thus transforming the object module, or several
object modules, into an executable image.
To allow the value of an expression to be fixed at link time, MACR0-11
outputs certain instructions in the object file, together with other
required parameters.
For relocatable expressions in the object
module, the base of the associated relocatable program section is
added to the value of the relocatable expression provided by MACR0-11.
For external expression values, the value of the external term in the
expression (since the externol symbol must be defined in one of the
other object modules being linked together) is determined and then
added to the absolute portion of the external expression, as provided
by MACR0-11.
All instructions that require modification at link time are flagged in
the assembly listing, as illustrated in the example below. The
apostrophe (')
following the octal expansion of the instruction
indicates that simple relocation is required; the letter G indicates
that the value of an external symbol must be added to the absolute
portion of an expression;
and the letter C indicates that complex
relocation analysis at link time is required in order to fix the value
of the expression.
EXAMPLE:
00S06S CLR
000040'

RELOC (RS)

;Assuming that the value of the
;symbol "RELOC", 40, is relocatable
;the relocation bias
;will be added to this value.

00S06S CLR
000000G

EXTERN (RS)

;The value of the symbol "EXTERN" is
;assembled as zero and is
;resolved at link time.

4-1

RELOCATION AND LINKING
005065 CLR
000006G

EXTERN+6(R5)

005065 CLR
000000C

-<EXTERN+RELOC>(R5) ;This expression is complex
;relocatable because it requires
;the negation of an expression
;that contains a global "EXTERN"
;reference and a relocatable term.

;The value of the symbol "EXTERN"
;is resolved at link time
;and added to
;the absolute portion (+6) of
;the expression.

For a complete description of object records output by MACR0-11, refer
to the applicable system manual (see Section 0. 3 in the Preface) •

4-2

CHAPTER 5
ADDRESSING MODES

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."
The PC always contains the address of the next word to be fetched.
This word will be either the address of the next instruction to be
executed, or the second or third word of the current instruction.
Table 5-1 lists the address modes, and Table 5-2 lists the symbols
used in this chapter to describe the address modes. Each mode of
address in the chapter is illustrated using either the single operand
instruction CLR or the double operand instruction MOV.
Table 5-1
Addressing Modes

Form

Mode
Register mode*
Register deferred mode*
Autoincrement mode*
Autoincrement deferred mode*
Autodecrement mode*
Autodecrement deferred mode*
Index mode**
Index deferred mode**
Immediate mode**
Absolute mode**
Relative mode**
Relative deferred mode**
Branch
*

@R or (ER)
(ER)+
@(ER)+
-(ER)
@-(ER)
E (ER)
@E(ER)
#E
@#E
E

@E
Address

Section
Reference
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9

5.10
5 .11
5.12
5.13

Does not increase the length of an instruction.

** Adds one word to the instruction length for each occurrence of
operand of this form.

5-1

ADDRESSING MODES
Table 5-2
Symbols Used in Chapter 5
Explanation

Symbol
E

Any expression, as defined in Chapter 3.

A register
expression;
in
other
words,
any
expression containing
a
term preceded by a percent
sign (%) or a symbol previously equated
to
such a
term, as shown below:

R0=%0
Rl=R0+1
R2=1+%1

;General register 0.
;General register 1.
;General register 2.

This symbol may also
represent any of the normal
default register definitions (see Section 3.4).
A register expression or an absolute
the range 0 to 7, inclusive.

expression

Format:
R

The register itself (R)
instruction.

contains the operand to be manipulated by

the

Example:
CLR

5.2

;Clears register 3.

Format:
@R or

(ER)

The register (R) contains the address of the operand to be manipulated
by the instruction.
Examples:
CLR
CLR
CLR

@Rl
(Rl)
(%1)

;All these instructions clear
;the word at the address
;contained in register 1.

5-2

ADDRESSING MODES
5.3

AUTOINCREMENT MODE

Format:

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

(R0)+
(R4)+

CLR
CLR
CLR

(R2)+

;Each instruction clears
;the word at the address
;contained in the specified
;register and increments
;that register's contents
;by two.
NOTE

Certain special instruction/address mode
combinations, which are rarely or never
used, do not operate the same on all
PDP-11 processors, as described below.
In the autoincrement mode, both the JMP
and JSR instructions autoincrement the
register before its use on the PDP-11/40
but not on the PDP-11/45 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
the PDP-11/40,
Register 0 originally
contains 100(8):
MOV

R0,(R0)+

;The quantity 102 is moved
;to location 100.

MOV

R0,-(R0)

;The quantity 76 is moved
;to location 100.

The use of these forms
should
be
since they are not compatible
avoided,
with the entire
family
of
PDP-11
processors.
An error code
(Z)
is printed
in the
assembly listing with each instruction
which is not compatible
among
all
members of the PDP-11 family.

5-3

after

ADDRESSING MODES

5.4

AUTOINCREMENT DEFERRED MODE

Format:
@(ER)+
The register (ER) contains a pointer to the address of the operand.
The contents of the
register are
incremented after being used as
pointer.
Example:
CLR

5.5

@(R3)+

;The contents of register 3 point
;to the address of a word to be
;cleared before the contents of the
;register are incremented by two.

AUTODECREMENT MODE

Format:
-(ER)
The contents of the reqister (ER} are decremented before being used as
the address of the operand (see Note in Section 5.3).
Examples:

5.6

CLR

-(R0)

CLR
CLR

-(R3)
- (R2)

;Decrement the contents of the speci; f ied register (0, 3, or 2) by two
;before 11sing its contents
;as the address of the word to be
;cleared.

AUTODECREMENT DEFERRED MODE

Format:
@-(ER)
The contents of the register (ER) are decremented before being used as
a pointer to the address of the operand.
Example:
CLR

@- (R2)

;Decrement the contents of
;register 2 by two before
;using its contents as a pointer
;to the address of the word to be
;cleared.

5-4

ADDRESSING MODES
5.7

INDEX MODE

Format:
E(ERj

An expression (E), plus the contents of a register
(ER), yields the
effective address of the operand.
In other words, the value E is the
offset of the instruction, and the contents of register ER form the
base.
(The value of the expression (E) is stored as the second or
third word of the instruction.)
Examples:

5.8

CLR

X+2(Rl)

MOV

R0,-2(R3)

;The effective address of the word
;to be cleared is X+2, plus the
;contents of register 1.
;The effective address of the
;destination location is -2, plus
;the contents of register 3.

INDEX DEFERRED MODE

Format:
@E(ER)
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.
(The value of the expression (E) is stored as the
second or third word of the instruction.)

Example:
CLR

@114(R4)

;If register 4 contains 100, this
;value, plus the offset 114, yields
;the pointer 214.
If location 214
;contains the address 2000, location
;2000 would be cleared.
NOTE

The expression @(ER) may be used, but it
will be assembled as if it were written
@0(ER), and a word will be used to store
the 0.

5-5

ADDRESSING MCDES
5.9

IMMEDIATE MODE

Format:
#E
Immediate mode allows the operand itself
(E)
to be stored as the
second or third word of the instruction.
The number sign (#) is an
addressing mode
indicator.
Appearing
in the operand field
this
character specifies the
immediate addressing mode,
indicating to
MACR0-11 that the operand itself immediately follows
the
instruction
word.
This mode is assembled as an autoincrement of the PC.
Examples:
MOV
MOV

#100, R0
#X, R0

;Move the value 100 into register 0.
;Move the value of symbol X into
; reg i st er 0 •

The operation of this mode can be shown through
MOV #100, R0, 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,

The source operand (the value 100) is assembled immediately following
Upon execution of the
instruction,
the
the
instruction word.
processor fetches the first word (MOV) and increments the PC by 2 so
that
it points to the second word, 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
nov+-

.1..1.'-.n.'-

5.10

iP"\c+-Y"11,..+-;,......1"""11\

.&..1.1."-''-.LU'-'-_.,VlJ./

•

ABSOLUTE MODE

Format:
@#E
Absolute mode is the equivalent of
immediate mode deferred.
The
address expression @#E 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.
Examples:
MOV

@#100, R0.

CLR

@#X

;Move the contents of absolute
;location 100 into register R0.
;Clear the contents of the location
;whose address is specified by
;the symbol X.

5-6

ADDRESSING MODES
The operation of this mode can be shown through
MOV @#100, R0, which assembles as two words:
Location 20:

0 1 3 7 0 0

Location 22:

0 0 0 1 0 0

Lo cat ion 24:

Next instruction

the

first

example,

The absolute address 100 is assembled
immediately following
the
instruction word.
Upon execution of the instruction, the processor
fetches the first word (MOV) and increments the PC by 2 so that it
points to the second word, 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).

RELATIVE MODE

5.11

Format:
E

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.
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 to yield the address of the source operand.
Examples:
CLR
MOV

100

R0,Y

;Clear absolute location 100
;Move the contents of register 0
;to location Y

The 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 0. 0 5 4

Location 24:

NEXT INSTRUCTION

shown

with

the

statement

The offset, the constant 54, is assembled
immediately following
the
instruction word.
Upon execution of the instruction, the processor
fetches the first word (MOV) and increments the PC by 2 so that
it
points to the second word,
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.

5-7

ADDRESSING
The index mode statement:
MOV

100-.-4 (PC) ,R3

is equivalent to the relative mode statement:
MOV

100, R3

100-.-4 is the offset for

the
index mode statement.
The current
location counter
(.)
holds the address of the first word of the
instruction (20, in this case) and the PC has to move down 4 bytes to
reach location 24
(the next instruction).
So, the offset could be
written as 100-20-4 or 54(8).
Therefore, for the
index mode,
the offset
(54(8))
added to the
PC(24(8))
yields the effective address
(54 + 24 = 100 (8)) of the
operand.
Thus, both statements move the contents of location 100 into
3.

NOTE
The addressing
form @#E differs from
form E in that the second or third word
of the instruction contains the abso~u~e
address of the operand, rather than the
relative distance between the operand
and
the PC
(see Section 5.10).
Thus,
the
instruction
CLR
@#100
clears
absolute
location 100,
even if the
instruction is moved from the point at
which
it
was
assembled.
See the
description of the .ENABL AMA function
in
Section ~.2.1,
which causes all
relative mode addresses to be assembled
as absolute mode addresses.

5.12

RELATIVE DEFERRED MODE

Format:
@E
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.
Example:
MOV

@X, R0

;Relative to the current value of
;the PC, move the contents of the
;location whose address is pointed
;to by location X into register 0.

5-8

ADDRESSING MODES
5.13

BRANCH INSTRUCTION ADDRESSING

The branch instructions are 1-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

MACR0-11 performs the reverse operation to form the word
the specified address.
Word offset=

form

the

offset

from

word

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

When the offset ii added to the PC, the PC is moved to the
(PC=.+2).
Hence the -2 in the following calculation.
Word offset=

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

The following conditions generate an error code (A)
listing:

5.14

rather

the

assembly

external

Branching from one program section to another

Branching to a
location
(global) symbol

Specifying a branch address that is out of range,
meaning
that the branch offset is a value that does not lie within
the range -12 8 ( 10) to + 12 7 ( 10) .

that

defined

USING TRAP INSTRUCTIONS

Since the EMT and TRAP instructions do not use the low-order byte of
the instruction word, information is 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.
Expressions greater than 377(8)
are truncated to
eight bits, and an error code
(T)
is generated in the assembly
listing.
For more information on traps see the PDP-11 Processor Handbook
the applicable system manual (see Section 0.3 in the Preface).

5-9

and

CHAPTER 6
GENERAL ASSEMBLER DIRECTIVES

A MACR0-11 directive is placed in the operator field of a source line.
Only one directive is allowed per source line. Each directive may
have a blank operand field or one or more operands.
Legal operands
differ with each directive.
General assembler
categories:

directives

are

divided

Listing control

Function control

Data storage

Radix and numeric control

Location counter control

Terminator

Program sectioning and boundaries

Symbol control

Conditional assembly

10.

File control

into

the

following

Each is described in its own section of this chapter (see Table 6-1
for an alphabetical listing of the directives and the associated
section reference) •
Table 6-1
Directives in Chapter Six
Directive

Section
Reference

Function

.ASCII

Stores delimited string as a sequence
of the 8-bit ASCII code of their
characters.

n.3.4

.ASCIZ

Same as .ASCII except the
followed by a zero byte.

6.3.5

string

(continued on next page)
6-1

GENERAL ASSEMBLER DIRECTIVES

Table 6-1 (Cont.)
Directives in Chapter Six
Section
Reference

Function

Directive
.ASECT

Similar to .PSECT.

6.7.2

• BLKB

Allocates bytes of data storage •

6.5.3

• BLKW

Allocates words of data storage •

6.5.3

.BYTE

Stores successive bytes of data.

6.3.1

• CROSS

Enables cross reference •

6.2.2

• CSECT

Similar to .PSECT •

6.7.2

.DSABL

Disables
functions.

assembler

fi.2.1

• ENABL

Enables specified assembler functions .

6.2.1

• END

Indicates end of source input.

6.6

• ENDC

T_;:I.; __ .._ __

.J..llU.J..1...Cl~C.::>

specified

--..:.t

c;uu

-&
V.L

---...:J.;.&-.;-- .... 1

1...V!1U.J..~.J..V!1Cl..L

- - - - - \ . . . . 1 ....
Cl.::>.::>CillU..Lj

u.::;. J_

block .
• EVEN

Ensures that current value
location counter is even •

the

n.5.1

. FLT2

Generates 2 words of storage for
floating-point number argument.

each

6.4.2.1

• FLT4

Generates 4 words of storage for
floating-point number argument.

each

6.4.2.1

.GLOBL

Defines listed symbols as global.

6.8.1

. I DENT

Provides additional means of
an object module.

labeling

6 .1. 4

.IF

Assembles block if
tions are met.

condi-

6.9.1

.IFF

Assembles
false.

block

condition

tests

6.9.2

.IFT

Assembles
true.

block

condition

tests

6.9.2

.IFTF

Assembles block regardless of whether
condition tests true or false.

6.9.2

• IIF

Permits writing a one line conditional
assembly block.

6.9.3

• INCLUDE.

Includes another MACR0-11 source file .

6.10.2

. LIBRARY

Adds file to MACR0-11
1 ist.

() .10 .1

specified

library

{continued on next page)
6-2

GENERAL ASSEMBLER DIRECTIVES
Table Fi-1 (Cont.)
Directives in Chapter Six
Directive

Section
Reference

Function

.LIMIT

Allocates two words for storage. At
link time puts the lowest address of
the load image in the first of the
saved words and the address of the
first free word following the image
in the second.

6.5.4

.LIST

Increments
listing count
certain types of code.

lists

6 .1.1

.NL I ST

Decrements listing count or suppresses
certain types of code.

n. i.1

.NOCROSS

Disables cross reference.

6.2.2

.ODD

Ensures that the current value
location counter is odd.

.PACKED

Generates packed
digits per byte.

. PAGE

Starts a new listing page .

6.LS

.PSECT

Declares names for program sections
and establishes their attributes.

6.7.1

.RAD50

Generates
format.

.RADIX

Changes radices throughout
or
portions of the source program.

. REM

Delimits a section of comments •

.RESTORE

Retrieves a previously .SAVEd
section.

program

6.7.4

.SAVE

Places the current program section on
top of the program section context
stack.

6.7.3

.SBTTL

Produces
a
table
of contents
immediately
preceding the assembly
1 is ting and puts subheadings on each

6 .1. 3

.TITLE

Assigns a name to the
and puts headings on
the assembly listing.

object module
each page of

6 .1. 2

.WEAK

Defines listed symbols as WEAK.

.WORD

Generates successive words of
the object module.

data

decimal

6-3

of the

data,

Radix-50

two

packed
in

6.5.2
6.3.8

6.3.6
6.4.1.1
6 .1. 6

6.8.2

data in

6.3.2

GENERAL ASSEMBLER DIRECTIVES
6.1

LISTING CONTROL DIRECTIVES

Listing control directives control the content, format, and pagination
of all line printer {see Figure 6-1) and teleprinter (see Figure 6-2)
assembly listing output. On the first line of each page, MACR0-11
prints the following {from left to right):
1.

Title of the object module, as established through the .TITLE
directive {see Section ~.1.2).

Assembler version identification.

Day of the week.

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
6.1.3).
In the line printer format {Figure 6-1)
binary extensions
statements generatinq more than one word are listed horizontally.

for

In the teleprinter format
(Figure 6-2) binary
extensions
for
statements generating more than one word are listed vertically. There
is no explicit truncation of output to 80 characters by the assembler.

6-4

1
2
3
4
5
6
7

GETSYM
; Scan off a RAD50 S!:llflbol t L1~ave with scar1 Poir1ter set at r1ext non-blar1k
; char Past end of S!:IRtbOl o s~abol buffer clear and z set if no S!:IRtbOl
; seen; In this case scan Pointer is unaltered.
;-

8 000126
9 000130

°'I
U1

10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37

000136
000142
000144
000146
000154
000156
000162
000164
000166
000172
000176
000202
000204
000206
000212
000216
000222
000224
000226
000232
000236
000240
000244
000246
000252
000254
000260

010146
016767
012701
005041
005041
136527
001436
116500
003431
006300
016011

00 0000G OOOOOOG
000004G
1

GETSYM::MOV
MOV
HOV
CLR
CLR
BITD

000000' 000040
000262'

000462'

116500
003421
006300
066011

000262'

116500
003411
060021

00 0262 I

020127
001347
105765
003370

000004G

000602'
1

2t:
000262'
3$:
012601
016700
000207

4$:

OOiOOOOG

Figure 6-1

BEG
MOVB
BLE
ASL
MOV
GE TC HR
HOVI!
BLE
ASL
ADD
GETCHR
HOVB
DLE
ADD
GETCHR
CMP
BNE
TSTB
BGT
SET ND
MOV
HOV
RETURN

;save work resister
Rlr-<SP>
; Save scar1 Pointer in case of rescan
CHRPNTrSYMBEG
;Poir1t at er1d of S!:lltbOl buffer
tSYMBOL+4rR1
;Now clear it
-<R1>
-<Rl>
CTTBL<RS)rtCT.ALP ;rest first char for alPhabetic
4$
;Exit if not with z set
;HaP to RADSO
CTTBL2 (RS), RO
3$
;Exit if not
;Hake word inde>~
RO
RSOTB1 <RO>,< FU> ;Load the hi!=Jh char
;Get another char
CTTBL2 (RS), R()
;Handle it as above
3$
RO
R50TB2<RO>, ml>
;Now set low order char
;MaP and test it
CTTBL2<R5>rR<>
3$
;Just add in the low char• advance Pointer
RO,< Rl >+
;Get followins char
R1.tSYHBOLt4
He st if at er1d of s1:1mbol buffer
;Go aSain if no
u
CTTBL2<R5>
ff lush to end of S!:llftbOl if !:les
2$
;Now scan to a non b 1 ar1'1, char
;Restore work resister
CSP>trR1
;set z if no s1:1mbol found
SYHBOLrRO
; E:<i t

Example of Line Printer Assembly Listing

G1
tz:I

tz:I

t"t
;J:il

en
en
tz:I
~

tXI

t"t
tz:I

!:t:I
0

!:t:I
tz:I

()

t-3

<
tz:I

°'I
°'

38
39
40
41
42
43
44
45 000262
46 000272
47 000302
48 000312
49 000322
50 000332
51 000342
..i..,. 000352
53 000362
54 000372
55 000402
56 000412
57 000422
I::"'")

; Table CTTBL2
; Ir1de>: with 7 bit ASCII value to set corresF-ondin5l RAD50 value
; If EQ 0 then space, if LT 0 then not RAD50; Other bits reserved.
;-

200
200
200
200
000
200
036
046
200
010
020
030
200

200
200
200
200
200
200
037
047
001
011
021
031
001

200
200
200
200
200
200
040
200
002
012
022
032
002

CTTBL2:

Figure 6-1

• NLIST
• BYTE
.BYTE
.BYTE
• BYTIE
, BYTE
.BYTIE
.BYTE
• BYTIE
.BYT:E
.BYTE
• BYTIE
• [1YTE:
.BYTE

(Cont.}

I1EX
200,200,200.200,200.200.200.200
200.200.200.200,200,200,200,200
200,200,200,200,200,200,200,200
200,200,200,200,200.200,200,200
$
000,200,200.200,033,200,200,200
200,200,200.200,200.200,034,200
036r037r040r041r042r043r044r045 ;01234567
046r047r200r200,200r200,200,200 ;99
200,001,002~003,004,oos,006,007 ; ABCDEFG
010r011r012r013,014r015~016r017 ;HIJKLMNO
020.021,022.023,024,025,026,027 ;pQRSTUVW
030,031,032,200.200.200.200.200 ;xvz
200,001.002,003,004,oos,006,007 ; abcdef 5'

Example of Line Printer Assembly Listing

G:
tr:
2

tr:

~·

t"'

:'.!>

u:
u:

tr:
3:
tt:
t""
tr:;
~

tr::

1-3
H

tr:1
00

;+
2

3
4
5

; ·-

6
7
8 000126
9 000130

10 000136
11 000142
12 000144
13 000146

"'I
-.I

14 000154
15 000156
16 000162
17 000164
18 000166
19 000172
20 000176
21 000202
22 000204
23 000206

GETSYM
Scan off a RAD50 s~mbol. Leave with scan Pointer set at next non-blank
char Past end of s~mbol. S~~bol buffer clear and Z set if no s~•bol
seen; In this case scan Pointer is unaltered.

010146 Gl::TSYM: : MDV
016767
MDV
OOOOOOG
OOOOOOG
012701
MDV
000004G
005041
CLR
005041
CLR
I1ITB
136527
000000'
000040
001436
BEG
116500 u:
MDVB
000262'
E:LE
003431
006300
ASL
016011
MDV
000462'

Rl,-<SP>
CHRPNT,SYMBEG

;save work resister
;save scan Pointer in case of rescan

tSYMBDLt4, R1

;Point at end of

-CR1>
- ( R1>

;Now clear it

buffer

CTTBL<RS>,tCT.ALP ;rest first char for alPhabetic
4$

CTTBL2<R5> ,RO

; E>:i t if not with
iMoP to RAD50

z set

;[;dt if not
3$
;Make word index
RO
R50H11 <r~o >, <r~1 > ;Load the hi sh char

GETCHf~

116500
000262'
003421
006300
066011
000602'

s~mbol

MOVB

CTHIL2 (RS) , RO

BLE
ASL
ADD

3$
RO
R50TB2<RO>, <R1 >

Figure 6-2

;Get another char
;Handle it as above

Example of Teleprinter Assembly Listing

24 000212
25 000216
26
27
28
29

000222
000224
000226
000232

30 000236
31 000240

°'ex>I

32
33
34
35

000244
000246
000252
000254

36 000260
37
38
39
40
41
42
43
44

116500
000262'
003411
060021

GETCHR
MOVD

CTTBL2 <R5 >, f;:O

BLE

3$
RO,<R1>t

ADD

2$:

GETCHR
CMF'

020127
000004G
001347
105765
000262'
003370

BNE
TSTB
BGT
SET NII
HOV
HOV

3$:
012601 4$:
016700
OOOOOOG
000207

R1,tSYMBOLt4
1$

CTTBL2<R5>

;Now set low ordur char
;MaP and test it
;Just add in the low char, advance pointer
;Get followins char
;rest if at end Clf si:>mbol buffer
;Go asain if no
;Fl•Jsh to end of S':llllbol if ':les

RE TU rm

;

>
Cf.l
t%)

;Now scan to a nc1n bl ar1k char
;Restore work re:=li ste r
;set z if no s':lmt•ol found
;Exit

;+
; Table CTTBL2
; I ride>: with 7 bit ASCII value to set co r resPor1dtr1S RAD50 value

t'i

Cf.l

2$
<SP>+. Rl
SYMDOL,RO

~
t%)

zt%)

0 then space, if LT 0 then not RADSO; Other bits reserved.

3:
tp

t'i

t%)

H
~

t%)
(')

t-:3

<
t%)
Cf.l

.NLIST

BEX

Figure 6-2 (Cont.)

Example of Teleprinter Assembly Listing

GENERAL ASSEMBLER DIRECTIVES

j .LIST I
6.1.1

1.NLIST

.LIST and .NLIST Directives

Formats:
.LIST
.LIST
. NLIST
.NLIST
where:

arg
arg

arg

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

symbolic

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
incremented; at each occurrence of an .NLIST directive, the listing
level count is decremented.
When the level count is negative, the
listing is suppressed
(unless the
line
contains
an
error).
Conversely, when the level count is greater than zero, the listing is
generated regardless of the context of the line.
Finally, when the
count is zero, the line is either listed or suppressed, depending on
the listing controls currently in effect for the program.
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
F-this line should list
G-this line should list
• ENDM

.LIST

LTEST
A-this line
E-this line
F-this line
G-this line

;Listing level count is -2.
;Listing level count is -1.
;Listing
;Listing
;Listing
;Listing

level
level
level
level

,• r i.... c +- m.:arer,..._,

should
should
should
should

;List test
;Listing level count is 0 ..
;Listing level count is -1.

'-'

list
list
list
list

ILi~"

.L. 'V

count
count
count
count

is
is
is
is

0.
0.
0.
0.

OVT"\.:at""'t.C!', l"'\.l"'\

'-At"'UlJW.LVJ.l•

;Call the macro
;Listing level count
;Listing level count
;Listing level count
;Listing level count

is
is
is
is

0.
0.
0.
0.

Note that the lines following line E will list because the listing
level count remains 0.
If a .LIST directive is placed at the
beginning of a program, all macro expansions will be listed unless a
.NLIST directive is encountered.

6-9

GENERAL ASSEMBLER DIRECTIVES
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,
c.s shown
in the example
below:
.MACRO

.LIST

;List next line.

.NLIST

;Do not list remainder of macro
;expansion .

X=.

. ENDM
. NLIST

X-

;Do not list macro expansions •

The symbolic arquments allowed for use with the listing directives are
described
in Table n-2.
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 the 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 ~-2.
Table 6-2
Symbolic Arguments of Listing Control Directives
Argument

Default

Function

SEQ*

List

Controls the listing of the sequential
numbers assigned to the source lines.
If this number
field
is suppressed
through
an
.NLIST SEQ directive,
MACR0-11 generates a tab,
effectively
allocating blank space for the field.
Thus, the positional relationships of
the other fields in the listing remain
undisturbed.
During
the
assembly
process, MACR0-11 examines each source
line for
possible error conditions.
For any line in error, the error code
is printed preceding the number field.

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

6-10

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

Default

Function
MACR0-11 does not assign line numbers
to
files
that have had such numbers
assigned by other programs (an editor
program, for instance).

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, MACR0-11
does not generate a tab, nor does it
allocate space for the field,
as
is
the case with the SEQ field described
above.
Thus, the suppression of the
current location counter (LOC) field
effectively
left-justifies
all
subsequent fields
(while preserving
positional
relationships)
to
the
position
normally occupied by the
counter's 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 LOC
field.

BEX

List

Controls
the
listing
of
binary
extensions
(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
definitions
expansions.

List

Controls the listing of macro
and repeat range expansions.

1 is ting
of
and
repeat

macro
range
calls

If the .NLIST ~rguments SEO, LOC, BIN, and SRC are in effect at
the same time,
that is,
if ~11
four significnnt fields in the
listing ~re to be suppressed, the printing of the resulting blank
line is inhibited.
(continued on next page)

n-11

GENERAL ASSEMBLER

~IRECT!VES

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

Argument

Default

No list

Controls
the
expansions.

MEB

No list

of
macro
Controls
the
listing
expansion binary code.
A .LIST MEB
directive lists only
those
macro
expansion
statements that generate
binary code.
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.
A .NLIST CND directive lists
only satisfied conditional coding.

No list

Controls the listing of all listing
directives having
no arguments,
in
other words, the directives that alter

"""'kL-11C

listing

,_.,._,

__ ... _.&...

.LJ..::>L-J.l!'::j

.LCVC.L

l...VUlll.o

i.;-~.;

macro

TOC

List

Controls the listing of the table of
contents during assembly pass 1 (see
Section 6.1.3 describing
the
.SBTTL
directive).
This argument does not
affect the printing of
the
full
assembly listing during assembly pass
2.

SYM

List

Controls the

table

listing

resulting

the

symbol

from the assembly of

the source program.
TTM

No list

Controls the listing output format.
The default
is set to line printer
format.
Figure 6-1
illustrates the
line printer output format.
Figure
6-2 illustrates the teleprinter output
format.

Any argument specified in a .LIST/.NLIST directive other than those
listed
in Table 6-2 causes the directive to be flagged with an error
code (A} in the assembly listing.
The listing control options can also be specified at assembly time
through switches
included
in the command string to MACR0-11 (see
Section 8.1.3 and/or the appropriate system manual}.
The use of these
switches overrides all corresponding listing control (.LIST or .NLIST}
directives specified in the source program.
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.

6-12

LISTING CONTROL EXAMPLE

MACR-0 V05.00 Saturday 08-Jan-83

2
3
4
5

9:~s

Page 1

• TITLE

LISTING CONTROL EXAMPLE

.LIST

;ust 11acro exPar1sions

; Listin.:l cor1trol test 1r1ac rcJ
;-

6
7
8
9

10
11
12
13
14
15 000000
000001
000004

000002

• NLIST
.WORD

000003

.MACRO
.NLIST
.WORD
.LIST
.ENDM

LSTMAC
ARG

LSTMAC
LOC

LOC

LSTMAC
,. NLIST

.worrn

000010

ARG

t'1

;Location counter test
Hh:is is a test co1111ent

LDC

)JI

tll
tll

tzJ

3:
t1'
t'1
tzJ

000001

000002

000003

20
21 000030
000001
000004

000002

Ic IN

;Generat.ed binar1:1 test

BIN
1,2,3,4

18
19 000020

000030
000036

tzJ

ztzJ

16
17 000010

000020

Hhis is a test co111tent

1,2,3,4

.LIST

ARG

H
~

Hhi s is a test co1111111ent
.LIST
BIN

tzJ

()

LSTMAC
.NLIST
.WORD
.LIST

I:EX
BEX

;Binar1:1 extensions test

1,2,3,4

Hhis i 5> a test co11ment

LSTMAC

SRC

IcEX
;source lines test

000003

Figure 6-3

t:J

Listing Produced with Listing Control Directives

<
tzJ

tll

22
23 000040

000040
000046

000001
000004

000002

000003

24
25 000050
000050

000001

000002

.LIST

sr~c

LSTMAC
• NLI ST
• WORD

COM
COM

• LIST

COM

LSTMAC
• NLIST

<COM,BEX>
COM,BEX

.woRr:

000003

.LIST

O"I

26
27
28
29 000060

000060
000062
000064
000066

1--'
~

000001
000002
000003
000004

30
31 000070
000070
32
33

000001
000001

; Co11111ent lines test

1,2,3,4

;comment lines and e:-:ter1ded

binar~

test

t;"l

I.Tl

:z:
tTj

1,2,3,4

•o

COM, I1EX

;J::ll

;En.3ble narrow

.LIST

TTM

LSTMAC
.NLIST
• WORD

SEQ
SEQ

; Se•lUence r1u111be rs test

!fl
!fl

1,2,3,4

;This is a test co111ment

III

listin~

;pi
fTj

:1:

1:-4
IT.I

'.U
t:j
1-1

.LIST

SEQ

LSTMAC
• NLIST
.WORD
.LIST

BEX
BEX

;Ecinar~

extensions test

(")

1,2,3,4

;This is a test comment

•j
t;j

BEX

!fl

• END

Figure 6-3

(Cont.)

tTJ

Listing Produced with Listing Control Directives

GENERAL ASSEMBLER DIRECTIVES

6.1.2

I.TITLE

.TITLE Directive

Format:
.TITLE string
where:

string

represents an identifier of l
or more
Radix-50
characters. Appendix A.2 contains a table of Radix-50
characters.

The .TITLE directive assigns a name to the object module.
The name
assigned is the first six non-blank, Radix-50 characters following the
.TITLE directive.
All
spaces and/or tabs
up
to
the
first
non-space/non-tab character following the .TITLE directive are ignored
by MACR0-11 when evaluating the text string.
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.
This
6-character name bears no relationship to the filename of the object
module, as specified in the command string to MACR0-11.
The name of
an object module
(specified in the .TITLE directive) appears in the
load map produced at link time.
This is also the module name which
the Librarian will recognize.
If the .TITLE directive is not specified, MACR0-11 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 during assembly pass 1 establishes the name for the entire
object module.
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 Radix-50 character, the directive is flagged with an error code
(A) in the assembly listing.

6.1.3

1.senL I

.SBTTL Directive

Format:
.SBTTL string
where:

string

represents an identifier of 1 or more printable
characters.

6-15

ASCII

The
.SBTTL directive is used to produce a table of
contents
immediately preceding the assembly listing and to print the text
following the .SBTTL directive on the second line of the header of
each page in the listing. The subheading in the text will be listed
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 containing the line
sequence number,
the page number,
and the text accompanying each
.SBTTL directive is printed for the assembly listing. The listing of
the table of contents is suppressed whenever an .NLIST TOC directive
is encountered in the source program (see Table 6-2). An example of a
table of contents listing is shown in Figure 6-4.

"TTE"T - RT-11 MULTI-TTY E"T SE

"ACRO V05.00 Saturday 08-Jan-83 10:00

TABLE OF CONTENTS
.MTOUT - Sinsle character output EMT
.MTRCTO - Reset CTRL/O EMT
.MTATCH - Attach to terainal EMT
.MTDTCH - Detach fro• a terainal EMT
.MTPRNT - Print aessase EMT
.MTSTAT - Return aulti-ter•inal s~stea status EMT
MTTIN - Sinsle character inPut
HTTGET - Get a charact~r fro- the rin~ buffer
TTRSET - Reset terainal status bits
MTTPUT - SinSle character outPut
MTRSET - StoP and detach all ter•inals attached to a Job
ESCAPE SEQUENCE TEST SUBROUTINE

5051-

52545556575859606263-

Figure 6-4

6.1.4

Assembly Listing Table of Contents

1.IDENT

.!DENT Directive

Format:
• !DENT
where:

/string/

string

represents a string of six
or
fewer
Radix-50
characters which establish the program identification
or version number.
This string
is included in the
global symbol directory of the object module and is
printed in the link map and librarian listing.

6-16

GENERAL ASSEMBLER DIRECTIVES

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 (see Note
in Section 6.3.4).
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.

In addition to the name assigned to .the object module with the
.TITLE
directive (see Section 6.1.3), the .IDENT directive allows the user to
label the object module with the program version number.
An example of the .IDENT directive is shown below:
.IDENT

/V01.00/

The character string
is converted to Radix-50
representation and
This
included
in the global symbol directory of the object module.
character string aiso appears in the iink map produced at link time
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.
The RT-11 linker allows only one .IDENT string
in a
program.
The
linker uses
the
first .IDENT directive encountered during the first
pass to estabiish the character string that will
be
identified with
all of the object modules.
The RSX-llM task builder allows an .IDENT string for each module
in
the program.
The TASK Builder uses the first .IDENT directive in each
moduie to establish the character string that will be identified with
that module.
Like the RT-11 Linker, the RSX-llM Task Builder uses the
.IDENT directives encountered on the first pass.

5.1.5

.PAGE Directive/Page Ejection

I.PAGE

Format:
.PAGE
The .PAGE directive is used within the source program to perform a
page eject at desired points in the listing.
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 and the
line sequence counter to be cleared.
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.

6-17

GENERAL ASSEMBLER DIRECTIVES
Page ejection is accomplished in three other ways:
1.

After reaching a count of 58 lines in the listing,
MACR0-11
automatically performs a
page eject to skip over page
perforations on line printer paper
and
to
formulate
teleprinter output
into pages.
The page number
is not
changed.

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 causes the page number to be incremented
and the line sequence counter to be cleared.

A page eject is performed when encountering a
new source
file.
In this case the page number is incremented and the
line sequence count is reset.

6.1.6

.REM

.REM Directive/Begin Remark Lines

Format:
.REM comment-character
where:

comment-character

represents a character
end
of the comment
character reoccurs.

that marks
block when

the
the

The .REM directive allows a programmer to insert a block of comments
into a MACR0-11 source program without having to precede the comment

lines with

rho
'-""•-

~Ammonr
__ a,•a"'-"'''-

~h~r~~rOY
- .............. .._.._.. _ _ _ ....

(o\
\I I•

The text between the

delimiting characters
is treated as comments.
any number of lines.
For example:

specified

The comments may span

.TITLE
Remark example
.REM
&
All the text that resides here is interpreted by MACR0-11
to be comment lines until another ampersand character is
found.
Any character may be used in place of the ampersand.&
CLR
PC
• END

6.2

FUNCTION DIRECTIVES

The following function directives are included in a source program to
invoke or inhibit certain MACR0-11 functions and operations incidental
to the assembly process itself.

'1-18

GENERAL ASSEMBLER DIRECTIVES

I
I.DSABL I

j.ENABL
6.2.1

.ENABL and .DSABL Directives

Formats:
.ENABL
.DSABL
where:

arg

arg
arg
represents one or more of the
arguments defined in Table 6-3.

optional

symbolic

Specifying any argument in an .ENABL/.DSABL directive other than those
listed
in Table 6-3 causes that directive to be flagged with an error
code (A) in the assembly listing.
Table 6-3
Symbolic Arguments of Function Control Directives
Argument

Default

ABS

Disable

Enabling this function produces absolute
binary
output
in FILES-11
format.
To
convert this output to Formatted Binary
format
(as
required
by the Absolute
Loader), use the FLX utility.

AMA

Disable

Enabling this function causes all
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.

CDR

Disable

Enabling this
function
causes
source
columns from 73 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.

Enable

Disabling this function causes MACR0-11 to
convert
all ASCII
input to upper-case
before processing it.

Function

(continued on next page)

6-19

GENERAL ASSEMBLER DIRECTIVES
Table 6-3 (Cont.)
Symbolic Arguments of Function Control Directives
Argument

Default

Function

LCM

Disable

This argument,
if enabled,
causes
the
MACR0-11
conditional assembly directives
.IF IDN and .IF DIF to be alphabetically
case
sensitive.
By
default,
these
directives are not case sensitive.

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, a .PSECT directive or a
.RESTORE
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,
.PSECT
directive
or
.RESTORE
directive
is
encountered
following
a paired .DSABL LSB
directive.
The basic function of this directive with
regard
to
.PSECTS
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.
This temporary dismissal
of the current program section may also be
accomplished through the .SAVE and .RESTORE
directives (see Sections 6.7.3 and 6.7.4).
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.

MCL

Disable

This argument, if enabled, causes MACR0-11
to search all known macro libraries for a
macro definition that matches any undefined
symbols appearing in the opcode field of a
MACR0-11 statement.
By
default,
this
option is disabled.
If MACR0-11 finds an
unknown symbol
in the opcode field,
it
either declares a
(U)
undefined symbol
error, or declares the symbol an external
symbol,
depending on the
.ENABL/.DSABL
option setting of GBL (described below).

PNC

Enable

Disabling this
function
inhibits binary
output until an
.ENABL PNC statement is
encountered within the same module.
(continued on next page)

6-20

GENERAL ASSEMBLER DIRECTIVES
Table 6-3 (Cont.)
Symbolic Arguments of Function Control Directives
Argument

Default

REG

Enable

Function
When specified, the .DSABL REG directive
inhibits
the
normal
MACR0-11 default
register definitions; if not disabled, the
default definitions listed below remain in
effect.
R0=%0
Rl=%1
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.

Enable

GBL

• ENABL/.DSABL

This argument, if disabled, causes MACR0-11
to
mark
all
undefined references in
assembly pass 2 with a (U) error in the
assembly listing.
The default for this
option is enabled, which causes MACR0-11 to
treat all undefined symbol references as
global, allowing the linker to resolve
them •

MACRO VOS.00 Saturday_08-Jan~a3 1o:zs
.TITLE

2
3

Pase 1
.ENABL/,DSABL

; ILLUSTRATE .ENABL/,DSABL LC

5
6
7
8

.MACRO

10
11
12
13
14
15

'STORE MACRO IN LOWER CASE

.ENABL

TEXT
.ASCII

$$$

.LIST
.NLIST

HE
BEX

TEXT
.ASCII

is
;Invoke a•cro in lower c•se
/Thi;; is a lower c•se strinsl/

.DSADL

TEXT
.ASCII

WAS
;RE-INVOKE MACRO IN UPPER CASE
/THIS WAS A LOWER CASE STRING/

/This Ufa lower case strinsl/

.ENDH

16 000000

000000

124

150

151

17
18
19

20 000033
000033
21
22

124
000001

Figure 6-5

110

111

;Now disable lower case

.END

Example of .ENABL and .DSABL Directives

6-21

GENERAL

ASSEMbL~H DlH~CT!VES

I
1.NOCROSS I
1.CROSS

6.2.2

Cross-Reference Directives:

.CROSS and .NOCROSS

Formats:
.CROSS
.CROSS syml,sym2, •.• symn
.NOCROSS
.NOCROSS syml,sym2, ••• symn
where:

syml,
sym2, ••.
symn

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

The .CROSS and the
.NOCROSS directives control which symbols are
included
in the cross-reference listing produced by the MACR0-11
assembler.
These directives have an effect only if the /C[R]
or the
/CROSS
qualifier was used
in the command line to select the
cross-reference capability.
By defauit, the cross-reference listing includes the definition and
all
the
references
to every user symbol
in the module.
The
cross-reference listing can be disabled for all
symbols or for a
specified list of symbols.
When the .NOCROSS directive
is used without a
symbol list,
the
cross-reference listing of all the symbols in the module is disabled.
The cross-reference listing of all
the symbols
in the module
is
reenabled when the
.CROSS directive is used without a symbol list.
Any symbol definition or reference that appears after a
.NOCROSS
directive that
is used without a symbol list and before the next
.CROSS directive that is used without a symbol list, is exciuded from
the cross-reference listing.
The
.NOCROSS directive,
used with a symbol
list, disables the
cross-reference listing
for
the listed symbols.
When the .CROSS
directive is used with a symbol list, the cross-reference listing of
the listed symbols is reenabled.
In the following example, the definition of LABELl and the reference
to LOCI and LOC2 are not included in the cross-reference listing.
Example:
LABELI:

.NOCROSS
MOV
.CROSS

LOCl,LOC2

;Stop cross reference
;Copy data
;Reenable cross reference

In the next example, the definition of LABEL2 and the reference to
LOC2 are included in the cross reference, but the reference to LOCI is
not included.
Example:
LABEL2:

.NOCROSS
MOV
.CROSS

LOCl
LOC1,LOC2
LOCl

;Do not cross reference LOCl
;Copy data
;Reenable cross reference
;of LOCI.
6-22

GENERAL ASSEMBLER DIRECTIVES
The .CROSS directive, used without a symbol list, cannot be used to
reenable the cross-reference listing of a symbol specified in the
symbol
list of a
.NOCROSS directive.
In
addition,
if
the
cross-reference listing of all the symbols in a module is disabled,
the .CROSS directive used with a symbol list will have no effect until
the cross-reference listing is reenabled by the .CROSS directive used
without a symbol list.
The .CROSS directive, with no symbol list, is equivalent to the .ENABL
CRF directive,
and the
.NOCROSS directive, with no symbol list, is
equivalent to the .DSABL CRF directive.

6.3

DATA STORAGE DIRECTIVES

A wide range of

data and data types can be generated with the
directives, ASCII conversion characters, and radix-control operators
described in the following sections.

6.3.1

1.BYTE

.BYTE Directive

Format:
.BYTE

exp

;Stores the binary value of the
;expression in the next byte.

.BYTE

expl,exp2,expn

;Stores the binary values of the list
;of expressions in successive bytes.

where:

exp,
expl,

expn

represent expressions that must be reduced to 8 bits
of data or less.
Each expression will be read as a
16-bit word
expression,
the high-order byte to be
truncated.
The high-order byte must be either all
zeros
or
a truncation
(T)
error results.
Multiple expressions must be separated by commas.

The .BYTE directive is used to generate
data in the object module.

successive

bytes

binary

Example:
SAM=5
• =410
.BYTE

"D48,SAM

;The value 060 (octal equivalent of 48
;decimal} is stored in location 410.
·'l"h.::>
...... .a -

u.::11110
.., -

_.,, ~ -

Oll?lt;

J.' J.J _,

ic-

.&. ~

c-+- ......... ,.....:i
..::;i

V .L

II;;;;" U

~ .....

~ I I

i~~-.i-~~ .....
U JJ

.J.. V \,,Cl l.. i

; 411.
The construction "D 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.

6-23

GENERAL ASSEMBLER DIRECTIVES
At link time, it is likely that a relocatable expression will
result
in a value having more than eight bits, in which case the task builder
or linker issues a truncation (T)
error
for
the object module
in
question.
For exc.mple,
the following statements create such a
possibility:
.BYTE

;Stores octal 23 in next byte.

.BYTE

;Relocatable value A will probably
;cause truncation error.

If an expression following
interpreted as a zero:

the

.BYTE

directive

null,

.=420
.BYTE

;Zeros are stored in bytes 420, 421,
;422, 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.

n.3.2

-WORD Directive

Formats:
.WORD

exp

;Stores the binary equivalent of the
;expression in the next word.

.WORD

expl,exp2,expn

;Stores the binary equivalents of the
;list of expressions in successive
;words.

where:

exp,
expl,

represent expressions that must reduce to 16 bits of
data or less.
Multiple expressions must be separated
by commas.

expn
The .WORD directive is used to generate successive words
the object module.

data

Example:
SAL=0
.=500
.WORD

177535,.+4,SAL

;Stores the values 177535, 506, and
;0 in words 500, 502, and 504,
;respectively.

6-24

GENERAL ASSEMBLER DIRECTIVES
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

c;.

;Stores the values 0, 5, and 0 in
;location 500, 502, and 504,
;respectively.

f ...; I

A statement with a blank operator field (one that

contains a symbol
other
than a macro call,
an instruction mnemonic,
a MACR0-11
directive, or a semicolon)
is
interpreted during assembly as an
implicit .WORD directive, as shown in the example below:
.=440
LABEL:

100,LABEL

;Stores the value 100 in location 440
;and the value 440 in location 442.
NOTE

You should not use this technique to
generate .WORD directives because it may
not be
included
in
future
PDP-11
assemblers.

6.3.3

ASCII Conversion Characters

The single quote (') and the double quote
(")
characters are unary
operators that can appear
in any MACR0-11 expression.
Used in
MACR0-11 expressions, these characters cause a 16-bit expression value
to be generated.
When the single quote is used, MACR0-11 takes the next character in
the expression and converts it from its 7-bit ASCII value to a 16-bit
expression value.
The high-order byte of the resulting expression
value is always zero
(0).
The 16-bit value is then used as an
absolute term within the expression.
For example, the statement:
MOV

#'A,R0

moves the following 16-bit expression value into register 0:

00000000 01000001
Binary Value of ASCII A
Thus the expression 'A results in a value of 101(8).
The single quote
(')
character must not
be
followed
by
carriage-return,
null, RUBOUT, line-feed, or form-feed character;
it is, an error code (A) is generated in the assembly listing.

6-25

a
if

GENERAL ASSEMBLER DIRECTIVES
When the double quote is used, MACR0-11 takes the next two characters
in the expression and
converts them to a 10-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,R0

moves the following

l~-bit

expression value into register 0:

01000010 01000001

t__Binary Value Of ASCII A
Binary Value of ASCII B
Thus the expression "AB results in a value of 041101(8).
The double quote (") character, like the single quote
(')
character,
must not be followed by a carriage-return, null, RUBOUT, line-feed, or
form-feed character;
if it is, an error code (A) is generated in the
assembly listing.
The ASCII character set is listed in Appendix A.l.

6.3.4

j.Asc11 I

.ASCII Directive

Format:
.ASCII
where:

/string l/ ••• /string n/

string

is a
string of printable ASCII characters.
The
vertical-tab,
null,
line-feed, RUBOUT, and all other
non-printable ASCII characters, except carriage-return
and
form-feed,
cause an error code (I) if used in an
.ASCII string.
The carriage-return and
form-feed
characters are flagged with an error code (A) because
these characters end the scan of the line,
preventing
MACR0-11
from detecting 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 (;) (see Note at end of section), 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.

6-26

GENERAL ASSEMBLER DIRECTIVES

The .ASCII directive translates character strings into their 7-bit
ASCII
equivalents
and
stores them in the object module.
A
non-printing character can be expressed only by enclosing
its
equivalent octal value within angle brackets.
Each set of angle
brackets so used represents a single character. For example,
in the
following statement:
.ASCII

<15>/ABC/<A+2>/DEF/<5><4>

the expressions <15>, <A+2>, <5>, and <4> represent the values of
non-printing characters.
Each bracketed expression must reduce to
eight bits of absolute data or less.
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/

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:
;Stores the binary representation
;of the letters HELLO in five
;consecutive bytes •

.ASCII

/HELLO/

• ASCII

/ABC/<15><12>/DEF/ ;Stores the binary representation
;of the characters A,B,C,carriage
;return,line feed,D,E,F in eight
;consecutive bytes.

.ASCII

/A<l5>B/

;Stores the binary representation
;of the characters A, <, 1, 5, >,
;and B in six consecutive bytes.
NOTE

The semicolon (;) and equal sign (=) can
be used as delimiting characters in the
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

;ABC;/DEF/

;Stores the binary
;representation of
;the characters
;A, B, C, D, E, and
;F in six
;consecutive bytes;
;not recommended
;practice.

6-27

GENERAL ASSEMBLER DIRECTIVES
.ASCII

/ABC/;DEF;

;Stores the binary
;representations of
;the characters A,
;B, and C in three
;consecutive bytes;
;the characters D,
;E, F, and ; are
;treated as a
;comment •

• ASCII

/ABC/=DEF=

;Stores 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

;The direct
; assignment

=DEF=

•An.::>r~t-inn
·-i:-------··

; . ASCII=DEF is
;performed, and a
;syntax error (Q)
;is generated upon
;encountering the
;second = sign.

6.3.5

, .ASCIZ

.ASCIZ Directive

Format:
.ASCIZ
where:

/string l/ .•. /string n/

string

6-28

GENERAL ASSEMBLER DIRECTIVES

represent delimiting characters.
These delimiters may
be any paired printing characters, other than the
equal sign (=), the left angle bracket
(<),
or the
semicolon
(;) (see Note in Section 6.3.4), 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
.ASCIZ directive is flagged with an error
code (A) in the assembly listing.
The .ASCIZ directive is similar 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=l5
LF=l2
HELLO:

10$:

~.3.6

.ASCIZ
.EVEN

MOV
MOV
MOVB
BNE

#HELLO,Rl
#LINBUF, R2
(Rl)+, (R2)+
HJ$

;Introductory message

;Get address of message.
;Get address of output buffer.
;Move a byte to output buffer.
;If not null, move another byte.

1.RADSO

.RAD50 Directive

Format:
.RAD50
where:

/string l/ •.. /string n/

string

represents a series of characters to be packed.
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.

6-29

GENERAL ASSEMBLER DIRECTIVES
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, cause an error code (I)
if used in a .RAD50 string.
The carriage-return and
form-feed
characters result
in an error code (A)
because these characters end the scan of the
line,
preventing
MACR0-11
from detecting
the 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
(;)
(see Note in Section 6.3.4), 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
.RAD50 directive is flagged with an error
code (A) in the assembly 1 isting.

The .RAD50 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);
therefore, any 6-character symbol
can
be stored in two consecutive words.
Examples of .RAD50 directives are
shown below:

/ABC/

; PBC'kS

• RAD50
. RAD50

/AB/
/ABCD/

;Packs AB (SPACE) into one word .
; Packs ABC into first word and
;D (SPACE) (SPACE) into second word.
; Packs ABC into first word, DEF into
;second word.

• RAD50 /ABC DEF/

Each

ABC into one word=

.RAD50

character

translated

into

its

Radix=50

equivalent,

(Cl,C2,C3)

are

indicated in the following table:
Character

Radix-50 Octal Equivalent
0

(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(8)

Refer to Appendix A.2 for a table of Radix-50 equivalents.

6-30

GENERAL ASSEMBLER DIRECTIVES
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

/AB/<35>

;Stores 3255 in one word •

CHRl=l
CHR2=2
CHR3=3

.RAD50 <CHRl><CHR2><CHR3> ;Equivalent to .RAD50 /ABC/.

6.3.7

Temporary Radix-50 Control Operator

Format:
"'Rccc
where:

CCC

represents a maximum of three characters to
be
If more than
converted to a 16-bit Radix-50 value.
three characters are specified,
any following
the
third character are ignored.
If fewer than three are
specified, it is assumed that the trailing characters
are blanks.

The "'R 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 following example shows how the "'R operator might be
used
to pack a
3-character file type specifier (MAC) into a single
16- bi t WO rd •
MOV #"'RMAC,FILEXT

;Store RAD50 MAC as file extension

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

~.3.8

I.PACKED!

.PACKED Directive

Format:
.PACKED decimal-string[,symbol]
where:

decimal-string

represents a decimal number
from 0
to
31(10) digits long.
Each digit must be in
the range 0 to 9.
The number may have a
sign,
but it
is not required and is not
counted as a digit.

symbol

is assigned a value equivalent to the
number of decimal digits in the string.

6-31

GENERAL ASSEMBLER DIRECTIVES
The .PACKED directive generates packed decimal data, 2 digits per
byte.
Arithmetic and operational properties of packed decimals are
similar to those of numeric strings.
Below is an example of the
.PACKED directive.
.PACKED -12,PACK
.PACKED +500
.PACKED 0
.PACKED -0,SUM
.PACKED 1234E6

6.4

;PACK gets value of 2
;500 is packed
;0 is packed
;SUM gets value of 1
;Illegal packed decimal number
;E6 will be treated as a variable
;and given a value of 4

RADIX AND NUMERIC CONTROL FACILITIES

6.4.1

Radix Control and Unary Control Operators

Any numeric or expression value in a MACR0-11 source program is read
as an octal value by default. Occasionally, however, an alternate
radix would be useful. By using the MACR0-11 facilities described
below, a programmer may declare a radix to affect a term or an entire
program depending on his needs.
NOTE
When two or more unary operators appear
together, modifying the same term, the
operators are applied to the term from
right to left.

6.4.1.1

1.RADiX

.RADIX Directive

Format:
.RADIX
where:

n
represents one of the three radices:
2, 8 and 10.
Any value other than null or one of the three
acceptable radices will cause an error code (A) in the
assembly listing.
If the argument n is not specified,
the octal default radix is assumed. The argument (n)
is always read as a decimal value.

Numbers used in a MACR0-11 source program are initially considered to
be octal values; however, with the .RADIX directive you can declare
alternate radices applicable throughout the source program or within
specific portions of the program.

'1-32

GENERAL ASSEMBLER DIRECTIVES
Any alternate radix declared in the source program through the
.RADIX
directive remains in effect until altered by the occurrence of another
such directive, for example:
• RADIX 10

;Begins a section of code having a
;decimal radix.

• RADIX

;Reverts to octal radix •

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
source program, 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

Formats:
"D"number"
"O"number"
"B"number"

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

These three unary operators allow the user to establish an alternate
radix for
a
single term.
An alternate is useful because after you
have specified a radix for a section of code or have decided
to use
the default octal radix, you may discover a number of cases where an
alternate radix is more convenient or desirable
(particularly within
macro definitions).
Creating a mask word (used to check bit status),
for example, might best be accomplished through the use of a
binary
radix.
Thus an alternate radix can be declared temporarily to meet a
localized requirement
in the source program.
The temporary radix
control operator may be used any time
regardless of the
radix
in
effect or other
radix declarations within the program.
Because the
operator affects only the term immediately following
it,
it may be
used anywhere a numeric value
is legal.
The term (or expression}
associated with the temporary radix control operator will be evaluated
during assembly as a 16-bit entity.
The expressions below are representative of the methods of
temporary radix control operators:
"Dl23

"o 47
"B 00001101
"O<A+l3>

specifying

Decimal Radix
Octal Radix
Binary Radix
Octal Radix

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
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.

6-33

GENERAL ASSEMBLER DIRECTIVES
The following example also illustrates the use of angle brackets to
delimit an expression that is to be interpreted in an alternate radix.
When using the temporary radix control operator only numeric values
are affected.
Any symbols used with the operator will be evaluated
with respect to the radix in effect at their declaration •
• RADIX 10
A=l0
.WORD

AO<A+l0>*10

When the temporary radix expression in the .WORD directive
evaluated, it yields the following equivalent statement:
.WORD

above

180

MACR0-11 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)

The above expression forms are equivalent in function to:
AD100

AD137n
.-.Dl28

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-11.
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.
The string
(which can be a single digit in length) may contain an
optional 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 and/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
3.0E0
3E0
.3El
300E-2
As can be inferred, the list could be extended indefinitely (3000E-3,
.03E2, and so on). A leading plus sign is optional (3.0 is considered
to be +3.0). A leading minus sign complements the sign bit. No other
operators are allowed (for example, 3.0+N is illegal).

6-34

GENERAL ASSEMBLER DIRECTIVES
All floating-point numbers are evaluated as 64 bits in
format:
63

EEEEEEEE

the

following

MMM ••••• MMM
Mantissa
Exponent
Sign

(55 bits)
(8 bi ts)
(1 bit)

MACR0-11 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 6.2.1).
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
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 ~-3).
Bit

Bit

~-3-2~~---..........,,...-~~~-~ ~-3_2~~---........,,..--~~~-~
Retained
field

Unretained
field

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:
ADDF
where:

#"041040 ,F0

F0 is assumed to represent floating accumulator zero.

Floating-point numbers are described
applicable PDP-11 Processor Handbook.

greater

detail

the

I.FL12 I
6.4.2.1

Floating-Point Storage Directives

Formats:
• FLT2
.FLT4

a rg 1 , a rg 2 , •••
argl,arg2, •••

6-35

l.FLT4J

GENERAL ASSEMBLER DIRECTIVES
where:

argl,arg2, .••

represent one or
more
floating-po int
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. As in
the .WORD directive, the arguments are evaluated and the results are
stored in the object module.

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
The "F unary
complemented as it is evaluated during assembly.
operator
allows you to specify an argument that is a 1-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
MACR0-11 as a 16-bit binary value before being complemented. For
example, the following statement:
TAG4:

.WORD

"Cl51

causes the l;s complement of the value l~l (octal) to be stored as a
16-bit value in the program. The resulting value expressed in octal
form is 177626 (8) •
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:
"C"D25
causes the decimal value 25 to be complemented during

resulting binary
177746(octal).

value,

when

expressed

assembly.

The

octal form, reduces to

The term created through the use of the temporary numeric control
operator can be used alone or in combination with other expression
elements. For example, the following construction:
"C2+6
is equivalent in function to:
<"C2>+6
This expression is evaluated during assembly as a l's 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.

6-36

GENERAL ASSEMBLER DIRECTIVES
AF, as stated above, is a unary operator for numeric control which
allows you to specify an argument that is a 1-word floating-point
number. For example, the following statement:
A:

MOV

creates a 1-word floating-point number at location A+2 containing
value 3.7 formatted as shown below.
14

BIT 15

EEEEEEEE

MMMMMMM

Sign (1 bit)

Exponent (8 bits)

Mantissa (7 bits)

The importance of ordering with respect to unary
below.
AFl.0
AF-1.0
-AFl.0
-AF-1.0

the

operators

shown

040200
140200
137600
037600

The value created by the AF unary operator and its 3rgument is, like
Ac and its argument, a term that can be used by itself or in an
expression. For example:

is equivalent to:

Again, the use of angle brackets is advised.
Expressions used
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.
Several MACR0-11 statements (listed below) may cause an odd number
bytes to be allocated:

.BYTE directive

.BLKB directive

.ASCII or .ASCIZ directive

.ODD directive

.PACKED directive

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

6-37

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

I.EVEN I

.EVEN Directive

Format:
.EVEN
The .EVEN directive ensures that the current location counter contains
an even value by adding 1 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:
.ASCIZ /This is a test/
.EVEN
;Ensures that the next statement will
;uegin on a word boundary •
• WORD XYZ

6.5.2

1.0001

.ODD Directive

Format:
.ODD
The .ODD directive ensures that the current location counter contains
an odd value by adding 1 if the current value is even.
If the current
location counter is already odd, no action is taken.
Any operands
following an .ODD directive are also flagged with an error code {Q) in
the assembly listing.

l.BLKB)
6.5.3

f .BLKWI

.BLKB and .BLKW Directives

Formats:
.BLKB
.BLKW

exp
exp

6-38

GENERAL ASSEMBLER DIRECTIVES
where:

exp

represents the specified number of bytes or words to be
reserved in the object program.
Any expression that is
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 (A) in the
assembly listing.
Furthermore,
if the
expression
contains a
forward reference (a reference to a symbol
that is not previously defined),
MACR0-11 generates
incorrect object file code and may cause statements
following the .BLKB/.BLKW directive to be flagged with
phase
(P) errors.
These directives should not be used
without arguments.
However, if no argument is present,
a default value of 1 is assumed.

The .BLKB directive reserves byte blocks in the object module;
the
.BLKW directive reserves word blocks.
Figure 6-6 illustrates the use
of the .BLKB and .BLKW directives.
;+

2
3
4
5
6
7
8
9
10

; Illustrate use of .BLKB and .BLKW directives

.PSECT

000000

Il1PURErDrGBLrRW

000000

COUNT:

.BLKW

000002

11ESSAG: .BLKB

000122

CHRSAV: .BLKB

;saved character

12 000123

FLAG:

;na~

13
14 000124

MSGPTR: .BLKW

;character counter

so.

;Hessase text buffer

Figure 6-6

.DLKB

bl:lte

;Hessase buffer Pointer

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.

6.5.4

I.LIMIT

.LIMIT Directive

Format:
.LIMIT
To know the upper and lower address boundaries of the image is often
desirable.
When the
.LIMIT directive
is specified in the source
program, MACR0-11 generates the following instruction:
.BLKW

causing two storage words to be reserved in the object module.
Later,
at link time, the lowest address in the load image (the initial value

6-39

GENERAL ASSEMBLER DIRECTIVES
of SP) is inserted into the first reserved word, and the address of
the first free word following the image is inserted into the second
reserved word.
During linking, the size of the image is rounded upward to the nearest
2-word boundary.

6.6

TERMINATING DIRECTIVE:

1.END

.END DIRECTIVE

Format:
.END
where:

exp

(exp]
represents an optional expression value which, if
present,
indicates the program-entry point, which is
the transfer address where the program begins.

When MACR0-11 encounters a valid occurrence of the .END directive, it
terminates the current assembly pass. Any text beyond this point in
the current source file, or in additional source files identified in
the command line, will be ignored.
When creating an image consisting of several object modules, only one
object module may be terminated with an .END exp statement (where exp
is the starting address). All other object modules must be terminated
with an .END statement (where .END has no argument); otherwise, an
error message will be issued at link time. If no starting address is
specified in any of the object modules, image execution will begin at
location 1 of the image and immediately fault because of an odd
addressing error.
The .END statement must not be used within
---...:J.:.a....; ___ ,

~Vl!UJ.1.-J.Vl!Cl.1.

-----i....1 ...

Cl.::>.::>ClllU.1..Y

\.....1--1..rU.1.V~"-i

.;,t=

.;,...

J..L

J.1.-

.;,_.

J.;;:)

,_._
;:,v

a
,.,...,-..~

macro

u;:,cu,

.;~
.I.I.-

expansion
.;,-.

..1..:J

.f:1"!'1.,...,.,,-..~

.1...1.U'::j':;IC\.A

T.T.:~k
W..1.1...ll

~'t""IJ

error code (O) in the assembly listing. The .END statement may be
used, however,
in an immediate conditional statement (see Section
6.9.3).

If the source program input is not terminated with an .END
an error code (E) results in the assembly listing.

n.7

directive,

PROGRAM SECTIONING DIRECTIVES

The MACR0-11 program sectioning directives are used to declare names
for program sections (p-sections)
and to establish certain program
section attributes essential to linking.

6-40

GENERAL ASSEMBLER DIRECTIVES

6.7.1

I.PSECT I

.PSECT Directive

Format:
.PSECT
where:

name,argl,arg2, ..• argn

name

represents
the symbolic name of
section, as described in Table 6-4.

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 6-4.
The slash
separating each pair of symbolic arguments listed
in the table indicates that 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 6-4
will cause that statement to be flagged with an
error code (A) in the assembly listing.
Table 6-4
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 Appendix
A.2.

RO/RW

Defines which type of access is
the program section:

permitted

RO=Read-Only Access
RW=Read/Write Access
NOTE
RSX-llM and RT-11
access.

use

only

Read/Write

(continued on next page)

6-41

GENERAL ASSEMBLER DIRECTIVES
Table 6-4 (Cont.)
Symbolic Arguments of .PSECT Directive
Argument

Default

I/D

Meaning
Defines the contents of the program section:
!=Instructions.
If a p-section has the I
attribute and
the program is overlaid, all
calls to the p-section
are
referenced
through a body of overlay code stored in the
root.
If a
concatenated p-section has the
I
attribute,
code
is concatenated on even
bytes.
D=Data.
If a p-section has the D attribute,
all calls to the p-section are referenced
directly.
If a
concatenated p-section has the
D
attribute,
code is concatenated on the next
byte regardless of whether the byte
is odd
or even.

GBL/LCL

LCL

Defines the scope of the program section,
it will be interpreted at link time:

NOTE
The GBL/LCL arguments apply only in the
case
of
overlays;
in
building
single-segment nonoverlaid programs, the

GBL/LCL

arguments

have

meaning,

because the total memory allocation for
the
program will go
into the
root
segment of the image.
LCL=Local.
If an object module contains a
local program section,
then the storage
allocation for that module will
remain
in
the segment containing
the module. Many
modules can contribute (allocate memory)
to
this
same program section;
the memory
allocation for each contributing module
is
either concatenated or overlaid within the
segment,
depending
on
the
allocation
argument of the program section (see CON/OVR
below) •
(continued on next page)

6-42

GENERAL ASSEMBLER DIRECTIVES
Table 6-4 (Cont.)
Symbolic Arguments of .PSECT Directive
Argument

Default

Meaning
GBL=Global.
If a global program section is
used
in more than one segment of a program,
all
references
to
the
p-section
are
collected across segment boundaries.
The
program sections are then stored
in the
segment
(of those originally containing the
p-sections) that is nearest the root.
NOTE
RT-11 stores the collected p-sections in
the root.

ABS/REL

REL

Defines the relocatability
program section:

attribute

the

ABS=Absolute
(non-relocatable).
The
ABS
argument causes the linker or task builder
to treat the p-section as an
absolute
module;
therefore,
no
relocation
is
required.
The program section is assembled
and loaded,
starting at absolute virtual
address 0.
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 during assembly, may generate
an error message
(A)
if virtual location
100000 is outside the segment's virtual
address space:
.PSECT

ALPHA,ABS

.WORD

.=.+100000
REL=Relocatable.
The REL argument causes the
linker
or
task
builder to treat the
p-section as a
relocatable module and a
relocation bias
is added
to all location
references within the program section making
the references absolute.
(continued on next page)

n-43

CENEI1AL ASSEMDLER DIRECTIVES
Table 6-4 (Cont.)
Symbolic Arguments of .PSECT Directive
Argument

Default

CON/OVR

CON

Meaning
Defines the allocation
program section:

requirements

the

CON=Concatenated.
All
references
to
program
section
are
concatenated
determine the total memory space needed
the p-section.

one
to
for

OVR=Overlaid.
All references to one program
section are overlaid;
the total memory
space
needed
equaling
the
largest,
individual p-section.
SAV/NOSAV

NOSAV

Determines where the linker allocates
for the program section:

storage

SAV=Save.
The linker
is forced
to always
allocate the program section to the root of
the image.
NOSAV=No Save.
The linker
program section normally.

allocates

the

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
argument.
Default values
unspecified arguments.

blank

name

argument

and

the

GBL

{see Table 6-4) are assumed for all other

The .PSECT directive may be used without
Section 6.7.1.1).

name

arguments

(see

The .PSECT directive allows a user to create program sections
(see
Section n.7.1.l) and to share code and data among the sections he has
created (see Section n.7.1.2).
In declaring
the program sections
(also called p-sections),
you may declare the attributes of the
p-sections. This allows you to control memory allocation and at the
same time increases program modularity.
(For a discussion of memory
allocation, refer to the applicable system manual - see Section 0.3 in
the Preface.)
MACR0-11 provides for 256(10) program sections, as listed below:
1.

One default absolute program section {. ABS.)

One default relocatable program section (. BLK.)*

* In RT-11 this program section is unnamed.
6-44

GENERAL ASSEMBLER DIRECTIVES
3.

Two-hundred-fifty-four named program sections.

For each program section specified or implied, MACR0-11 maintains
following information:
1.

Program section name

Contents of the current location counter

Maximum location counter value encountered

Program section attributes (described in Table 6-4 above).

the

6.7.1.1 Creating Program Sections - The first statement of a source
program is always an implied .PSECT directive;
this causes MACR0-11
to begin assembling source statements at relocatable zero of the
unnamed program section.
The first occurrence of a .PSECT directive with a given name assumes
that the current location counter is set at relocatable zero.
The
scope oI ~n1s a1rect1ve tnen extends unt11
a directive declaring a
different program section is specified.
Subsequent .PSECT directives
cause assembly to resume where the named section previously ended.
For example:

A:
B:
C:
X:

Y:
D:

.PSECT
.WORD
.WORD
.WORD
.PSECT
.WORD
.WORD
.PSECT
.WORD

0
0
0
ALPHA
0
0

;Declares unnamed relocatable program
;section assembled at relocatable
;addresses 0 through 5.
;Declares relocatable program section
;named ALPHA assembled at relocatable
;addresses 0 through 3.
;Returns to unnamed relocatable
;program section and continues assem; bly 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.
If arguments are specified, they must be
identical
to the ones previously declared for the p-section.
If the
arguments differ, the arguments of the first
.PSECT- will
remain in
effect, and an error code (A) will be generated as a warning.

6-45

GENERAL ASSEMBLER DIRECTIVES
By maintaining separate location counters for
each program section,
MACR0-11 allows you to write statements that are not physically
sequential but that can be loaded sequentially following assembly,
as
shown in the following example.
A:
B:
C:
ST:

.PSECT
.WORD
.WORD
.WORD
CLR
CLR
CLR
.PSECT

SECl,REL,RO
0
0

;Start a relocatable program section
;named SE Cl assembled at relocatable
;addresses 0 through 5.

A
B

;Assemble code at relocatable
;addresses 6 through 21 ( 8) •

SECA,ABS

;Start an absolute program section
;named SECA. Assemble code at
;absolute addresses 0 through 3.
;Resume relocatable program section
; SECl. Assemble code at relocatable
;addresses 22 through 27 (8) •

.WORD
.PSECT
INC
BR

.+2,A
SE Cl
A
ST

All labels in an absolute program section are absolute;
likewise, all
labels in a relocatable section are relocatable.
The current location
counter symbol (.) is 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
program section.

zero

the

unnamed

Since it is not known during
assembly where
relocatable
sections will
be loaded,
all
references
to
relocatable
sections are assembled as references
relative to
the base
referenced section.

program
program
of the

In the following example, references
to the symbols X and Y are
translated
into
references
relative to the base of the relocatable
program section named SEN.
.PSECT ENT ,ABS
.=.+1000
A:
CLR
x

Y:
X:

JMP

. PSECT
MOV
JMP
HALT
.WORD

SEN,REL
R0,Rl
A

;Assembled as CLR base
;relocatable section +
;Assembled as JMP base
;relocatable section +
;Assembled as JMP 1000.

6-46

of
10(8).

of
6 ( 8) •

GENERAL ASSEMBLER DIRECTIVES
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 dicussion of the
current location counter.
Thus, MACR0-11 provides the linker or task builder 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.

6.7.1.2 Code or Data Sharing - Named relocatable program sections
with the arguments GBL and OVR operate in the same manner as FORTRAN
COMMON, that is, program sections of the same name with the arguments
GBL and OVR from different assemblies are all loaded at the same
location at link time.
All other program sections
(those with the
argument CON) are concatenated.
A single symbol could name both an internal symbol and a program
section.
Considering FORTRAN again, using the same symbolic name is
necessary to accommodate the following statement:
COMMON

/X/ A,B,C,X

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

6.7.1.3 Memory Allocation Considerations - The assembler does not
generate an error when a module ends at an odd location.
You can,
therefore, 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 6-4;
CON/OVR), odd length modules (except the last) may cause succeeding
modules to be linked starting at odd locations,
thereby making the
linked program unexecutable.
To avoid this problem, separate code and
data from each other and place them in separately named program
sections
(see Table 6-4;
I/D).
The linker or task builder can then
begin each program section on an even address.
Refer to the
applicable system manual for further information on memory allocation
of tasks (see Section 0. 3 in the Preface) .

6-47

6.7.2

1.ASEcrl
1.CSECTI

.ASECT and .CSECT Directives

Formats:
.ASECT
.CSECT
.CSECT
where:

symbol

represents one or more of the arguments in Table

n-4.

IAS and RSX-llM assembly-language programs use the .PSECT and
.ASECT
directives exclusively, because the .PSECT directive provides all the
capabilities of the
.CSECT directive defined
for other
PDP-11
assemblers.
MACR0-11 will accept both .ASECT and .CSECT directives,
but assembles them as though they were
.PSECT directives with the
default attributes listed in Table 6-5.
Compatibility exists between
other MACR0-11 p ograms and the IAS/RSX-llM Task Builders, because the
Task Builders a so treat the .ASECT and .CSECT directives like .PSECT
directives with he defauit vaiues iisted in Table 6-5.
Table 6-5
Program Section Default Values
Default Value
Attribute
.ASECT
.,._,

__..,,..._

l'<CllllC

.CSECT (named)

.CSECT (unnamed)

r"ti:)mO

. BLK. *

7\0C

C"1.UIJ •

J..l'""'l.Ll"""

ACCESS

Type

Scope

GBL

LCL

Relocation

ABS

REL

Allocation

OVR

CON

In RT-11 this program section has no default name.

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.
~-48

OVR

are

assumed

for

the

GENERAL ASSEMBLER DIRECTIVES

1.sAVE

.SAVE Directive

6.7.3

Format:
.SAVE
.SAVE stores the current program section context on the top
program section context stack, while leaving the current
section context in effect.
If the stack is full when .SAVE is
an error
(A)
occurs.
The stack can handle 16 .SAVEs.
The
section context includes the values of the current location
and the maximum value assigned to the location counter in the
program section.

to the
program
issued,
program
counter
current

See Figure 6-7 for an example of .SAVE.

I.RESTORE!

.RESTORE Directive

6.7.4
Format~

.RESTORE
The .RESTORE directive retrieves the program section from the top of
the program section context stack.
If the stack is empty when
.RESTORE is issued, an error (A) occurs. When
.RESTORE retrieves a
program section, it restores the current location counter to the value
it had when the program section was saved.
See Figure 6-7 for an example of .RESTORE.

6-49

.11AIN.

MACRO V05.00 Honda\:1 17·-·Jan-83 08:52

Pasf~

EXAHPLE OF .SAVEi.RESTORE USAGE
.SBTTL
2
3

; Hacro DS
; Define local imPure storase

5
6
7
8
9
10

NAHE:

'"
I

Vl
~

of .SAVEi.RESTORE usase

12
13
14
15
16
17
18
19
20
21 000000
22 000004
23 000010
24 000014
25
26
27
28 000020
29
30
31
32
33
34 000022
35 000022
36 000022
37 000022
38

Exa~Ple

.HACRO DS
NAHErSIZE
.SAVE
;save the current PSECT
.PSECT IHPURErDr6BL
;store the data in the i•Pure PSECT
;set aside the space
.BLKW
SIZE
;Reenter the current PSECT
.RESTORE
.ENDH

; SCANSY
; Scan the hash table for valid entries
;-

016701
010167
066701
010167

000000'
000002'
000004'
000006'

SCANSY: HOV
HOV
ADD
HOV

SYHBASrRl
RlrCURSYH
SYHSIZrR1
RlrSYHTOP

;Get base of table
;Initialize Pointer to table
;Point Past the table
;save end address

Rest of ro•Jtine.,,
RETURN

000207

;Table is scannedr exit

; Local data
;-

DS
DS
DS
DS

SYHBAS
CURSYH
SYMSIZ
SYHTOP

;ease address of s~•bol table
;current s~~bol Pointer durins scan
;size of tabler b~tes
;set to end address of table

40
41
42
43
44 000022
45
46
47
48

; SSORT
; Perfor~ shell sort on

sY~bol

table Prior to listins

;016701

000006'

SSOF:T:

HOV

SYMTOPrRl

;Get end of table

Additional code •••
000001

.END

Figure 6-7

Example of .SAVE and .RESTORE Directives

GENERAL ASSEMBLER DIRECTIVES
6.8

SYMBOL CONTROL DIRECTIVES

The symbol control directives are used to set
symbol.

~.8.1

the

type

given

j.GLOBLI

.GLOBL Directive

Format:
.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

label

field

and/or

comment

field.
The .GLOBL directive is provided to define (and thus provide linkage
to)
symbols not otherwise defined as global symbols within a module.
In defining global symbols the directive .GLOBL A,B,C is similar to:
A==:expression
B==:expression
C==:expression

A==expression
B==expression
C==expression

A·.

B ••

C::

Because object modules are linked by global symbols, these symbols are
vital
to
a
program.
The following
paragraph, describing
the
processing of a program from assembly to linking,
explains the
global's role.
In assembling a source program, MACR0-11 produces a relocatable object
module and a listing file containing the assembly listing and symbol
table.
The linker or task builder joins separately assembled object
modules
into a single executable
image.
During linking, object
modules are relocated relative to the base of the module and linked by
global symbols.
Because these symbols will be referenced by other
program modules, they must be singled out as global symbols in the
defining modules.
As shown above,
the
.GLOBL directive, global
assignment operator, or global label operator will define a symbol as
global.
All internal symbols appearing within a given program must be defined
at the end of assembly pass 1 or they will be assumed to be default
global references.
Refer to Section 6.2.1
for a description of
enabling/disabling of global references.

6-51

GENERAL ASSEMBLER DIRECTlVEb
In the following example, 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
( g 1oba1 ) r e fer enc e i f • ENA BL GB L i s i n e f f e ct.
Define a subroutine with 2 entry points which calls an
external subroutine
• PSECT
• GLOBL
MOV
MOV
JSR
RTS
MOV
CLR
BR

A:
X:
B ••

@(R5)+,R0
#X,Rl
PC,C
RS
(R5)+,Rl
R2

;Declare the unnamed program section .
;Define A as a global symbol •
;Define entry point A.

;Call external subroutine
; Exit.
;Define entry point B.

External symbols can appear in the operand field of an instruction or
MACR0-11 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)

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 3.3, 6.9.1 and 6.9.3}.

6.8.2

I .WEAK I

.WEAK Directive

Format:
.WEAK syml,sym2, ••• symn
where:

syml
sym2, •••
symn

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

.WEAK

SUB1,SUB2

Example:

The .WEAK directive may also embody a label
field.

6-52

field

and/or

comment

GENERAL ASSEMBLER DIRECTIVES
The .WEAK directive is used to specify symbols that are either defined
externally in another module or defined globally in the current
module.
This directive suppresses object library searches
for
specified external symbols.
When the .WEAK directive specifies a symbol that is externally
defined, it is considered a global symbol. If the linker finds the
symbol's definition in another module, it uses that definition.
If
the linker does not find an external definition, the symbol is given a
value of 0. The linker does not search a library for the global
symbol, but if a module brought in from a library for another reason
contains the symbol's definition, the linker uses that definition.
If a symbol that is defined in the current module is specified by the
.WEAK directive, the symbol is considered globally defined. However,
if the current module is inserted in an object library, the symbol is
not inserted in the library's symbol table. Consequently, the module
is not found when the library is searched at link time to resolve the
symbol.
NOTE
The .WEAK directive is only supported by
the RT-11 V5.0 LIBRARIAN (LIBR) and
LINKER (LINK).
Support is not
yet
implemented in the RSX-11 taskbuilder
( T KB ) o r 1 i bra r i an ( LB R) •

6.9

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.

6.9.1

Conditional Assembly Block Directives

1.ENDC

Format:
.IF

cond,argument(s) ;Start conditional assembly block.

range

;Range of conditional assembly block •

• ENDC

;End of conditional assembly block.

6-53

GENERAL ASSEMBLER DIRECTIVES
where:

cond

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 6-6.
represents any legal
and/or tab) •

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
condition to be tested (see Table 6-6).

range

represents the body of code that
is either
included in the assembly, or excluded, depending
upon whether the 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 6-6,
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 6-6
Legal Condition Tests for Conditional Assembly Directives
Conditions
Positive

Complement

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
defined).

Macro-type
argument

Argument
is
non-blank) •

(or not

blank

(or

(continued on next page)

6-54

GENERAL ASSEMBLER DIRECTIVES
Table 6-6 (Cont.)
Legal Condition Tests for Conditional Assembly Directives
Conditions
Positive

Complement
DIF

IDN

Assemble Block If:

Arguments
Two macro-type
arguments

Arguments are
identical
(or different) • The
• IF
IDN/.IF DIF conditional
directives
are
not
alphabetically
case
sensitive
by
default.
The user may enable these
directives
to be case
sensitive by using
the
.ENABL
option
(.ENABL
LCM) •

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 793) 9
<A,B,C>
"/124/
An example of a conditional assembly directive follows:
.IF EQ

ALPHA+l

;Assemble block if ALPHA+l=0

• 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.

6-55

symbols

SYMl

GENERAL ASSEMBLER DIRECTIVES
Nested conditional directives take the form:
Conditional Assembly Directive
Conditional Assembly Directive

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

SYMl
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.
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.
MACR0-11 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 (O) in the assembly listing.

I. IFF I

Q!!J
1.1FTFI

Subconditional Assembly Block Directives

6.9.2

Formats:
.!FF
.IFT
.IFTF
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.

6-56

within

body

code

within

GENERAL ASSEMBLER DIRECTIVES
The subconditional directives are described in detail in Table 6-7.
If a subconditional directive appears outside a conditional assembly
block, an error code (O) is generated in the assembly listing.
Table 6-7
Subconditional Assembly Block Directives
Subconditional
Directive

Function

.IFF

If the
condition
tested
upon
entering
the
conditional
assembly block is 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 program.

.IFT

If the
condition
tested
upon
entering
the
conditional
assembly
block is 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
below • .C...onditional or subconditi~nal dir~ctive~ in nested conditigq~~
~ssembly bJ..Q..ds....ru:iL~ni::f~i.~].l.XIf?I~1I ·u -~~.~·--,:e;::~~i~u·~.- ·:ro.r __o~t!:F) .:~.on9Jtiq_n_
in the block is_ not satisfied. Examples 3 and 4 below illustrate
nested dire'c'tfv'es· - 'that . a.re n;t evaluated because
of
previous
unsatisfied conditional coding.
EXAMPLE 1:
• IF DF

Assume that symbol SYM is defined.
SYM

;Tests TRUE, SYM is defined. Assemble
;the following code.

.IFF

;Tests FALSE. SYM is defined. Do not
;assemble the following code.

.IFT

;Tests TRUE. SYM is defined. Assem;ble the following code.

.IFTF

;Assemble following code uncondition;ally.

6-57

GENERAL ASSEMBLER DIRECTIVES
.IFT

;Tests TRUE. SYM is defined. Assem;ble remainder of conditional assem;bly block .

• ENDC
EXAMPLE 2:
. IF DF
• IF DF
• IFF

Assume that
defined •

symbol X is defined and that symbol Y is not

X
Y

;Tests TRUE, symbol X is defined .
;Tests FALSE, symbol Y is not defined •
;Tests TRUE, symbol Y is not defined,
;assemble the following code.

• IFT

;Tests FALSE, symbol Y is not defined .
;Do not assemble the following code •

. ENDC
.ENDC
EXAMPLE 3:

Assume that
defined .

symbol A is defined and that symbol B is not

• IF DF

;Tests TRUE. A is defined.
;Assemble the following code.

MOV

A,ARl

.IFF
MOV

;Tests FALSE. A is defined. Do not
;assemble the following code.
Rl, R0

• IF NDF B

;Nested conditional directive is not
;evaluated •

. ENDC
.ENDC
EXAMPLE 4:

Assume that symbol X is not defined
defined.

• IF DF

and that symbol Y is

;Tests FALSE. Symbol X is not defined •
;Do not assemble the following code.
;Nested conditional directive is not
;evaluated.

.IFF

;Nested subconditional directive is
;not evaluated.

.IFT

;Nested subconditional directive is
;not evaluated •

. ENDC
.ENDC

6-58

GENERAL ASSEMBLER DIRECTIVES

6.9.3

Immediate Conditional Assembly Directive

Format:
• IIF

where:

cond,arg,statement

cond

represents one of the legal condition tests defined
for conditional assembly blocks in Table 6-6.
represents any
and/or tab) •

arg

legal

separator

(comma,

space,

represents the argument
associated
with
the
immediate conditional directive;
an expression,
symbolic argument, or macro-type argument,
as
described in Table 6-6.
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
if the condition is satisfied.

assembled

An immediate conditional assembly directive provides a means for
writing a 1-line conditional assembly block.
The use of this
directive requires no terminating .ENDC statement and the condition to
be tested is completely expressed within the line containing the
directive.
For example, the immediate conditional statement:
• IIF

DF FOO,BEQ ALPHA

generates the code
BEQ

ALPHA

if the symbol FOO is defined within the source program.
As with the .IF directive, a condition test other than those listed in
Table 6-6, an illegal argument, or a null argument specified in an
.IIF directive results in an error code (A) in the assembly listing.

'5-59

FILE CONTROL DIRECTIVES

6.10

The MACR0-11 file control directives are used to add
file
names to
macro library lists and to insert a source file into the source file
being currently used.

6.10.1

1.LIBRARY

.LIBRARY Directive

Format:
.LIBRARY string
where:

represents a delimited string
that
specification of a macro library.

string

the

file

The .LIBRARY directive adds a file name to a macro library list that
is searched.
A library list
is searched whenever a .MCALL or an
undefined opcode
is encountered within a MACR0-11 program.
The
libraries that make up the list are searched in the reverse order in
which they were specified to the MACR0-11 assembler.
If any information was omitted
from
the macro library argument,
default values are assumed.
The default
file
specification for
MACRO-ll/RT-11 is DK:.MLB, and for other systems it is SY:.MLB.
The .LIBRARY directive is used as follows:
• LIBRARY
• LIBRARY
. LIBRARY

/DBl: [SMITH]USERLIB/
?DK:SYSDEF.MLB?
\CURRENT.MLB\

MACR0-11 searches all macro libraries if it finds an unknown symbol in
.1-1.-..l...llt:

----...::1-

UJ::-1\....VUC'

S:!-1...J

.LJ.C'..LU

--...::J

ClllU

.a-\...-

l-llC:

..... ,,.a--

....... -""'11

0Ul-V-1LL\..o0.J....1.

-""'~;-r'\

vp\....1.V.LJ

h:.~

.LJUio...l

ha.or'\

·"•··"'-"""'''

1""'\YOt7ir'\11c1,1
I:"'.&.. '-"'V ""-VV."'-J"""-I

by . ENABL MCL.
NOTE
If you are using MACR0-11 with an RT-11
operating system, you should be aware of
the
following
two
restrictions.
The
device driver
for the specified device
that the .LIBRARY file resides on must
already be loaded,
either explicitly
with
the
KMON
LOAD
command,
or
implicitly by reference to the device on
the original MACR0-11 command line.
The
second
restriction is that there is a
limit on the number of
.LIBRARY files
that may be specified.
The limit is
twelve minus the
number
of
files
specified
in the MACR0-11 command line.
Since there can be a maximum of eight
files
on a MACR0-11/RT-ll command line,
there are at least four available slots
for .LIBRARY files.

6-60

on.:::.h1or1
'"""""""".....,..1."-""

GENERAL ASSEMBLER DIRECTIVES

6.10.2

l.1NCLUDEI

.INCLUDE Directive

Format:
.INCLUDE string
where:

string

represents a delimited string that is
specification of a macro source file.

the

file

The .INCLUDE directive is used to insert a source file within the
source file currently being used. When this directive is encountered,
the current source file is stacked and the source file specified by
the directive is read into memory. When the end of the specified
source file is reached, the original source file is popped from the
stack and assembly resumes at the line following the directive. A
source file can also be inserted within a source file that has already
been specified by the .INCLUDE directive. In this case the original
source file and the first source file specified by the .INCLUDE
directive are stacked and the second specified source file is read
into memory. When the end of the second source file is reached, the
first specified source file is popped from the stack and assembly
resumes at the line following the directive, and when the end of the
first specified source file is reached, the original source file is
popped from the stack and assembly of that file is started again at
the line following the .INCLUDE directive.
The maximum nesting level
of source files specified by the .INCLUDE directive is five.
If any information is omitted from the source file argument, default
values are assumed. The default file specification for MACRO-ll/RT-11
is DK:.MAC, and for other systems it is SY:.MAC.
The .INCLUDE directive is used as follows:
• INCLUDE
• INCLUDE
. INCLUDE

/DR3: [l,2]MACROS/
?DK:SYSDEF?
\CURRENT.MAC\

; File MACROS. MAC

NOTE
If you are using MACR0-11 with an RT-11
operating system, the device driver for
the specified device that the .INCLUDE
file resides on must already be loaded,
either explicitly with the KMON LOAD
command, or implicitly by reference to
the device on the origlnal MACR0-11
command line.

6-61

CHAPTER 7
MACRO DIRECTIVES

7.1

DEFINING MACROS

By using macros a programmer can use a
sequence of lines into a source program.

single

line

insert

A macro definition is headed by a .MACRO directive (see Section 7.1.1)
followed by the source lines. The source lines may optionally contain
dummy arguments. If such arguments are used, each one is listed in
the .MACRO directive.
A macro call (see Section 7.3) is the statement used by the programmer
to call the macro into the source program.
It consists of the macro
name followed by the real arguments needed to replace any dummy
arguments used in the macro.
Macro expansion is the insertion of the macro source lines into the
main program.
Included in this insertion is the replacement of the
dummy arguments by the real arguments.
Macro directives provide the means to manipulate the macro expansions.
Only one directive is allowed per source line. Each directive may
have a blank operand field or one or more operands.
Legal operands
differ
with
each directive.
The macros and their associated
directives are detailed in this chapter.

7.1.1

(.MACRO

.MACRO Directive

Format:
[label:]
where:

.MACRO

name, dummy argument list

label

represents an optional statement label.

name

represents the user-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) .

7-1

legal

separator

(comma,

space,

MACRO DIRECTIVES
dummy
argument
list

where:

represents a number of legal symbols (see Section
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 (A) to be generated.

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

ABS

A,B

;Defines macro ABS with two arguments •

The first statement of a macro definition must be a .MACRO directive.
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 applies to
and • REPT.

7.1.2

.IRP,

.IRPC,

j.ENDMI

.ENDM Directive

Format:
.ENDM
where:

[name]
represents an optional argument specifying
name
of
the macro being terminated by
directive.

name

Example:
;Terminates the current
;macro definition.

.ENDM
.ENDM

ABS

;Terminates the current
;macro definition named ABS.

7-2

the
the

MACRO DIRECTIVES
If specified, the macro name in the oENDM 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.
In
either
case,
the current macro definition is terminated.
Specifying the macro name in the .ENDM statement thus permits MACR0-11
to
detect missing .ENDM statements or improperly nested macro
definitions.
If a legal label is
The .ENDM directive must not have a label.
If an illegal label is attached, the
attached, it will be ignored.
directive will be ignored.
The .ENDM directive may be followed
below:
.MACRO
JSR
.WORD
• ENDM

TYPMSG MESSGE
R5,TYPMSG
MESS GE

comment

field,

shown

;Type a message.
;End of TYPMSG macro •

The final statement of every macro definition must be an .ENDM
directive.
The .ENDM directive is also used to terminate indefinite
repeat blocks (see Section 7.6) and may be used to terminate repeat
blocks (see Section 7. 7) •

7.1.3

j .MEXITI

.MEXIT Directive

Format:
.MEXIT
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
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

;Start conditional assembly block •

. MEXIT
• ENDC

;Terminate macro expansion •
;End conditional assembly block •

. ENDM

;Normal end of macro .

In an assembly where the dummy symbol N is replaced by zero (see Table
6-6), the .MEXIT directive would assemble the conditional block and
terminate the macro expansion.
When macros are nested, a .MEXIT
directive causes an exit to the next higher level of macro expansion.
7-3

MACRO DIRECTIVES
A .MEXIT directive encountered outside a macro definition
with an error code (O) 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 used in 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

Format:
[label:]
where:

name

real arguments

label

represents an optional statement label.

name

represents the name of the macro, as specified
the .MACHO directive (see Section 7.1.l).

real
arguments

represent symbolic arguments which replace the
dummy arguments listed in the
.MACRO directive.
When multiple arguments occur, they are separated
by any legal separator.
Arguments to the macro
call are treated as character strings, their usage
is determined by the macro definition.

A macro definition must be established by means of the
.MACRO
directive
(see Section 7.1.1)
before the macro can be called and
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 call;
the appearance of the symbol in the operand field designates it as a
label, as shown below:
ABS:

7.3

MOV

(R0) ,Rl

;ABS is defined as a label.

ABS

;ABS is considered to be a label.

ABS

#4,ENT,LAR

;ABS is a macro call.

ARGUMENTS IN MACRO DEFINITIONS AND MACRO CALLS

Multiple arguments within a macro definition or macro call must be
separated by one of the legal separating characters described in
Section 3 .1.1.

7-4

MACRO DIRECTIVES
Macro definition arguments (dummy) and mocro 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
call contain multiple arguments:
.MACRO

REN A,B,C

REN

ALPHA,BETA,<Cl,C2>

associated

macro

Arguments which themselves contain separating characters must
enclosed in paired angle brackets.
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.
The up-arrow (A) construction allows 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>/,#44,WEV

causing the entire character string <MOV
argument.

X,Y>

passed

Because of the use of the up-arrow (A) shown above, care must be taken
when passing an argument beginning with a unary operator (AO, AD, AB,
AR, AF ••• ).
These arguments must be enclosed in angle brackets
(as
shown below)
or MACR0-11 will
read
the character
following the
up-arrow as a delimiter.

The following macro call:
REN

#44,WEVA/MOV X,Y/

contains only two arguments
(#44 and
up-arrow is a
unary operator
(see
preceded by an argument separator.

WEVA/MOV x I Y/) ,
because the
Section 3.1.3)
and it is not

As shown in the examples above, spaces can be used within bracketed
argument constructions to increase the legibility of such expressions.

7-5

7.3.1

Macro Nesting

Macro nesting occurs where the expansion of one macro includes a call
to another.
The depth of nesting allowed depends upon the amount of
dynamic memory used by the source program being assembled.
To pass an argument containing legal argument delimiters to nested
macros,
enclose the argument
in the macro definition within angle
brackets, as shown in the coding sequence below.
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 DUM1,DUM2
<DUMl>
<DUM2>

.MACRO
DUM3
ADD
MOV
• ENDM

LEVEL2 DUM3
#10,40
R0, (Rl)+

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

<MOV

X,R0>,<MOV

R2,R0>

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

X, R0
#10 ,R0
R0, (Rl) +
R2,R0
#10, R0
R0,(Rl)+

When macro definitions are nested,
,.. .... ,

the

, -...:J

'-Cl.1..1.CU

outer

macro

the

inner

been

definition

cannot

called and expanded.

For

example, in the following coding:
.MACRO

LVl A,B

.MACRO

LV2 C

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

7-6

MACRO DIRECTIVES
7.3.2

Special Characters in Macro Arguments

If an argument does not contain spaces, tabs, semicolons, or commas it
may include special characters without enclosing them in a bracketed
construction. For example:
.MACRO
MOV
.ENDM

PUSH ARG
ARG I - (SP)

PUSH

X+3(%2)

causes the following code to be generated:
Mov

7.3.3

x + 3 ( %2 ) I - ( s p)

Passing Numeric Arguments as Symbols

If the unary operator backslash (\) precedes an argument, the macro
treats that argument as a numeric 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=0

.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 7.3.7.

x,c

The above macro call (INC) would thus expand to:
X0:

.WORD

In this expanded code, the label X0:
results from the concatenation
of two real arguments. The single quote (') character in the label
A1 B: concatenates the real arguments X and 0 as they are passed
during the expansion of the macro. This type of argument construction
is described in more detail in Section 7.3.7.
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 (that is, C}
cannot be updated in the CON macro definition, because the character 0
has replaced C in the argument string (INC X, C) • 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 0 character is
passed to the macro expansion.}

7-7

MACRO DIRECTIVES
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

IDT SYM
/V01.'SYM/

IDT

\ID

;Assume that the symbol ID takes
;on a unique 2-digit value.
;Where V01 is the update
;version of the program.

The above macro call would then expand to:
.IDENT

/V01.6/

where 6 is the numeric value of the symbol ID.

7.3.4

Number of Arguments in Macro Calls

A macro can be defined with or without arguments.
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 B and
.IF
NB
(see Table 6-6)
can be used within the macro to detect missing
arguments. The number of arguments can also be determined using
the
.NARG directive (Section 7.4.1).

7.3.5

Creating Local Symbols Automatically

A label is often required in an expanded macro.
In the conventional
macro facilities
thus far described,
a label must be explicitly
specified as an argument with each macro call.
The user must be
careful
in issuing
subsequent calls to the same macro in order to
avoid duplicating labels. This concern can be eliminated through a
feature of MACR0-11 that creates a unique symbol where a label is
required in an expanded macro.
As noted in Section 3.5, MACR0-11 can automatically create local
symbols of the form n$, where n is a decimal integer within the range
30000 through 65535, inclusive.
Such local symbols are created by
MACR0-11 in numerical order, as shown below:
30000$
30001$

65534$
65535$

7-8

MACRO DIRECTIVES
This automatic generation 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
#5,A

;Contains dummy argument B preceded by
;question mark.

. ENDM
A local symbol is created automatically by MACR0-11 only when a real
argument of the macro call is either null or missing, as shown in
Example 1 below. If the real argument is specified in the macro call,
however, MACR0-11 inhibits the generation of a local symbol and normal
argument replacement occurs, as shown in Example 2 below.
(Examples 1
and 2 are both expansions of the Alpha macro defined above.)
EXAMPLE 1:

Create a Local Symbol for the Missing Argument:

ALPHA
TST
BEQ
ADD

Rl
Rl
30000$
#5,Rl

;Second argument is missing.
;Local symbol is created.

30000$:

EXAMPLE 2:

Do Not Create a Local Symbol:

ALPHA
TST
BEQ
ADD

R2,XYZ
R2
XYZ
#5,R2

;Second argument XYZ is specified.
;Normal argument replacement occurs.

XYZ:
Automatically created local symbols are restricted to the first 16(10)
arguments of a macro definition.
Automatically created local symbols resulting from the expansion of a
macro, as described above, do not establish a local symbol block in
their own right.
When a macro has several arguments earmarked for automatic local
symbol generation, substituting a specific label for one such argument
risks assembly errors because MACR0-11 constructs its
argument
substitution list at the point of macro invocation. Therefore, the
appearance of a label, the .ENABL LSB directive, or the .PSECT
directive,
in the macro expansion will create a new local symbol
block. The new local symbol block could leave local symbol references
in the previous block and their symbol definitions in the new one,
causing error codes in the assembly listing.
Furthermore, a later
macro exoansion that creates local svmbols in the new block may
duplicate~ one of the symbols in question, causing an additional error
code (P) in the assembly listing.
4

7-9

MACRO DIRECTIVES
7.3.6

Keyword Arguments

Format:
name=string
where:

name

represents the dummy argument,

string

represents the real symbolic argument.

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

argument

separators

Macros may be defined with, and/or called with,
keyword arguments.
When a keyword argument appears in the dummy argument list of a macro
definition, 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,
however,
the specified string becomes the real
argument for
the dummy argument that 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.
A keyword argument may also be specified anywhere
in the
real
argument list of a macro ca11 out, in this case, does not affect
the positional ordering of the arguments.
1
2
3
4
5
6
7
8

.LIST

Define a macro having keywords in dummy argument
list
.MACRO
.WORD
.WORD

•

t.rf"lon
ftVJ.\..L.I

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

• ENDM

10
11

12
13
14
15
16
17 000000
000000
000002
000004
18
19 000006
000006
000010
000012
20
21 000014
000014
000016
000020

Now invoke several times
TEST
.WORD
.WORD
.WORD

A,B,C

000000G
000000G
000000G
000040
000030
000020

TEST
.WORD
.WORD
.WORD

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

TEST
.WORD
.WORD
.WORD

BLOCK=5

000001
000005
000000G

A
B

5
TEMP

7-10

MACRO DIRECTIVES
22
23 000022
000022
000024
000026
24
25 000030
000030
000032
000034
26
27 000036
000036
000040
000042
28
29
30

7.3.7

000005
000000
000000G

TEST
.WORD
.WORD
.WORD

000001
000000
000000G

TEST
.WORD
.WORD
.WORD

000001
000000
000000C

TEST
.WORD
.WORD
.WORD

000001

• END

CONTRL=5,ADDRES=VARIAB
5
VARI AB
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,C,
/C/
I I A, I I B

when the macro DEF is called through the statement:
DEF

X,Y,<MACR0-11>

it is expanded, as follows:
XY:

.ASCIZ
.BYTE

/MACR0-11/
I x , 'Y

In expanding the first line, the scan for the first argument
terminates upon finding the first apostrophe (') character.
Since A
is a dummy argument, the apostrophe (') is removed.
The scan then
resumes with B;
B is also noted as another dummy argument. The two
real arguments X and Y are 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 MACR0-11 to be substituted in
this field.

7-11

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

7.4

MACRO ATTRIBUTE DIRECTIVES:

.NARG, .NCHR, AND .NTYPE

MACR0-11 has three directives that allow the user to determine certain
attributes of macro arguments:
.NARG, .NCHR, and .NTYPE. 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 below.

7.4.1

1.NARG

.NARG Directive

Format:
[label:]
where:

.NARG symbol

label

represents an optional statement label.

symbol

represents any legal symbol.
This symbol is
equated ~u the number of non-keyword 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.

The .NARG directive is used to determine the number of non-keyword
arguments in the macro call currently being expanded. Hence, the
.NARG directive can appear only within a macro definition;
if it
appears elsewhere, an error code (O) is generated in the assembly
1 is ting.
An example of the .NARG directive is shown in Figure 7-1.

7-12

MACRO DIRECTIVES

• TITLE

NARG

.ENABL
.LIST

LC
HE

2
3

4
5

; Exa11Ple of the .NARG directive
;-

14
15

NULL
.NARG
,IF EQ
.MEX IT
.IFF
.REPT
NOP

.END!'!

.HACRO

9
10
11
12
13

NUH
SYH
SYM
NUM

.ENDC

17
18
19
20 000000

.ENDM
NULL
.NARG
, IF EQ
.MEX IT
.IFF
.REPT
NOP
.ENDH
.ENDC

000000

SYM
SYM

NULL
.NARG
.IF EQ
.HEX IT
.IFF
.REPT
NOP
.ENDH
NOP
NOP
NOP
NOP
NOP
NOP
.ENDC

22 000000

000001

000006
000000
000002
000004
000006
000010
000012
23
24

000240
000240
000240
000240
000240
000240

7.4.2

.END

000001

Figure 7-1

SYH
SYH

Example of .NARG Directive

j .NCHR I

.NCHR Directive

Format:
[label:]
where:

.NCHR symbol,<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.
represents any
and/or tab) •

7-13

legal

separator

(comma,

space,

MACRO DIRECTIVES

represents a string of printable characters.
If
the character string contains a legal separator
(comma, space, and/or tab) the whole string must
be
enclosed
within
angle brackets (<>) or
up-arrows (A).
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.

The .NCHR directive, which can appear anywhere in a MACR0-11 program,
is used to determine the number of characters in a specified character
string. This directive is useful in calculating the length of macro
arguments.
An example of the .NCHR directive is shown in Figure 7-2.
1
2
3
4
5

• TITLE

NCHR

.ENABL
.LIST

LC
HE

; Illustrate the .NCHR directive
;-

8
9

.MACRO

STRING

HES SAG

.NCHR

iii,iiESSAG

11
12
13
14
15
16 000000

.WORD
• ASCII
.EVEN

$$$
/HESSAG/

STRING
.NCHR
.WORD
• ASCII

<Hello>
U$,Hello
$$$
/Hello/

000000
000002
000003
000004
000005
000006

.ENDM
HSG1:
000005
000005
110
145
154
154
157

,EVEN

Figure 7-2

7.4.3

.END

000001

Example of .NCHR Directive

I.NTYPEI

.NTYPE Directive

Format:
[label:]
where:

.NTYPE symbol,aexp

label

represents an optional statement label.

symbol

represents any legal symbol.
This symbol is
equated to the n-bit addressing mode of the
following expression (.=:1exp).
If a symbol is not
specified, the .NTYPE directive is flagged with an
error code (A) in the assembly listing.

7-14

MACRO DIRECTIVES

represents any
and/or tab) •
a exp

legal

separator

(comma,

space,

represents any legal address expression, as used
with an opcode. If no argument is specified, an
error code (A) will appear in the
assembly
1 ist ing.

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 {O) in the assembly listing.
An example of the use of an .NTYPE directive in a macro definition
shown in Figure 7-3.

2
3
4

• TITLE

NTYPE

.ENABL
.LIST

LC
HE

5
6
7
8
9

.HACRO

; Illustrate the .NTYPE directive
;-

• IF Ell

12
13

14
15
16
17

.ENDC
.ENDM

SAVE
.NTYPE
UU70

ARG
Ut,ARG

HOV
.IFF
MOV

ARGr-CSP>

;save in resister aode

tARGr-CSP>

;save in non-resister •ode

SAVE
.NTYPE

Rl
ttt,Rl

19 000000
000001
• IF EG

000000

010146

UU70

MOV
.IFF
HOV

Rl•-<SP>

;save in resister •ode

tRl,-<SP>

;save in non-resister •ode

SAVE
.NTYPE

TEMP
SU.TEMP

.ENDC
20
21 000002
000067

.IF ED
000002

012746
000006'

fU&70

MDV
.IFF
MOV

TEHP,-<SP>

;save in resister •ode

tTEMP,-CSP>

;save ir1 non-resister aode

.WORD

.ENDC
22
23 000006
24
25

000000
000001

Figure 7-3

TEMP:

.END

Example of .NTYPE Directive in Macro Definition

For additional information
Chapter 5 and Appendix B.2.

concerning

7-15

addressing

modes,

ref er

MACRO DIRECTIVES

7.5

!.ERROR I

.ERROR AND .PRINT DIRECTIVES

Format:
[ 1abe1:]
where:

.ERROR (expr]

;text

label

represents an optional statement label.

ex pr

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 message associated with the
directive.

.ERROR

The .ERROR directive is used to output messages to the listing file
during assembly pass 2. A common use of this directive is to alert
the user to a rejected or erroneous macro call or to the existence of
an illegal set ot conditions in a conditional a~~embly. If the
listing file is not specified, the .ERROR messages are output to the
command output device.
Upon encountering an .ERROR directive anywhere in
MACR0-11 outputs a single line containing:

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.

program,

For example, the following directive:
.ERROR

;Invalid macro argument

causes a line in the following form to be output to the listing file:
Seq. Loe.
No.
p

No.

s12 005642

Exp.
Value
00007n

Text
.ERROR

7-16

;Invalid macro argument

MACRO DIRECTIVES

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

7.6

INDEFINITE REPEAT BLOCK DIRECTIVES:

.IRP AND .IRPC

An indefinite repeat block is similar to a macro definition with only
one dummy argument.
At each expansion of the indefinite repeat range,
this dummy argument is replaced with successive elements of a
real
argument list.
Since the repeat directive and its associated 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 the conventional macros previously
described in this chapter.
An indefinite repeat o~ock can appear either w1tn1n or outside another
macro definition, indefinite repeat block, or repeat block.
The rules
for specifying indefinite repeat block arguments are the same as for
specifying macro arguments (see Section 7.3).

7.6.1

.!RP Directive

Format:
[label:]

.IRP sym,<argument list>

(range of indefinite repeat block)

.ENDM
where:

label

represents an optional statement label.
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 applies to .IRPC and .REPT.

7-17

MACRO DIRECTIVES
sym

represents a dummy argument that is replaced with
successive real arguments from 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,
repeat ranges and/or the .MEXIT
directive (see Section 7 .1. 3).

.ENDM

indicates the end of the indefinite
range.

repeat

block

The .IRP directive is used to replace a dummy argument with success1ve

real arguments specified in an argument string. This replacement
process occurs during the expansion of an indefinite repeat block
range.
An example of the use of the .IRP directive is shown in Figure 7-4.

7.6.2

I 'R"" I
I·'
.-"'I

.IRPC Directive

Format:
[label:]

• IRPC

sym,<string>

(range of indefinite repeat block)

.ENDM
where:

label

represents an optional statement label
in Section 7.~.l).

sym

represents a dummy argument that is replaced with
successive real arguments from within the angle
brackets. If no dummy argument is specified, the
.IRPC directive is flaqqed with an error code (A)
in the assembly listing:-

7-18

(see

Note

MACRO DIRECTIVES
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, repeat
ranges and/or the .MEXIT directive (see Section
7.1.3).

.ENDM

indicates the end of the indefinite
range.

repeat

block

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
successive characters in the specified string.
An example of the use of the .IRPC directive is shown in Figure 7-4 •

2
3

• TITLE

IRPTST

• LIST

; Illustrate the .IRP and .IRPC directives
; b~ creatins a Pair of RADSO tables
;-

6
7
B

.IRP
.RADSO
.ENDR
.RADSO
.RADSO
.RADSO
.RADSO
.RADSO
.RADSO
.RADSO
.RADSO

REG.<Pc,sP.RS.R4.R3.R2,Rt.RO>
/REG/

NUH,<76543210>
/R'NUH/

073110
073040
072770
072720
072650
072600
072530
072460

.IRPC
.RADSO
.ENDR
.RADSO
.RAD50
.RADSO
.RADSO
.RADSO
.RADSO
.RADSO
,RADSO

000001

.END

REGS:

9 000000

10
11
000000
000002
000004
000006
000010
000012
000014
000016
12
13 000020
14
15
000020
000022
000024
000026
000030
000032
000034
000036
16
17

062170
074500
072770
072720
072650
072600
072530
072460
REGS2:

Figure 7-4

/PC/
/SP/
/RS/
/R4/
/R3/
/R2/
/R1/
/RO/

/R7/
/R6/
/RS/
/R4/
/R3/
/R2/
/Rt/
/RO/

Example of .IRP and .IRPC Directives

7-19

MACRO DIRECTIVES

t.REPTI
7.7

REPEAT BLOCK DIRECTIVE:

l.l:NoRI

.REPT, .ENDR

Format:
[label:]

.REPT

exp

(range of repeat block)

.ENDR
where:

label

represents an optional statement label
in Section 7.o.l).

(see

Note

exp

represents any legal expression.
This
value
controls the number of times the biock 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
repeat
block
may contain macro definitions,
indefinite·· repeat blocks,
other
repeat blocks
and/or the .MEXIT directive (see Section 7.1.3).

• ENDM
or

indicates the end of the repeat block range •

.ENDR
The .REPT directive is used to duplicate a block of
number of times, in line with other source code.

7.8

MACRO LIBRARY DIRECTIVE:

.MCALL

code,

certain

I.MCALL I

Format:
.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 names must be separated by any legal
separator (comma, space, and/or tab).

7-20

MACRO DIRECTIVES
The .MCALL directive allows you to indicate in advance those system
and/or user-defined macro definitions that are 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:
•

Auto-Mcall mode is disabled

(the default)

or
•

The name of the macro being called is one of MACRO'S permanent
symbols or directives, such as SUB, .ERROR, or .PRINT.

The /ML switch (see Section 8.1.3)
under RSX-llM and
the /LIBRARY
qualifier
(see Section 8.2.2) under IAS and RT-11, used with an input
file specification, indicate to MACR0-11 that the
file
is a macro
library.
Additional macro libraries to be searched may also be
specified in the MACR0-11 program itself, using the MACR0-11
.LIBRARY
directive.
See Section 6.10.1
for
a description of the .LIBRARY
directive. When a macro call is encountered in the source program,
MACR0-11 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.
For
multiple library files,
the search for the named macros begins with
the last such file specified.
The files are searched in reverse order
until
the
required macro definitions are found,
finishing,
if
necessary, with a search of the system macro library.
If any named macro is not found upon completion of the search,
the
.MCALL statement
is flagged with an error code {U) in the assembly
listing.
Furthermore, a statement elsewhere
in the source program
that attempts to expand such an undefined macro is flagged with an
error code (O) in the assembly listing.
The command strings to MACR0-11, through which file specifications are
supplied, are described in detail in the applicable system manual (see
Section 0.3 in the Preface).

7.9

MACRO DELETION DIRECTIVE:

.MDELETE

Format:
.MDELETE namel,name2, .•. ,namen
where:

namel,
name2, ...
namen

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

The .MDELETE directive deletes
the definitions of the specified
macro(s),
freeing
virtual memory.
If references are made to deleted
macros, the referencing line is flagged with an opcode (O) error.
An example of the .MDELETE directive is shown below •
.MDELETE

. EXIT,EXIT$S

7-21

CHAPTER 8
IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES

MACR0-11 assembles one or more ASCII source files containing MACR0-11
statements into a single relocatable binary object file. This binary
object file contains 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 H.

8.1

RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES

On RSX-llM and RSX-llM-PLUS systems, two command languages are
available:
the Monitor Console Routine (MCR) and the DIGITAL Command
Language (DCL). When you log onto the system, you are given either
MCR or DCL as the default command language. Your default command
language is contained in your account file.
By typing CTRL/C (AC)
from the monitor prompt, you can see
explicit prompt for the command language you are currently using.

the

> Ac
MCR>

You can switch from one command language to the other.
DCL to MCR, type the following command:

To switch from

DCL> SET TERMINAL MCR
To switch from MCR to DCL, type the following command:
MCR> SET /DCL=TI:
In addition to switching from one command language to the other, you
can type a DCL command from a terminal set to MCR, and an MCR command
from a terminal set to DCL, as shown below:
MCR> DCL crnd-string
DCL> MCR crnd-string

8-1

T A~/R~X-11 M/R~X-11 M-PT.ll~

8.1.1

nPF.RA'l'TN~

PRnrRntTRR~

Initiating MACR0-11 Under RSX-llM/RSX-llM-PLUS

The following sections describe those MACR0-11 operating procedures
that apply to both the Monitor Console Routine and the DIGITAL Command
Language.
Any one of the four methods shown below may be employed to
initiate MACR0-11.

8.1.1.1

Method 1 - Direct MACR0-11 Call

MCR Format:
MCR>MAC
MAC>cmd-string
The Monitor Console Routine
(MCR)
accepts MAC as input, causing
MACR0-11 to be activated.
Since a command string is not present with
the MCR line, MACR0-11 then solicits input with the prompting sequence
MAC> and waits for command string input. After the assembly of the
indicated files has been completed, MACR0-11 again solicits command
string
input with the MAC> prompting sequence.
This process will be
repeated until CTRL/Z ("Z) is entered.
DCL Format:
MAr~n r /n11ril
i fi Pr t ~\ l
-.J.--·-----File(s)? filespec[/qualifiers] •••

nrr.'>
---~

···----·-

L.I

,._I

DCL accepts MACRO as
input.
In addition,
you may include the
qualifiers contained
in Table 8-3.
Since no file specifications are
included in the DCL command line, MACR0-11 solicits input with the
File(s)
prompt.
You can then enter the name of one or more source
ivhen you press
files plus any of the qualifiers listed in
RETURN, MACR0-11 performs the assembly.

8.1.1.2

Method 2 - Single Assembly

MCR Format:
MCR>MAC cmd-string
DCL Format:
DCL> MACRO cmd-string
In method 2, no prompting from MACR0-11 occurs,
since the command
string input is included in the command line.
MACR0-11 then assembles
the source files in the command string and exits when finished.

8.1.1.3

Method 3 - Install, Run Immediately, and Remove On Exit

Format:
>RUN $MAC
MAC>cmd-string

8-2

IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES
This method is used when the MACR0-11 assembler
is not permanently
installed in the system.
On RSX-llM, the system must be generated for
this type of call support.
MAC is run
from the system directory.
MACR0-11 solicits command string input.
The command string must have
the MCR format even if run from a DCL terminal. When MACR0-11 exits,
it is automatically removed from the system.
If the system has the "flying install"
format is not needed.

8.1.1.4

feature,

the

RUN

calling

Method 4 - Using the Indirect Command Processor

MCR Formats:
MCR>MAC
MAC>@filespec

or
MCR>MAC @f ilespec
or
MAC>RUN $MAC[/UIC=[g,m]]
MAC>@f ilespec

These forms use the indirect command processor, which effectively
accomplishes the substitution of "@filespec"
for the "cmd-string"
input employed in methods 1 through 3.
In these formats, the indirect
command processor
is passing commands to the assembler.
The file
specified as "@f ilespec" contains MACR0-11 command strings.
After
this file
is opened, command lines are read from the file until the
end-of-file is detected.
Three nested levels of indirect files are
permitted in MACR0-11.
MCR and DCL Format:
DCL> @f ilespec
These forms use the indirect command processor to pass commands to the
command language.
This is the only form you can use with DCL. The
indirect command file "@f ilespec" must contain one of the command
lines to initiate MACR0-11 as listed in methods 1 through 3.
NOTE
MACR0-11 can be terminated by entering a
CTRL/Z
(AZ)
at any time a request for
rtnmm.:in~

""""u1u1\.A1.1'-.A

8.1.2

Co+-r;nrt

t..:J\....L.L11'j

inn11+-

.Lllt''-''-

ic:-

.J..J

,.....,..,..oc-o""'+-

f:'.l..~a:::n;;:;11\-e

Default File Specifications

MACR0-11 accepts as input or creates as output up to six types of
files.
When using the MACR0-11 assembler, you should keep in mind the
default device, directory, name, and types listed in Table 8-1.
Table
8-1 lists the default values for each file specification.

8-3

IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES
Table 8-1
File Specification Default Values
Default Values
Directory

Filename

Type

Your default
volume

Current

None

.OBJ

Listing
File

Device used
for object
file

Directory
used in
Object file

None

.LST

Source
Files

Your default
volume

Current;
used for
source 1
or device of
last source
file specified

None

.MAC

user
Macro
Library

Your default
volume

Current, if
macro file
is specified
first; if not,
directory of
last source
file

None

.MLB

System
Mocro
Library

Library
device

(1,1]

RSXMAC

.SML

Indirect
Command
File

Your default
volume

Current

None

.CMD

File

Device

Object
File

8.1.3

MCR Command String Format

In response to the MAC> prompting sequence printed by MACR0-11, type
the output and input file specifications in the form shown below:
MAC>object,listing=srcl,src2, ••• ,srcn
where:

object

represents the binary object (output) file.

listing

represents the assembly listing
(output)
file
containing the table of contents, the assembly
listing, and the symbol table.
separates output file
file specifications.

srcl,
src2, •••
srcn

specifications

from

input

represent the ASCII source (input) files
containing the MACR0-11 source program or the
user-supplied macro library files to be assembled.

8-4

IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES
Only two output file specifications in the command string will be
recognized by MACR0-11; any more than two such files will be ignored.
No limit is set on the number of source input files.
If the entire
command string is longer than 80 characters and less than or equal to
132 characters, a hyphen can be placed at the end of the first line as
a continuation character.
A null specification in either of the output file specification fields
signifies that the associated output file is not desired. A null
specification in the input file field, however, is an error condition,
resulting in the error message "MAC -- Illegal filename" on the
command output device (see Section 8.5). Note that the absence of
both the device name (dev:) and the name of the file (filename.type)
from a file specification is the equivalent of a null specification.
NOTE
When no listing file is specified, any
errors encountered in the source program
are printed on the terminal from which
MACR0-11 was initiated.
When the /NL
switch is used in the listing file
specification without an argument, the
errors and symbol table are output to
the tile specified.
Each file specification contains the following information:
filespec /switch:value •••
where:

filespec

is the standard file specification.

/switch

represents an ASCII name identifying a switch
option.
This switch option may be specified in
three forms, as shown below, depending on the
function desired:
/switch
/noswitch
/-switch

Invokes the specified
switch
action.
Negates the specified
switch
action.
Also
negates
the
specified
switch action.

In addition, the switch
identifier
may
be
accompanied by any number of the following values:
ASCII character strings, octal numbers, or decimal
numbers.
The default assumption for a numeric
value is octal. Decimal values must be followed
by a decimal po int (.) •
Any numeric value preceded by a number sign (i) is
regarded as an explicit octal declaration; this
option is provided for documentation purposes and
ready identification of octal values.
Also, any numeric value can be preceded by a plus
s i g n ( + ) or a minus ( - ) s i g n •
The po s i t iv e
specification is the default assumption.
If an

8-5

IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES
explicit octal declaration is specified
(#), the
sign indicator,
if included,
must precede the
number sign.
All switch values must be preceded by a colon (:).
The switch specifications are interpreted in
context of the program to which they apply.
switch options
applicable
to
MACR0-11
described in Table 8-2 below.

the
The
are

A syntax error detected in the command string causes MACR0-11
output the following error message to the command output device:

MAC

Command syntax error

followed by a copy of the entire command string.
At assembly time, you may want to override certain MACR0-11 directives
appearing
in
the source program or to provide MACR0-11 with
information establishing how certain files are to be handled during
assembly.
You can do so through one or more switches, which may be
selectively invoked
as
additional
parameters
in
each
file
specification.
The available switches for
use
in MACR0-11 file
specifications under RSX-llM/RSX-llM-PLUS are listed in Table 8-2.
Table 8-2
MACR0-11 File Specification Switches
Function

Switch
/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 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.

/EN:arg
/DS:arg

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
6.2.1).
This switch overrides the arguments and
remains
in effect for the entire assembly
process.

/ML (see Note)

The /ML switch, which takes no accompanying
switch values,
indicates to MACR0-11 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,
MACR0-11
initiates a fixed search algorithm,
(continued on next page)

8-6

IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES
Table 8-2 (Cont.)
MACR0-11 File Specification Switches
Switch

Function

/ML (Cont.)

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.
This means that a user macro library file must
be specified in the command line or by using the
MACR0-11 .LIBRARY directive (see Section 6.10.1)
prior to the source file(s)
that use macros
defined in the library file.
MACR0-11 does not pre-scan the command line
for
macro libraries;
when a
new source file is
needed,
it
parses
the
next
input
file
specification.
If
that file specification
contains the /ML switch, it is appended
to the
front of the library file list.
As a result, a
user macro library file must be specified in the
command line prior to the source files which
macro
in
order
to
resolve
require it,
definitions.

/SP

Spool listing output (default value).

/NOSP

Do not spool output.

/CR: [arg]

Produce a cross-reference listing
8. 3) •

(see

Section

Switches for the object file are limited to /EN and /DS;
when
specified,
they apply throughout the entire command string.
Switch
options for the listing file are limited to /LI, /NL,
/SP,
/CR,
and
/NOSP.
Switches for
input files are limited to /ML, /EN, and /DS;
the option /ML applies only to the file
immediately preceding the
option so specified, whereas the /EN and /DS options, as noted above,
are also applicable to subsequent files in the command string.
Multiple occurrences of the same switch following a file specification
must be avoided, because the accompanying values of a subsequent like
switch specification override any previously-specified values.
If two
such switch values are desired,
they can be specified in the form

shown below:
/LI:SRC:MEB

8-7

IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES
8.1.4

DCL Operating Procedures

RSX-llM/RSX-llM-PLUS indicates its readiness to accept a command by
prompting with the DCL prompt.
In response to the prompt, enter the
command string in one of the formats shown below:
>MACRO[/qualifiers]
FILE? filespec(/qualifier(s]] (,filespec(/qualifier(s]] ••• ]
or

[rx:L]>MACRO[/qualifiers] filespec[/qualifier[s]] [,filespec[/qualifier[s]]~ •• ]
where:

qualifiers

affect either the entire command
string
(command
qualifiers)
or
the
filespec
(parameter qualifiers).
See Table 8-3 for a
description of the command qualifiers and
Table 8-4 for a description of the parameter
qualifiers.

f ilespec

is the standard file specification
Section 8.4.

You use the comma
(,)
to separate file specifications.
concatenates all the files and then performs the assembly.

shown

MACR0-11

Table 8-3
DCL Command Qualifiers
Function

Qualifier
/[NO]CROSS REFERENCE

Suppresses
or
generates
a
cross-reference listing
(see Section
8.3).
When the cross-reference is
generated,
a listing file
is also
generated, whether or not the /LIST
qualifier
is present in the command
string.
/NOCROSS_REFERENCE is the default.

/DSABLE:arg
/ENABLE:arg
/DSABLE:(arg,arg ••• )
/ENABLE:(arg,arg ••• )

Overrides
the
.DSABLE
or
.ENABLE
assembler directives in the source
program. When more than one argument
is entered, arguments must be enclosed
in
parentheses
and
separated by
commas.
You can specify any of the following
arguments with the /DSABLE or /ENABLE
qualifier.

Argument
ABSOLUTE

Enabling this function causes all
relative addresses {address mode
67) to be assembled as absolute
addresses (address mode 37).
(continued on next page)

8-8

IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES
Table 8-3 (Cont.)
DCL Command Qualifiers
Function

Qualifier
ABSOLUTE

(Cont.)

AUTO MCALL

By default, the ABSOLUTE argument
is disabled.
Enabling the AUTO MCALL argument
causes MACR0-11 -to search all
known macro libraries for a macro
definition
that
matches
any
undefined symbols appearing
in
the opcode field of a MACR0-11
statement.
By default,
this
option
is
disabled and if MACR0-11 finds an
unknown symbol
in the
opcode
field,
it either declares a (U)
undefined
symbol
error,
or
declares
the
symbol
as
an
external symbol, depending upon
the
GLOBAL argument described
below.

BINARY

Enabling this

function

absolute
binary
FILES-11 format.
By default, the
is disabled.
CARD FORMAT

produces

output

BINARY

argument

This function,
when
enabled,
treats columns 73 through the end
of the line as comments.
By
default,
the
CARD FORMAT
argument is disabled.

CASE MATCH

Enabling the CASE MATCH argument
causes the MACR0-11 conditional
assembly directives
.IF IDN/.IF
DIF to be alphabetically case
sensitive.
By default these directives
not case sensitive.

GLOBAL

Disabling

this

function

are

causes

MACR0-11 to flag
all undefined
symbol references with an error
code (U) on the assembly listing.
By default, the GLOBAL argument
is enabled and MACR0-11 treats
all symbols that are undefined at
the end of assembly pass 1 as
default global references.
(continued on next page)

8-9

IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES
Table 8-3 (Cont.)
DCL Command Qualifiers
Function

Qualifier
LOCAL

Enabling
the
LOCAL
argument
causes the assembler to treat all
symbols as local symbols.
When
enabled,
all global symbols are
flagged with the U (undefined
symbol) error message.
By default, the LOCAL argument is
disabled.

LOWER CASE

Enabling this function
causes
MACR0-11
to accept lower-case
ASCII input.
Disabling this function causes
MACR0-11 to convert lower-case
ASCII input to upper-case.
By
default,
the
argument is enabled.

LOWER CASE

Disabling this function
MACR0-11 to
ignore the
register definitions.

By default, register

causes
normal

definitions

are enabled.
TRUNCATION

When this function
is enabled,
MACR0-11 performs floating-point
truncation.
When this function is disabled,
MACR0-11 performs floating-point
rounding.
The
TRUNCATION
argument
disabled by default.

(continued on next page)

8-10

IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES
Table 8-3 (Cont.)
DCL Command Qualifiers
Function

Qualifier
/[NO]LIST[:filespec]

Specifies whether or not the assembler
should create and print a listing
file. You can include /LIST as a
qualifier for either a command or a
file
specification.
If
/LIST
qualifies the command, the listing
file is both entered in your directory
and printed on the line printer.
If
you
do
not
include
a
file
specification, the listing file has a
.LST file type and is named after the
last file named in the MACRO command.
The listing file cannot be a library
file.
(The LINK command and all other
language commands use the name of the
~~-~~
L
.1 l.;:, l-

L~~l.1..1 t::

----~
Ualllt::U

~.11!

~hl-l!t::

~-----~

~VllllllOlJU

-~
a;:,

~hl-llt::

default file name.) If /LIST qualifies
a file specification, the file is
entered in your directory but is not
printed on the line printer.
The
listing file is named after the file
it qualifies.
The default is /NOLIST.
/NOLIST is the default qualifier.
/[NO]OBJECT[:filespec]

Indicates whether or not the assembler
should create an object module.
If
you
do
not
include
a
file
specification in the command line, the
assembler creates an object file with
the same file name as the source file
and an .OBJ extension.
/OBJECT is the default qualifier.

/[NO]SHOW:arg
/[NO]SHOW: (arg,arg .•. )

Overrides
the
.LIST
and
.NLIST
assembler directives that
may
be
included in the source file. You can
use any of the following arguments
with the /SHOW qualifier.

Argument
BINARY

Controls the listing
expansion binary code.

macro

CALLS

Controls listing of macro calls
and repeat range expansions.
(continued on next page)

8-11

Table 8-3 {Cont.)
DCL Command Qualifiers
Function

Qualifier
COMMENTS

Controls listing of comments.

CONDITIONALS

Controls listing of
conditional coding.

CONTENTS

Controls listing of the table of
contents during assembly pass 1.

COUNTER

Controls listing of the
location counter field.

DEFINITIONS

Controls
listing
definitions
and
expansions.

of
repeat

macro
range

EXPANSIONS

Controls
listing
expansions.

macro

EXTENSIONS

Controls

binary

listing

unsatisfied

current

""""'f"""\~r\~, """'t"""
'.n. .t;'Vll.J.._, ..£.VJ.It.,.) e

LISTING DIRECTIVES

Controls
listing
of
listing
control
directives
without
arguments,
that
is,
directives
that
alter the listing
level
counter.

OBJECT BINARY

Controls listing of the generated
binary code.

SEQUENCE NUMBERS

Controls listing of
sequence numbers.

SOURCE

Controls listing of source lines.

SYMBOLS

Controls listing of
table
resulting
assembly.

I [NO]WIDE

source

line

the symbol
from
the

When set to WIDE,
the listing
is
printed
in 132 column format.
When
set to /NOWIDE, the listing is printed
in 80 column format.
/NOWIDE is the
default qualifier.

8-12

IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES
Table 8-4
DCL Parameter Qualifiers
Qualifier

Function

/LIBRARY

Specifies that an input file is a macro library file.
The assembler processes the files
listed
in the
command line in reverse order.
Therefore, a library
file cannot be the last file in the command line.

8.1.5

MACR0-11 Command String Examples
1.

The following command strings assemble the source file
FILNAM.MAC into a relocatable object module named FILNAM.OBJ.
MCR> MAC FILNAM=FILNAM
DCL> M/1.CRO

FILE?

FILNAM

DCL> MACRO FILNAM

The following command strings assemble the source
FILNAM.MAC
and
produce an object file with the
TESTA.OBJ.

file
name

MCR> MAC TESTA=FILNAM
DCL> MACRO/OBJECT:TESTA FILNAM
3.

The following command strings concatenate and assemble the
source files named FILNAM.MLB, TESTA.MAC, SPAN3.MAC, and
SHELL.MAC and create an object file named SHELL.OBJ.
MCR> MAC SHELL=FILNAM/ML,TESTA,SPAN3,SHELL
DCL> MACRO FILNAM/LIBRARY,TESTA,SPAN3,SHELL

The following command strings produce an object module and an
assembly listing.
Any .LIST TTM or .LIST COM directives in
the source file are ignored.
The listing produced by this
command has no comments
included and
is printed in wide
format.
MCR> MAC FILNAM,FILNAM/NL:TTM:COM=FILNAM
DCL> MACRO/LIST/NOSHOW:COMMENTS/WIDE FILNAM

8-13

TAS/RSX-llM/RSX-llM-PI.US OPF.'RA'J'TNG PROCF.DURES
8.2

IAS MACR0-11 OPERATING PROCEDURES

The following sections describe those MACR0-11
that apply exclusively to the IAS system.

8.2.1

operating

procedures

Initiating MACR0-11 Under IAS

The MACRO command, used under !AS, causes MACR0-11 to assemble one or
more
ASCII
source files
containing MACR0-11 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 8.3, below).
You can input a MACR0-11 program either directly from
the terminal
(interactive mode) or from a batch file (batch mode).
For interactive
mode use the MACRO command which can be
issued whenever the IAS
Program Development System
(PDS)
is at command level, a condition
signified by the appearance of the prompt:
PDS>
For batch mode use the $MACRO command.
When the assembly is completed,
MhCR0-11 terminates operations and
returns control
to PDS.
(Refer to the !AS User's Guide for further
information about interactive and batch mocre-operatio~

8.2.2

IAS Command String

Formats:
Interactive Mode
PDS> MACRO qualifiers

input
filespec /LIBRARY+ •..

or
PDS> MACRO qualifiers
FILES?

input
filespec /LIBRARY+ •••

Batch Mode
$MACRO qualifiers

input
filespec /LIBRARY+ ...

where:
input
f ilespec

is the specification of an input file
(see Section 8.4) that contains MACR0-11
source program code. When the program
consists of multiple files, a plus sign
(+) must be used to separate each file
specification from the next.
The "wild
card"
form of a file specification is
not allowed.

8-14

_ IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES

specifies that an input file is a macro
library file.
Library files hold the
definitions
of
externally
defined
macros.
As noted in Section 7.8, an
externally
defined
macro
must
be
identified in an .MCALL directive before
it can be retrieved and assembled with
the
user
program.
When
MACR0-11
encounters an .MCALL directive, a search
begins for the definitions of the macros
listed.

/LIBRARY

The search order is important because a
macro
might
have
two
different
definitions in library files LIBl and
LIB2.
For example,
if you need the
definition in LIBl, then you must place
LIBl after LIB2 in the command line
because MACR0-11 searches the last file
specified in the command line first,
then moves backwards through the files
given until all have been searched.
If a macro's definition is not found in
any of the files named by the user,
MACR0-11 automatically
searches
the
system macro library;
if the definition
is still not found, an error code (U) is
generated in the assembly listing.
specifies one or more of the following:

qualifiers
output
/OBJECT[:filespec]

produces
an
object
file
as
specified by filespec (see Section
8 •4) •
The def au 1 t i s a f i 1 e with
the same filename as the last named
source file and an .OBJ extension.
/OBJECT is always
the
default
condition.

/NOOBJECT

does not produce an object file.

output
/LIST [: f ilespec]

produces an assembly listing file
according to filespec (see Section
8.4).
If
filespec
is
not
specified, the listing is printed
on the line printer. The default
in interactive mode is /NOLIST and
in batch mode is /LIST.

/NOLI ST

does 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 t~e source program
are displayed at the terminal from which
MACR0-11 was initiated.

8-15

IAS/RSX-llM/RSX-llM-PLUS
/CROSSREFERENCE[:argl ••. arg4]
produces a cross-reference listing.
Argl
through arg4 are described in
Section 8.3.
This qualifier may be
abbreviated to /C.
A MACR0-11 command string can be specified using any one of the three
formats
shown above for
the
interactive and batch modes.
In
interactive mode, if the input file specification (filespec) does not
begin on the same line as the MACRO command and its qualifiers, PDS
prints the following prompting message:
FILES?
then waits for the user to specify the input file(s).
In batch mode, the $MACRO command and its arguments must appear on the
same line unless the PDS line continuation symbol (-) is used.

8.2.3

!AS Indirect Command Files

Format:

@f llespec;
where:
@

specifies that
indirect file.

the

name

that

follows

f ilespec

is the file specification (see Section 8.4)
of a
file
that contains a command string.
The default
extension for the file name is .CMD.

The indirect command file facility of PDS can be used with MACR0-11
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 8.2.2.
When an indirect command file reference
is used in a MACR0-11 command string, the contents of the specified
file are taken as all or part of the command string.
An indirect command file reference must always be the rightmost
in the command (see Section 8.2.4 for examples).

8.2.4

entry

!AS Command String Examples

The following examples show typical PDS MACR0-11 command strings.
1.

PDS> MACRO /NOLIST
FILES? A+BOOT.MAC;3
In this example, the source files A.MAC and BOOT.MAC;3 will
be assembled to produce an object file called BOOT.OBJ.
No
listing will be produced.

8-16

IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES

Where the indirect command file TEST.CMD contains the command
string:
MACRO/OBJECT:MYFILE

A+B

the command:
PDS>@TEST
causes MACR0-11 to assemble the two files
into an object file called MYFILE.OBJ.
3.

Where the indirect command
command string segment:

file

A.MAC

IND02.CMD

and

B.MAC

contains

the

ATEST/LIBRARY+BTEST+SRTl.021
the command:
PDS>MACRO/LIST:DKl:TST

@IND02

causes MACR0-11 to assemble the files BTEST.MAC and SRTl.021
using the macro library file ATEST.MAC to produce an object
file named SRTl.OBJ. A listing file named TST.LST is placed
on disk unit 1.
4.

$MACRO/LIST:DK0:MICR/NOOBJECT
LIB1/LIBRARY+MICR.MAC;002
In this example, the library file is assembled with the file
MICR.MAC;002.
The program listing file named MICR.LST is
placed on disk unit 0.

8.3

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

8-17

IAS/RSX-llM/RSX-llM-FLUS OPERATING PROCEDURES
To generate a cross-reference listing, specify the /CR switch in the
MACR0-11
command string.
Optional arguments can also be specified.
The form of the switch is:
/CR :

MAC
SYM}
REG
PST
{ SEC
ERR

where:
SYM

specifies user-defined symbols (default)

MAC

specifies macro symbols (default)

REG

specifies register symbols

PST

specifies permanent symbols

SEC

specifies program sections

ERR

specifies error lines (default)

If you wish to generate listings for user-defined
only, use /CR.
No argument is necessary.

and

macro
that

However,
if an argument
is
specified,
only
cross-reference listing is generated.
For example:

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 six types
of cross-reference listings, you must specify all six arguments
(the
order in which they are specified is not significant).
Use a colon to
separate arguments.
For example:
/CR:REG:SYM:MAC:PST:SEC:ERR
The CREF processor (CRF) is more
fully described
Reference Manual supplied with your system.

the

Utilities

Figure 8-1 illustrates a complete cross-reference listing.
In the
listing,
references are made
in the
form page-line.
To make the
listing more informative the CREF processor uses the following signs:
Meaning
somewhere in the source program the symbol
defined by a direct assignment statement.

listed

destructive reference;
at the line referenced by the
processor the value of the symbol
is changed (its
previous contents destroyed).

at the line referenced by the processor the symbol
listed
is defined by a direct assignment statement, a
colon sign (:) or a double colon sign (: :) .

8-18

IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES

R50UNP MACRO V05.00 Saturday 08-Jan-83 11:47
Cross reference table <CREF VOS.OO>

RSOUNP
SYMBOL

2-16#
2-17

2-25

RSOUNP MACRO V05.00 Saturday 08-Jan-83 11:47
Cross reference table <CREF V05.00)

Pase R-1

2-48*

2-23*
2-50*
2-18*
2-52*
2-19*
2-16
2-16*

R1
R2
R3
R4
SP

2-32*
2-51*
2-23

2-33*
2-52

2-21*
2-17*
2-27

2-33
2-18

2-43

2-45

2-25

2-27*

RSOUNP MACRO V05.00 Saturday 08-Jan-83 11:47
Cross reference table <CREF V05.00>

• ABS.
PURE I

2-49*

Pase C-1

0-0
0-0
2-14

Figure 8-1

8.4

Pase S-1

Sample CREF Listing

IAS/RSX-llM/RSX-llM-PLUS FILE SPECIFICATION

Format:
dev: [g,m]name.ext;ver
where:
dev:

is the name of the 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
(for example, DPl:, DK0:, DT3:).
The
default device is specified in Table 8-1.
The default
device under IAS is established initially by the system
manager for each user and can be changed through the
SET command.

[ g ,m]

is the User File Directory (UFD)
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 UFD code is:
[200,30].

8-19

IAS/RSX-llM/kSX-llM-PLU~

UPEHATlNG PROCEDURES

The
default
UFD
is
equivalent
to
the
User
Identification Code (UIC) given at log-in time.
Under
IAS, the UFD can be changed through the SET DEFAULT
command.
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
.LST
.MAC
.OBJ
.CRF

;ver

Indirect command (input) file
A listing (print format} file
MACR0-11 source module (input file)
MACR0-11 object module (output file}
Intermediate CREF input file.created by
MACR0-11.

is an octal number between 1 and 77777 that is used to
differentiate between versions of the same file.
This
number is prefixed by a semicolon (;}.
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.
This
is
the
general
form
for
a
file
specification
in
IAS/RSX-llM/RSX-llM-PLUS systems.
Detailed information is provided in
the applicable system user's guide or operating procedures manual (see
Section 0.3 in the Preface}.

8.5

MACR0-11 ERROR MESSAGES UNDER IAS/RSX-llM/RSX-llM-PLUS

MACR0-11 outputs an error message to the command output device when
one of the error conditions described below is detected.
MACR0-11
prints below the error message the command line that caused the error.
If the error
is a
.INCLUDE or a
.LIBRARY directive file error,
MACR0-11 prints below the error message both the source line and the
command line that caused the error.
MAC -- Error message
MACR0-11 source line
MACR0-11 command line
These error messages reflect operational problems and should not be
confused with the error codes (see Appendix D} produced by MACR0-11
during assembly.

8-20

IAS/RSX-llM/RSX-llM-PLUS OPERATING PROCEDURES
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;
MACR0-11 then attempts to
restart by reading
another command line.
In the case of a command I/O error, however,
MACR0-11 exits, since it is unable to obtain additional command line
input.
MAC

Command file/open failure
Either the file from which MACR0-11 is reading a
command could
not be opened initially or between assemblies;
or the indirect
command file specified as "@filename"
in the MACR0-11 command
line could not be opened.
See "OPEN FAILURE ON INPUT FILE".

MAC -- Command I/O error
An error was
returned by the
file system during MACRO-ll's
attempt to read a command line.
This is an unconditionally fatal
error,
causing MACR0-11 to exit.
No MACR0-11
restart
is
attempted when this message appears.
MAC -- Command syntax error
An error was detected in the syntax of the MACR0-11 command line.
MAC -- Illegal filename
Neither the device name nor the filename was present in the input
file specification (the input file specification was null), or a
wild card convention (asterisk)
was employed
in an input or
output file specification.
Wildcard options
specifications.

(*)

are

not

permitted

MACR0-11

file

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 MACR0-11.
MAC -- .INCLUDE directive file error
The file specified in the
.INCLUDE statement either does not
exist or is invalid, the device specified in the command line is
not available, or the .INCLUDE stacking depth exceeds five.
MAC -- Indirect command syntax error
The name of the indirect command file
(@filename)
specified
the MACR0-11 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.
(Three levels of indirect
command files are permitted in MACR0-11.)
MAC -- Insufficient dynamic memory
There is not enough physical memory available
for MACR0-11 to
page its symbol table.
Reinstall MACR0-11 in a larger partition,
or see Appendix F.3.
MAC -- Invalid format in macro library
The library file has been corrupted, or it was
the Librarian Utility Program (LBR).

8-21

not

produced

I~S/RSX

llM/RSX llM PLUS

MAC -- I/O error on input file
In reading a record from a source input file or macro library
file,
the
file
system detected an error;
for example, a line
containing more than 132(10) characters was 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
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 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
the device or by attempting to write to a device that is full.
MAC -- .LIBRARY directive file error
The file specified in the
.LIBRARY statement either does not
exist or
is
invalid,
the
file
specification in the .LIBRARY
directive is for a non-random access devie:e, U1t! uev.i<..:e ::;pe<..:ified
in the command line is not available, or the .LIBRARY stacking
depth exceeds the maximum depth allowed.
MAC -- Open failure on input file
1. Specified device does not exist.
2.
The volume is not mounted.
3. A problem exists with the device.
4.
Specified directory file does not exist.
5. Specified file does not exist.
~.
User does not have access privilege to the file directory

to the file itself.
MAC

Open failure on output file
Specified device does not exist.
The volume is not mounted.
A problem exists with the device.
Specified directory file does not exist.
5. User does not have access privilege to the file directory.
6. The volume is full or the device is write protected.
7. There is insufficient space for File Control Blocks.

1.
2.
3.
4.

MAC -- 64K storage limit exceeded
64K words of work file memory are available to MACR0-11.
This
message
indicates that the assembler has generated so many
symbols (about 13,000 to 14,000) that it has run out of space.
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.

8-22

CHAPTER 9
RSTS/RT-11 OPERATING PROCEDURES

9.1

MACR0-11 UNDER RSTS

The only way a MACR0-11 program can run on a RSTS
either the RT-11 or RSX run-time systems.

9.1.1

system

through

RT-11

run-time

RT-11 Through RSTS

There are two ways to run a MACRO program
system:

under

the

Use the RT-11 Emulator.
This is done by typing:
SW RTll.
The terminal will
respond with the RT-11 prompt (a dot
printed by the keyboard monitor).
You can then use the RT-11
commands (see Section 9.2).

The terminal will respond
Type the command: RUN $MACRO.SAV.
(*) prompt. You can then enter a command
with an asterisk
string of the form:
OBJFIL,LSTFIL=SRC ••. SRC6
where:

9.1.2

OBJFIL

is an object (output)
extension .OBJ.

file with

the

default

LSTFIL

is a listing (output)
extension .LST.

file with

the

default

SRC •••
SRC6

are source
(input)
files with the default
extension .MAC.
Six input files are allowed
in this command.

RSX Through RSTS

To run a MACRO program
command: RUN $MAC.TSK.

under the RSX run-time system,
The terminal will respond with:

MAC>
In answer you enter a command string of the form:
OBJFIL,LSTFIL=SRC ••• SRCN

9-1

type

the

RSTS/RT-11 OPERATING PROCEDURES
where:

OBJFIL

is an object (output)
extension .OBJ.

file with

the

default

LSTFIL

is a listing (output)
extension .LST.

file with

the

default

SRC •••
SRCN

are source
(input)
extension .MAC.

files

with the default

NOTE
There are other commands that can be
used
to call RT-11 and RSX but they are
site dependent and so are not mentioned
here.

9.2

INITIATING MACR0-11 UNDER RT-11

The following sections describe those
that apply only to the RT-11 system.

MACR0-11

operating

procedures

To call the MACR0-11 assembler from the system device, respond to
system prompt (a dot printed by the keyboard monitor) by typing:

the

R MACRO
When the assembler responds with an
accept command string input.

9.3

asterisk

(*),

ready

RT-11 COMMAND STRING

Format:
[dev:obj,dev:list,dev:cref/s:arg]=dev:srcl,src2, ••• ,dev:srcn/s:arg
where
dev

is any
legal
RT-11
device
file-structured device for input

obj

is the file specification of the binary object file
that the assembly process produces;
the device for
this file should not be TT or LP

list

is the file specification of the assembly
listing that the assembly process produces

cref

is the file specification of the CREF
temporary
assembly process
cross-reference
file
that
the
produces.
(Omission of dev:cref does not preclude a
cross-reference listing, however.)

/s:arg

is a set of file specification
(see Table 9-2).

9-2

for

options

output;

and

any

symbol

arguments

RSTS/RT-11 OPERATING PROCEDURES
srcl,
src2, ...
srcn

represent the ASCII source {input) files containing the
MACR0-11 source program or the user-supplied
macro
library files to be assembled.
You can specify as many
as six source files.

The following command string calls for an assembly that uses one
source file
plus the system MACRO library to produce an object file
BINF.OBJ and a listing.
The listing goes directly to the line
printer.
*DK:BINK.OBJ,LP:=DK:SRC.MAC
All output file specifications are optional.
The system does not
produce
an
output file
unless
the command string contains a
specification for that file.
The system determines the file type of an output file specification by
its position in the command string, as determined by the number of
commas in the string.
For example, to omit the object file, you must
begin the command string with a comma.
The following command produces
a listing, including cross-reference tables,
but not binary object
files.
*,LP:/C=(source file specification)
Notice that you need not include a comma after the final
specification in the command string.

output

file

Table 9-1 lists the default values for each file specification.
Table 9-1
Default File Specification Values
Default
Device

File

Default
File Name

Default
File Type

Object

DK:

Must specify

• OBJ

Listing

Same as for object
file

Must specify

.LST

Cref

DK:

Must specify

.TMP

First source

DK:

Must specify

.MAC

Additional source

Same as for preceding
source file

Must specify

.MAC

System MACRO
Library

System device SY:

SYSMAC

.SML

User MACRO
Library

DK: if first file,
otherwise same as for
preceding source file

Must specify

.MLB

9-3

RSTS/RT-11 OPERATING PROCEDURES
NOTE
Some assemblies need more symbol
table
space than available memory can contain.
When
this
occurs
the
system
automatically creates a temporary work
file called WRK.TMP to provide extended
symbol table space.
The default device for WRK.TMP is DK.
To
cause
the
system to assign a
different device,
enter the following
command:
.ASSIGN dev:
where:

9.4

dev

is the
file-structured
device
that will hold
WRK.TMP.

FILE SPECIFICATION OPTIONS

At assembly time you may need to override certain MACRO directives
aooearina
in the source oroarams.
You may also need to direct
MACRO-llJon the handling of certain files during assembly.
You can
satisfy these needs by using the switches described in Table 9-2.
Table 9-2
File Specification Options
Option

Usage

/L:arg
/N:arg

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

/E:arg
/D:arg

Function control switches; these options accept ASCII
switch values (arg) which are equivalent in function
and name to the arguments of the .ENABL and .DSABL
directives specified
in the source program
(see
Section 6.2.1).
This switch overrides the arguments
of the directives and remains
in effect for
the
entire assembly process.
(continued on next page)

9-4

RSTS/RT-11 OPERATING PROCEDURES
Table 9-2 (Cont.)
File Specification Options
Usage

Option

Indicates input file is MACRO library file. When the
assembler encounters an .MCALL directive in the
source code, it searches macro libraries according to
their order of appearance in the command string.
When it locates a macro record whose name matches
that given in the .MCALL, it assembles the macro as
indicated by that definition. Thus, if two or more
macro libraries contain definitions of the same macro
name, the macro library that appears rightmost in the
command string takes precedence.

Consider the following command string:
*(output file specification)=ALIB/M,
BLIB/M,XIZ
Assume that each of the two macro libraries, ALIB.MLB
and BLIB.MLB, contain a macro called .BIG, but with
different definitions.
Then, if source file XIZ
contains
a macro call .MCALL .BIG, the system
includes the definition of .BIG in the program as it
appears in the macro library BLIB.
If the command string does not include the standard
system
macro
library
SYSMAC.SML,
the
system
automatically includes it as the first source file in
the
command string.
Therefore,
if macro library
ALIB.MLB contains a definition of a macro called .READ,
that definition of .READ overrides the standard .READ
macro definition in SYSMAC.SML.
/C:arg

Controls contents of cross-reference listing.

The /M switch affects only the source file to which
The other options affect the entire command string.

9.5

appended.

CROSS-REFERENCE (CREF) TABLE GENERATION OPTION

A cross-reference (CREF) table lists all or a subset of the symbols in
a source program,
identifying the statements that define and use
symbols.

9.5.1

Obtaining a Cross-Reference Table

To obtain a CREF table you must include
command string.
Usually you include
assembly listing file specification.

9-5

the
the

/C:arg
/C:arg

option in the
option with the

RSTS/RT-11 OPERA.TING
If the command string does not include a cref file specification,
the
system automatically generates a temporary file on device DK:.
If you
need to have a device other than DK:
contain the temporary cref file,
you must include the dev:cref field in the command string.
A complete CREF listing contains the following six sections:

A cross reference of program symbols--labels
program ~nd symbols followed by an operator.

A cross reference of register symbols.
Rl, R2, R3, R4, R5, SP, and PC.

A cross reference of MACRO symbols--those symbols defined
.MACRO and .MCALL directives.

A cross reference of permanent
mnemonics and assembler directives.

A cross reference of program sections--the names you
as operands of .CSECT or .PSECT directives.

A cross reference of errors--the system groups and lists
flagged errors from the assembly by error type.

used

the

These symbols are R0,

symbols--all

operation
specify
all

You can include any or all of these
six
sections
on
the
cross-reference listing
by specifying the appropriate arguments with
the /C option. These arguments are listed and described in Table 9-3.
Table 9-3
/C Option Arguments
Argument

CREF Section

User defined symbols
R

MACRO symbolic names

Permanent symbols including instructions and directives

Control and program sections

Error code grouping
NOTE
Specifying /C with no arguments
is
equivalent to specifying /C:S:M:E.
That
special
case
excepted,
you
must
explicitly request each CREF section by
including
its
arguments.
No
cross-reference file occurs if the /C
option is not specified,
even if the
command string
includes a CREF file
specification.

9-6

RSTS/RT-11 OPERATING PROCEDURES
9.5.2

Handling Cross-Reference Table Files

When you request a cross-reference listing by means of the /C option,
you cause the system to generate a temporary file, DK:CREF.TMP.
If device DK:
is write-locked or if it contains insufficient free
space for the temporary file, you can allocate another device for the
file.
To allocate another device, specify a third output file in the
command string;
that is, include a dev:cref specification.
(You must
still include the /C option to control the form and content of the
listing.
The dev:cref specification is ignored if the /C option is
not also present in the command string.)
The system then uses the dev:cref file
instead of DK:CREF.TMP
deletes it automatically after producing the CREF listing.

and

The following command string causes the system to use RK2:TEMP.TMP
the temporary CREF file.

*,LP:,RK2:TEMP.TMP=SOURCE/C
Another way to assign an alternative device for the CREF.TMP
to enter the following command prior to entering R MACRO:

file

.ASSIGN dev:CF
This method is preferred if you intend to do several assemblies, as it
relieves you from having to include the dev:cref specification in each
command strinq.
If you enter the ASSIGN dev:
CF command!
and later
include a cref specification in a command string, the specification in
the command string prevails for that assembly only.
The system lists requested cross-reference tables following the
assembly listing.
Each table begins on a new page.

MACRO

The system prints symbols and also symbol values, control sections,
and error codes,
if applicable, beginning at the left margin of the
page. References to each symbol are listed on the same line,
left-to-right across the page.
The system lists references in the
form P-L; where P is the page in which the symbol,
control section,
or error code appears, and L is the line number on the page.
A number sign (#) next to a reference indicates a symbol definition.
An
asterisk
(*)
next to a
reference indicates a destructive
reference--an operation that alters the contents of the addressed
location.

9.5.3

MACR0-11 Error Messages Under RT-11

MACR0-11 outputs an error message to the command output device when
one of the error conditions described below is detected.
MACR0-11
prints below the error message the command line that caused the error.
If the error
is a
.INCLUDE or a
.LIBRARY directive file error,
MACR0-11 prints below the error message both the source line and the
command line that caused the error.
?MACRO-s-Error message
MACR0-11 source line
MACR0-11 command line

9-7

RSTS/RT-ll OPERATING PROCEDURES

The s in the error message represents the letter code
the severity level of the error.

that

indicates

These error messages reflect operational problems and should not be
confused with the error codes (see Appendix D) produced by MACR0-11
during assembly.
Meaning

Error Message
?MACRO-F-Device full DEV:

The output volume does not
room for an output file
command string.

have sufficient
specified in the

?MACRO-F-File not found DEV:FILNAM.TYP
An input file in the command line
exist on the specified device.

does

not

?MACRO-F-.INCLUDE directive file error
The file specified in the .INCLUDE statement
either does not exist or is invalid, the
device specified in the command line is not
available, or the .INCLUDE stacking depth
exceeds five.
?MACRO-F-Insuff icient memory
MACRO does not: nave Lne Hl.l.IIJ.ILIUHl aff1uunt
memory {l~K words) necessary to run.

- ~

Ul..

?MACRO-F-Invalid command
The command line contains a syntax error
specifies more than six input files.

?MACRO-F-Invalid device
A device specified in the command
not exist on the system.
?MACRO-F-Invalid macro

line

does

, !'----··

.J.J.ULCILY

The library file has been corrupted or it was
not produced by the RT-11 librarian, LIBR.
?MACRO-F-Invalid option:

/x
The specified option was
the program.

not

recognized

?MACRO-F-I/O error on DEV:FILNAM.TYP
A hardware error occurred while attempting to
read from or write to the device on the
specified file.
?MACRO-F-I/O error on work file
MACRO failed to open, read, or write
work file, WRK.TMP.

9-8

its

RSTS/RT-11 OPERATING PROCEDURES

?MACRO-F-.LIBRARY directive file error
The file specified in the .LIBRARY statement
either does not exist or is invalid, the file
specification in the
.LIBRARY directive is
for a non-random access device, the device
specified
in the command
line
is
not
available,
or the
.LIBRARY stacking depth
exceeds the maximum depth allowed.
?MACRO-F-Protected file already exists DEV:FILNAM.TYP
An attempt was made to create a
file having
the same name as an existing protected file.
?MACRO-F-Storage limit exceeded (64K)
MACRO's Virtual Symbol Table
can
store
symbols
and
macros
up to 64K
in any
combination.
Your program contains more than
64K worth of one or both of these elements.
?MACRO-W-I/O error on CREF file:
CREF aborted
MACRO ran out of device space while writing
the cref file,
or a hardware error has
occurred.
The cref file is aborted but
assembly continues.

9-9

APPENDIX A
MACR0-11 CHARACTER SETS

A.I

ASCII CHARACTER SET
Even
Parity
Bit

7-Bit
Octal
Code

0
1

000

001

NUL
SOH

002

STX

003

ETX

004

EOT

005

ENQ

006

ACK

007
010

BEL

0
1
1

Character

011

012

HT
LF

1
0

013
014

VT
FF

015

016

eJ 1 7

1
0

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/a.
Data link escape. CONTROL/P (DC0).
Device
control
l;
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-1

MACR0-11 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
0 L\ 4
045
046
047
050
051
052
053

CAN
EM
SUB
ESC
FS
GS
RS

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.

0
1

1
0
1
0
0
1
1
0
0

1
0
l

1
0
0
1
1
0

0
0
1
0

1
1
0

1
0
0

1
1
0
0
1
0

1
1
0
1
0
0
1
0

1
l
0
0
0

us
SP

Remarks

*$
%
&

Accent acute or apostrophe.
(
)

I
0
1
2
3

4
5
6
7
8
9

<
>

077

100

101

A
B

102

103
104
105
106
107
110
111

file

054

055
056
057
060
061
0e;2
063
064
065
066
067
070
071
072
073
074
075
076

also

c
D
E
F
G
H
I

A-2

ERR,

(SYNC) •

MACR0-11 CHARACTER SETS
Even
Parity
Bit

7-Bit
Octal
Code

112
113
114
115
116
117
120
121
122
123
124
125
126
127
130
131
132

0
1
0

1
1
0
1
0
0

1
1
0
0
l
0

Remarks

Character
J
K

L
M
N

0
p
Q

s
T

w
x
y

z
r

sh ft/l.
sh ft/m.

134
135
136
137
140
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

DEL

') ')

.J. .J .J

1
1
0

0
1
l

0
1
0
0
l

1
0
0

1
0
1
1
0

1
0
0

1
0
1
1
0
0
l

1
0

Appears as :ff: or

Appears as

]

f=+- It.

..... '-/ r... •

*
**
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

z
This code generated by ALTMODE.
This code generated by pref ix key
(if present) •
Delete, Rubout.

on some machines.

< on some machines.

A-3

MACR0-11 CHARACTER SETS
A.2

RADIX-50 CHARACTER SET
ASCII
Octal
Equivalent

Character

Radix-50
Equivalent

Space

A-Z

101-132

1-32

34
35

Unused

36-47

60-71

0-9

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
p~rformed in o~tal) ~
X=ll3000
2=002400
8=000002
X2B=ll5402
c::; nnl o
--••"::J--

f"l-. ::a,..
V•.a."'""'.&..

or
First Char.

Second
Character

Third
Character

Space
A
B

Space
A

Space

E
F
G

H
I
J
K
L

M
N

0
p
Q

000000
003100
006200
011300
014400
017500
022600
025700
031000
034100
037200
042300
045400
050500
053600
056700
062000
065100
070200
073300
076400
101500

c
D

E
F
G

H
I
J
K
L

M
N

0
p
Q

000000
000050
000120
000170
000240
000310
000360
000430
000500
000550
000620
000670
000740
001010
001060
001130
001200
001250
001320
001370
001440
001510

A-4

E
F
G

H
I
J
K
L

M
N

0
p
Q

000000
000001
000002
000003
000004
000005
000006
000007
000010
000011
000012
000013
000014
000015
000016
000017
000020
000021
000022
000023
000024
000025

MACR0-11 CHARACTER SETS
Single Char.
or
First Char.

Second
Character

Third
Character

v
w
x

104600
107700
113000
y
116100
z
121200
$
124300
127400
Unused 132500
135600
0
140700
1
2
144000
147100
3
152200
4
155300
5
160400
6
163500
7
8
166600
171700
9

z
$

Unused
0
1
2
3
4
5
6

7
8
9

001560
001630
001700
001750
002020
002070
002140
002210
002260
002330
002400
002450
002520
002570
002640
002710
002760
003030

A-5

z
$

Unused
0
1
2
3
4
5
6
7
8
9

000026
000027
000030
000031
000032
000033
000034
000035
000036
000037
000040
000041
000042
000043
000044
000045
000046
000047

APPENDIX B
MACR0-11 ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

B.l

SPECIAL CHARACTERS
Function

Character

%
tab
space
#
@
(

, (comma)
+

(apostrophe)

<
>
\
vertical tab

B.2

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

Symbols used in the table:
n

is an integer, 0 to 7, representing a register number

is a register expression

is an expression

is either a register expression or an expression whose value
is in the range 0 to 7.
B-1

MACR0-11 ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

Format

Address
Mode
Name

Address
Mode
Number

@R or
(ER)

(ER)+

Auto increment

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

@(ER)+

Au to increment
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 ( E'R)

Index

fin

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-2

Meaning
R

contains

the
the ad-

MACR0-11 ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES
B.3

ASSEMBLER DIRECTIVES

The MACR0-11 assembler directives are summarized in the following
table.
For a detailed description of each directive, the table
contains references to .the appropriate sections in the body of the
manual.
Form

qection
Reference

Operation

5.3.3
7.3.7

Followed by one ASCII character
a
single
quote
(apostrophe)
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.

6.3.3

Followed by two ASCII characters a
double quote 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

A temporary radix
the value n to
binary number.

"cexpr

6.4.2.2

A temporary numeric control, causes
the expression's value to be onescomplemented.

"on

6.4.1.2

A temporary radix
the value n to
decimal number.

"Fn

6.4.2.2

A temporary numeric control, causes
the value n to be treated as a
sixteen-bit floating-point number.

"on

6.4.1.2

A temporary radix
the value n to
octal number.

"Rccc

6.3.7

Converts 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

control, causes
be treated as a

control, causes
be treated as an

MACR0-11 ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

Form

Section
Reference

Operation

• AS C IZ Is tr i ng I

6.3.5

Generates
a
block
of
data
containing
the ASCII equivalent of
the character string
(enclosed
in
d el im it i ng
characters) ,
one
character per byte, with a
zero
byte
terminating
the specified
string.

.ASECT

6.7.2

Begins or
resumes
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.

.CROSS syml,sym2, ...

6.2.2

Enables the cross-reference listing
for
the specified symbol list.
If
a symbol
list
is not specified,
this
directive
reenables
the
cross-reference listing
for
all
symbols in the program.

. CSE CT [name]

5.7.2

Begins or resumes named or unnamed
relocatable program section.
This
directive
is
provided
for
compatibility
with other PDP-11
assemblers.

• DSABL a rg

6.2.1

Disables the function specified
the argument.

• ENABL a rg

6.2.1

Enables
(invokes)
the
function
specified by the argument.

• END [exp]

6.6

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

• ENDC

6.9.1

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.

B-4

the

absolute

conditional

MACR0-11 ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

Form

Section
Reference

,ENDR

7.7

Indicates the end of the current
repeat block.
This directive is
provided for compatibility
with
other PDP-11 assemblers.

.ERROR exp;text

7.5

A 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.l

Generates
successive
2-word
floating-point equivalents for the
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,syrn2, ••.

6.8.1

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

.!DENT /string/

6. l . 4

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

6.9.1

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

6.9.2

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

.IFT

6.9.2

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.

Operation

B-5

MACR0-11 ASSF.MBl.Y LANGUAGE ANO ASSEMBLER DIRECTIVES

Form

Section
Reference

.IFTF

6.9.2

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

.IIF cond,arg,
statement

6.9.3

Acts
as
a 1-line
conditional
assembly block where the condition
is
tested
for
the
argument
specified.
The
statement
is
assembled only if the condition
tests true.

• INCLUDE string

6.10.2

Inserts a specified source file
within the source file currently
being used.

• 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.

.LIBRARY string

6.rn.1

Adds a specified
macro
library
searched.

.LIMIT

6.5.4

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,
arg2, ••.

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.

Operation

B-6

file name to a
list
that
is

the start of a macro
having the
specified
the
following dummy

MACR0-11 ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

Form

Section
Reference

Operation

.MDELETE namel,name2, ••. 7.9

Deletes ~ne definitions of
the
specified macro(s), freeing virtual
memory.

.MEXIT

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

7 .1. 3

.NARG symbol

Appearing 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

Appearing 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, decrements the
listing
level count by 1. With an
argument, this directive suppresses
that
portion
of
the
listing
specified by the argument.

.NOCROSS syml,sym2, •••

6.2.2

Disables
the
cross-reference
listing for the listed symbols.
If
a symbol
list is not specified,
this
directive
disables
the
cross-reference listing
for
all
symbols in the program •

• NTYPE symbol,aexp

7.4.3

Appearing only within
a
macro
definition,
equates the symbol to
the 6-bit addressing mode of the
specified address expression.

.ODD

6.5.2

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

.PACKED

6.3.8

Causes a decimal number of 31 (10)
digits or less to be packed 2
digits per byte.

.PAGE

f) .1. 5

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

n.7.1

Begins or
resumes
a named
unnamed program section having
specified attributes.
8-7

or
the

Form

Section
Reference

Operation

.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.

.REM comment-character

6.1.6

Allows a programmer to insert a
block of comments into a MACR0-11
source program without having to
precede the comment lines with the
comment character (;).

.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.

.RESTORE

6.7.4

Retrieves
a
previously
.SAVEd
program section from the top of the
program
section
context
stack
leaving the current program section
in effect.

.SAVE

6.7.3

Stores the current program section
on the top of the program section
context stack leaving the current
program section in eifecL.

.SBTTL string

6.1.3

Causes the specified string to be
printed as part of the assembly
1 istina oaae header.
The strino
compon~nt Jof each .SBTTL directive
is collected into a
table
of
contents at the beginning of the
assembly listing.

.TITLE string

~.1.2

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.

.WEAK syml,sym2, •..

().8.2

Specifies symbols that are either
defined
externally
in
another
module or are defined globally in
the current module.

.WORD expl,exp2, ••

6.3.2

Generates successive words of data;
each word contains the value of the
corresponding specified expression.

B-8

APPENDIX C

PERMANENT SYMBOL TABLE (PST)

The mnemonics for the PDP-11 operation
(op)
codes and MACR0-11
assembler directives are stored in the Permanent Symbol Table (PST).
The PST contains the symbols that are automatically recognized by
MACR0-11.
For a detailed description of the op codes, see the
Handbook.

C.l

PDP-11

Processor

OP CODES
Instruction
Mnemonic

Octal
Value

Operation

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
BMI
BNE
BPL
BPT
BR

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
100400
001000
100000
000003
000400

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 I f Equal
Branch If Greater Than Or Equal
Branch I f Greater Than
Branch I f 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 I f Lower
Branch I f Lower Or Same
Branch I f Less Than
Branch I f Minus
Branch I f Not Equal
Branch I f Plus
Breakpoint Trap
Branch Unconditional

C-1

PERMANENT SYMBOL TABLE (PST)
Instruction
Mnemonic

Octal
Value

BVC
BVS
CALL
CALLR
CCC
CLC
CLN
CLR
CLRB
CLV
CLZ
CMP

102000
102400
004700
000100
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

IOT
JMP
JSR
MARK
MED6X
MED74C
MFPI

000004
000100
004000
006400
076600
076601
006500

MFPS

106700

MFPT
MOV
MOVB

000007
010000
110000

MTPI

000600

MTPS

106400

MUL
NEG
NEGB
NOP
RESET
RETURN
ROL
ROLB
ROR
RORB

070000
005400
105400
000240
000005
000207
006100
106100
006000
106000

Operation
Branch If Overflow Is Clear
Branch If Overflow Is Set
Jump To Subroutine (JSR PC,xxx)
Jump (JMP add r)
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
PDP-11/60 Maintenance
PDP-11/74 CIS Maintenance
Move From Previous
Instruction Space
Move from PS
(LSI-11, LSI-11/23, LSI-11/2)
Move From Processor Type
Move Source To Destination
Move Source To Destination
(Byte)
Move To Previous
Instruction Space
Move to PS
(LSI-11, LSI-11/23, LSI-11/2)
Multiply
Negate Destination
Negate Destination (Byte)
No Operation
Reset External Bus
Return From Subroutine (RTS PC)
Rotate Left
Rotate Left (Byte)
Rotate Right
Rotate Right (Byte)
C-2

PERMANENT SYMBOL TABLE (PST)
Instruction
Mnemonic

Octal
Value

RTI

000002

RTS
RTT

000200
000006

SBC
SBCB

005600
105600
000277
000261
000270
000262
000264
077000
160000

sec
SEC
SEN
SEV
SEZ
SOB
SUB
SWAB
SXT
TST
TSTB
TSTSET
WAIT
WRTLCK

000300
006700
i04400
005700
105700
007200
000001
007300

XFC
XOR

076700
074000

TRAP

Operation
Return From Interrupt
(Permits a trace
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)
Test Destination And Set Low Bit
Wait For Interrupt
Read/Lock Destination. Write/Unlock
R0 Into Destination
Extended Function Code
Exclusive OR

COMMERCIAL INSTRUCTION SET (CIS) OP CODES
Every operation listed in the CIS table has two instruction mnemonics.
The suffix "I", attached to every second mnemonic, indicates that the
addresses are inline.
The inline instructions require two arguments;
the other instructions (excepting L2DN and L3DN) require no arguments.
Instruction
Mnemonic
ADDN
ADDNI
ADDP
ADDPI
ASHN
AS HNI
ASHP
.Z\SHPI

CMPC
CMPC I
CMPN
CMPNI
CMPP
CMPPI
CVTLN
CVTLNI
CVTLP

Octal
Value
076050
076150
076070
076170
076056
076156
010010
07617n
076044
076144
07'5052
076152
076072
076172
076057
076157
076077

Operation
Add Numeric
Add Numeric
Md Packed
Add Packed
Arithmetic Shift Numeric
Arithmetic Shift Numeric
Arithmetic Shift Packed

Arithmetic
Compare
Compare
Compare
Compare
Compare
Compare
Convert
Convert
Convert

C-3

Cl-.; .f:+.

.. JlJ .L .1. l-

n-,....1 ... -...:i

.LGl~l'-CU

Character String
Character String
Numeric
Numeric
Packed
Packed
Long To Numeric
Long To Numeric
Long To Packed

PERMANENT SYMBOL TABLE {PST)

Instruction
Mnemonic

Octal
Value

CVTLPI
CVTNP
CVTNPI
CVTPN
CVTPNI
DIVP
DI VP I
LOCC
LOCCI
L2DN*
L3DN*
MATC
MATCI
MOVC
MOVCI
MOVRC
MOVRCI
MOVTC
MOVTCI
MULP
MULPI
SCANC
SCANCI
SKPC
SKPCI
SPANC
SPANCI
SUBN
SUB NI
SUBP
SUB PI

076177
076055
076155
076054
076154
076075
076175
076040
076140
07602N
07606N
076045
076145
076030
076130
076031
076131
076032
076132
076074
076174
076042
076142
076041
076141
076043
076143
076051
076151
076071
076171

Operation
Convert Long To Packed
Convert Numeric To Packed
Convert Numeric To Packed
Convert Packed To Numeric
Convert packed To Numeric
Divide Decimal
Divide Decimal
Locate Character
Locate Character
Load 2 Descriptors @(RN)+
Load 3 Descriptors @(RN)+
Match Character
Match Character
Move Character
Move Character
Move Reverse Justified Character
Move Reverse Justified Character
Move Translated Character
Move Translated Character
Multiply Decimal
Multiply Decimal
Scan Character
scan Character
Skip Character
Skip Character
Span Character
Span Character
Subtract Numeric
Subtract Numeric
Subtract Packed
Subtract Packed

where N=0 .•. 7

FLOATING POINT PROCESSOR OP CODES
Instruction
Mnemonic

Octal
Value

ABSD
ABSF
ADDD
ADDF
CFCC

170600
170600
172000
172000
170000

CLRD
CLRF
CMPD
CMPF
DIVD
DIVF
LDCDF

170400
170400
173400
173400
174400
174400
177400

LDC FD

177400

LDC ID

1 77000

Operation
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

C-4

PERMANENT SYMBOL TABLE (PST)
Instruction
Mnemonic

Octal
Value

LDC IF

177000

LDCLD

1 77000

LDCLF

177000

LDD
LDEXP
LDF
LDFPS
MFPD

172400
176400
172400
170100
106500

MODD

171400

MODF

1 71400

MTPD
MULD
MULF
NEGD
NEGF
SETD
SETF
SETI
SETL
SPL
STA0
STB0
STCDF

106600
171000
171000
170700
170700
170011
170001
170002
170012
000230
170005
170006
176000

STCDI

175400

S'T'CDL

175400

STCFD

176000

STCFI

175400

STCFL

175400

STD
STEXP
STF
ST FPS
STST
SUBD
SUBF
TSTD
TSTF

174000
175000
174000
170200
170300
173000
173000
170500
170500

Operation
Load And Convert Integer To
Floating
Load And Convert Long
Integer To Double
Load And Convert Long
Integer To Floating
Load Double
Load Exponent
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
Diagnostic Floating Point
Diagnostic Floating Point
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

C-5

PERMANENT SYMBOL TABLE {PST)
C.2

MACR0-11 DIRECTIVES

The MACR0-11 directives that follow are described in greater detail in
Appendix B.
Function

Directive
• ASCII
.ASCIZ
.ASECT
.BLKB
.BLKW
.BYTE
.CROSS
.CSECT
• DSABL
• ENABL

• rnD
• ENDC
.ENDM
.ENDR
.ERROR
• EVEN
.FLT2
.FLT4
• GLOBL
.!DENT
• IF
.!FF
.!FT
.IFTF
.IIF
.INCLUDE
.IRP
.IRPC
.LIBRARY

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-11 assemblers).
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.
Enables cross-reference listing
for
specified
symbols;
enables cross-reference for all symbols.
Begins relocatable program section
(provided
for
compatibility with other PDP-11 assemblers).
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-11 assemblers).
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).
Inserts specified source file within source file
currently being used.
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.
Adds a specified file name to a macro library list
that is searched.

C-6

PERMANENT SYMBOL TABLE (PST)
Function

Directive
.LIMIT
.LIST
.MCALL
. MDELETE
.MEXIT
.NARG
.NCHR
.NLIST
.NOCROSS
.NTYPE
• ODD
• PACKED
. PAGE
• PRINT
.PSECT
. RADIX
.RAD50
.REM
.REPT
.RESTORE
.SAVE
.SBTTL
.TITLE
.WEAK
.WORD

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.
Deletes the definitions of the specified macro(s) •
Exit from current macro definition or
indefinite
repeat block.
Equates specified symbol
to
the
number
of
non-keyword 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.
Disables cross-reference listing
for
specified
symbols;
disables cross-reference listing for all
symbols.
Equates specified symbols to the addressing mode
of the specified argument.
Byte-aligns the current location counter •
Generates packed decimal data, 2 digits per byte .
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.
Inserts a block of comments into a MACR0-11 source
program without having to precede comments lines
with the comment character (;).
Begins repeat block and replicates it according to
the value of the specified expression.
Stores the current program section context on the
top of the program section context stack.
Retrieves the program section from the top of the
program section context stack.
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.
Specifies symbols
that
are
either
defined
externally
in another module or are defined
globally in the current module.
Generates successive word data in accordance with
specified arguments.

C-7

APPENDIX D
ERROR MESSAGES

An error code is printed as the first character in a source line
containing an error. This error code identifies the error condition
detected during the processing of the line. Example:
Q

26 000236

MOV Rl,R2,A

010102

The extraneous argument A in the MOV instruction above causes the line
to be flagged with a Q (syntax) error.
Meaning

Error Code
A

Assembly error. Because many different conditions
produce this error message, the directives which
may yield a general assembly error have been
categorized
below
to
reflect
these
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 6-2) is specified with the
directive •
• ENABL/.DSABL -- Other than a legally defined
argument
(see Table n-3) is specified with the
directive, or the attribute arguments of a
previously declared program section •
• PSECT -- Other than a legally defined argument
(see
Table
6-4)
is
specified with the
directive, or the attribute arguments of a
previously declared program section change (see
Section 6.7.1.1) •
• IF/.IIF -- Other than
a
legally
defined
conditional test (see Table 6-6) or an illegal
argument expression value is specified with the
directive •
• MACRO -- An illegal or duplicate symbol
in dummy argument list.

D-1

found

ERROR MESSAGES
Meaning

Error Code
A

(cont.)

.TITLE -- Program name is not specified in the
directive,
or
first
non-blank
character
following
the directive
is a
non-Radix-50
character .
. IRP/.IRPC -- No dummy argument is specified in
the directive .
. NARG/.NCHAR/.NTYPE -- No symbol
in the directive.

specified

.IF/.IIF -- No
conditional
specified in the directive.

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 ae11m1L.ers do nuL
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
(from -128(10) to +127(10) words) has been
exceeded.

A statement makes
invalid
use of
the
current location counter.
For example, a
".=expression" statement attempts to force
the
current location counter to cross
program section (.PSECT) boundaries.

A statement
expression:

contains

invalid

address

In cases
where
an
absolute
address
expression is required, specifying a global
complex
symbol, a relocatable value, or a
relocatable value (see Section 3.9) results
in an invalid address expression.
If an undefined symbol is made a default
global
reference
by
the
.ENABL GBL
directive (see Section 6.2.1) during passl,
any attempt to redefine the symbol during
pass 2 will result in an
invalid address
expression.

D-2

ERROR MESSAGES
Meaning

Error Code
A

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
results in an invalid address expression.

(cont.)

For example:
.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.

.SAVE -- The stack is full when
directive is issued.

the

.RESTORE -- The stack is empty
.RESTORE directive is issued.

when

the

two

CATEGORY 5:

.SAVE

ILLEGAL FORWARD REFERENCE.

This type of error results from either
possible conditions:
1.

A
global
assignment
statement
(symbol==expression
or symbol==:expression} contains a forward reference
to
another symbol.

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, MACR0-11 generates
this
error code, ends assembly p~ss 1, and
proceeds with assembly pass 2.
Also caused by
assembler-stack overflow.
In this case MACR0-11
will place a question mark (?} into the line at
the point where the overflow occurred.

D-3

ERROR MESSAGES
Meaning

Error Code
I

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 during macro expansion when longer
real
arguments,
replacing the dummy arguments, cause a
line to exceed 132(10) characters.

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.

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.
An error code
P 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.
bits
in a word,
than 8 significant
.BYTE directive or

D-4

A number generated more than 16
or an expression generated more
bits during
the use of the
trap (EMT or TRAP) instruction.

ERROR MESSAGES
Error Code

Meaning

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
or
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-11
family.
See Section 5.3 for details.

D-5

APPENDIX E
SAMPLE CODING STANDARD

Local user requirements must be met in a coding standard, but
following this model as closely as possible helps you and DIGITAL by
simplifying communication and software maintenance.
Remember that
this is a sample and may not entirely apply to your system.

E.l

LINE FORMAT

Source lines are from one to eighty
following format:

E.2

characters

length

with

the

Label Field - i f present, begins in
should be coded in uppercase only.

(tab
Operation field - begins in column 9
field should be coded in uppercase only.

Operand field - begins in column 17 (tab stop 2).
should be coded in uppercase only.

Comment field - begins in column 33 (tab stop 4).
If the
operand field extends beyond column 33 (tab stop 4) leave a
space and start the comment. This field should be coded in
uppercase and lowercase to increase readability.

column

stop

This

field

1) •

This

This field

COMMENTS

To make the program easier to understand, comments should be used to
explain the logic behind the instructions.
In general this will
consist of a comment per line of code.
However, if a particularly
difficult or obscure section of code is used, precede that section
with a longer explanation.
Comments that are too long for the comment field may be continued on
the following line. Begin the new line with a semicolon, space over
to the column the comment began in and continue writing. All comments
should be written in uppercase and lowercase to increase readability.

E-1

SAMPLE CODING STANDARD
If a lengthy text is needed for an explanation, begin the comment with
a
line containing only the characters
;+ and end it with a line
containing only the characters ;-.
The lines between these delimiters
should each begin with a semicolon and a space.
For example:

;+
The invert routine accepts
a list of random numbers and
applies the Kolmogorov Algorithm
to alphabetize them.
;-

E.3

NAMING STANDARDS

E.3.1
E.3.1.1

Registers
General Purpose Registers - Use the default name:
;REG 0
;REG 1
;REG 2
;REG 3
;REG 4
;REG 5
;Stack pointer (K~u bJ
;Program counter (REG 7)

R0=%0
Rl=%1
R2=%2
R3=%3
R4=%4
R5=%5
SP=%6
PC=%7

NOTE
These register names are defined within
the assembler;
other standard symbols
must be put in a file and linked with
the program.

E.3.1.2 Hardware Registers - Use the hardware
definition.
example, PS (Program Status Register) and SWR (Switch Register).

E.3.1.3 Device Registers - Use the hardware notation.
the control status register for the RK disk is RKCS.

E.3.2

For

example,

Processor Priority

Testing or altering the processor priority is done using the symbols
PR0, PRl, PR2,

••...• PR7

which are equated to their corresponding priority bit pattern.

E-2

SAMPLE CODING STANDARD
E.3.3

Symbols*

The following chart diagrams the syntax of the 5 major types of symbol
names:
symbol

pos-1

pos-2

pos-3

pos-4

pos-5

pos-6

length

non-global
symbol

letter

a-num/ a-num/
null
null

a-num/
null

>=l

$/.

a-num
null

a-num/
null

>=l

***

global
offset

letter

$/.

a-num

a-num/
null

>=3

global bit
pattern

letter

$/.

a-num/

a-num/ a-num/
null
null

>=4

local
symbol

number

global
symbol

***
a-num

***

>=2

**
is an alphanumeric character.

where:

a-num

E.3.3.1

Symbol Examples

Non-Global Symbols
AlB

ZXCJl
INS RT

Global Address Symbols
$JIM
.VECTR

$SEC
Global Absolute Offset Symbols
A$JIM
A$XT
A. ENT

Symbols that are branch targets are also called labels, but we will
always use the term "symbol".

** Number is in the range 0<number<~5535.
*** The use of $ or •

for global names is reserved

software.
E-3

for

DEC-supplied

SAMPLE CODING STANDARD
Global Bit Pattern Symbols
Al$20
B3.6
JI.M
Local Symbols
37$
271$

E.3.3.2 Local Symbols - Target symbols for branches that exist solely
for positional reference will use local symbols of the form
<num>$:
Local symbols are formatted such that the numbers proceed sequentially
down the page and from page to page.

E.3.3.3 Global Symbols - Use of global symbols is restricted,
within
reason,
to those cases where reference to the code occurs external to
the code.
A program never contains a .GLOBL statement without showing cause.

E.3.3.4 Macro Names - In a macro name the last two characters
(last
character possibly being null) have special significance;
the next to
last character is a$, the last character specifies the mode of the
macro.
For example, in the three macro forms in-line, stack,
and p-section,
the in-line
form has no suffix, the stack has an <S> suffix, and the
p-section a <C>.
Thus the Queue I/O macro can be written as any of
QIO$
QIO$S
QIO$C
depending on the form required.

These are not reserved letters.

E.3.3.5 General Symbols - Make frequently used bit patterns such as
carriage return
(CR) and line feed (LF) conventional symbols as they
are needed.

E-4

SAMPLE CODING STANDARD
E.4

PROGRAM MODULES

There are no limits on program size.
However,
since the virtual
memory capacity of a computer is finite keep programs as compact as
possible by:
1.

creating them for a single function

writing them in accordance
guidelines in Appendix F.

with

the

memory

allocation

Code areas are different than data areas. Code is read-only but data
can be read-only or read-write;
read-only data should be segregated
from read-write data.
Both areas, code and data,
should have
explanatory comments.

E.4.1

The Module Preface

Put each program module in a separate file.
For easy reference the
file name should be similar to the name of the module. The file type
is of the form 'NNN' where 'NNN' is the edit or the version number
(see Section E.8). The availability of File Control Services and File
Control Primitives will greatly simplify version number maintenance.

E.4.2

The Module

Below is a list of the information that is
MACR0-11 module (see Section E.4.3).

included

the

example

The information is formatted as follows:
1.

The first six items appear on the same page and do
explicit headings.

A .NLIST statement,
followed by any
.ENABL/.DSABL
or
.NLIST/.LIST options that are relevant to the assembly of
this module, followed by a matching
.LIST statement.
The
.NLIST statement has a comment appended to it specifying the
module edit level.

A .TITLE statement that specifies the name of the module.
If
a module contains more than one routine, .SBTTL statements
are used.

Several .SBTTL statements giving the name, general function,
and version number of the module.
The .SBTTL directive
inserts this information in the table of contents for quick
reference.

An .IDENT statement that specifies the version number of
module (see Section E.8).

A copyright statement, and a disclaimer, followed by a
form
feed.
Note that the copyright, even though a comment, should
be all uppercase.
This insures that the copyright will be
presented correctly, even on a terminal that only has
uppercase.

E-5

not

have

the

3AMPI..E

The name of the facility, that the module is a part of.

The name of the author.

The date of module creation.

10.

A one or two line abstract of the function(s) of the module.

11.

A description of all external references made by the
one per line, in alphabetical order.

module,

12.

A chronological edit trail of
that includes the following:

module

•
•
•
•

modifications

the

Edit number
Editor's identification
Edit date
Description of the modification made
NOTE
Items 6 through 12 should appear on
same page.

13.

Any references to external
.INCLUDE directives.

14.

.MCALL's to any externally defined macros.

15.

A list of the definitions of all equated symbols used in the
module.
These definitions should appear one per line and in
alphabetical order.

16.

All local
order.

17.

All local data.
•
•
•

macro

files,

definitions,

using

the

preferably

.LIBRARY

alphabetical

The comments in this section should include:

Description of each element (type, size, and so forth)
Organization (functional, alphabetical, adjacent, and
forth)
Adjacency requirements (if any)

18.

A form feed, followed by an .SBTTL statement
routine that follows.

19.

A routine header, giving the following information:
•
•
•
•
•
•

and

describing

Routine name
Description
Inputs
Calling sequence
Outputs
Side effects, register usage, and so forth
NOTE
Items 18 and 19 are repeated
routine within the module.

E-6

for

every

the

SAMPLE CODING STANDARD
E.4.3

Module Example
.NL I ST
.ENABL
.LIST
.LIST
• TITLE
• SBTTL
.SBTTL
.SBTTL
.SBTTL
.IDENT

GBL
MEB
MAC INI
MAC INI

- Once-only code for the MACR0-11 assembler
- Once-only code for the MACR0-11 assembler
.IDENT

/Y05.01/

,·**************************************************************************
,. *
*
COPYRIGHT (c)
1982, 1983
,. *
*
BY DIGITAL EQUIPMENT CORPORATION, MAYNARD, MASS.
,. *
*
ALL RIGHTS RESERVED.
,. *
*
,. *
*
,. *

,. *
,. *
,. *
,. *

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 THAT IS NOT SUPPLIED BY DIGITAL.

*
*

*
*
*

*
*
*
*
*
*
*

,·**************************************************************************
<FF>
;++
Facility:

MACR0-11

The PDP-11 macro assembler for RT/RSX/VMS and RSTS/E

Author:

Joe Worrall

Created:

21-Aug-82

Abstract:

MACINI contains code only executed once per task invocation.

Externals

Description

$LIB ID
$POSID
$STABF

File-ID of the system library account (LB: [1,1])
File-ID of the P/OS library account
(LB: [1,5])
Workfile statistics buffer

Edit

Who

Date

Description of modification

001
002
003

Jrw
Jrw
Jrw

25-Aug-82
05-Sep-82
10-Nov-82

Handle P/OS .PARSE module.
Allow recursive FINIT$'s.
Setup statistics buffer.

;--

<FF>
External file references
• LIBRARY
• INCLUDE

/MAC LIB/
/MAC PRE/

;Add MACLIB.MLB to macro library list
;Include MACPRE.MAC in assembly

E-7

External library ".MCALL's" for this module
.MCALL

FINIT$

Equated symbols
••• Equated symbols •••
Local macros
••• Local macros •••
Local data
••• Local data
<FF>
.SBTTL

$!NIT - Handle once only code for MACR0-11 assembler

$!NIT
This routine is a collection of all the code, only executed
once in any one run of the MACR0-11 task.
It's collected
here because:
0
0

It's logical to keep it in one place
It keeps the code out of the root, keeping
the assembler SMALL.

INPUTS:

n/a

CALL:

CALL

$!NIT

OUTPUTS:
Record management, statistics, and FCS buffers
are setup. If the system contains EIS support,
the DIV and MUL routine vectors are setup to
point to the hardware instructions.
EFFECTS:

R0 - RS

Destroyed!

;••• Begin module code •••

E.4.4

Modularity

No other characteristic has more impact on the ultimate engineering
success of a system than does modularity. Adherence to a set of call
and return conventions helps achieve this modularity.

E-8

SAMPLE CODING STANDARD
E.4.4.1

Calling Conventions (Inter-Module/Intra-Module)

Transfer of Control
Macros exist for call and return.
The actual transfer is via a
JSR PC instruction.
For register save routines, a JSR RN,SAVE is
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.
(The preservation of the register state 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, use R0, Rl and
R2 rather than R0, R2, RS.
Saving and restoring occurs in one
place.

E.4.4.2
macro.

Exiting - All subroutine exits occur through a single

RETURN

E.4.4.3 Success/Failure Indication - The C bit is 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.4.4.4 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-9

SA~PL!

E.5
E.5.1

CODING STANDARD

CODE FORMAT
Program Flow

Programs are organized on the listing so that they flow down the page,
even at the cost of an extra branch or jump. All unconditional branch
and jump instructions should be followed by a blank line. This causes
these instructions to stand out in the source code, allowing the code
to be traced more easily.
For example:

PROCESS

AAA

BBB

COMMON

appears on the listing as:

TST
BNE

BBB

CMN

AAA:

BBB:
CMN:

E-10

SAMPLE CODING STANDARD
rather than:
TST
BNE

BBB

CMN

AAA:

CMN:

BBB:

E.5.2

Common Exits

A common exit appears as the last code sequence on the listing.
the flow chart:

II
EXIT

appears on the 1 isting as:
PRl:
BR

EXIT

PR2:

PR3:

PR4:

E-11

Thus

S~MPLE

ST~ND~RD

CODING

EXIT:
and not as:
PRl:

EXIT:

PR2:
BR

EXIT

PR3:

PR4:

E.5.3

Code with Interrupts Inhibited

Code that is executed with interrupts inhibited, is flagged by a three
semicolon (;;;) comment delimiter.
For example:
•• ERTZ:

;Enable by returning
;by system subroutines,
BIS
BIT
BEQ
RTT

#PR7,PS
#PR7,2(SP)
HJ$

HJ$:

E.5.4

;;; inhibit
;;; c
; ; ;

; ;;
;;;

m
m

; ; ;
; ; i
; ; ;
; ; ;

e
n

t
s

Code in System State

RSX-llM executive subroutines and other privileged code that is
executed
in system state is flagged by a two semicolon (;;) comment
delimiter.
For example:
Switch to system state, •••
and exit.

E-12

SAMPLE CODING STANDARD
CALL

$SWSTK,EXIT

; Inhibit context switching
;; Return in system state
; ;
; ;

;; Go back to user state (EXIT)
; User state code

RETURN
EXIT:

E.6

INSTRUCTION USAGE

E.6.1

Forbidden Instructions
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
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, other operations such as ADD and SUB
from PC should be discouraged.
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.
4.

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-13

bAM~L~

E.n.2

CUUlNG STANDARD

Conditional Branches

When using the PDP-11 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

(BCC)

BLOS

(BCS)

A common pitfall is to use a signed branch (for example, BGT) when
comparing two memory addresses. This works until the two addresses
have opposite signs; that is, one of them goes across the 16K
{100000 (8))
bound.
This type of coding error usually results from
re-linking the program at different addresses and/or changing the size
of the program.

E.7

PROGRAM SOURCE FILES

Source creation and maintenance is done in base levels. A base level
is the point at which the program source files have been frozen. From
the freeze point to the next base level, corrections are not made
directly to the base level itself, rather a file of corrections is
accumulated for each file in the base level.
Whenever an updated
source file is desired, the correction file is applied to the base
file.
,
The accumulation of corrections proceeds until a logical breaking
point has occurred
(a milestone or significant implementation point
has been reached). At this time all accumulated corrections are
applied to the previous base level to create a new base level and
correction files are started for the new base level.

E.8

PDP-11 VERSION NUMBER STANDARD

The PDP-11 Version Number Standard applies to all modules, parameter
files, complete programs, and libraries which are written as part of
the PDP-11 Software Development effort. It is used to provide unique
identification of all released, pre-released, and in-house software.
The version number is limited in that only six characters of
identification
are
used.
Future implementations of the Macro
Assembler, linker, and librarian should provide for at least nine
characters, and possibly twelve.
It is expected that this standard
will be improved as the need arises.
Version Identifier Format:
<version> <edit> <patch>
where:

consists of
two
decimal
digits
which
represent the release number of a program.
The version number starts at 00 and is
incremented to reflect the number of major
changes in the program.

E-14

SAMPLE CODING STANDARD
<edit>

consists of
two
decimal
digits
which
represent the number of alterations made to
the source program. The edit number begins
at 01 (is null if there are no edits) and is
incremented with each alteration.

<patch>

is a letter between B and z which represents
the number of alterations made to the binary
form of the program. The patch number begins
at B (is null if there are no patches) and
changes alphabetically with each patch.

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.8.1

Displaying the Version Identifier

The visible output of the version identifier should appear as:
Program
Name <key-letter> <version> . <edit> <patch>,
where the following Key Letters have been identified:

in-house experimental version

field test, pre-release, or in-house release version

released or frozen version

'X' corresponds roughly to individual support, 'Y' to
and 'V' to company support.

group

support,

The dot (.) which separates <version> from <edit> is not used if both
<edit> and <patch> are null. When a version identifier is displayed
as part of program identification, then the format is:
Program
Name <space><key-letter><version> • <edit><patch>
Examples:
PIP V05.00
LINK V08.00
MACRO V05.00

E.8.2

Use of the Version Number in the Program

All sources must contain the version number in an .IDENT directive.
In programs (or libraries) which consist of more than one module, each
module must have a version number. The version number of the program
or library is not necessarily related to the version numbers of the
constituent modules; it is perfectly reasonable, for example, that
the first version of a new FORTRAN library, V00, contain an existing
SIN routine, say V05.01.

E-15

SAMPLE CODING STANDARD
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
(for
example,
the FORTRAN Library)
have an identification module in the
first position.
An identification module exists solely to provide
identification.
For example:
;OTS identification
.TITLE FTNLIB
.IDENT /V02.00/
.END
is an identification module.

E-16

APPENDIX F
ALLOCATING VIRTUAL MEMORY

This appendix is intended for the MACR0-11 user who wants to avoid the
problem of thrashing, by optimizing the allocation of virtual memory.
Users of smaller systems should become thoroughly familiar with the
conventions discussed herein. This appendix discusses the following
topics:
1.

General hints and space-saving guidelines

Macro definitions and expansions

Operational techniques.

The user is assumed to have pursued a policy of modular programming,
as advised in Appendix E. Modular programming results in bodies of
code that are small, distinct and highly functional. Using such code,
which presents many advantages, one can usually avoid the problem of
insufficient dynamic memory during assembly.

F.l

GENERAL HINTS AND SPACE-SAVING GUIDELINES

Work-file memory is shared by a number of MACRO-ll's tables, each of
which is allocated space on demand (64K words of dynamically pageable
storage are available to the assembler).
The tables and their
corresponding entry sizes are as follows:
1.

User-defined symbols - five words.

Local symbols - three words.

Program sections - six words.

Macro names - five words.

Macro text - nine words.

Source files - six words.

In addition, several scratch pad tables are used during
process, as follows:
1.

Expression analysis - five words.

Object code generation - five words.

F-1

the

assembly

ALLOCATING VIRTUAL MEMORY
3.

Macro argument processing -

three words.

.MCALL argument processing -

five words.

The above information can serve as a guide
for estimating dynamic
storage
requirements
and
for determining ways to reduce such
requirements.
For example, the use of local symbols whenever possible
is highly
encouraged,
since their
internal representation requires 25 percent
less dynamic storage than that
required
for
regular user-defined
symbols.
The usage of local symbols can often be maximized by
extending the scope of local symbol blocks through the
.ENABL
LSB/.DSABL LSB MACR0-11 directives see Sections 3.5 and 6.2.1).
Since MACR0-11 does not support a purge function,
once a symbol
is
defined,
it permanently occupies
its dynamic memory allocation.
Numerous instances occur during
conditional assemblies and
repeat
loops when a temporarily assigned symbol is used as a count or offset
indicator.
If possible, the symbols so used should be re-used.
In keeping with the same principle, special treatment should be given
to the definition of commonly used symbols.
Instead of simply
appending a prefix file which defines all possibly used symbols for
each assembly,
users are encouraged to group symbols into logical
classes.
Each class can then become a shortened prefix
file or a
macro in a library (see Section F.2 below).
In either case, selective
definition of symbolic assignments is achieved,
resulting
in fewer
defined (but unreferenced) symbols.
An example of this idea is seen in the definition of IAS/RSX-llM
standard symbols.
The RSX system macro library, for example, supplies
several macros used to define distinct classes of symbols.
These
groupings and associated macro names are as follows:
DRERR$ - Directive return status codes
FILIO$ - File-related I/O function codes
IOERR$ - I/O return status codes
SPCIO$ - Special I/O function codes

F.2

MACRO DEFINITIONS AND EXPANSIONS

Dynamic storage is used most heavily for the storage of macro text.
Upon macro definition or the
issuance of an .MCALL directive, the
entire macro body is stored, including all comments appearing
in the
macro definition.
For this reason, comments should not be included as
part of the macro
text.
A librarian function
switch
(/SZ)
is
available to compress macro source text by removing all trailing
blanks and tabs, blank lines, and comments.
The system macro library
(RSXMAC.SML)
has
already been compressed.
User-supplied macro
libraries (.MLB) and macro definition prefix
files
should also be
compressed.
For additional
information regarding these two utility
tasks, consult the applicable RSX-llM or RSX-llM-PLUS Utilities Manual
(see Section 0.3 in the Preface).

F-2

ALLOCATING VIRTUAL MEMORY

It often seems practical to include a file of commonly used macro
definitions in each assembly. This practice, however, may produce the
undesirable allocation of valuable dynamic storage for unnecessary
macros.
This waste of memory can be avoided by making the file of
macro definitions a user-supplied macro library file (see Table 8-1).
This means that the names of desired macros must be listed as
arguments in the .MCALL directive (see Section 7.8), or the automatic
MACRO call, .ENABL MCL, must be enabled (see Section 6.2.1).
Certain types of macros can be redefined to null after they have been
invoked.
This practice not only frees storage space, it also
eliminates the overhead and the dynamic memory wasted by calling a
useless macro.
The practice of redefining macros to null applies
mainly to those that define symbolic assignments, as shown in the
example below.
The redefinition process may be accomplished as
follows:
SYMl
SYM2

.MACRO
VALl
VAL2

OFFl
OFF2
OFF3

SYMBOL
OFFl+SIZl
OFF2+SIZ2

OFFN

OFFM+SIZM

DEFIN
;Define symbolic assignments.

;Define symbolic offsets.

• MACRO
.ENDM

DEFIN

.ENDM

DEF IN

;Macro null redefinition •

Macros exhibiting this redefinition property should be defined (or
read via the .MCALL directive)
and invoked before all other macro
definition and/or .MCALL processing, a practice that ensures more
efficient use of dynamic memory.

F.3

OPERATIONAL TECHNIQUES

When, despite adhering to the guidelines discussed above, performance
still falls below expectations, several additional measures may be
taken to increase dynamic memory.

F-3

ALLOCATING VIRTUAL MEMORY
The first measure involves shifting the burden of symbol definition
from MACR0-11 to the linker or task builder. In most cases, the
definition of system I/O and File Control Services (FCS) symbols (and
user-defined symbols of the same nature) is not necessary during the
assembly process, since such symbols are defaulted
to
global
references (Appendix D.l, category 4 of error code A). The linker or
task builder attempts to resolve all
global
references
from
user-specified default libraries and/or the system object library
(SYSLIB). Furthermore, by applying the selective search option for
object modules consisting only of global symbol definitions, the
actual additional burden to the linker is minimal.
(either object or
The second way is to produce only one output file
listing), as opposed to two.
The additional memory required to
support the second output file is allocated from available dynamic
memory at the start of each assembly.

F-4

APPENDIX G
WRITING POSITION-INDEPENDENT CODE

G.l

INTRODUCTION TO POSITION-INDEPENDENT CODE

The output of a MACR0-11 assembly is a relocatable ob ect module. The
Task Builder or Linker binds one or more modules toge her to create an
executable task image.
Once created, if the program s to run it must
be loaded at the virtual address specified at link time.
This is
because the Task Builder or Linker has to modify some instructions to
reflect the memory locations in which the program is to run.
Such a
body of code is considered position-dependent
(dependent on the
virtual addresses to which it is bound).
All PDP-11 processors offer addressing modes ~hat make it possible to
write code that does not depend on the virtual addresses to which it
is bound. Such code is termed position-independent and to run can be
loaded at any virtual address.
Position-independent code can improve
system efficiency, both in use of virtual address space and in
conservation of physical memory.
In multiprogramming systems like IAS, RSX-llM and RSX-llM-PLUS, it is
important that many tasks be able to share a single physical copy of
common code, for example, a library routine.
To make the optimum use
of
a
task's
virtual
address
space,
shared code should be
position-independent.
Position-dependent code can also be shared, but
it must appear in the 'same virtual locations in every task using it.
This restricts the placement of such code by the Task Builder or
Linker and can result in the loss of virtual addressing space.
The construction of position-independent code is closely linked to the
proper usage of PDP-11 addressing modes.
The remainder of this
Appendix assumes you are familiar with the addressing modes described
in Chapter 5.
All addressing modes
involving
only
register
position-independent. These modes are as follows:

R
(R)

(R)+
@(R)+
-(R)
@-(R)

references

are

register mode
register deferred mode
autoincrement mode
autoincrement deferred mode
autodecrement mode
autodecrement deferred mode

When
using
these
addressing
modes,
you
are
guaranteed
position-independence, provided the contents of the registers have
been supplied such that they are not dependent upon a particular
virtual memory location.

G-1

WRITING POSITION-INDEPENDENT CODE
The relative addressing modes are position-independent
when
a
relocatable address is referenced from a relocatable instruction.
These modes are as follows:
relative mode
relative deferred mode

Relative modes are not position-independent when an absolute address
(that is a non-relocatable address) is referenced from a relocatable
instruction. In this case, absolute addressing (@#A) may be used to
make the reference position-independent.
Index modes can be either position-independent or position-dependent,
according to their use in the program. These modes are as follows:
X(R)
@X(R)

index mode
index deferred mode

If the base, X, is an absolute value (for example, a control
offset), the reference is position-independent. For example:
MOV

2 (SP) ,R0

;Position-independent

MOV

N(SP) ,R0

;Position-independent

block

N=4
If,

however,

position-dependent.
CLR

rP.locntable

address~

the

reference

For example:

ADDR (Rl)

;Position-dependent

Immediate mode can be either position-independent or not, according to
its usage. Immediate mode references are formatted as follows:
#N

immediate mode

When an absolute expression defines the value of N, the code is
position-independent.
When a relocatable expression defines N, the
code is position-dependent. That is, immediate mode references are
position-independent only when N is an absolute value.
Absolute mode addressing is position-independent only in those cases
where an absolute virtual location is being referenced. Absolute mode
addressing references are formatted as follows:
@#A

absolute mode

An example of a position-independent absolute reference is a reference
to the directive status word {$DSW) from a relocatable instruction.
For example:
MOV

G.2

@#$DSW,R0

;Retrieve directive status

EXAMPLES

The RSX-llM library routine, PWRUP, is a FORTRAN callable subroutine
that establishes or removes a user power failure Asynchronous System
Trap (AST) entry point address. Imbedded within the routine is the
AST entry point that saves all registers, effects a call to the
user-specified entry point, restores all registers on return, and
executes an AST exit directive. The following examples are excerpts
G-2

WRITING POSITION-INDEPENDENT CODE
from this routine.
The first example, Figure G-1 has been modified to
illustrate position-dependent references.
The second example, Figure
G-2, is the position-independent version.
;+

; Position dependent code exaaPle
;-

PWRUP:: CLR

-<SP>

Perfora further initialization •••

10s:

MOV

$0TSVrR4

HOV
BNE
CLR
BR
MOV
MOV

CSP>+ rR2
10$
-<SP>
20$
R2rF.PFCR4>
tBAr-CSP>

;Point R4 at obJect tiae svstea save area
; the ebove reference to $0TSV i§ ~osition
; dePer1dent
;Retrieve AST entrv Point address
;Branch if one was sPecified
;rf noner specifv no Power fail routine
;BvPass AST setuP
;set the AST entrv Point
;Push our AST service address
; the above reference to BA is Position; dePendent

20$:

; Continue Processins,.,

;+
; AST service routine
;-

MOV

ROr-CSP>

;Preserve RO

; Rest of routine follows.,.

Figure G-1

Example of Position-Dependent Code

; Position indePendent code exaaPle
;PWRUP:: CLR

-CSP>

;Assuae success

Perfora necessarv initialization •••

10$i

MOV

@HOTSVrR4

HOV
BNE
CLR
BR
HOV
HOV
ADD

CSP>+,R2
10$
-<SP>
20$
R21F.PF<R4>
pc,-csP>
tBA-.,<SP>

;Point R4 at obJect tiae svstea save area
; the above reference to SOTSV is Position; independent
;Retrieve AST entry Point address
;Branch if one was specified
;If none, sPecifY no Power fail routine
;BvPass AST setuP
;set the AST entry Point
;Push our PC to relocate our AST service addr
;Relocate our AST service address now
; the above reference to BA is Position; independent; this costs one word to relocate

20s:

; AST service routine
;BA:

MOV

ROr-CSP>

;Preserve RO

; Rest of routine follows •••

Figure G-2

Example of Position-Independent Code

G-3

WRITING POSITION-INDEPENDENT CODE
The position-dependent version of the subroutine contains a relative
reference to an absolute symbol ($0TSV) and a literal reference to a
relocatable symbol (BA).
Both references are bound by the Task
Builder to fixed memory locations. Therefore, the routine will not
execute properly as part of a resident library if its location in
virtual memory is not the same as the location specified at link time.
In the position-independent version, the reference to $0TSV has been
changed to an absolute reference. In addition, the necessary code has
been added to compute the virtual location of BA based upon the value
of the program counter. In this case, the value is obtained by adding
the value of the program counter to the fixed displacement between the
current location and the specified symbol. Thus, execution of the
modified routine is not affected by its location in the image's
virtual address space.
The MACR0-11 Assembler provides a way of checking whether the code is
position-independent.
In an assembly listing, MACR0-11 inserts a '
character following the contents of any word which requires the Task
Builder or Linker to perform a relocation operation and, therefore,
may not be position independent code.
The cases which cause an
apostrophe to be inserted in the assembly listing are as follows:
1.

Absolute mode references when the reference is relocatable.
References are not flagged when they are absolute. For
example:
MOV

ADDR(Rl) ,R5
@ADDR(Rl) ,R5

offset

;Non-pie if ADDR is relocatable.
;Non-pie if ADDR is relocatable.

Relative and relative deferred mode references when the
address specified is relocatable with respect to another
program section. For example:
MOV
MOV

;Pie only if ADDR is absolute.

Index and index deferred mode references when the
relocatable. For example:
MOV
MOV

@#ADDR,Rl

ADDRl,Rl
@ADDRl,Rl

;Non-pie when ADDRl is absolute.

Immediate mode references to relocatable addresses.
MOV

#ADDR,Rl

;Non-pie when ADDR is relocatable.

In one case, MACR0-11 does not flag a potential position-dependent
reference.
This occurs where a relative reference is made to an
absolute virtual location from a relocatable instruction (see the MOV
$OTSV,R4 instruction in Figure G-1).
References requiring more than simple relocation at link time are
indicated in the assembly listing.
Simple global references are
flagged with the letter G. Statements which contain multiple global
references or require complex relocation, are flagged with the letter
C (see Section 3.9 and Chapter 4).
It is difficult to positively
state whether or not a C-flagged statement is position-independente
However, in general, position dependence can be decided by applying
the guidelines discussed earlier in this Appendix to the resulting
address value produced at link time.

G-4

APPENDIX

SAMPLE ASSEMBLY AND CROSS REFERENCE LISTING

>R50UNP

MACRO V05.00 Saturday 08-Jan-83 11:47

TABLE OF CONTENTS
2-

RSOUNP

RADSO unpack routine

MACRO V05.00 Saturday 08-Jan-83 11:47

.TITLE
.IDENT

2
3
4
6
7

11
12
13

RSOUNP
/02/

9
10

Pase

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,

14
15
16
l7
18
19
20
21

THE INFORMATION IN THIS SOFTWARE IS SUBJECT TO CHANGE WITHOUT NOTICE
AND SHOULD NOT BE CONSTRUED AS A COMMITMENT BY DIGITAL EQUIPMENT
CORF'ORATION,
DIGITAL ASSUMES NO RESPONSIBILITY FOR THE USE OR RELIABILITY
SOFTWARE ON EQUIPMENT WHICH IS NOT SUPPLIED BY DIGITAL.
UF'DATE

HISTORY:

24
25

D.N. CUTLER

10-FEB-73

H-1

ITS

SAMPLE ASSEMBLY AND CROSS REFERENCE LISTING

R50UNP MACRO V05.00 Saturday 08-Jan-83 11:47
RAD50 UNPACK ROUTINE
.SBTTL

Pase 2

RAD50 unpack routine

2
;+

4
5
6
7
8

R50UNP
UnPack a 6 char RAD50 SY•bol to ASCII
Enter with R2 -> Output ASCII strins
SYHBOLr SYHBOL+2 = RAD50 sY•bol to unpack

10
11
12
13
14 000000
15
16 000000
17 000002
18 000006
19 000010
20 000014
21 000020

22 000024

:.d 0000:50
24 000032
25 000036
26
27
28
29
30
31
32
33
34
35

000042
000044
000046

Return with R2 -> Past output strins
ROr Rlr R3 Destroyed

;.PSECT
010446 R50UNP::HOV
012704
HOV
OOOOOOG
012401
u:
HOV
012703
HOV
003100
004767
CALL
000030
012703
HOV
000050
004767
CALL
000020

46 000064
47 000066
48 000070

;save R4
;Point at RAD50 SY•bol buffer

CR4>trR1
t50*50rR3

;Get next RAD50 word
;set divisor for hi Sh character

10$

;UnPacl'-. and store the character

t50rR3

;Now set divisor for 111iddle character

10$

;UnPack and store the character

OiO:i.00

MC'.'

P!~RO

CALL

11$

;Copy re•aininS character
;Translate and store i t

CMP

R4rtSYHBOL+4

He st if last word done

BNE
HOV
RETURN

;Branch if no
;Restore R4
;Return to caller

005000
071003

10$!

RAD~O

CLR
DIV

CSf'>trR4

word and convert char to ASCII
RO
R3rRO

Translate RAD50 character code to ASCII
0 =
SP ace
1-32
A-Z
$
33
34 =
35 =
•Jn used code
36-47 = 0-9

37
38
39
40
41
42
43 000054
44 000056
45 000060

R4r-CSP>
tSYHBOLrR4

004767
000016
020427
0000048
001361
012604
000207
Divid~

000050
000052

PUREirI

005700
001412
020027
000033
002405
001402
062700
000011

1u:

TST
BEG
CHP

RO
23$
ROrt33

Hest if sF-ace
;Branch if so
Hest if •iddle

BLT
BEQ
ADD

22$
21$
t22-11rRO

;Branch if alF-habetic
;Branch if dollar sisn
;Dot or disits 0-9

R50UNP MACRO V05.00 Saturday 08-Jan-83 11:47
RAD50 UNPACK ROUT WE
49 000074
50 000100
51 000104

52 000110
53 000112
54
55

062700
177711
062700
000040
062700
000040
110022
000207
000001

Pase 2-1

2u:

ADD

t11-100rRO

;Dollar

22$:

ADI•

tl00-40' RO

;AlPhabetic

23$:

ADD

t40rRO

;Space

HOVB
RETURN

RO, CR2 >+

;store ASCII char in buffer

.END

H-2

SAMPLE ASSEMBLY AND CROSS REFERENCE LISTING

Pase 2-2

RSOUNP MACRO V05.00 Saturday 08-Jan-83
SYMbol Table
RSOUNP

OOOOOORG

ABS. 000000
000000
PURE I
000114
Errors detected:

002 SYMBOL= ****** GX
000
001
002
0

*** AsseMbler statistics
Work File reads: 0
Work File writes: 0
Size oF work File: 7936 Words
< 1 Pases)
Size oF core Pool:
16158 Words (72 Pases)
Qperatins sYsteM:
RT-11
ElaPsed tiMe: 00:00:04.34
DK:RSOUNP,DK:R50UNP/C=DK:SUM

RSOUNP MACRO VOS.00 Saturday 08-Jan-83 11:47
Cross reference table <CREF V05.00>
RS OU NP
SYMBOL

2-16#
2-17

2-25

RSOUNP MACRO V05.00 Saturday 08-Jan-83 11 :47
Cross reference table <CREF V05.00l
RO

Rl
R2
R3
R4
SP

2-23*
2-50*
2-18*
2-52*
2-19*
2-16
2-16*

2-32*
Z-51*
2-23

2-33*
2-52

2-21*
2-17*
2-27

2-33
2-18

Z-43

2-45

2-25

2-27*

RSOUNP MACRO VOS.00 Saturday 08-Jan-83 11:47
Cross reference table CCREF VOS.OOl

• ABS.
PURE I

Pase S-1

Pase R-1

2-48*

Pase C-1

0-0
0-0
2-14

H-3

2-49*

APPENDIX I
OBSOLETE M..~CR0-11 DIRECTIVES, SYNTAX, AND COMMAND LINE OPTIONS

I.l

OBSOLETE DIRECTIVES AND SYNTAX

The following directives and syntax, although supported in this
release of the assembler, will NOT be supported in future assembler
releases. The following table shows the old directives and syntax,
and the new syntax to use. All MACR0-11 code that contains the old
directives and syntax should be updated to use the new syntax.
Table I-1
Old and New Directives and Syntax
Syntax no longer supported

New syntax to use

• EOT
.IFZ xxx or .IFEQ xxx
.IF Z,xxx
.IFNZ xxx or .IFNE xxx
.IF NZ,xxx
.IFL xxx or .IFLT xxx
.IF L,xxx
.IFG xxx or .IFGT xxx
.IF G,xxx
.IFLE xxx
.IFDF xxx
.IFNDF xxx

None
.IF EQ,xxx
.IF EQ,xxx
.IF NE,xxx
.IF NE,xxx
.IF LT,xxx
.IF LT,xxx
.IF GT,xxx
.IF GT,xxx
.IF LE,xxx
.IF DF,xxx
.IF NDF,xxx

I.2

OBSOLETE COMMAND LINE OPTION

The MACR0-11 command line option /P[ASS] :n is no longer supported by
DIGITAL. This switch was originally created to speed up assemblies in
some cases by only scanning a given file with one pass of the
assembler.
It has been found that the /P[ASS] :n switch has many side effects, and
has caused more problems than can be documented reasonably.
Although the syntax of the /P[ASS] :n switch is still allowed to appear
on a MACR0-11 command line, no SPRs will be accepted relating to the
switch. All documentation for the /P[ASS] :n switch has been removed.
Any assembly command files containing the /P[ASS] :n switch
updated by removing this switch.

I-1

should

APPENDIX J
RELEASE NOTES

This appendix explains the changes that have been made to MACR0-11
since the last version release. The new features mentioned are fully
documented in chapters one through nine of this manual.

J.l

CHANGES -- ALL VERSIONS OF MACR0-11
1.

The opcode, CALLR addr (Call-Return), has been added to the
permanent symbol table (PST). This opcode is equivalent to
the JMP addr opcode. The CALLR addr opcode was added to
complement the CALL addr opcode -- which is equivalent to the
JSR PC,addr opcode.

The previous version of MACR0-11 used a range of 64$ to 127$
for automatic local symbol generation. MACR0-11 now uses a
range of 30000$ to 65535$ when generating local symbols.

Most
assembler
generated
listing
text
is
now
in
upper/lowercase.
This change was made to increase the
readability of MACR0-11 code.
Lines of code that include the
.SBTTL
or
the .TITLE directive are not converted to
uppercase.

Lines of code that include the .SBTTL directive are listed in
the table of contents of an assembly listing, even if a
.NLIST statement is in effect at the time the .SBTTL lines
are encountered.
You may specify the .NLIST directive with
the TOC argument to prevent the table of contents from being
printed.

The symbol table is printed at the end of an assembly, even
if the .NLIST directive is in effect. You may specify the
.NLIST directive with the SYM argument to prevent the symbol
table from being printed.

All page headers include the day of the week.

J-1

The assembler statistics information that appears at the end
of the assembly listing file has been updated to include the
following additional information:
•

Total number of virtual work file reads

•

Total number of virtual work file writes

•

Maximum amount of virtual memory used (in words and pages)

•

Size of physical memory freespace (in words and pages)

•

Operating system and environment
running under

•

Total elapsed assembly time

•

MACR0-11 command line

that

the

assembler

The PSECT synopsis that is printed in the listing file, after
the symbol table, includes the psect attributes.

The maximum number of relocatable terms in a
complex
expression has been changed.
The maximum size of an .OBJ
record that MACR0-11 can produce was increased from 42. bytes
to 128. bytes.
Do not compare .OBJ files that have been created by different
versions
of MACR0-11 when verifying whether your code
generation is correct. Changes that have been made for
this
version of MACR0-11
(mentioned above)
will
invalidate a
direct comparison of assembler
.OBJ output.
Verify code
generation by linking or taskbuilding the .OBJ files involved
and then comparing the .SAV or the .TSK image files.
NOTE
Because the .OBJ files produced by this new version
of MACR0-11 are different, users of the PAT (object
file patch utility) are warned that checksums must be
recomputed on any object patches assembled with this
new version of MACR0-11.

10.

The default for the LC argument has been changed from
LC to • ENABL LC.

11.

The following .ENABL/.DSABL options have been added:
1.
2.

12.

.ENABL LCM/.DSABL LCM
.ENABL MCL/.DSABL MCL

The following directives have been added to MACR0-11.
new directives are documented in this manual.
1.
2.
3~
4.
5.

n.
7.

.DSABL

• CROSS
• INCLUDE
• LIBRARY
.MDELETE
• NOCROSS
. REM
• WEAK
J-2

These

RELEASE NOTES
J.2

CHANGES -- MACR0-11/RSX VERSION ONLY
1.

The cross-reference options SEC and ERR have been added.
NOTE
The RSX-11 CREF program (CRF)
has been updated to
include
support
for
these
two
new
macro
cross-reference options.
Only the new RSX-11 CRF
version
(V2)
distributed with RSX-llM V4.l and
RSX-llM-PLUS V2.l should be used with this version of
MACR0-11.

J.3

The default for the command line option /[-]SP has been
modified from /SP to /-SP.
The new default may be modified
by the system manager using the TKB GBLPAT option described
in the MACR0-11/RSX task build command file.

CHANGES -- MACR0-11/RT-ll VERSION ONLY
1.

The message:
Errors detected:
has been removed.
terminal only if
being assembled.

MACR0-11 prints this message on the
errors have been detected in the module

If the first character in a MACRO-ll/RT-11 command line is a
semicolon
(;),
the line is treated as a comment and is
ignored.
This change was made to maintain compatibility with
the RSX-11 version of MACR0-11.

RSX-11 sty~e command line switches may be used in addition to
the one-character switches:

/M
/E
/D
/P
/L
/N
4.

may
may
may
may
may
may

be
be
be
be
be
be

represented
represented
represented
represented
represented
represented

as
as
as
as
as
as

/M [LIB]
/E [NABL]
/D [SABL]
/P[ASS]
/L[IST]
/N [LIST]

The default file extension for macro libraries has been
changed to
.MLB, to conform with RSX-11.
The RT-11 VS LIBR
program defaults its macro library output to the .MLB
extension also.

J-3

HELEASE NOTES
5.

Prior to this release of MACR0-11, if you specified more than
one .MLB file on a command line, and each file had a
definition of the same macro, the first macro library
specified would be used for the macro definition if called in
the source program. This has been modified to work like the
RSX-11 macro assembler. The RT-11 macro assembler now scans
.MLB files from the last file specified (either in the
MACR0-11 command line or by using the .LIBRARY directive) to
the first file specified.
The assembler then scans the
system default macro library, SY:SYSMAC.SML.

The default for the GBL argument has been changed from .DSABL
GBL to .ENABL GBL.

J-4

INDEX
A error, 3-10, 3-13, 5-10, 6-15,
6-25, 6-26, 6-28, 6-29, 6-32,
6-33, 6-38, 6-40, 6-42. 6-44,
6-47, 6-56, 7-2, 7-12 to 7-14,
7-16' 7-1 7' 7-20
Absolute address, D-2
Absolute binary output, 6-19
Absolute expression, 3-17
Absolute mode, 5-1, 5-7,
B-2, G-2, G-4
Absolute module, 6-42
Absolute program section, 6-42 to
6-45, B-4. See also .ASECT
directive
ADD instruction, E-12, G-3: H-2
Addition operator, 3-2, 3-5, B-1
Address boundaries, 6-39
Addressing modes, 5-1
Apostrophe, G-4
ASCII
character set, A-1
conversion characters, 6-23 to
6-26
.ASCII directive, 6-1, 6-21, 6-26
to 6-28, 6-36, B-3
.ASCIZ directive, 6-1, 6-28,
6-36, B-4
.ASECT directive, 3-11, 3-13,
3-14, 6-2, 6-44 to 6-47, B-4
Assembler directives. See Permanent
symbol table
version number, 6-4
Assembly
See A error
error.
listing symbols, 4-1
pass 1, 1-1, 1-2, 6-12, 6-15,
6-16, 6-49, 8-10, 8-12,
D-3
pass 2, 1-2, 6-12, 6-21, 7-15,
D-3
Assignment operator. See Direct
assignment operator
Assignment statement. See Direct
assignment statement
Autodecrement deferred mode, 5-1,
5-5, B-2, G-1
Autodecrement indicator, 3-2
Autodecrement mode, 5-1, 5-4,
B-1, B-2, G-1
Autoincrement deferred mode, 5-1,
5-4, B-2, G-1

Autoincrement indicator, 3-2
Autoincrement mode, 5-1, 5-3,
B-2, G-1
Base level, E-14
BCC instruction, E-13
BCS instruction, E-14
BEQ instruction, H-2
BGE instruction, E-13
BGT instruction, E-14
BHI instruction, E-14
BHIS instruction, E-13
BIC instruction, E-13
Binary operator, 3-4, 3-5, 3-16
Blank line, 2-1
BLE instruction, E-14
.BLKB directive, 3-14, 6-2, 6-36
to 6-38, B-4, D-3
.BLKW directive, 3-14, 6-2, 6-36,
6-38, 6-48, B-4, D-3
BLO instruction, E-13
BLOS instruction, E-13
BLT instruction, E-13, H-2
BNE instruction, E-10, G-3, H-2
BR instruction, E-10, E-11, G-3
Branch instruction
addressing, 5-9, D-2
use of, E-13
.BYTE directive, 6-2, 6-23, 6-36,
B-4, D-4
C bit, E-9
CALL instruction, H-2
Calling convention, E-8
Character set
ASCII, A-1 to A-3
legal, 3-1 to 3-3
Radix-50, A-5, A-6
CLR instruction, G-3, G-3, H-2
CMP instruction, E-13, H-2
Coding standard, E-1
Comment, E-1, E-5
delimiter, 3-2, B-1, E-12
field, 2-1, 2-4, 2-5, E-1
Commercial instruction set, C-3
Common exit, E-11
Complex relocatable expression,
3-18
Complex relocation, 4-1, G-4

Index-!

Concatenation indicator, 3-3,
B-1, B-3
Conditional assembly, 6-51 to
6-56, 7-8, 7-16, D-4
immediate, 6-56
Conditional assembly block, 7-3,
B-4, B-5
Conditional assembly directive,
6-49
Copyright statement, E-5
.CROSS directive, 6-2, 6-22,
B-4, C-5
Cross-reference listing, 3-12,
6-19, 8-8, 8-9, 8-14, 8-16 to
8-18, 9-2, 9-3, 9-5 to 9-7
.CSECT directive, 3-11, 3-13,
6-2, 6-44 to 6-47, 9-6, B-4
Current location counter, 2-2,
3-2, 3-12 to 3-14, 3-17, 5-8,
6-11, 6-36 to 6-38, 6-43 to
6-44, B-5, B-7, D-2, D-3
D error, 2-3
Data
sharing, 6-45
storage, 6-2
storage directives, 6-23
Default radix, 3-14
Default register definitions,
3-10, 6-21
Deferred addressing indicator,
3-2, B-1
Delimiting characters, 3-3, 6-17,
6-29, B-3 to B-5, B-8
Device register, E-2
Direct assignment
operator, 3-1, 3-2, 3-9, B-1
statement, 3-6 to 3-9, 3-13,
6-37
Directives. See Permanent
symbol table
DIV instruction, H-2
Division operator, 3-2, 3-5, B-1
Double ASCII character indicator,
3-2, B-1
.DSABL directive, 6-2, 6-19 to
6-21, 8-6, 8-8, 9-4, B-4,
D-1
Dummy argument, 7-2, 7-11, 7-17
E error, 6-40
EMT instruction, 5-9, D-4
.ENABL directive, 6-2, 6-19 to
6-21, 8-6, 8-8, 9-4, B-4,
D-1, D-2, D-4, F-2
.END directive, 6-2, 6-40, B-4,
D-3, H-2
.ENDC directive, 6-2, 6-12, 6-53
to 6-56, 6-59, 7-3, B-4
.ENDM directive, 6-13, 6-21, 7-2,
7-3, 7-6 to 7-8, 7-10, 7-11,
7-17 to 7-19, B-4, s~8, F-3

.ENDR directive, 7-19, 7-20, B-5,
B-8
Entry point symbol, 6-52
.ERROR directive, 7-16, B-5, D-4
Error messages, D-1 to D-5
.EVEN directive, 6-2, 6-29, 6-38,
B-5
Expression, evaluation of, 3-16
Expression indicator, immediate,
3-2, B-1
External expression, 3-17
External symbol, 6-52. See also
Global symbol
Field terminator, 3-2, B-1
FILES-11, 6-19
Floating-point directives, B-5.
See also .FLT2 directive
Floating-point indicator, B-3
Floating-point processor, 3-14,
6-34, 6-35, C-4
Floating-point rounding, 6-19,
6-32
Floating-point truncation, 6-19,
6-35
/
.FLT2 directive, 6_;2, 6-35, B-5
.FLT4 directive, 6-2, 6-35, B-5
FLX, 6-19
Forbidden instructions, E-13
Format control, 2-5
Formatted binarv, 6-19
FORTRAN, 6-47, E-15, G-2
Forward reference, 3-8, 3-9,
3 -10 , 3 -1 3 , D- 4
i 11ega1 , D-3
Function control switches. See
Switches, function control
Function directive, 6-18
Global expression evaluation,
3-17
Global label, 6-51
Global reference, 6-21, 6-51,
F-4, G-4
Global symbol, 1-2, 3-7, B-5,
D-2, D-3, E-4
Global symbol definition, 2-2,
3-1, 3-2, 3-8, 6-51. See
also .GLOBL directive
Global symbol directory, 1-2
.GLOBL directive, 3-7, 6-2, 6-51,
B-5, E-4
Hardware register, E-2
I error, 6-28; 6-30
IAS, 6-48, 7-21, 8-14 to 8-17,
8-19 to 8-22, G-1
.IDENT directive, 6-2, 6-16, B-5,
D-2, E-5, E-7, E-15, H-1

Index-2

INDEX

.IF directive, 6-2, 6-12, 6-53 to
6-59, 7-3, 7-8, B-5, D-1, D-2
.IFF directive, 6-2, 6-56 to
6-58, B-5
• IFT directive, 6-2, 6-56 to
6-58, B-5
.IFTF directive, 6-2, 6-56, 6-57,
B-6
.IIF directive, 6-2, 6-59, B-6,
D-1, D-2
Illegal characters, 3-3, D-2, D-3
Illegal forward reference, D-3
Immediate conditional assembly,
6-59
Immediate expression indicator,
3-2, B-1
Immediate mode, 5-1, 5-6, B-2,
G-2, G-4
Implicit eWORD directive, 2-1,
2-4, 6-25
.INCLUDE directive, 6-2, 6-61,
9-8, B-6, C-6
Indefinite repeat block. See
Repeat block, indefinite
Index deferred mode, 5-1, 5-5,
B-2, G-2, G-4
Index mode, 5-1, 5-5, 5-7, B-2,
G-2, G-4
Initial argument indicator, 3-2,
B-1
Initial expression indicator, 3-2
Initial register indicator, 3-2,
B-1
Instruction set
commercial, C-3
PDP-11, C-1
Interrupts, E-12
.IRP directive, 7-2, 7-17 to
7-19, B-6, D-2
.IRPC directive, 7:2, 7-17 to
7-19, B-6, D-2
Item terminator, 3-2, B-1
JMP instruction, 5-3, E-13
JSR in~truction, 5-3, E-9
L error, 2-1
Label
field, 2-1 to 2-3, E-1
multiple definition, 2-3
terminator, 3-1, B-1
.LIBRARY directive, 6-2, 6-60,
9-9, B-6, C-6
.LIMIT directive, 6-3, 6-39, B-6
Line format, E-1
Line printer listing format, 6-5,
6-6, 6-12. See also Listing
control
Linker, 1-2, 2-2, 6-17, 6-43,
6-47, 6-51, F-4, G-1, G-4
Linking, 4-1, 6-40

.LIST directive, 6-3, 6-9 to
6-14, 6-21, 8-6, 8-11,
8-13, 9-4, B-6, D-1
Listing control, 6-4 to 6-14 •
See also .LIST directive,
.NLIST directive
Listing control switches. See
Switches, listing control
Listing level count, 6-9, 6-10,
6-12, B-6, B-7
Local symbol, 3-11, 3-12, 7-8,
7-9, D-4, E-4, F-2
Local symbol block, 3-11, 3-12,
6-20 I D-4, F-2
Location counter. See Current
location counter
Location counter control, 6-34 to
6-36
Logical AND operator, 3-2, 3-5,
6-55, B-1
Loqical inclusive OR operator,
- 3-2, 3-5, 6-55
Logical OR operator, B-1
M error, 2-3, 3-1, 3-2, 3-8
Macro
argument, 7-7, 7-14, 7-15, B-3
argument concatenation, 7-11
attribute directive, 7-12
definition, 6-33, 7-1 to 7-13,
7-15, 7-17, 7-18, 7-20, B-4,
B-6, B-7, E-6, F-2
directive, 7-1, 7-2, 7-4. See
also .MACRO directive
expansion, 7-1, 7-3, 7-5 to
7-7, 7-9, 7-11, 7-17, B-7,
D-4, F-2
expansion listing, 6-9, 6-12
keyword argument, 7-4, 7-10
keyword indicator, 3-1
name, 7-1, 7-2, 7-4, D-4, E-4
nesting, 7-2, 7-3, 7-6, 7-17
numeric argument, 7-7
redefinition, F-3
symbol, 3-6
Macro call, 7-1, 7-4 to 7-11,
7-12, 7-20, B-1, B-6. See
also .MCALL directive
Macro call argument, 7-4
Macro call numeric argument, 3-3
.MACRO directive, 6-13, 6-21, 7-1
to 7-9, 7-10, 7-11, 9-6, B-6,
D-1, F-3
Macro library directive. See
.MCALL directive
Macro symbol table, 3-6, 3-7
MACR0-11 character set. See
Character set, legal
.MCALL directive, 7-20, 8-6,
8-15, 9-5 to 9-6, B-6,
D-4, F-1 to F-3
.MDELETE directive, 7-21, B-7, C-7

Index-3

I Nu EX

Memory
allocation, 6-42, 6-47, F-1,
F-2
conservation, F-1
.MEXIT directive, 7-3, 7-18 to
7-20, B-7
Modularity, 6-44, E-8, F-1
Module checking routine, E-9
Module preface, E-5
Monitor console routine, 8-1, 8-2
MOV instruction, 3-13, 3-14,
6-37, 6-58, D-1, E-13, G-2 to
G-4, H-2
MOVB instruction, H-2
Multiple definition. See M error
Multiple expression, 2-4
Multiple label, 2-2
Multiple symbol, 2-4
Multiplication operator, 3-2,
3-5, B-1
N error, 3-15
Naming standard, E-2
.NARG directive~ 7-8, 7-12, 7-1~,
B-7, D-2
.NCHR directive, 7-12, 7-13, B-7,
D-2
Nested conditional directive,
6-55, 6-58, 7-3
.NLIST directive, 6-3, 6-9 to
6-14, 6-16, 6-21; 8-6; 8-11;
8-13, 9-4, B-7, D-1
.NOCROSS directive, 6-3, 6-22,
B-7, C-6
.NTYPE directive, 7-12, 7-14,
B-7, D-2
Number of arguments. See .NARG
directive
Numeric argument indicator, B-1
Numeric control
operator, 6-33
temporary, 6-36, B-3
Numeric directive, 6-34
O error, 6-40, 6-56, 6-57, 7-4,
7-12, 7-15, 7-21
Object module name, 1-2
.ODD directive, 6-3, 6-37, 6-38,
B-7
Operand field, 2-1, 2-4, E-1
Operand field separator, 3-2, B-1
Operation field, E-1
Operator field, 2-1, 2-3, 2-4
Overlay, 6-42, 6-44
P error, 6-20, 7-16
.PACKED directive, 6-3, 6-31,
6-37, B-7, C-7
.PAGE directive, 6-3, 6-17, 7-4,
B-7

Page
header, 6-4
number, 6-17
Patch, E-15
Permanent symbol table, C-1 to
C-3, 3-6, 3-7
Position-independent code, G-1 to
G-4
.PRINT directive, 7-17, B-7
Processor priority, E-2
Program counter, 5-1, E-2, G-4
Program counter definition, 3-10
Program development system, 8-14
Program module, E-5
Program section directive. See
.PSECT directive
Program section name, 6-41
Program section table, 1-1
Program version number. See
Version identifier, program
Programming standard, E-1
.PSECT directive, 3-12, 3-14,
6-2, 6-3, 6-20, 6-41 to
6-48, 7-9, 9-6, B-7, D-1,
D-2, H-2

Q error, 6-29, 6-34, 6-38
R error, 3-10
,RAD50 directive: fi-3, 6-29,
B-8, H-2
Radix control, 3-15, 6-32, 6-34,
B-8
temporary, 6-31, 6-33, B-3
.RADIX directive, 3-15, 6-3,
6-32, B-8, D-1
Radix-50, 3-5, 6-30, 6-41, B-3,
B-5, B-8
.
character set, A-4
temporary operator, 6-31
Read-only access, 6-41
Read/write access, 6-41
Register
conventions, E-9
definitions, default, 3-10,
6-21
expression, 5-2, B-1
symbol, 3-10, D-4
term indicator, 3-2, B-1
Register deferred mode, 5-1, 5-2,
B-2, G-1
Register mode, 5-1, 5-2, B-2, G-1
Relative deferred mode, 5-1, 5-8,
B-2, G-2, G-4
Relative mode, 5-1, 5-7, 5-8,
B-2, G-2, G-4
Relocatable expression, 3-17
Relocatable module, 6-43
Relocatable program section, 6-44
to 6-47, B-4
Relocation, 4-1, 6-43

Index-4

INDEX

Relocation bias, 2-2, 3-17, 3-18,
4-1, 6-43
.REM directive, 6-3, 6-18, B-8,
C-7
Repeat block
directive. See .REPT directive
indefinite, 7-3, 7-17 to 7-20,
B-4, B-6
.REPT directive, 7-2, 7-17, 7-20,
B-8, D-3
Reserved symbols, 2-3, 3-1, 3-7
.RESTORE directive, 3-11, 3-14,
6-3, 6-20, 6-49, B-8, C-7,
D-3
.RETURN directive, H-2
RSTS, 9-1 to 9-9
RSX run-time system, 9-1, 9-2
RSX-llM, 6-17, 6-41, 6-48, 7-21,
8-1 to 8-13, 8-19 to 8-22,
E-12, F-3, G-1
RSX-llM-PLUS, 8-1 to 8-13, 8-19
to 8-22, G-1
RT-11, 6-17, 6-41, 6-43, 7-21,
9-1 to 9-9
RT-11 run-time system, 9-1

T error, 3-15, 6-24
Table of contents, 6-12, 6-16,
B-8
Task builder. See Linker
Teleprinter listing format, 6-7,
6-13. See also Listing
control
Temporary numeric control. See
Numeric control, temporary
Temporary radix control. See
Radix control, temporary
Temporary Radix-50 operator, 6-31
Term, definition of, 3-15
Terminal argument indicator, 3-2,
B-1
Terminal expression indicator,
3-2
Terminal register indicator, 3-2,
B-1
Terminating directive. See .END
directive
Thrashing, F-1
.TITLE directive, 6-3, 6-4, 6-13,
6-15, 6-21, B-8, D-2, E-5,
E-7, E-16, H-1
TRAP instruction, 5-9, D-4
TST instruction, E-10, E-11, H-2

.SAVE directive, 6-3, 6-20, 6-49,
6-50, B-8, C-7, D-3
.SBTTL directive, 6-3, 6-4, 6-15,
B-8, H-2
Separating characters, 3-3
Sequence number, 6-19
Single ASCII character indicator,
3-3, B-1, B-3
Source line format, 2-5
Source line terminator, B-1
Special characters, 3-1 to 3-3,
7-7
Stack pointer, E-2
definition, 3-10
Statement format, 2-1
SUB instruction, E-13
Subconditional assembly, 6-56 to
6-59
Subtraction operator, 3-2, 3-5,
B-1
Success/failure indicator, E-9

U error, 3-8, 3-9, 3-15, 6-21,
7-21, 8-7, 8-9, 8-15
Unary operator, 3-4, 3-16, 7-5,
7-7
control, 6-32, 6-34
universal, 3-3, 3-5, B-1
Unconditional assembly, 6-56
Undefined symbol, 3-8, 6-21, D-2,
D-4. See also U error
Universal unary operator. See
Unary operator, universal
Upper-case ASCII, 6-19
User-defined symbol, 3-6 to 3-8
User-defined symbol table, 2-2,
3-6 to 3-8, 3-15

Switches
file specification, 8-6
function control, 8-6, 9-4
listing control, 8-6, 8-7,
9-4
Symbol name syntax, E-3
Symbol table, 1-1, 1-2, F-1
Symbolic argument, 6-41
SYSLIB, F-4
System macro library, 1-1, 7-20,
8-4, 8-14, 9-3, 9-5. See
also .MCALL directive

Version identifier
assembler, 6-4
file, 8-20
program, 6-17, B-5
standard, c-~q to ~io. See also
.!DENT directive
.WEAK directive, 6-3, 6-52, B-8,
C-7
.WORD directive, 3-13, 3-14, 6-3,
6-24, 6-34, 6-36, B-8. See
also Implicit .WORD directive
Z error, 5-3

Index-5

HOW TO ORDER
ADDITIONAL DOCUMENTATION
From
Chicago

Call

Write

312-640-5612

Digital Equipment Corporation
Accessories & Supplies Center
1050 East Remington Road
Schaumburg, IL 60195

8:15 A.M. to 5:00 P.M. CT

San Francisco

408-734-4915
8:15 A.M. to 5:00 P.M. PT

Alaska, Hawaii

603-884-6660
8:30 A.M. to 6:00 P.M. ET

Digital Equipment Corporation
Accessories & Supplies Center
632 Caribbean Drive
Sunnyvale, CA 94086

or 408-734-4915
8:15 A.M. to 5:00 P.M. PT

New Hampshire

603-884-6660
8:30 A.M. to 6:00 P.M. ET

Rest of U.S.A.,
Puerto Rico*

1-800-258-1710

Digital Equipment Corporation
Accessories & Supplies Center
P.O. Box CS2008
Nashua, NH 03061

8:30 A.M. to 6:00 P.M. ET

*Prepaid orders from Puerto Rico must be placed with the local DIGITAL subsidiary (call 80~754-7575)

Canada
British Columbia

1-800-267-6146
8:00 A.M. to 5:00 P.M. ET

Ottawa-Hull

613-234-7726
8:00 A.M. to 5:00 P.M. ET

Elsewhere

Digital Equipment of Canada Ltd
940 Belfast Road
Ottawa, Ontario K1G 4C2
Attn: A&SG Business Manager

112-800-267-6146
8:00 A.M. to 5:00 P.M. ET

Elsewhere

Digital Equipment Corporation
A&SG Business Manager*
*clo DIGITAL's local subsidiary or approved distributor

PDP-11 MACR0-11
Language Reference
Manual
AA-V027A-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. If you require a written reply and are eligible to receive one under Software
Performance Report (SPR) service, submit your comments on an SPR form.

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

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

Please indicate the type of user/reader that you most nearly represent.
Assembly language programmer
Higher-level language programmer
Occasional programmer (experienced)
User with little programming experience
Student programmer
Other (please s p e c i f y ) - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Organization--------------------Telephone - - - - - - - - - - - - - - Street-----------------------------------------~

C i t Y - - - - - - - - - - - - - - - - - - - - - - - State - - - - - - Z i p Code _ _ _ _ _ __
or Country

Do Not Tear - Fold Here and Tape

11111

BUSINESS REPLY MAIL
FIRST CLASS PERMIT N0.33 MAYNARD MASS.
POSTAGE WILL BE PAID BY ADDRESSEE

SSG/ML PUBLICATIONS, ML05-5/E45
DIGITAL EQUIPMENT CORPORATION
146 MAIN STREET
••Av ... aion

••A ft"41'7CA

ml"\ I l'tl"t.nlJ, IYlr'l v

Do Not Tear - Fold Here

I I -.l"T

No Postage
Necessary
if Mai led in the
United States