Digital PDFs

DEC-11-OIMRA-B-D

December 1976

192 pages

Original

7.9MB

Document:	IAS/RSX-11 MACRO-11 Reference Manual MACRO-11 Reference Manual
Order Number:	DEC-11-OIMRA-B-D
Revision:	0
Pages:	192
Original Filename:	http://bitsavers.org/pdf/dec/pdp11/rsx11m_s/RSX11M_V3.2_Jun79/3A_ProgramDevelopment/DEC-11-OIMRA-B-D_MACRO-11_Reference_197612.pdf

OCR Text

IAS/RSX-11
MACR0-11 Reference Manual
Order No. DEC-11-0IMRA-B-D

digital equipment corporation • maynard. massachusetts

First Printing, December 1975
Revised:
Decemb·er 1976

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.
Digital Equipment Corporation assumes no responsibility for the use
or reliability of its software on equipment that is not supplied by
DIGITAL.

(§) 1975, 1976 by Digital Equipment Corporation

The postage prepaid READER'S COMMENTS form on the last page of this
document requests the user's critical evaluation to assist us in preparing future documentation.
The following are trademarks of Digital Equipment Corporation:

DIGITAL
DEC
PDP
DECUS
UNIBUS
COMPUTER LABS
COMT EX
DDT
DEC COMM

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

7/80-14

MAS SB US
OMNIBUS
OS/8
PHA

RSTS
RSX
TYPESET-8
TYPESET-10
TYPESET-11

CONTENTS

Page

PREFACE
0.1
0.2
0.3
0.4

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

CHAPTER

x
x

INTRODUCTION TO MACR0-11

MACR0-11 FEATURES

1-1

1.1
1.1.1
1.1. 2

OVERVIEW OF MACR0-11
Assembly Pass 1
Assembly Pass 2

1-1
1-1
1-2

SOURCE PROGRAM FORMAT

2-1

2.1
2.2
2.2.1
2.2.2
2.2.3
2.2.4
2.3

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

2-1
2-1

PART II
CHAPTER

ix
ix

2-Z
2-4
2-4
2-5
2-6

PROGRAMMING IN MACR0-11 ASSEMBLY LANGUAGE

SYMBOLS AND EXPRESSIONS

3-1

3.1
3 .1.1
3.1. 2
3 .1. 3
3.2
3.2.1
3.2.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9

CHARACTER SET
Separating and Delimiting Characters
Illegal Characters
-Unary and Binary Operators
MACR0-11 SYMBOLS
Permanent Symbols
User-Defined and Macr:o Symbols
DIRECT ASSIGNMENT STATEMENTS
REGISTER SYMBOLS
LOCAL SYMBOLS
CURRENT LOCATION COUNTER
NUMBERS
TERMS
EXPRESSIONS

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

CHAPTER

RELOCATION AND LINKING

4-1

CHAPTER

ADDRESSING MODES

5-1

5.1
5.2

5-1
5-2

iii

CONTENTS (CONT.)
Page
5.3

5.4
5.5

5.6
5.7

5.S
5.9
5.10
5.11

5.12
5 .13
5.14
5.15

AUTOINCREMENT MODE
AUTOINCREMENT DEFERRED MODE
AUTODECREMENT MODE
AUTODECREMENT DEFERRED MODE
INDEX MODE
INDEX•DEFERRED MODE
IMMEDIATE MODE
ABSOLUTE MODE
RELATIVE MODE
RELATIVE DEFERRED MODE
SUMMARY OF ADDRESSING FORMS
BRANCH INSTRUCTION ADDRESSING
USING TRAP INSTRUCTIONS
PART III

CHAPTER

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

5-7
5-8

5-8

MACR0-11 DIRECTIVES

GENERAL ASSEMBLER DIRECTIVES

6-1

6.1

LISTING CONTROL DIRECTIVES
.LIST and .NLIST Directives
Page Headings
.TITLE Directive
.SBTTL Directive
.!DENT Directive
.PAGE Directive/Page Ejection
FUNCTION DIRECTIVES:
.ENABL and .DSABL
DATA STORAGE DIRECTIVES
.BYTE Directive
.WORD Directive
ASCII Conversion Characters
.ASCII Directive
.ASCIZ Directive
.RAD50 Directive
Temporary Radix-SO Control Operator: tR
~~DIX AND NUMERIC CONTROL FACILITIES
Radix Control and Unary Control Operators
.RADIX Directive
Temporary Radix Control Operators: tD,
to, and tB
Numeric Directives and Unary Control
Operators
.FLT2 and .FLT4 - Floating Point Storage
Directives
Temporary Numeric Control Operators: tC
and tF
LOCATION COUNTER CONTROL DIRECTIVES
.EVEN Directive
.ODD Directive
.BLKB and .BLKW Directives
TERMINATING DIRECTIVES
.END Directive
.EOT Directive
PROGRAM BOUNDARIES DIRECTIVE:
.LIMIT
PROGRAM SECTIONING DIRECTIVES
.PSECT Directive
Creating Program Sections
Code or Data Sharing

6-1
6-1
6-8
6..,.11
6-11
6-12

6 .1.1
6 .1. 2
6 .1. 3
6 .1. 4
6 .1. 5
6 .1. 6
6.2

6.3
6.3.1
6.3.2
6. 3. 3
6.3.4
6.3.5

6.3.6
6.3.7
6.4

6.4.1
6.4.1.1
6.4.1.2

6.4.2

6.4.2.1
6.4.2.2
6.5
6.5.l
6.5.2
6.5.3
6.6

6.6.1
6.6.2
6.7
6.8
6.8.1
6.8.1.1
6.8.1.2

6-13
6-13
6-17
6-17
6-18
6-19

6-20
6-21
6-22
6-23

6-24
6-24
6-24

6-25
6-26

6-27

6-27
6-29
6-29
6-29

6-30
6-31
6-31

6-31
6-31
6-32

6-32
6-36
6-38

CONTENTS (CONT.)

Page
6.8.1.3
6.8.2
6.9
6.10
6.10.1
6.10.2
6.10.3

•

Memory Allocation Considerations
.ASECT and .CSECT Directives
SYMBOL CONTROL DIRECTIVE:
.GLOBL
CONDITIONAL ASSEMBLY DIRECTIVES
Conditional Assembly Block Directives:
• J:F I • ENDC
Subconditional Assembly Block Directives:
• IFF I • IFT, • IFTF
Immediate Conditional Assembly Directive:
.IIF

6.10.4
CHAPTER

PAL~llR

6-41
6-43
6-46
6-46

MACRO DIRECTIVES

7-1

7.1
7 .1.1
7 .1. 2
7 .1. 3
7 .1. 4
7.2
7.3

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

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

PART IV
CHAPTER

Conditional Assembly Directives

6-38
6-38
6-39
6-41

7-4
7-5
7-6
7-6
7-7
7-7
7-9
7-10
7-11
7-11
7-12
7-13
7-14

7-15
7-15
7-16
7-17
7-18

OPERATING PROCEDURES

8-1

8.1
8 .1.1
8.1.1.1
8.1.1.2
8.1.1.3
8.1.1.4

RSX-llD AND RSX-llM OPERATING PROCEDURES
Initiating MACR0-11 Under RSX-llM/RSX-llD
Method 1 - Direct MACR0-11 Call
Method 2 - Using RUN Facility
Method 3 - Single Assembly
Method 4 - Install, Run Immediately, and
Remove On Exit
Method 5 - Using Indirect Filename
Facility
RSX-11 Command String Format
RSX-11 File Specification Switches

8-1
8-1
8-1
8-2
8-2

8.1.1.5
8 .1. 2
8 .1. 3

8-2

8-2
8-3
8-5

CONTENTS (CONT • )

Page

APPENDIX

•

8 .1. 4
8.2
8.2.1
8.2.2
8.2.3
8.2.4
8.3
8.4

Cross-Reference Processor (CREF)
IAS MACR0-11 OPERATING PROCEDURES
Initiating MACR0-11 Under IAS
IAS Command String Format
IAS Indirect Command Files
IAS Command String Examples
IAS/RSX-11 FILE SPECIFICATION FORMAT
MACR0-11 ERROR MESSAGES

8-8
8-10
8-10
8-10
8-12
8-12
8-13
8-14

MACR0-11 CHARACTER SETS

A-1

A. l
A. 2

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-2

PERMANENT SYMBOL TABLE (PST)

C-1

C.l
C.2

OP CODES
MACR0-11 DIRECTIVES

C-1
C-4

DIAGNOSTIC ERROR MESSAGE SUMMARY

D-1

D.l

MACR0-11 ERROR CODES

D-1

SAMPLE CODING STANDARD

E-1

E.l
E.2
E.3
E.4
E.4.1
E.4.1.1
E.4.1.2
E.4.1.3
E.4.2
E.4.3
E.4.4
E.4.5
E.4.5.1
E.4.5.2
E.4.5.3
E.4.5.4
E.5
E.5.1
E.5.2
E.5.3
E.5.4
E.5.4.1
E.5.4.2
E.5.4.3
E.5.4.4
E.5.4.5

INTRODUCTION
LINE FORMAT
COMMENTS
NAMING STANDARDS
Register Standards
General Purpose Registers
Hardware Registers
Device Registers
Processor Priority
Other Symbols
Using the Standard Symbolics
Symbols
Global Symbols
Symbol Examples
Program-Local Symbols
Macro Names
PROGRAM MODULES
General 'Comments on Programs
The Module Pref ace
Formatting the Module Preface
Modularity
Calling Conventions (Inter-Module)
Exiting
Intra-Module Calling Conventions
Success/Failure Indication
Module Checking Routines

E-1
E-1
E-2
E-2
E-2
E-2
E-2
E-2
E-3
E-3
E-3
E-3
E-3
E-4
E-4
E-5
E-5
E-5
E-5
E-7
E-8
E-8
E-9
E-9
E-9
E-9

CONTENTS (CONT • )

Page

APPENDIX

APPENDIX
APPENDIX

APPENDIX

E.6
E.6.1
E.6.2
E.6.3
E.7
E.8
E.9
E.9.1
E.10
E.10.l
E.10.2

FORMATTING STANDARDS
Program Flow
Common Exits
Code with Interrupts Inhibited
PROGRAM SOURCE FILES
FORBIDDEN INSTRUCTION USAGE
RECOMMENDED CODING PRACTICE
Conditional Branches
PDP-11 VERSION NUMBER STANDARD
Displaying the Version Identifier
Use of the Version Number in the Program

E-9
E-9
E-11
E-12
E-12
E-12
E-13
E-13
E-13
E-14
E-15

ALLOCATING VIRTUAL MEMORY

F-1

F.l
F.2
F.3

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

F-1
F-2
F-4

FEATURES/FUNCTIONS NOT SUPPORTED BY THE
RSX-llM SK ASSEMBLER

G-1

WRITING POSITION INDEPENDENT CODE

H-1

H.l
H.2

INTRODUCTION TO POSITION INDEPENDENT CODE
EXAMPLES

H-1
H-2

SAMPLE ASSEMBLY AND CROSS REFERENCE LISTING

I-1

INDEX

Index-1

FIGURES
FIGURE

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

8-1
H-1
H-2

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

3-11

3-12
6-6
6-7

6-9
6-12
6-16

6-30
7-17
8-9
H-3
H-3

TABLES
TABLE

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

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

3-1
3-3
3-3

3-4
3-5

CONTENTS (CONT.)
TABLES (CONT • )
Page
TABLE

6-1

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

6-6
8-1
8-2

Symbolic Arguments of Listing Control
Directives
Sym~olic Arguments of Function Control
Directives
Symbolic Arguments of .PSECT Directive
Non-IAS/RSX-11 Program Section Default
Values
Legal Condition Tests for Conditional
Assembly Directives
Subconditional Assembly Block Directives
File Specification Default Values
MACR0-11 File Specification Switches for
RSX-11

viii

6-3
6-14
6-33

6-39
6-41
6-44
8-5
8-6

PREFACE

0.1

MANUAL OBJECTIVES AND READER ASSUMPTIONS

The intent of this manual is to enable users to develop programs coded
in the MACR0-11 assembly language. No prior knowledge of the MACR0-11
Relocatable Assembler is assumed.
Although the description of the assembly
language
is
wholly
self-contained within this manual, the reader is assumed to be
familiar with the PDP-11 processors and related terminology, as
presented in the PDP-11 Processor Handbooks. No attempt is made in
this document to describe the PDP-11 hardware or the functions of the
various PDP-11 instructions.
Since the development of programs necessarily involves the use of the
Task Builder to create an executable task image, the reader is
encouraged to become familiar with this system program, as presented
in the applicable Task Builder reference manual (see Section 0.3).
In presenting MACR0-11, a tutorial bias has been adopted to enlarge
upon the reference material.
This posture is reflected in the
examples and the accompanying commentary describing MACR0-11 language
elements in typical applications.

0.2

STRUCTURE OF THE DOCUMENT

This manual contains four parts. Part I, consisting of two chapters,
briefly introduces MACR0-11.
Chapter 1 lists the key features of
MACR0-11, and Chapter 2 identifies the advantages of following
programming standards and conventions. Also described is the format
used in coding MACR0-11 source programs.
Part II, consisting of three chapters, presents general information
essential to programming with the MACR0-11 assembly language. Chapter
3 describes the symbols, terms, and expressions that form the elements
of MACR0-11 instructions. The ch~racter set is listed, and the types
of programming symbols that may be defined by the user are discussed.
Chapter 4 describes the output of MACR0-11 and presents concepts
essential to the proper relocation and linking of object modules by
the Task Builder.
Chapter 5 briefly describes how data stored in
memory can be accessed and manipulated using the addressing modes
recognized by the PDP-11 hardware.
Part III, consisting of two chapters, describes
the
MACR0-11
directives that control the processing of source statements during
assembly. Chapter 6 discusses directives which accomplish generalized
MACR0-11 functions, while Chapter 7 deals with directives used in the
definition and expansion of macros.
ix

Part IV, consisting
procedures essential
MACR0-11 programs.

only of Chapter S, presents the operating
to the assembly,
linking, and initiating of

Finally, several appendixes are provided,
supplying
information of interest to the MACR0-11 programmer.

additional

Appendix A lists the ASCII and Radix-50 character sets that may be
used in MACR0-11 programs.
Appendix B lists the special characters
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.
The permanent symbols that
have been defined for use with MACR0-11 are listed alphabetically in
Appendix C.
The diagnostic error codes produced by MACR0-11 to identify various
types of errors detected during the assembly process are listed
alphabetically in Appendix D.
Appendix E contains a sample coding
standard that is recommended practice in preparing MACR0-11 programs.
Appendix F discusses several methods of conserving dynamic memory
space for
users of small systems who may experience difficulty in
assembling MACR0-11 programs.
Appendix G lists the features and functions that are not supported in
the SK RSX-llM version of MACR0-11.
MACR0-11 is available in two
versions under RSX-llM.
One is a 14K version which is functionally
identical to the RSX-llD assembler and which has all the features
described in this manual.
The other assembler is an SK version which,
because of size limitations, supports an extensive subset of MACR0-11
features.
Appendix H is a discussion of position independent code
(PIC).

0.3

ASSOCIATED DOCUMENTS

The reader should refer to the
listed below for descriptions
manual.

applicable documentation
of documents associated

directory
with this

!AS Documentation Directory
RSX-llD Documentation Directory
RSX-llM/RSX-llS Documentation Directory

0.4

DOCUMENT CONVENTIONS

The symbols defined below are used throughout this manual.
Symbol

Definition

[]

Brackets indicate
optional.

that

the

enclosed

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

argument
choice

must

is
be

Ellipsis indicates optional continuation of an argument
list in the form of the last specified argument.

UPPER-CASE
CHARACTERS

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

lower-case
characters

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

(n)

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

PART I
INTRODUCTION TO MACR0-11

CHAPTER 1
MACR0-11 FEATURES

The MACR0-11 Assembler provides the following features:
1.

Program and command string control of assembly functions

Device and filename specifications for input and output files

Error listing on command output device

Alphabetized,
formatted
symbol
table
cross-reference listing of symbols

Relocatable object modules

Global symbols for linking independent object modules

Conditional assembly directives

Program sectioning directives

User-defined macros and macro libraries

10.

Comprehensive system macro library

11.

Extensive program
functions

12.

An indirect command
assembly process.

1.1

and

command
file

string

facility

listing;

control
for

optional

listing

controlling

the

OVERVIEW OF MACR0-11

MACH0-11 is a 2-pass assembler.
The functions and operations relevant
to E~ach assembly pass are described in the following sections.

1.1.1

Assembly Pass 1

The main purpose of assembly pass 1 is to locate and read all required
macros from libraries;
to build symbol tables and program section
tables for the program; while also performing a rudimentary assembly
of each source statement.
The first stage of assembly pass 1 is the initialization of all impure
data areas that MACR0-11 uses internally for the assembly process.
These areas include all dynamic storage areas and buffer areas used as
file storage regions.
1-1

MACR0-11 FEATURES
After initializing.memory areas, MACR0-11 issues a call to a system
subroutine which transfers a command line into memory.
This command
line contains the specifications of the files to be used during
assembly. After scanning the command line for proper syntax, 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. They are then closed to minimize requirements for
active file space.
As the assembly process begins, MACR0-11 initiates a routine which
retrieves source lines from the input file.
If no such file is
currently open, as is the case at the beginning of assembly, MACR0-11
opens the next input file specified in the command line previously
read and begins to assemble the source
statements.
MACR0-11
determines the length of each instruction and assembles it accordingly
as one word, two words, or three words.
At the end of assembly pass 1, MACR0-11 reopens the output files
described above and writes out information that is to be used later by
the Task Builder in linking the object modules.
Such information as
the object module name,
the program version number, and the global
symbol directory (GSD) entries for each program section are output to
the object file.
After writing out the GSD entries for a given
program section, MACR0-11 scans through the symbol tables to find all
the global symbols that are bound to that particular program section.
MACRd-11 then writes out GSD records to the object file
for
these
symbols.
This process continues for each program section, bringing to
a close assembly pass 1.

1.1.2

Assembly Pass 2

As an integral part of pass 2, MACR0-11 simultaneously writes the
object records to the output file and generates the assembly listing,
followed by the symbol table
listing
for
the
program.
A
cross-reference listing may also be generated.
See Section 8.1.4.
Basically, assembly pass 2 consists of the same steps performed in
assembly
pass 1, except that all source statements containing
MACR0-11-detected errors are flagged with an erLor code as the
assembly listing file is created. The object file that is created as
the final consequence of pass 2 contains all the object records,
together with relocation records containing information necessary for
subsequent Task Builder linking of the object file.
The information thus passed to the Task Builder 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 Task
Builder Reference Manual (see Section 0.3 in the Preface).

1-2

CHAPTER 2
SOURCE PROGRAM FORMAT

2.1

PROGRAMMING STANDARDS AND CONVENTIONS

Assembly level programming deals directly with the host hardware.
Hence, great care must be exercised in establishing programming
standards and conventions to enable code written by one group to be
interchanged easily with another group.
Standards provide a number of
advantages.
When applied to the program
development
process,
standards make the programming effort easier to:
Plan
Comprehend
Test
Modify
Conve~rt.

Even though standards must accommodate local requirements, many
aspects
of
the
program
development
process
have· universal
applicability.
The standards common to all of DIGITAL's PDP-11
software products are presented in Appendix E as a model for users.
Observance of these standards is beneficial to DIGITAL and its users,
by simplifying both communications and the continuing task of software
maintenance and enhancement.

2.2

STATEMENT FORMAT

A source program is composed of a sequence of source coding lines.
Each line contains a single assembly-language statement. MACR0-11
will accept a source line of 132 characters, but 80 characters is the
recommended length,
because of constraints imposed by listing format
and terminal line size.
A MACR0-11 statement may consist of as many as four
fields.
These
fields are identified by their order of appearance within the
statement and/or by specified separating characters between fields.
The general format of a MACR0-11 statement is:
Label:

Operator Operand

;Comment(s)

The label and comment fields are optional.
The operator and operand
fields are interdependent,
i.e., when both fields are present in a
source statement, each field is evaluated by MACR0-11 in the context
of the other.
A statement may contain an operator field and no operand field,
but
the reverse is not true.
A statement containing an operand with no
operator does not conform to established MACR0-11 coding conventions;
such a statement is currently interpreted by MACR0-11 during assembly
as an implicit .WORD directive (see Section 6.3.2).
2-1

SOURCE PROGRAM FORMAT
MACR0-11 interprets and processes source program statements one by
one, generating one or more binary instructions or data words, or
performing a specified assembly process. Blank lines, although legal,
have no significance in the~~ource prograri~
An assembly-language statement must be completed on one
no continuation lines are allowed in MACR0-11.

source

line;

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

REGTST: BIT

(columns)

#MASK,VALUE

;COMPARES BITS IN OPERANDS.

'rhe above formatting conventions are not mandatory in coding MACR0-11
programs
(free-field
coding
is permissible).
However, it is
recommended that source programs be prepared in accordance with these
conventions for consistency and clarity.

2.2.1

Label Field

A label is a means of
program.

symbolically

referring

location

A label is a user-defined symbol which is assigned the value of the
current location counter and entered into the user-defined symbol
table. The current location counter is the means by which MACR0-11
assigns memory addresses to the source program statements as they are
encountered during the assembly process. The address value of the
label is absolute or relocatable, depending on whether the current
program section being assembled is absolute or relocatable.
(The
concept of program sections and the attributes that may be specified
for them are discussed in detail in Section 6.8.)

In the case of an absolute program section, the value of the current
location counter is likewise absolute, i.e., its value references an
absolute virtual memory address (such as location 100).
Similarly,
the value of the current location counter in a relocatable program
section is also relocatable; however, a relocation bias calculated by
the Task Builder will be added to the apparent value of the current
location counter to establish its effective absolute virtual address
at execution time.

2-2

SOURCE PROGRAM FORMAT
If present, a label always appears as the first field in a source
statement and must h.e terminated by a.colon. For example, if the
current location counter value is a6solu€e 100(9), the statement:
ABCoij

MOV

A,B

assigns the value 100(8) to the label ABCD. Subsequent references to
this label would then yield a value of absolute 100(8). In this
example, if the location counter value were relocatable, the final
value of ABCD would be 100(8)+K, where K represents the relocation
bias of the program section, as calculated by the Task Builder at link
time.

ABC:

$DD:

A7.7:

MOV

A,B

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

MOV

For example, the

A,B

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

assigned

Of the two methods of assigning multiple labels shown above, the
\.second _!.s prefe;;,,red, because consistency of field positioning within
tm~ soatce program improves readability.
lobal s mbol.
Such a
A ouble colon
independently-assembled o Ject modules.
labe
can
e referenced
References to this label in other modules will be resolved by the Task
Builder when the modules are linked as a composite executable task.
For example, the statement
ABCD::

MOV

A,B

establishes the label ABCD as a global symbol.
The distinguishing
attribute of a global symbol is that it can be referenced from within
an object module other than the module in which the symbol is defined
(sE~e Section 6.9).
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.

2-3

SOURCE PROGRAM FORMAT

co e (M) is ge
in
characters in two or more
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 (see Appendix D). Furthermore, any statement in
the source program which references a multi-defined label results in
an additional diagnostic message: in this case, an error code (D) is
generated in the assembly listing (see Appendix D).

2.2.2

Operator Field

The operator field specifies the action to be performed.
It may
consist of an instruction mnemonic {op code), an assembler directive,
or a macro call.
The operator field follows the label field in a source statement.
Chapters 6 and 7 describe these three types of operator field entries.
When the operator is an instruction mnemonic, the mnemonic op code
specifies the machine instruction to be generated. MACR0-11 then·
continues with the evaluation of the address(es) of the operand(s)
which follow(s).
When the operator is a directive, the directive
causes MACR0-11 to perform 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.
The operator field need not be preceded by a label:
but it may be
preceded by one or more labels and followed by one or more operands
and/or a comment. Furthermore, leading and trailing spaces or tabs in
the operator field have no significance: such characters serve only
to separate the operator field from the preceding and following
fields.
An operator is terminated by a space, tab, or any non-RAD50 character,
as in the following examples:
MOV A,B
MOV

:THE SPACE TERMINATES THE OPERATOR
:MOV.
A,B

MOV@A,B

:THE TAB TERMINATES THE OPERATOR MOV.
:THE @ CHARACTER TERMINATES THE
:OPERATOR MOV.

Although the statements above are ~11 equivalent in
second statement is the recommended form because
MACR0-11 coding conventions.

2.2.3

function, the
it conforms to

Operand Field

When the operator field contains an instruction mnemonic (op code) ,
the operand field specifies those program variables that are to be
2-4

SOURCE PROGRAM FORMAT
evaluated/manipulated by the operator. The operand field may also be
used to supply arguments to MACR0-11 directives and macro calls, as
described in Chapters 6 and 7, respectively.
Ope1cands may be expressions or symbolic arguments (within the context
of the specified operation). Multiple expressions used in the operand
field of a MACR0-11 statement must be separated by a comma;
multiple
symbolic arguments similarly used may be delimited by any legal
separator, i.e., a comma, tab, and/or space.
An operand should be
preceded by an operator field; if it is not, the statement is treated
by MACR0-11 as an implicit .WORD directive (see Section 6.3.2).
When the operator field contains an op code, associated operands
always expressions, as shown in the following statement:
MOV

are

RO,A+2(Rl)

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

ALPHA

ARG1,ARG2

Refer to the description of each MACR0-11 directive to determine
and number of operands required in issuing the directive.

the

The operand field is terminated by a semicolon when the field
followed by a comment. For example, in the following statement:

typE~

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
nul
RUBOUT,
a e- turn,
line-feed_.,..
vertical-tab or
characters appearrti9 in the
comment fief~, even specia characters reserved for use in MACR0-11,
are checked only for ASCII legality and then included in the assembly
listing as they appear in the source text.
All comment fields must begin with the GemicoJop qharaQter(·).
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 le~dipg semicgJon, 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 lin~ within the code body.
Comments do not affect assembly processing or program execution.
However, comments are useful in source listings for later analysis,
debugging, or documentation purposes.

2-5

SOURCE PROGRAM FORMAT
2o3

FORMAT CONTROL

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

MOV

(SP)+,TAG

:POP VALUE OFF STACK.

This statement conforms to the standard
horizontal
formatting
conventions, i.e., the statement elements are separated into four
distinct fields and are therefore easily discernible.
Page formatting and assembly listing considerations are discussed in
Chapter 6 in the context of MACR0-11 directives that may be specified
to accomplish desired formatting operations. Appendix E describes the
coding conventions used in all DIGITAL PDP-11 operating system
software.

2-6

PART II
PROGRAMMING
IN MACR0-11 ASSEMBLY
LANGUAGE

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 are discussed as
they relate to MACR0-11 programming.

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, .ENABL LC).

The digits 0 through 9.

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

The special characters listed in Table 3-1.

(dollar sign) .
These
as Digital Equipment

Table 3-1
Special Characters Used in MACR0-11
Character

Designation

Colon

Label terminator.

Double colon

Label terminator; defines the
label as a global label.

Equal sign

Direct assignment
operator;
and macro keyword indicator.

Double equal
sign

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

Percent sign

Tab

Item or field terminator.

Space

Item or field terminator.

Function

3-1

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

Designation

Number sign

Immediate expression
indicator.

At sign

Deferred addressing indicator.

(

Left parenthesis

Initial register indicator.

)

Right parenthesis

Terminal register indicator.

Period

Current location counter

Comma

Operand field separator.

Semicolon

Comment field indicator.

Left angle
bracket

Initial argument or expression
indicator.

Right angle
bracket

Terminal argument or expression indicator.

Plus sign

Arithmetic addition operator
or autoincrement indicator.

Minus sign

Arithmetic subtraction operator or autodecrement indicator.

Asterisk

Arithmetic multiplication operator.

Slash

Arithmetic division operator.

Ampersand

Logical AND operator.

Exclamation point

Logical inclusive OR operator.

Double quote

Double ASCII character indicator.

Single quote

Single ASCII character indicator;
or
concatenation
indicator.

Up arrow or
circumflex

Universal unary operator or
argument indicator.

Backslash

Macro call numeric argument
indicator.

3.1.1

Function

Separating and Delimiting Characters

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

3-2

defined

SYMBOLS AND EXPRESSIONS

Table 3-2
Legal Separating Characters
Character
Space

Definition

Usage

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 ~~:i;;n:eilid QI:!§ ~'~
J;]po;ea (see Section 3. 9) •

Comma

A comma is a legal separator
between
symbolic
arguments
operand
field.
within the
Multiple expressions used in
the operand field must
be
separated by a comma.

Table 3-3
Legal Argument Delimiters
Character

< •.•• >

Definition

Usage

Paired angle brackets

Paired angle brackets may be
used anywhere in a program to
enclose an
expression
for
treatment as a singte 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

3.1.2

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.

Illegal Characters

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

A character is not an element of the recognized MACR0-11
character set.
A character of this kind is replaced in the
.Ustipg by a c;pJe§tion mark, and an error code (I) is printed
in the assembly listing (see Appendix D). The exception to
this is an embedded null which, when detected, terminates the
scan of the current line.

3-3

SYMBOLS AND EXPRESSIONS
2.

3.1.3

A legal MACR0-11 character is illegal in the context of its
usage within the source statement, i.e., its syntax is
illegal or questionable. Such a character causes an error
code (Q) to be printed in the assembly listing.

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.
A term preceded by a unary operator is considered to
contain that operator. The term so specified thus becomes a value
which can be used alone or as an element of an expression.
Table 3-4
Legal Unary Operators
Unary
Operator

Explanation

Example

Plus sign

Produces
the
value of A.

Minus sign

-A

Produces
negative
the
( 2 IS complement) value of
A.

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).

"0127

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

"F3. 0

Effect
positive

"034

Interprets 34 as an octal
number.

"BllOOOlll

Interprets 11000111 as
binary number.

"RABC

Evaluates ABC in Radix-50
f o.rm.

Unary operators can be used adjacent to each other or in constructions
involving multiple terms, as shown below:
(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
shows the relationships that can be established between expression
terms through the use of binary operators.

3-4

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

Explanation

Example

Addition

A+B

-·

Subtraction

A-B

Multiplication

A*B (16-bit product returned)

Division

A/B (16-bit quotient returned)

Logical AND

A&B

Logical inclusive OR

A!B

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

3.2

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

; EQUALS 11 ( 8} •
; EQUALS 7 ( 8} •

MACR0-11 SYMBOLS

Three types of symbols may be defined for use within MACR0-11 source
programs: permanent symbols, user-defined symbols, and macro symbols.
MACR0-11 maintains three types of symbol tables:
the Permanent Symbol
Table
(PST},
the User Symbol Table (UST}, and the Macro Symbol Table
(MST}.
The PST contains all the permanent symbols defined within (and
thus automatically recognized by} MACR0-11 and is part of the MACR0-11
task image. The UST and MST are constructed as the source program is
assembled.

3.2.1

Permanent Symbols

Permanent symbols consist of the instruction mnemonics
(see Appendix
C}
and MACR0-11 directives
(see Chapters 6 and 7 and Appendix B} •
These symbols are a permanent part of the MACR0-11 task image and need
not be defined before 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 treated by the programmer as
labels
(see Section 2.2.l}
or that are equated to a specific value
through a direct assignment statement (see Section 3.3) or appear as
macro names or dummy arguments.
These symbols are added to the User
Symbol Table as they are encountered during assembly.
Macro symbols
are those symbols used as macro names (see Section_7.l}.
Similarly,
these symbols are added to the Macro Symbol Table as they are
encountered during assembly.
3-5

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

~rhe

creation

user-defined

and

macro
(except

The first character of a symbol must not be a ppmher
in the case of local symbols; see sed1on 3.5).

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.

The value of a symbol depends upon its use in the program.
When a
symbol appears in the operator field, it may be any one of the three
symbol types described above i.e., permanent, user-defined, 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 redefinitiQn of Permanent Symbol Table
entries as macro symbols. That is, permanent symbols may be used as
macro symbols. But the user must keep in mind the sequence in which
the search for symbols is performed in order to avoid incorrect
interpretation of the symbol's use.
When a symbol appears in the operand field, the User-Defined Symbol
Table is searched first, then the Permanent Symbol Table is searched.
Depending on their use in the source program, user-defined symbols
have either a local
(internal) attribute or a global (external)
attribute.
Normally, 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:
3-6

SYMBOLS AND EXPRESSIONS
1.

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

Use of the double colon (::)
2.2.1).

in defining a label

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

sign

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

(see Section

direct
at

assignment
the

end

NOTE
Undefined symbols at the end of assembly are
assigned
a value of 0 and placed into the
user-defined symbol table as undefined default
global references.
If the .DSABL GBL directive is
in effect, however,
(see Section
6.2),
the
automatic global reference default function of
MACR0-11 is inhibited, causing the
statement
containing the undefined symbol to be flagged with
an error code (U) in the assembly listing
(see
Appendix D) .
Global symbols provide linkages between independently-assembled object
modules within the task image. A global symbol defined as a label,
for example, may serve as an entry-point address to another section of
code within the task image.
Such symbols are referenced from other
source modules in order to transfer control throughout the task's
execution.
These global symbols are resolved by the Task Builder at
link time, ensu~ing that the resulting task
image is a logically
coherent and complete body of code.

3.3

DIRECT ASSIGNMENT STATEMENTS

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

expression

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

reference

- cannot contain an undefined global reference.

A direct assignment statement embodying the double equal {==) sign, as
shown above, defines the symbol as global (see Section 6.9).

3-7

SYMBOLS AND EXPRESSIONS
The following examples illustrate
statements:

the

coding

direct

assignment

A=l

;THE SYMBOL A IS EQUATED TO THE
;VALUE 1.

B=A-l&MASKLOW

;THE SYMBOL B IS EQUATED TO THE
;VALUE OF THE ENTIRE EXPRESSION
;WHICH FOLLOWS.

C:
D=.

MOV

#1,ABLE

;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 last of the three examples above is provided only to illustrate
the performance of MACR0-11 in such situations. See Section 3.6 for a
description of the period (.) as the cur~ent location counter symbol.
The following conventions apply to the
statements:

coding

direct

assignment

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

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

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

assignment

A direct assignment statement
comment field.

only

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

may

X=Y

(Illegal forward reference}

y=z

(Legal forward reference}

followed

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

3-8

SYMBOLS AND EXPRESSIONS
3.4

The eight general· registers of the PDP-11 processor are numbered O
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.
Use standard symbolic names for all register references.
The register definitions listed below are automatically assigned by
MACR0-11, i.e., these definitions 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

;REGISTER 0 DEFINITION.
;REGISTER 1 DEFINITION.
;REGISTER 2 DEFINITION.
;REGISTER 3 DEFINITION.
;REGISTER 4 DEFINITION.
;REGISTER 5 DEFINITION.
;STACK POINTER DEFINITION.
;PROGRAM COUNTER DEFINITION.

PC=%7

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

their

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. Although you can reassign the
standard register symbols through the use of the .DSABL REG directive
(see Section 6.2), this practice is not recommended.
An attempt to
redefine a default register symbol without first specifying the .DSABL
REG directive to override the normal register definitions causes that
assignment statement to be flagged with an error code (R) in the
assembly listing.
The symbolic default names assigned to , the
registers, as listed above, are the conventional names used in all
DIGITAL-supplied PDP-11 system programs. For this reason, you are
well advised to follow these conventions.
All non-standard register symbols must be defined before they are
referenced in the source program. A register expression less than O
or greater than 7 is flagged with an error code (R) in the assembly
listing.
The % character may be used with any legal term
specify a register. For example, the statement
CLR

%3+1

is equivalent in function to the statement
CLR

and clears the contents of register 4.

3-9

expression

SYMBOLS AND EXPRESSIONS
In contrast, the statement
CLR

clears the contents of virtual memory location 4.

3.5

LOCAL SYMBOLS

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

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

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

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

.ENABL LSB

Ending delimiter:

.ENABL

delimited

through

(see Section 6. 2)

LSB

or
.DSABL

LSB (see Section 6. 2)

followed by one of:
.PSECT
.CSECT
.ASECT

Symbolic label

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

Local symbols provide a c6nvenient means of generating labels for
branch instructions and other such references within a local symbol
block.
Using local symbols reduces the possibility of symbols with
multiple definitions appearing within a user program.
In addition,
the use of local symbols differentiates entry-point labels from local
labels,
since local symbols cannot be referenced from outside their
respective local symbol block. Thus, local symbols of the same name
can appear
in other local symbol blocks without conflict.
Local
symbols do not appear in cross-reference listings.

3-10

SYMBOLS AND EXPRESSIONS
Local symbols require less symbol table space than other types of
symbols.
Their use is
When defining local symbols, use
the range fro
the range from 128$ to 65535$.
Local symbols within
,
nc us
generated automatically as a
of MACR0-11.
Such local symbols
are useful
in the expansion of macros during assembly and are
described in detail in this context in Section 7.3.5.
Be sure to avoid multiple definitions of local symbols within the same
local symbol block.
For example, if the local symbol 10$ is defined
two or more times within the same local symbol block, each symbol
represents a different address value.
Such a multi-defined symbol
causes an error code (P) to be generated in the assembly listing.
For examples of local symbols and local symbol blocks as
in a source program, see Figure 3-1.
121
122
123
124
125 000000
12i;; 01110000
127 1/HHH1!0•1
128 000006
129 000012
1::rn
131 00000,~
132 000000
133 000V'104
134 000006
135 00011112
136
137 0001110,,
138 00000i~
139 01110Ql04
14111 000Vl06
141 000012
142

appear

PROGRAM IlllITIALIZATIOlll CODE

012700

0Ql000"1'

00~H'l20

02270111
101374

0~0000 1

01270"11
005"12111
022700
101374

(11(110000 I

I PSEC T
XC TPRG I I MOV
CLR
!SI
CMP

Bi"! 1

(110(110001

,PSECT
XC TPAS 11 MQV
IS 1
CLR
CMP

BHI
2'1110Ql00 I

012700
01115020
02270'11
I 111374

011111111101'1 I

,PSECT
XCTLIN1 IMOV
IS 1
CLR
C:MP

BHl

Figure 3-1

3.6

they

XCTPRG,GBL
#lMPUR!::,R0
(R0)•
#IMPURT,R"'
1S

JIMPURE DATA

IlllITIALIZATIO~

)(C TPAS I GBL
#IMPPAS,~0

JPASS INITIALIZATION

( R0) •
#IMPPAT,R0
1S
XCTLIN,Gl:!L
#lMPLIN,R0
(R0) +
#!MPUT,R0

ILI"<E INITIALIZATIO"'

Assembly Listing Showing Local Symbol Block

CURRENT LOCATION COUNTER

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

MOV

#.,RO

;THE PERIOD (.) REFERS TO THE ADDRESS
;OF THE MOV INSTRUCTION.

(The function of the # symbol is explained in Section 5.9.)
SAL=O
.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
3-11

SYMBOLS AND EXPRESSIONS
value 177535 is thus stored in location 500. The value represented by
.+4 is stored in location 502; this value is derived as the current
value of the location counter (which is now 502) , plus the absolute
value 4, thereby depositing the value 506 in location 502. Finally,
the value of SAL, previously equated to O, 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
Similar to other MACR0-11 symbols, the current location counter symbol
(.) has an attribute of relocatability associated with it: it is
either absolute or relocatable, depending on the specific such
attribute of the current program section.
(A program section and its
attributes are defined through the use of the .PSECT directive
described in Section 6.8.1.) The existing attribute (or mode) of the
current location counter cannot be changed by specifying a defining
expression having a different attribute.
Furthermore, such a defining expression must not force the location
counter into another program section (.PSECT area), even though the
program sections so involved may both be absolute or relocatable. The
expression defining the location counter value must not contain a
forward reference, i.e., the expression must not contain a reference
to
a symbol that is not previously defined.
Such violations
constitute a general assembly error, resulting in an error code (A) in
the assembly listing.
Thus, the attribute (or mode) of the current location counte~ takes on
the
attribute of the current program section.
Therefore, its
attribute from program section to program section can be changed only
through
the program sectioning directives (.PSECT, .ASECT, and
.CSECT), as described in Section 6.8.
The following coding illustrates
counter:

the

3-12

use

the

current

location

SYMBOLS AND EXPRESSIONS
.ASECT
.=500
FIRST:

MOV

.+10,COUNT

.=520
SECOND: MOV

.,INDEX

;SET LOCATION COUNTER TO
;ABSOLUTE 500(0CTAL).
;THE LABEL "FIRST" HAS THE VALUE
; 500 (OCTAL) •
;.+10 EQUALS 510(0CTAL). THE
;CONTENTS OF THE LOCATION
;510(0CTAL) WILL BE DEPOSITED
;IN THE LOCATION "COUNT."
;THE ASSEMBLY LOCATION COUNTER
;NOW HAS A VALUE OF
;ABSOLUTE 520(0CTAL).
;THE LABEL SECOND HAS THE
;VALUE 520(0CTAL).
;THE CONTENTS OF LOCATION
;520(0CTAL), 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 following statements:
.=.+40
or
.BLKB 40
or
.BLKW 20
reserves 40(8) bytes of storage space in the source program.
The
.BLKB and .BLKW directives, however, are recommended as the preferred
ways to reserve storage space (see Section 6.5.3).

3. 7

NUMBE:RS

MACR0-11 assumes that all numbers in the source program are to be
interpreted in octal radix, unless otherwise specified. An exception
to this is that operands associated with Floating Point Processor
instructions and Floating Point Data directives are treated as decimal
(see Section 6.4.2). This default radix can be altered with the
.RADIX directive (see Section 6.4.1.1). 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.

3-13

SYMBOLS AND EXPRESSIONS
MACR0-11
Negative numbers must be preceded by a minus sign;
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, i.e., 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,
not relocatable.

i.e.,

they

are

Single-word floating-point numbers may be generated with the nF
operator (see Section 6.4.2.2) and are stored in the following format:
15

Sign
Bit

8-bit
Exponent

7 6

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 3.2.
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.)

An undefined symbol is assigned a value of zero and
inserted in the User-Defined Symbol Table as an undefined
default global reference. If the .DSABL GBL directive
(see Section 6.2) is in effect, the automatic global
reference default function of MACR0-11 is inhibited, in
which case, the statement containing the undefined symbol
is flagged with an error code (U) in the assembly
listing.

User-Defined

used, with zero
{Appendix C lists

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

3-14

SYMBOLS AND EXPRESSIONS
4.

3.9

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

EXPRESSIONS

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

An
An expression is absolute if its value is fixed.
expression whose terms are numbers and ASCII conversion
A relocatable
characters will reduce to an absolute value.
3-15

SYMBOLS AND EXPRESSIONS

expression or term minus a relocatable term, where both
elements being evaluated belong to the same program section,
are also absolute, since such an expression is reduced to a
single term by MACR0-11 upon completion of the expression
scan. For example, the expression TAG2-TAG1, where both TAGl
and TAG2 are defined in the same program section, is an
absolute expression.
2.

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

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

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

following

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

terms

belonging

different

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

addition

specified

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

3-16

and

complex

relocatable

CHAPTER 4
RELOCATION AND LINKING

The output of MACR0-11 is an object module which must be processed by
the Task Builder before it can be loaded and executed. Essentially,
the Task Builder fixes (i.e., makes absolute) the values of external
or relocatable symbols in the object module, thus transforming the
object module, or several such object modules, into an executable task
image. This process is called linking.
To enable the Task Builder to fix the value of an expression, MACR0-11
issues certain directives to the Task Builder, together with other
required parameters. In the case of relocatable expressions in the
object module, the Task Builder adds the base of the associated
relocatable program section to the value of the relocatable expression
provided by MACR0-11. In the case of external expression values, the
Task Builder determines the value of the external term in the
expression (since the external symbol must be defined in one of the
other object modules being linked together) and then adds it to the
absolute portion of the external expression, as provided by MACR0-11.
All instructions that require modification by the Task Builder 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 by the Task Builder is required in order to fix
the value of the expression.
EXAMPLE:
005065 CLR
OOOOOOG

EXTERN (R5)

;THE VALUE OF THE "EXTERN" SYMBOL IS
;ASSEMBLED AS ZERO AND IS TO BE
;RESOLVED BY THE TASK BUILDER.

005065 CLR
000006G

EXTERN+6 (RS)

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

005065 CLR
000040'

RELOC (R5)

;ASSUMING THAT THE VALUE OF THE
;SYr-.1.BOL "RELOC" IS RELOCATABLE
;40, THE TASK BUILDER WILL ADD A
;RELOCATION BIAS TO THIS VALUE.

005065

-<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.

ooooooc

CLR

4-1

RELOCATION AND LINKING
For a complete description of object records output by MACR0-11, refer
to the applicable Task Builder Reference Manual (see Section 0.3 in
the Preface).

4-2

CHAPTER 5
ADDRESSING MODES

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

used

describing

E is any expression, as defined in Chapter 3.

Risa register expression, i.e., any expression containing a
term preceded by a percent sign (%) or a symbol previously
equated to such a term, as shown in the examples below:
R0=%0
Rl=RO+l
R2=1+%1

;GENERAL REGISTER 0.
;GENERAL REGISTER 1.
;GENERAL REGISTER 2.

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

5.1

modes

addressing

the

normal

default

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

the

A is a general addressing specification which produces a
6-bit
mode address field,
as described
in the PDP-11
Processor Handbooks.
The addressing specification, A,
is
described
in terms of E, R, and ER, as defined above.
Each
addressing specification within this section is illustrated
using either the single operand instruction CLR or the double
operand instruction MOV.

The register itself (R) contains the operand to be manipulated by
instruction.
5-1

the

ADDRESSING MODES
Format for A:

Example:
CLR

5.2

;CLEARS REGISTER 3.

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

@R or (ER)

Examples:
CLR
CLR
CLR

5.3

@Rl
(Rl)
(1)

;ALL THESE INSTRUCTIONS CLEAR
;THE WORD AT THE ADDRESS
;CONTAINED IN REGISTER 1.

AUTOINCREMENT MODE

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

(ER)+

Examples:
CLR
CLR
CLR

(RO)+
(R4)+

(R2)+

;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 exactly 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 autoincremented
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):

5-2

after

~DDRESSING

MODES

MOV

RO, (RO)+

;THE QUANTITY 102 IS MOVED
;TO LOCATION 100.

MOV

RO.,- (RO)

;THE QUANTITY 76 IS MOVED
;TO LOCATION 100.

The use of these forms should be avoided, since
they are not compatible 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.

5.4

AUTOINCREMEN~ DEFERRED

MODE

The~
The~

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

@(ER)+

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

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

- (ER)

Examples:
CLR

- (RO)

CI~R

- (R3)
- (R2)

CLR

5.6

;DECREMENT THE CONTENTS OF THE SPECI; FIED REGISTER (0, 3, OR 2) BY TWO
;BEFORE USING ITS CONTENTS
;AS THE ADDRESS OF THE WORD TO BE
;CLEARED.

AUTODECREMENT DEFERRED MODE

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

@-(ER)

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.
5-3

ADDRESSING MODES
5.7

INDEX MODE

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

E(ER)

Examples:

5.8

CLR

X+2(Rl)

MOV

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

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

@E(ER)

Example:
CLR

5.9

...

@114(R4)

;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.

IMMEDIATE MODE

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

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

Examples:
MOV
MOV

#100,RO
#X,RO

;MOVE THE VALUE 100 INTO REGISTER 0.
;MOVE THE VALUE OF SYMBOL X INTO
;REGISTER O.

The number sign
(#)
in the MACR0-11 character set has special
significance as an addressing mode indicator. When this character
appears in the operand field,
as shown above,
it specifies the
immediate addressing mode,
indicating to MACR0-11 that the operand
itself immediately follows the instruction word.

5-4

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

0 1 2 7 0 0

Location 22:

0 0 0 1 0 0

Location 24:

Next instruction

the

first

example,

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

5.10

ABSOLUTE MODE

Absolute mode is the equivalent of immediate mode deferred.
The
address expression @#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.
Format for A:

@#E

Examples:
MOV

@#100,RO

CLR

@#X

;MOVE THE CONTENTS OF LOCATION 100
;INTO REGISTER RO.
;CLEAR THE CONTENTS OF THE LOCATION
;WHOSE ADDRESS IS SPECIFIED BY
;THE SYMBOL X.

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

0 1 3 7 0 0

Location 22:

0 0 0 1 0 0

Location 24:

Next instruction

the

first

example,

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

5.11

RELATIVE MODE

Relative mode is the normal mode for memory references within your
program.
It is assembled as index mode, using the PC as the index
register.
5-5

ADDRESSING MODES
Format for A:

Examples:
CLR

lf)O

MOV

RO,Y

;CLEAR LOCATION 100, RELATIVE TO
;THE CONTENTS OF THE PC.
;MOVE THE CONTENTS OF REGISTER 0
;TO LOCATION Y, RELATIVE TO THE
;CONTENTS OF THE PC.

In relative mode, the offset for the address calculation is assembled
as the second or third word of the instruction. This value is added
to the contents of the PC (the base register) to yield the address of
the source operand.
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

Loc·ation 24:

Next instruction

shown

with

the

statement

Note that the constant 54 is assembled immediately following the
instruction word, i.e., as the second word in the instruction. Upon
execution of the instruction, the processor fetches the first word
(MOV) and increments the PC by 2 so that it points to location 22
(containing the value 54). After the next fetch and increment cycle,
the processor calculates the effective address of the source operand
by taking the contents of location 22 (the offset) and adding it to
the current value of the PC, which now points to location 24 (the next
instruction). Thus, the source operand address is the result of the
calculation
OFFSET+PC = 54+24 = 100(8),
causing the contents of
location 100 to be moved into register 3.
Since MACR0-11 considers the contents of the current location counter
(.) as the address of the first word of the instruction, an equivalent
index mode statement is shown below:
MOV

100-.-4(PC) ,R3

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

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

Example:
MOV

@X,RO

;RELATIVE TO THE CURRENT VALUE OF
;THE PC, MOVE THE CONTENTS OF THE
;LOCATION WHOSE ADDRESS IS POINTED
;TO BY LOCATION X INTO REGISTER O.
5-6

ADDRESSING MODES
5.13

SUMMARY OF ADDRESSING FORMS

Each PDP-11 instruction takes at least one word. Operands of the form
listed below do not increase the length of an instruction.
Form

Meaning

@R or (ER)

(ER)+

Autoincrement mode

@(ER)+

Autoincrement deferred mode

- (ER)

Autodecrement mode

@- (ER)

Autodecrement deferred mode

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

Meaning

E (ER)

Index mode

@E (ER)

Index deferred mode

Immediate mode

@#E

Absolute mode (see Note below)

Relative mode

Relative deferred mode

The syntax of the addressing modes is
Additional
discussion
of addressing
applicable PDP-11 Processor Handbook.

summarized in Appendix B.
modes is provided in the

NOTE
An alternate form for @R is (ER).
However, the
form @(ER)
is only logically, but not physically
equivalent to
the
expression
@O(ER).
The
addressing form @#E differs from form E in that
the second or third word of the instruction
contains the absolute address of the operand,
rather than the relative distance between the
operand and the PC.
Thus, the instruction CLR
@#100 clears absolute location 100, even if the
instruction is moved from the point at which it
was assembled. See the description of the .ENABL
AMA function in Section 6.2, which causes all
relative mode addresses to be
assembled
as
absolute mode addresses.

5-7

ADDRESSING MODES
5ol4

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

rather

form

the

MACR0-11 performs the reverse operation to form the word offset from
the specified address. Remember that when the offset is added to the
current value of the PC, the PC is pointing to the word following the
branch
instruction;
hence,
the
factor -2 in the following
calculation:
Word offset

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

Since the value of the PC
Word offset

= .+2, we have:

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

In using branch instructions, you must
following error conditions:

exercise

avoid

the

Branching from one program section to another;

Branching to a location
(global) symbol; or

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

that

defined

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

5.15

care

generated

external

the

USING TRAP INSTRUCTIONS

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

5-8

PART III
MACR0-11 DIRECTIVES

Chapters 6 and 7 describe all the directives used with MACR0-11.
Directives are statements that cause MACR0-11 to perform certain
operations during assembly. Chapter 6 describes several types of
directives, including those which control symbol interpretation,
listing header material, program sections, data storage formats, and
assembly listings.
Chapter 7 describes those directives concerning
macros, macro arguments, and repetitive coding sequences.
MACR0-11 directives can be preceded by a label (subject to any
restrictions associated with specific directives) and followed by a
comment. A MACR0-11 directive occupies the operator field of a source
statement.
Only one directive can be included in any given source
line. The operand field may be occupied by one or more operands or
left blank; legal operands differ with each directive specified.

CHAPTER 6
GENERAL ASSEMBLER DIRECTIVES

This category of directives includes:
1.

Listing control

Function control

Data storage

Radix and numeric

Location counter control

Terminators

Program boundaries

Program sectioning

Symbol control

10.

Conditional assembly

11.

PAL-llR conditional assembly.

co~trol

Each is described in its own section of this chapter.

6.1

LISTING CONTROL DIRECTIVES

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

6.1.1

.LIST and .NLIST Directives

Listing control options can be specified in the text of a MACR0-11
program through the .LIST and .NLIST directives. These directives are
of the form:
.LIST
.LIST
.NLIST
.NLIST

arg
arg

6-1

GENERAL ASSEMBLER DIRECTIVES
where:

arg

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

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 .~LIST directive,
the listing
level count is decremented.
hen th listin 1
· negative,
t.b H s+ j no is ,suppressed ~un ess
e
ine contains an
rror).
Conversely, when the listing level count is greater than zero, the
listing is always generated.
Finally, when the count is zero,
the
line is either listed or suppressed, contingent upon the other listing
controls currently in effect for
the program.
For example,
the
following macro definition employs the .LIST and .NLIST directives to
selectively list portions of the macro body when the macro is
e.xpanded:
2

.MACRO LTEST
A-THIS LINE SHOULD LIST
.NLIST
B-THIS LINE SHOULD NOT LIST
• NLIST
C-THIS LINE SHOULD NOT LIST
.LIST
D-THIS LINE SHOULD NOT LIST
.LIST
E-THIS LINE SHOULD LIST
.ENDM

;LIST TEST
; LISTING LEVEL COUNT IS o.
;LISTING LEVEL COUNT IS -1.

• LIST
ME
LT EST
A-THIS LINE SHOULD LIST
E-THIS LINE SHOULD LIST

;LIST MACRO EXPANSION .
;CALL THE MACRO
;LISTING LEVEL COUNT IS 0.
;LISTING LEVEL COUNT IS BACK TO O.

; LISTING LEVEL COUNT IS -2 •
;LISTING LEVEL COUNT IS -1.
;LISTING LEVEL COUNT IS o.
; LISTING LEVEL COUNT IS BACK TO O.

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

.LIST

;LIST NEXT LINE.

.NLIST

;DO NOT LIST REMAINDER OF MACRO
;EXPANSION .

X=.

. ENDM
. NLIST

;DO NOT LIST MACRO EXPANSIONS .

X=.
The symbo1ic arguments allowed for use with the listing directives are
described in Table 6-1.
These arguments can be used singly or in
combination with each other.
If multiple arguments are specified in a
listing directive, each argument must be separated by a comma, tab, or

6-2

GENERAL ASSEMBLER DIRECTIVES
space. For any argument not specifically included in a listing
control statement, the associated default assumption (List or No list)
is applicable throughout the source program. The default assumptions
for the listing control directives also appear in Table 6-1.
Table 6-1
Symbolic Arguments of Listing Control Directives
Argument

Default

Function

SEQ*

List

Controls the listing of source line
sequence
numbers.
MACR0-11 assigns
sequence number 1 to the first source
line in a file, and increments the
sequence number for each additional line
in
the
file.
If
this field is
suppressed
through
an
.NLIST
SEQ
directive, MACR0-11 generates a tab,
effectively allocating space for the
field, but fills the field with blanks.
Thus, the inter-positional relationships
of subsequent fields in the listing
remain undisturbed. During the assembly
process, MACR0-11 examines each source
line for possible error conditions. For
any line in error, an appropriate error
flag is printed preceding the
line
sequence number field (see Appendix D).
MACR0-11 does
not
assign
sequence
numbers for files that have had sequence
numbers assigned by other programs, such
as an editor.

LOC*

List

Controls the listing of the current
location counter field. Normally, this
field is not suppressed. However, if it
is suppressed through the .NLIST LOC
directive, MACR0-11 does not generate a
tab, nor does it allocate space for the
field, as is the case with the source
line
sequence
number
field
(SEQ)
described above. Thus, the suppression
of the current location counter (LOC)
field effectively left-justifies
all
subsequent
fields
(while preserving
inter-positional relationships) to that
position otherwise normally occupied by
this field.

BIN*

List

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

BEX

List

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

SRC*

List

Controls the listing of source lines.
6-3

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

Default

Function

COM

List

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

List

Controls
the
listing
of
macro
definitions and repeat range expansions.

List

Controls the listing of macro calls
repeat range expansions.

No list

Controls
the
expansions.

MEB

No list

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

CND

List

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

No list

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

TOC

List

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

SYM

List

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

TTM

List

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

listing

and
macro

* If the .NLIST arguments SEQ, LOC, BIN, and SRC are in effect at the
same time, i.e., if all four significant fields in the listing are to
be suppressed, the printing of the resulting blank line is inhibited.
6-4

GENERAL ASSEMBLER DIRECTIVES
An example of an assembly listing, as sent to a 132-column line
printer, is shown in Figure 6-1. Note that binary extensions for
statements generating more than one word are formatted horizontally on
the source line.
An examplE~ of an assembly listing, as sent to a teleprinter (in the
same format as for an BO-column line printer), is shown in Figure 6-2.
Notice that binary extensions for statements generating more than one
word are printed on subsequent lines. There is no explicit truncation
of output to 80 characters by the assembler.
Any argument specified in a .LIST/.NLIST directive other than those
listed in Table 6-1 causes the directive to be flagged with an error
code (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
Chapter 8). The use of these switches overrides all corresponding
listing control (.LIST or .NLIST) directives specified in the source
program.

6-5

CSITST .... TEST OF CSI1 ANC CSI2 MACRO M0707
READ ANO PARSE COMMAl\IO LINES

°'I
O"I

209
210
211 11101230
212 001244
213 001246
214 001254
215 OH!l1J00
216 001324
217 001J26
218 001332
219 001336
220 001360
221 001404
222 001412
223 001416
224 001424
225 001450
226 001474
227
228 0014T6
229 001502
230 001504
231 001512
232 001520
233 001522
234 001~26
235 001552
236 001576
237 001602
238 001620
239 001622
240 001626
241 001634
242 001636
243 001662
244 001666
245 001104
246 001706
247 001712
248 001720
249 001722

09•JUL .. 74 15147

GETLN I
103003
Ur
103064
016046
166016

1211110020
00121004

066060
162660
166060

01110016
00'11002
111111001 b

01210020
000002

000655
005760
001652
112767
132760
001402
10!5267

0111011102
000060
11100040

251
176432
000001

176416
10Sr

, SB TTL

READ ANO PARSE COMM01D LINES

GCMLS
BCC
E.XIUS
TYPE
CSISl
BCC
MO\/
SUB
TYPE
TYPE
ADD
SUB
Sl.JB
TYPE
TYPEM
BR

#GCLBLI<
1S

TST
BE.Q
M0\18
BITB
BEQ
INCB
TYPEM

JCHECK LENGTH OF LINE
C,CMLDCR0)
GETLN
Jlf NULL, SKIP BACK FOR NEXT LINE
#•0,EQUBIT
JASSUME EQUAL SIGN NOT FOUND
•CS,EQU,C,STATCR0) JCriECK STATUS
JSKIP IF EQUAL SIGN NOT SE.EN
105
EQUBIT
I E.LSE, INDICATE EQUAL SIGN FOUND
JS~NO EQUAL SIGN STATUS MESSAGE
EQU,40
OPT,40
JS(NO OUTPUT SCAN MESSAGE
JINIT LOCNS FOR CSI2 CALL/TEST
INIT2
,OUTPUT,#SWTBL JPARSE OUTPUT SPEC
CS2ERR
JSl<IP ON ERROR
JE\IALUATE RESULTS Of SE~ANTIC PARSE
EVALU8
#CS,MOR,C,STAT(R0)
IADDITIONAL OUTPUT SPECS?
OPARSE
JYES, CONTINUE WITH OUTPUT SCAN
JSEND INPUT SCAN MESSAGE
IPT,40
JINIT LOCNS FOR CSI2 CALL/TEST
!NIT2
,INPUT,#SWTBL
IPARSE INPUT SPEC
JSl<IP ON ERROR
CS2E.RR
JEVALUATE RESULTS OF SEMANTIC PARSE
EVALU8
#CS,MOR,C,STAT(R0)
JADDITIONAL INPUT SPECS?
I PARSE
JYES, CONTINUE WITH INPUT SCAN
JMPGET
JGET ANOTHER COMMAND LINE

TYPEM

103441
132760
01111360

000020

000001

103407
132760
001360
000412

000020

Figure 6-1

000001

PAGE 5

OPARSE1 CALL
CSI52
BCS
CALL
BITS
BNE
TYPEM
lPARSEz CALL
CSIS2
BCS
CALL

ans
BNE
BR

JGET LINE \/IA GCML
JSKIP IF NO ERROR
J ELSI:., E:.XIT
G,CMLD•2(R0),G,CMLDCRl1!),#!0
JSENO OUT TME INPUT LINE
#CSIBLl<,GCLBLK•G,CMLD+2,GCLBLK•G,CMLO
JBRANCH IF NO ERROR DETECTED
25
JPUT STRING ERROR ADOR IN STK
c.FILD+2CR0),~CSP)
C,CMLD+2'.K0),CSP)
JCALCULATE LENGTH OF FIRST PART
C,CMLD+2(R0),(S~),#15
JSENO OUT FIRST PART OF STRING
JSENO OUT SECOND PART
C,FILD+2(R0),C,FILDCR0),•'5
C,fILOCR0),C,fILD+2CR0) JCALC AODR OF LAST PART OF STRING
(SP)+,C,CMLO(R0)
JDEDUCT LENGT~ OF FIRST PART
C,FILDCR0),C,CMLDCR0)
JCALC LENGTH OF LAST PART
C,FILD+2CR0),C,CMLDCR0),#40
JSEND OUT LAST PART
JSENO SYNTAX ERROR MESSAGE
STX,40
JTRY FOR MORE
GETLN

Example of Line Printer Assembly Listing

l:!j

z
l:!j

t"i

)lll

en
en

l:!j

s:
tJ:I

t"i

l:!j

0
H
~

l:!j
()

<:
l:!j
C/l

GENERAL ASSEMBLER DIRECTIVES
CSITST •• TEST OF CSil ANO CSI2 MACRO M0707
READ ANO PARSE COMMAND LINES
209
210
211 01111230
212 001244

GETL.N I
103003

213 001246
111

214 001254
21 !5 001300
216 001324

00132~

220 001360
221 00141114

READ ANO PARSE COMMAND L.INES

GCML.S
BCC
EXIT SS
TYPE
CSl Sl

•GCLBLK
1S

MOV

IGET LINE VIA GCML
JSK!P IF NO ERROR
IELSE, EXIT
G1 CML0•2CR0),G,CMLOCR0>,•'0
ISENO OUT THE INPUT LINE
*CSIBL.K,GCLBLK•G,CML0•2,GCLBLK•G,CMLD
2S
IBRANCH IF NO ERROR DETECTED
C,FIL0•2CR0),•(SP)
JPUT STRING ERROR AOOR IN STK

SUB

ICALCULATE LENGTH OF FIRST PART

TYPE
TYPE
AOO

066060

PAGE 5

,SBTTL.

ace

103064
016046
000020
218 001:332 ' 166016
000004
219 001336

217

09•JUL•74 15159

C,CML0•2CR0),(SP),*'S
ISEND OUT FIRST PART OF STRING
C,FIL0•2CR0) 1 C,FILDCR0) 1 #'$
JSEND OUT SECOND PART
C,FILD(R0),C,FILD•2CR0) ICALC AODR OF LAST PART OF STRING

000016
222 001412

22.3 001416

000020
162660
000002
166060

IOEOUCT LENGTH OF FIRST PART

SUB

ICALC LENGTH OF LAST PART

SUB

C,FILDCR0),C,CMLDCR0)

TYPE
TYPEM
BR

C,FILD•2CR0),C,CMLOCR0) 1 #40
JSEND OUT LAST PART
STX,40
JSENO SYNTAX ERROR MESSAGE
GETLN
JTRY FOR MORE

TST

C,CMLOCR0)

JCHECK LENGTH OF LINE

BEQ
MOVB

GETLN
#t0,EQUBIT

J!F NULL., SKIP BACK FOR NEXT LINE
JASSUME EQUAL SIGN NOT FOUND

BITB

•Cs,Eau,C,STAT(R0)

BEQ
INCB

105
EC.lUB IT

JSKIP IF EQUAL SIGN NOT SEEN
JELSE, INDICATE EQUAL SIGN FOUND

EQU,40
OPT,40
INIT2
,OUTPUT,#SwTBL
CS2E'RR
EVALU8

JSEND EQUAL SIGN STATUS MESSAGE
JSEND OUTPUT SCAN MESSAGE
JINIT LOCNS FOR CSI2 CALL/TEST
JPARSE OUTPUT SPEC
JSKIP ON ERROR
JE\IALUATE RESULTS OF SEMANTIC PARSE

000016
000002

224 001424
225 0014f>0
226 001414

000655

227

228 001476

01115760
000002

229 00151112
230 001504

001652

2$1

112767

000061'/1
231 001512
232 017.J 1520
233 001!522

176432
132760
000040
000001
001402
105267
176416

235 001552
236 11101576
237 0i"1602

238 001620

103441

239 001622
240 001626

132760
01110020
000001

241 001634

001360

242 001636
243 01111662
244 00166b
245 001104
246 001706
247 0CH712

TYPEM
TYPEM
OPARSEI C4LL
CSIS2
BCS
CALL
10$ I

234 01111526

103407
132760

BITB
BNE
TYPEM
IPARSE1 CALL
CSIS2
BCS
CALL
BITS

#CS,MOR,C,STAT(R0)

JCHECK STATUS

JAOOITIONAL OUTPUT SPECS?

OPARSE

JYES, CONTINUE wITH OUTPUT SCAN
IPT,4~
JSENO INPUT SCAN MESSAGE
lN?T2
JINIT LOCNS FOR CSI2 CALL/TEST
,INPUT,#SWTBL
IPARSE INPUT SPEC
CS2ERR
JSKIP ON ERROR
EVALU8
JEVALUATE RESULTS OF SEMANTIC PARSE
#CS,MOR,C,STATCR0) IADDITIONAL INPUT SPECS?

000020
248 001720

000001
001360

Figure 6-2

BNE

I PARSE

IYES, CONTINUE WITH INPUT SCAN

Example of Terminal Assembly Listing

6-7

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

6.1.2

Page Headings

MACR0-11 prints each assembly page in the format shown in either
Figure 6-1 or Figure 6-2, depending on the listing mode (see TTM,
Table 6-1). On the first line of each page, MACR0-11 prints the
following (from left to right):
1.

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

Assembler version identification.

Date.

Time-of-day.

Page number.

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

6-8

2;
28

29 00111062
000062
000010

000001
0!210011!4

000002
000005

000003

30
31
32 000074
01110074

,MAIN,

°'I
l.O

000001

"4ACRO M0707

000002

000003

09 .. JUL .. 74 16129

LSTMAC
,NLIST
,llljORO

COM
COM
1,2,3,4,5

,LIST

COM

LSTMAC
, NLIST
1 lliORD
,LIST

<COM,BEX>
COM,BEX
1,2,3,4,5
COM,BEX

ICOMMENT LINES TEST

JCOMMENT LINES AND EXTENDED BINARY TEST

PAGE. 1-1

33
34
35
36

,LIST

TT"4

JNARROW LISTING MOOE IS IN EFFECT

LSTMAC
, NLIST
llljQRO
1

SEQ
SEQ
1,2,3,4,5

JSEQUENCE NUMBERS TEST

,LIST

SEQ

0!1J0001

l.STMAC
,NL.IST
, 1111QRD
,LIST

BEX
BEX
l,2,3,4,5
BEX

000001'

,ENO

37
38 000106

000106
000110
000112
000114
000116

000001
000002

JTHIS IS A COMME.NT

~00003

000004
000005

40
41 000120
000120
42
43
44

JEXTENOED BINARY TEST
~THIS

IS A C0"4MENT

Figure 6-3 Listing Produced With Listing Control Directives

MACRO M0707

MA I...,.

09 .. JUL•74 16129

PAGE 1
,NLIST
,LIST

2
3
4

LISTING CONTROL TEST MACRO

6
1
8
9

•"'ACRO
,NL I ST
I lllORD
,LIST
,ENOM

10
11
12
13
14
15
16
17

°'I

LSTMAC ARG
ARG
l.2,3,4,5
ARG

JTHIS IS A COMMENT

Cl
tEl

2:
tEl

t""

>
tJl
0~0012

000001
000004

I-'
0

LISTING MOOE IS IN EFFECT
JLIST MACRO EXPANSIONS

1 Iii I DE

TT"'1

000002
1210121005

,NL I ST
,wORD

000003

LSTMAC LOC
LOC
1,2,3,4,5
,LIST

18
19

JLOCATION COUNTER TEST
JTHIS IS A COMMENT

tJl
tEl

tXJ

t""

tEl

LDC

20 000024

LST"'AC
,NLIST
,wORD

000024

BIN
l,2,3,4,5

21
22
23 01210036
0121012136

1210121001

000002

000003

24
25
26 0000!50
000050
000056

000001
000004

01i'10002
000005

BIN

JGENERATED BINARY TEST

JT!o!!S !S A COMMENT
BIN
,LIST
LSTMAC
I NLI ST
•"'ORD
,LIST

BEX
BEX
1,2,3,4,5

LSTMAC

SRC

IL. I ST

SRC

H
~

tEl

()

1-3
H

JEXTENDED BINARY TEST
JTHIS IS A COMMENT

BEX

JSOURCE LINES TEST

000003

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

<
tEl

tJl

GENERAL ASSEMBLER DIRECTIVES
6.1.3

.TITLE Directive

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

PROGRAM TO PERFORM DAILY ACCOUNTING

causes the assembled object modul~ to be named.PROGRA. Note that this
6-character name bears no relationship to the filename of the object
module, as specified in the command string to MACR0-11. The name of
an object module
(specified in the .TITLE directive) appears in the
Task Builder load map.
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 establishes the name for the entire object module.
All spaces and/or tabs up to the first non-space/non-tab character
following the .TITLE directive are ignored by MACR0-11 when evaluating
the text string.
If the .TITLE directive is specified without an object module name, or
if the first non-space/non-tab character in the object module name is
not a Radix-50 character, the directive is flagged with an error code
(A) in the assembly listing.
Section A.2 of Appendix A contains a table of Radix-50 characters.

6.1.4

.SBTTL Directive

The~

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

CONDITIONAL ASSEMBLIES

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

second

line

the

header

the

assembly

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

6-11

GENERAL ASSEMBLER DIRECTIVES

CS?TST •• TEST OF CS?t ANO CSl2 MACRO M0707
TABLE OF CONTENTS

MACRO DEFINITIONS

2• 55
3• 74
••153

MESSAGE STRINGS

DATA
READ ANO PARSE COMMAND LINES
EVALUATE THE SEMANTIC ANALYSIS
MISCE~LANEOUS

5•209
6•255
7•345

SUBROUT l NES'

Figure 6-4

6.1.5

091111JUL•74 15147

Assembly Listing Table of Contents

.!DENT Directive

The .IDENT directive provides an additional means of labeling the
object module produced by MACR0-11.
In addition to the name assigned
to the object module with the .TITLE directive (see Section 6.1.3), a
character string up to six Radix-50 characters can be specified
between paired printing delimiters to label the object module with the
program version number.
This directive takes the following form:
.IDENT
where:

/string/

string

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

represent delimiting characters.
These delimiters
may be any paired printing characters, other than
the equal sign (=), the left angle bracket (<), or
the semicolon
(;), 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
.!DENT directive is flagged with an
error code (A) in the assembly listing.

An example of the .!DENT directive is shown below:
• ID ENT

/VO SA/

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

6-12

GENERAL ASSEMBLER DIRECTIVES
6.1.6

.PAGE Directive/Page Ejection

Page ejection is accomplished in one of four 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.

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

6.2

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

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

FUNCTION DIRECTIVES:

.ENABL AND .OSABL

Several function control options are provided by MACR0-11 through the
.ENABL and .DSABL directives.
These directives are included in a
source program to invoke or inhibit certain MACR0-11 functions and
operations
incidental
to
the assembly process itself.
These
directives take the following form:
.ENABL
.DSABL
where:

arg

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

6-13

symbolic

GENERAL ASSEMBLER DIRECTIVES

Table 6-2
Symbolic Arguments of Function Control Directives
Argument

Default

Function

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

CRF

Enable

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

FPT

Disable

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

Disable

Enabling this function causes MACR0-11
to accept lower-case ASCII input instead
of converting it to upper-case. If this
function is not enabled, all text is
converted to upper-case.

LSB

Disable

This argument permits the enabling or
disabling
of a local symbol block.
Although a local
symbol
block
is
normally established by encountering a
new symbolic label or a .PSECT directive
in the source program, an .ENABL LSB
directive establishes a new local symbol
block which is not terminated until (1)
another .ENABL LSB is encountered, or
(2) another symbolic label or .PSECT
directive is encountered following a
paired .DSABL LSB directive.
Although the
.ENABL
LSB
directive
permits a local symbol block to cross
.PSECT boundaries, local symbols cannot
be defined in a program section other
than the one that was in effect when the
block was entered. The basic function
of this directive
with
regard
to
.PSECT's is limited to those instances
6-14

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

Default

Function

Disable

where it is desirable to leave a program
section temporarily to
store
data,
followed by a return to the original
program section.
Attempts to define
local symbols in an alternate program
section are flagged with an error code
(P) in the assembly listing.
An example of the .ENABL LSB and .DSABL
LSB directives, as typically used in a
source program, is shown in Figure 6-5.

PNC

Enable

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

REG

Enable

When specified, the .DSABL REG directive
inhibits the normal 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.

GBL

Enable

When the .ENABL
GBL
directive
is
specified, MACR0-11 treats all symbol
references that are undefined at the end
of assembly pass 1 as default global
references;
when
the
.DSABL
GBL
directive is specified, MACR0-11 treats
all
such
references
as
undefined
symbols.
In assembly pass 2, if the
.DSABL GBL function is still in effect,
these undefined symbols are flagged with
an error code (U)
in the
assembly
listing; otherwise, they continue to be
regarded
by
MACR0-11
as
global
references.

Specifying any argument in an .ENABL/.DSABL directive other than those
listed in Table 6-2 causes that directive to be flagged with an error
code (A) in the assembly listing.
6-15

SQUEEZE MACRO M0707

O"I

I
I-'
O"I

11J9•JUL•7 4 t 5 rt 3

272
273
274
275
276
277
278 f/Jf/JJ142 0UH11J3
279 003144 060203
280 003146 020311.11
281 003150 001422
282 003152 124327
283 003156· 001373
284 1111213160 010302
285 01113162 000412
286
287 003164 060102
288 003166 020201
289 003170 001412
290 003172 124227
291 003176 001773
292 003200 121227
293 011Jl204 001770
294 003206 005202
295 003210 160102
296 003212 000241
297 2103214 000401
298 ""3216 000261
299 003220
300

PAGE 4

.ENABL

0011Hi'J73

00011! 11
li!lfl0040

FNDSMI1 MOV
ADO
CMP
lSI
BEQ
CMPB
BNE
MOV
BR
SKPBLKI ADO
CMP
l0S1
BEQ
CMPB
BEQ
CMPB
BEQ
INC
2051
SUB
CLC
BR
3051
SEC
40Se
RETURN
8 0SABL

LSB
Rt,R3
R2,R3
RJ,R1
30S
•CR3) 1 #SEMIC
lS

RJ,R2
20S
Rt,R2
R2,Rl
30S
•CR2) ,#TAB
105

CR2),#BLANI<
l0S

R2
Rt,R2
40S

JPUT AOOR OF LINE IN R3
JPOINT R3 PAST LAST CHAR IN LINE
IOOES RJ POINT TO START OF LINE?
IIF SO, LEAVE INDICATING FAILURE
JIS THE LAST CHARACTER SEMICO~ON?
JNO, CONTINUE LOOKING
JYES, POINT R2 PAST NE~ ENO•OF•LINE
ILEAVE VIA COMMON SUCCESS CODE
JPOINl R2 PAST ENO•OF•LINE
JOOES R2 POINT TO START OF LINE?
JIF SO, LEAVE WIT~ FAILURE
JIS THE LAST CHARACTER A TAB?
JIF SO, IGNORE IT
JIS IT A BLANK?
JIF SO, IGNORE IT
JNON•BLANK CMARACTER••POINT PAST. IT
JRE•COMPUTE LINE LENGTH
JINDICATE SUCCESS
JBRANCH TO LEAVE.
JINDICATE FAILURE

LSB

301
302

Ci)

t:lJ

t"'

)it

m
m
t:lJ

3:
tD
t"'

t:lJ
~

ct-1
~

t:lJ

t-3

t-1

<
t:lJ

Figure 6-5 Example of .ENABL and .DSABL Directives

GENERAL ASSEMBLER DIRECTIVES
6.3

DATA STORAGE DIRECTIVES

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

.ASCII
.ASCIZ
.FLT2
.PLT4
.RAD50
"B

"C
"D
"F'

"O
"R

These MACR0-11 facilities are described in the following sections.

6.3.l

.BYTE Directive

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

binary

.BYTE

exp

;STORES THE BINARY VALUE OF THE
;EXPRESSION "EXP" IN THE NEXT BYTE.

.BYTE

expl,exp2,expn

;STORES THE BINARY VALUES OF THE LIST
;OF EXPRESSIONS IN SUCCESSIVE BYTES.

A legal expression must reduce to eight bits of data or less.
The
operands of a .BYTE directive are evaluated as word expressions before
being truncated to the low-order eight bits. The 16-bit value of the
specified expression must have a high-order byte (which is truncated)
that is either all zeros (0) or all ones (1). Each expression valqe
is
stored
in the next byte of the object module.
Multiple
expressions, which must be separated by commas, are stored in
successive bytes, as described below:
SAM.=5
.=410

.BYTE

"048,SAM

;THE VALUE 060 (OCTAL EQUIVALENT OF 48
;DECIMAL) IS STORED IN LOCATION 410.
;THE VALUE 005 IS STORED IN LOCATION
; 411.

If the high-order byte of the expression reduces to a value other than
0 or -1, the value is truncated to the low-order eight bits and
flagged with an error code (T) in the assembly listing.
The construction "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.
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
6-17

,.;\

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

;STORES OCTAL 23 IN NEXT BYTE.

.BYTE

;RELOCATABLE VALUE A WILL PROBABLY
;CAUSE TASK BUILDER TRUNCATION
;DIAGNOSTIC.

For

If an expression following the
.BYTE directive
interpreted as a zero, as described below:

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.

6. 3. 2

.WORD Directive

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

data

.WORD

exp

;STORES THE BINARY EQUIVALENT OF THE
;EXPRESSION EXP IN THE NEXT WORD.

.WORD

expl,exp2,expn

;STORES THE BINARY EQUIVALENTS OF THE
;LIST OF EXPRESSIONS IN SUCCESSIVE
;WORDS.

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

177535,.+4,SAL

;STORES THE VALUES 177535, 506, AND
;0 IN WORDS 500, 502, AND 504,
;RESPECTIVELY.

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

'5'

;STORES THE VALUES O, 5, AND 0 IN
;LOCATION 500, 502, AND 504,
;RESPECTIVELY.

A statement containing a blank operator field, i.e., a symbol that is
not recognized by MACR0-11 as a macro call, an instruction nmemonic, 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.

6-18

GENERAL ASSEMBLER DIRECTIVES
CAUTION
You should not use this technique ~o
generate .WORD directive~s because it ma_y
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. When so used,
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 16-bit value is then used as an absolute term
within the expression.
For example, the statement:
MOV

#'A,RO

results in the following 16-bit
register 0:

expression

value

being

moved

into

00000000 01000001
Binary Value of ASCII A
Thus, in the example above, the expression 'A results in a value of
101(8).
Note that the high-order byte is always zero (0) in the
resulting expression value when the single quote unary operator is
used.
The ' character must not be followed by a
RUBOUT,
line-feed, or form-feed character~
(A) is generated in the assembly listing.

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

When the double quote is used, MACR0-11 takes the next two characters
in the expression and converts them to a 16-bit binary expression
value from their 7-bit ASCII values.
This 16-bit value is then used
as
an absolute term within the expression.
For example,
the
statement:
MOV

#"AB,RO

results in the following 16-bit
register 0:

expression

value

being

moved

into

01000010 01000001

Binary Value of ASCII A
Binary Value of ASCII B
Thus, in the example above, the expression "AB results in a
041101 ( 8).
6-19

value

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

6u3.4

.ASCII Directive

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

/string l/ ... /string n/

string

is a string of printable ASCII characters. All
printable ASCII characters
are
legal.
The
vertical-tab, null, line-feed, RUBOUT, and all
other non-printable ASCII
characters,
except
carriage-return
and
form-feed,
are
illegal
characters.
Such
an
illegal
non-printing
character is flagged with an error code (I) in the
assembly
listing.
The
carriage-return
and
form-feed characters terminate the scan of the
source line. This premature termination of the
.ASCII statement results in the generation of an
error code (A) in the assembly listing, because
MACR0-11 is unable to complete the scan of the
matching delimiter at the end of the character
string.

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

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

only
Each
For

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

the expressions <15>, <A+2>, <5>, and <4> represent the values of
non-printing characters. Furthermore, the expressions must reduce to
eight bits of absolute data or less, subject to the same rules for
generating data as with the .BYTE directive (see Section 6.3.1).
Angle brackets can be embedded between delimiting characters in the
character string, but angle brackets so used do not take on their
usual significance as delimiters for non-printing characters.
For
example, the statement:
.ASCII

/ABC<expression>DEF/

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:
6-20

GENERAL ASSEMBLER DlRECTIVES
.ASCII

/HELLO/

;STORES THE BINARY REPRESENTATION
;OF THE LETTERS HELLO IN FIVE
;CONSECUTIVE BYTES .

. 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.

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

;ABC;/DEF/

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

. 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

6.3.5

=DEF=

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

.ASCIZ Directive

The .ASCIZ directive is equivalent to the .ASCII directive described
above, except that a zero byte is automatically inserted as the final
character of the string.
Thus, when a list or text string has been
created with an .ASCIZ directive, a search for the null character in
the last byte can effectively determine the end of the string, as
reflected by the coding below:
CR=l5
LF=l2
HELLO:

10$:

.ASCIZ
.EVEN

MOV
MOV
MOVB
BNE

#HELLO,Rl
#LINBUF,R2
(Rl)+,(R2)+
10$

;INTRODUCTORY MESSAGE

;GET ADDRESS OF MESSAGE.
;GET ADDRESS OF OUTPUT BUFFER.
;MOVE A BYTE TO OUTPUT BUFFER.
;IF NOT NULL, MOVE ANOTHER BYTE.

6-21

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

6.3.6

.RADSO Directive

The .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.
The form of the directive is:
.RAD50
where:

/string l/ ... /string n/

string

represents a series of characters to be packed
{three characters per word).
The string must
consist of the characters A through z,
0 through
9,
dollar sign { $) , period {.) and space { } .
An
illegal printing character causes an error
flag
{Q) to be printed in the assembly listing.
If fewer than three characters are to be packed,
the string is packed left-justified within the
word, and trailing spaces are assumed.
As with the .ASCII directive described in Section
6.3.4,
the vertical-tab, null, line-feed, RUBOUT,
and all other non-printing characters,
except
carriage-return
and
form-feed,
are
illegal
characters, resulting in an error code (I} in the
assembly listing.
Similarly, the carriage-return
and form-feed characters result in an error code
(A}
because these characters end the scan of the
line, preventing MACR0-11 from detecting
the
terminating matching delimiter.

represent delimiting characters.
These delimiters
may be any paired printing characters, other than
the equal sign (=}, the left angle bracket (<}, or
the semicolon
(;},
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 listing.

Examples of .RAD50 directives are shown below:
.RAD50
.RAD50
.RAD50

/ABC/
/AB/
/ABCD/

.RAD50

/ABCDEF/

;PACKS ABC INTO ONE WORD.
;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.

6-22

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

its

Radix-50

Character

Radix-50 Octal Equivalent

(space)
A-Z

0
1-32
33
34
35
36-47

(undefined)
0-9

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

through

equivalent,

(Cl,C2,C3)

are

((Cl*50)+C2)*50+C3

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

A.2

Appendix

for

table

Radix-50

Angle brackets (<>) must be used in the .RAD50 directive whenever
special codes are to be inserted in the text string, as shown in the
example below:
.RAD50

/AB/<35>

;STORES 3255 IN ONE WORD.

.. RAD50

<CHRl><CHR2><CHR3> ;EQUIVALENT TO .RAD50 /ABC/.

CHRl=l
CHR2=2
CHR3=3

6 .. 3.7

Temporary Radix-50 Control Operator:

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 ftR operator is coded as follows:

where ccc represents a maximum of three characters to be converted to
a 16-bit Radix-50 value. If more than three characters are specified,
any following the third character are ignored. If fewer than 3 are
specified, it is assumed that the trailing characters are blanks. The
following example shows how the "R operator might be used to pack a
3-character file type specifier (MAC) into a single 16-bit word.
MOV

#ARMAC,FILEXT

;STORE RAD50 MAC AS FILE EXTENSION

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

6-23

GENERAL ASSEMBLER DIRECTIVES
6.4

RADIX AND NUMERIC CONTROL FACILITIES

6.4.1

Radix Control and Unary Control Operators

The normal default assumption for numeric values or expression values
appearing in a MACR0-11 source program is octal.
However, numerous
instances may occur where an alternate radix is useful for portions of
a program or for variables within a given statement.
It may be
useful, for example,
to declare a given radix for applicability
throughout a program or to specify a numeric value or expression value
in a manner that causes it to be interpreted as a binary, octal, or
decimal value during assembly.
In other such instances, it may be
useful to complement numeric values or expression values.
These
MACR0-11 facilities are described in the following sections.
NOTE
When two or more unary operators appear together,
modifying
the
same term,
the operators are
applied, from right to left, to the term.

6.4.1.1
.RADIX Directive - Numbers used in a MACR0-11 source program
are initially considered to be octal values; however, you can declare
any one of the following radices for
applicability throughout the
source program or within specific portions of the program:
2, 8, 10

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

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

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

. RADIX

;BEGINS A SECTION OF CODE HAVING A
;DECIMAL RADIX.

;REVERTS TO OCTAL RADIX .

Any value other than null, 2, 8, or 10 specified as an argument in the
.RADIX directive causes an error code
(A) to be generated in the
assembly listing.
In general, macro definitions should not contain or rely on radix
settings established with the
.RADIX directive.
Rather, temporary
radix control operators should be used within a macro definition.
Where a possible radix conflict exists within a macro definition or in
possible future uses of that code, it is recommended that the user
6-24

GENERAL ASSEMBLER DIRECTIVES
specify numeric or expression values using the temporary radix control
operators described below.

6.4.1.2 Temporary Radix Control Operators: ~D, no, and nB - Once the
user has specified a given radix for a section of code or has decided
to use the default octal radix, he may discover a number of cases
where an alternate radix is more convenient or desirable (particularly
within macro definitions). The creation of a mask word, for example,
might best be accomplished through the use of a binary radix.
MACR0-11 has three unary operators that allow the user to establish an
alternate radix, as shown below:
"D 11 number 11
"0 11 number 11
"B 11 number 11

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

Thus, an alternate radix can be declared temporarily to meet a
localized requirement in the source program. Such a declaration can
be made at any time, regardless of the existence of the default octal
radix or andther specific radix declaration elsewhere in the program.
In other words, the effect of a temporary radix control operator is
limited to the term or expression immediately following the operator.
Any value specified in connection with a temporary radix control
operator is evaluated during assembly as a 16-bit entity. Temporary
radix control declarations can be included in the source program
anywhere a numeric value is legal.
The P.xpressions below are representative of the methods of
temporary radix control operators:
"0123
"O 47

"B 00001101
"O<A+13>

specifying

Decimal radix
Octal Radix
Binary Radix
Octal Radix

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

.WORD

"O<A+lO>*lO

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

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)
6-25

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

listed

"0100
"01376
"0128

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 POP-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.
'.rhe string (which can be a single digit in length) may optionally
contain a decimal point, and may be followed by an optional exponent
indicator in the form of the letter E and a signed decimal integer
exponent. The number may not contain embedded blanks, tabs or angle
brackets and may not be an expression. Such a string will result in
one or more errors (A or Q) in the assembly listing.
The list of numeric representations below contains seven
valid representations of the same floating-point number:

distinct,

3
3.

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

EEEEEEEE

the

following

MMM ..... MMM
Mantissa
Exponent
Sign

(55 bits)
(8 bits)
(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).
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
6-26

GENERAL ASSEMBLER DIRECTIVES
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 6-2).
Bit
32

Bit
0
Retained
field

Bit
32 31

lI
l

Bit
0
Unretained
field

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

#A041040,F0
FO is assumed to represent floating accumulator zero.

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

greater

detail

the

6.ii.2.1 .FLT2 and .FLT4 - Floating-Point Storage Directives - MACR0-11
supports two directives that evaluate successive
floating-point
numbers and
store
the
results
in the object module. These
directives are similar to the .WORD directive and are of the
form:
. FLT2
. FLT4
where:

argl,arg2, •..
argl,arg2, ••.

argl,arg2, ...

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

. F1~T2 causes two words of storage to be generated for each argument,
while .FLT4 generates four words of storag~ for each argument.

6.4.2.2 Temporary Numeric Control Operators:
ftc and ftF - The AC
unary operator allows you to specify an argument that is to be
complemented as it is evaluated during assembly.
The AF unary
ope~rator
allows you to specify an argument consisting of a 1-word
floating-point number.
As with the radix control operators described above,
the numeric
control operator (AC) 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

causes the l's complement of the value 151 (octal) to be stored as a
16-bit value in the program.
The resulting value expressed in octal
form is 177626(8).
6-27

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

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

is equivalent in function to:

This expression is evaluated during assembly as the 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.
MACR0-11 also supports a unary operator
for
numeric control which
allows
you
to
specify
an
argument consisting of 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.
BIT 15

EEEEEEEE

MMMMMMM

Sign (bit 15)

Exponent (bits 14-7)

Mantissa (bits 6-0)

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

the

operators

shown

020400
120400
157400
057400

The value created by the AF unary operator and its argument is then
term that can be used by itself or in an expression.
For example:

is equivalent to:

6-28

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

6.5

LOCA'rION COUNTER CONTROL DIRECTIVES

The directives used in controlling the value of the current location
counter and in reserving storage space in the object program are
described in the following sections.
In this connection, it should be noted that several
MACR0-11
statements may cause an odd number of bytes to be allocated, as listed
below:
1.

.BYTE directive

.BLKB directive

.ASCII or .ASCIZ directive

.ODD directive

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

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

6.5.1

.EVEN Directive

The .EVEN directive ensures that the current location counter contains
an even value by adding 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:

6.5.2

.ASCIZ
.EVEN

/THIS IS A TEST/

. WORD

XYZ

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

.ODD Directive

The .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.

6-29

GENERAL ASSEMBLER DIRECTIVES
6.5.3

.BLKB and .BLKW Directives

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

where:

exp
exp
represents the specified number of bytes or words
to be reserved in the object program. If no
argument is present, a default value of 1 is
assumed.
1'~es_e_
directives should not be used
without arguments.
Any
expression
that
is
completely
defined at assembly-time and that
reduces to an absolute value is legal.
If the
expression specified in either of these directives
is not an absolute value, the statement is flagged
with an error code (A) in the assembly listing.

Figure 6-6 illustrates the use of the .BLKB and .BLKW directives.

168 000000

PASS11

167 1211/JeJeJl!l 0

188
169 121000QJl!J
170
1''1
172
173

00001210
000004
000004
000005
11• 000006
11!i 000010

176
177
178
179
180
181
182
183
184
18!5
18&

000003

000020
000024
000025
00111026
llt00030

000032
00011134
181 IH0036
188 000000

,PSECT

lMPURE,O

EU.KW

,PSECT
SYMBOL11,BL.l<W
MOOE11
FL.AGS11 ,BL.KB
SECTOR11,BL.l<B
VALUEll ,BLKW
RELLVLI I, BL.KW
,REPT
,BLKW
,ENOR
CLCNAM11,BLKW
CLCFGS11 1 BLl<B
CLCSEC11,BL..KB
CLCLOC11,BL.KW
CLCMAX11,BL.KW
CHRPNT11,BLKW
SYMBEG11,BLKW
ENDFLG11,BL.KW
,PSECT

Figure 6-6

IMPPAS,O,GBL.

JPASS FLAG
INEXT GROUP MUST SUV TOGETHER
ISYMBOL ACCUMULATOR
IMOOE/FLAGS BYTE

I
IS YMBOL. /E XP·RE SS I ON

IEXPRESSION VALUE
JRELOCATION LEVEL
MAXXMT•<< 1 •SYMB0L>/2>

TYPE

JCURRENT LOCATION COUNTER NAME

2
1

'
OF GROUPED DATA
'JENO
JCHARACTER POINTER

1
1
1
1
1

JPOINTER TO START OF SYMBOL

Example of .BLKB and .BLKW Directives

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

same

effect

the

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

6-30

GENERAL ASSEMBLER DIRECTIVES
6.6

TERMINATING DIRECTIVES

6.6.1

.END Directive

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

and

exp

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

When MACR0-11 encounters a valid occurrence of the .END directive,
it
terminates the current assembly pass. Any additional text beyond this
point in the current source file, as well as in additional source
filE~S identified in the command line, will be ignored.
When creating a task image consisting of several object modules, only
one object module may be terminated with an
.END exp statement
specifying the starting address. All other object modules must be
terminated with an
.END statement without an address argument;
otherwise, the Task Builder will issue a diagnostic message.
If no
starting address is specified in any of the object modules, task
execution will begin at location 1 of the task and immediately fault
because of an odd addressing error.
The .END statement must not be used within a macro expansion or a
conditional assembly block;
if it is so used, it is flagged with an
error code (0) in the assembly listing.
The
.END statement may be
used, however,
in an immediate conditional statement (see Section
6.10.2).
If the source program input is not terminated with an .END
an error code (E) results in the assembly listing.

6.6.2

directive,

.EOT Directive

UndEer RSX-11 and IAS operating systems, the MACR0-11 .EOT directive is
ignored and simply treated as a directive without effect, i.e., as a
no-op.

6.7

PROGRAM BOUNDARIES DIRECTIVE:

.LIMIT

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

causing two storage words to be re~erved in the object module.
Later,
at link time,
the address of the bottom of the task's stack is
inserted into the first reserved word, and the address of the first
free word following the task image is inserted into the second
res~erved word.

6-31

GENERAL ASSEMBLER DIRECTIVES
During linking, the size of the task image is rounded
nearest 2-word boundary.

upward

the

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

6.8

PROGRAM SECTIONING DIRECTIVES

The MACR0-11 program sectioning directives are used to declare names
for
program sections and to establish certain program section
attributes essential to Task Builder processing.

6.8.1

.PSECT Directive

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

name,argl,arg2, .•• argn

name

represents the symbolic name of
section, as described in Table 6-3.
represents any legal separator
space) .

argl,
arg2, ...
argn

the

(comma, tab

program
and/or

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

6-32

GENERAL ASSEMBLER DIRECTIVES

Table 6-3
Symbolic Arguments of .PSECT Directive
Argume~nt

Default

Meaning

NAME

Blank

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

RO/RW

Defines which
type
of
access
permitted to the program section:

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

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

GBL/LCL

LCL

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

6-33

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

Default

Meaning

GBL/LCL
(cont'd)

LCL

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

ABS/REL

REL

Defines the relocatability attribute
the program section:

ABS=Absolute (non-relocatable).
When
the ABS argument is specified, the
program section is regarded by the
Task Builder as an absolute module,
thus requiring no relocation.
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
-1't during assembly, may generate a
Task Builder
error
message
if
virtual location 100000 is outside
the segment's virtual address space:
.PSECT
.=.+100000
.WORD

ALPHA,ABS
X

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

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

Default

CON/OVR

CON

Meaning
Defines the allocation
the program section:

requirements

CON=Concatenated. All program section
contributions are to be concatenated
with other references to this same
program
section
in
order
to
determine
the
total
memory
allocation
requirement
for
this
program section.

______._

OVR=Overlaid.
All
program
section
contributions are to be overlaid.
Thus,
the
total
allocation
requirement for the program section
is equal to the largest allocation
request
made
by any individual
contribution
to
this
program
section.
_......__

_ _ _ _ __

The only argument in the .PSECT directive that is position-dependent
is NAME.
If it is omitted, a comma must be used in its place.
For
example, the directive:
.PSECT

,GBL

shows a .PSECT directive with a blank name argument and the GBL
argument.
Default values
(see Table 6-3} are assumed for all other
unspecified arguments.
Once the attributes of a program section are declared through a .PSECT
MACR0-11 assumes that these attributes remain in effect for
all subsequent .PSECT directives of the same name that are encountered
within the module.

dirE~ctive,

MACR0-11 provides for 256(10) program sections, as listed below:
1.

One default absolute program section (. ABS.)

One default unnamed relocatable program section

Two-hundred-fifty-four named program sections.

The .PSECT directive enables the user to:
1.

Create program sections (see Section 6.8.1.1)

Share code and
6.8.1.2).

data

among

6-35

program

sections

(see

Section

GENERAL ASSEMBLER DIRECTIVES
For each program s~ction 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 6-3 above.

i.e.,

the

.PSECT

the

arguments

6.8.1.1 Creating Program Sections - MACR0-11 automatically begins
assembling source statements at relocatable zero of the unnamed
program section, i.e., the first statement of a source program is
always an implied .PSECT directive.
The first occurrence of a .PSECT directive with a given name assumes
that the current location counter is set at relocatable zero.
The
scope of this directive then extends until a directive declaring a
different program section is specified.
Further occurrences of a
program section name in subsequent
.PSECT statements cause the
resumption of assembly where that section previously ended.
For
example:

A··
B··

_c::
x··

y ..

.PSECT
.WORD
.WORD
.WORD
, . PSECT
1 • WORD
\.WORD

--.PSE-c-T

.WORD

0
0
0
ALPHA· (
0
0
0

;DECLARES UNNAMED RELOCATABLE PROGRAM
;SECTION ASSEMBLED AT RELOCATABLE
;ADDRESSES 0, 2, AND 4.
;DECLARES RELOCATABLE PROGRAM SECTION
;NAMED ALPHA ASSEMBLED AT RELOCATABLE
;ADDRESSES 0 AND 2.
;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.
By maintaining separate location counters for
each program section,
MACR0-11 allows the user to write statements that are not physically
contiguous within the program, but that can be loaded contiguously
following assembly, as shown in the following example.

6-36

GENERAL ASSEMBLER DIRECTIVES
.PSE:CT
.WOHD
.WOFD
.WORD
CLR
CLR
CLR
.PSECT

(}

A:
:L B:

q. c:

l<J ST:

't t>
,_..,)1

SECl,REL,RO
0
0

0
A
B

;ASSEMBLE CODE AT RELOCATABLE
;ADDRESSES 6 THROUGH 12.

SECA,ABS

;START AN ABSOLUTE PROGRAM SECTION
;NAMED SECA. ASSEMBLE CODE AT
;ABSOLUTE ADDRESSES 0 AND 2 .
;RESUME RELOCATABLE PROGRAM SECTION
;SECl. ASSEMBLE CODE AT RELOCATABLE
;ADDRESSES 14 AND 16.

.WORD
.PSECT

. +2,A
SECl

INC

;START A RELOCATABLE PROGRAM SECTION
;NAMED SEC! ASSEMBLED M' RELOCATABLE
;ADDRESSES O, 2, AND 4.
-

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

.PSECT

ALT,REL

the label A is assigned to relocatable address zero of the unnamed (or
blank) program section.
It is not known during assembly where relocatable program sections
will be loaded, therefore all references between relocatable sections
in a single assembly are translated by MACR0-11 to references relative
to the base of the referenced section.
Thus, MACR0-11 provides the
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.
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
. =. +10.00
. \ CLR
A:
\

.)_

..r·,.
\-

Y:
X:

ENT,ABS

J'MP

.PSECT
MOV
J'MP
HALT
.WORD

SEN,REL
RO,Rl
A

;ASSEMBLED AS CLR BASE OF
;RELOCATABLE SECTION + 10.
;ASSEMBLED AS JMP BASE OF
;RELOCATABLE SECTION + 6.
;ASSEMBLED AS JMP 1000.

0
NOTE

In the preceding example, using a constant in conjunction
with the GUrrent location counter symbol (.) in the form
.=1000 would result in an error, because constants are
always absolute and are always associated with the program's
.ASECT (. ABS.).
If the form .=1000 were used,
a program
section incompatibility would be detected.
See Section 3.6
for a discussion of the current location counter.
6-.37

GENERAL ASSEMBLER DIRECTIVES
6.8.1.2 Code or Data Sharing - Named relocatable program sections
with the arguments GBL and OVR operate in the same manner as FORTRAN
COMMON, i.e., program sections of the same name with the arguments GBL
and OVR from different assemblies are all loaded at the same location
by the Task Builder. All other program sections, i.e., those with the
argument CON, are concatenated.
Note that no conflict exists between internal symbolic names and
program section names, i.e., it is legal to use the same symbolic name
for both purposes. Considering FORTRAN again, using the same symbolic
name is necessary to accommodate the following statement:
COMMON

/X/ A,B,C,X

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

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

6.8.2

.ASECT and .CSECT Directives

IAS and RSX-11 assembly-language programs use the
.PSECT and
.ASECT
directives exclusively, since the .PSECT directive provides all the
capabilities of the
.CSECT directive defined for other
PDP-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 6-4. Also, compatibility exists
between other MACR0-11 programs and the IAS/RSX-11 Task Builders,
since the respective Task Builders recognize the .ASECT and .CSECT
directives that appear
in such programs and likewise assign the
default values listed in Table 6-4.

6-38

GENERAL ASSEMBLER DIRECTIVES
Table 6-4
Non-IAS/RSX-11 Program Section Default Values
Default Value
Attribute
.ASECT

.CSECT (named)

.CSECT (unnamed)

Name

. ABS.

name

Blank

Access

Type

Scope

GBL

LCL

Relocation

ABS

REL

Allocation

OVR

CON

The allowable syntactical forms of the .ASECT
are:
.. ASECT
,, CSECT
. CSECT

and

.CSECT

are

assumed

directives

symbol

Note that the statement:
, CSECT

JIM

is identical to the statement:
.PSECT

JIM,GBL,OVR

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

6.9

SYMBOL CONTROL DIRECTIVE:

OVR

for

the

.GLOBL

MACR0-11 produces a relocatable object module and a listing file
containing the assembly listing and symbol table.
The Task Builder
joins separately-assembled object modules into a single executable
task image.
During linking, object modules are relocated as a
function of the specified base of the module.
The object modules are
then linked via global symbols, such that a global symbol in one
module, defined either by a global assignment operator (==), a global
label operator
(::), or the .GLOBL directive can be referenced from
another module.
which will be referenced by ot
program modules must be sin led o
mc~tl

The .GLOBL directive is provided to define (and thus provide linkage
to)
symbols not otherwise defined as global symbols within a module.
For example, if the .DSABL GBL directive is in effect
{see Section
6 .. 2),
.GLOBL directives might be included in a source program to
effect linkage to library routines.
For a global symbol definition,
the directive .GLOBL A,B,C is equivalent to:
A==expression {or A::)
B==expression {or B::)
C==expression {or C::)
6-39

GENERAL ASSEMBLER DIRECTIVES
Thus, the general form of the .GLOBL directive is:
.GLOBL
where:

syml,sym2, .•. symn

syml,
sym2, •.•
symn

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

.A .GLOBL directive may also embody a

label

field

and/or

comment

field.
At the end of assembly pass 1, MACR0-11 determines whether a given
global symbol is defined within the current program module or whether
it is to be treated as an external symbol.
All internal symbols
appearing within a given program must be defined at the end of
assembly pass 1 or they will be assumed to be default global
references.
Refer
to
Section
6.2
for
a
description
of
enabling/disabling of global references.
In the example below, A and B are entry-point symbols. The symbol A
has been explicitly defined as a global symbol by means of the .GLOBL
directive, and the symbol B has been explicitly defined as a global
label by means of the double colon (::).
Since the symbol C is not
defined as a label within the current assembly,
it is an external
(global) reference.
DEFINE A SUBROUTINE WITH 2 ENTRY POINTS WHICH CALLS AN
EXTERNAL SUBROUTINE

A:
X:
B. •

. PSECT
. GLOBL
MOV
MOV
JSR
RTS
MOV
CLR
BR

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

;DECLARE THE UNNAMED PROGRAM SECTION •
;DEFINE A AS A GLOBAL SYMBOL .
;DEFINE ENTRY POINT A.
;CALL EXTERNAL SUBROUTINE C.
;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)

It should be noted that an undefined external symbol cannot be used in
the evaluation of a direct assignment statement or as an argument in a
conditional assembly directive (see Sections 6.10.l and 6.10.3).

6-40

GENERAL ASSEMBLER DIRECTIVES
6.10

CONDITIONAL ASSEMBLY DIRECTIVES

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

6.10.1

Conditional Assembly Block Directives:

.IF, .ENDC

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

wherie:

cond,argument(s)

;START CONDITIONAL ASSEMBLY BLOCK.

range

;RANGE OF CONDITIONAL ASSEMBLY BLOCK.

. ENDC

;END OF CONDITIONAL ASSEMBLY BLOCK .

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 6-5.
represents any
and/or tab).

legal

separator

(comma,

space,

argument(s)

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

range

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

.ENDC

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

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

Assemble Block If:

Expression

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

Expression

Expression
is
than 0 (or less
e_g_ual to 0) .

Positive

Complement

EQ
GT

6-41

greater
than or

GENERAL ASSEMBLER DIRECTIVES
Table 6-5 (Cont.)
Legal Condition Tests for Conditional Assembly Directives
Conditions
Arguments

Assemble Block If:

Expression

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

NDF

Symbolic
argument

Symbol is defined (or not
defined).

Macro-type
argument

Argument is blank (or
non-blank).

ION

DIF

Two macro-type
arguments

Arguments are identical
(or different).

Expression

Same as EQ/NE.

Expression

Same as GT/LT.

Positive

Complement

NOTE
A macro-type argument (which is a form of symbolic
argument), as shown below,
is enclosed within
angle brackets or denoted with
an
up-arrow
construction (as described in Section 7.3.1).
<A,B,C>
"/124/
An example of a conditional assembly directive follows:
. IF EQ

ALPHA+l

;ASSEMBLE BLOCK IF ALPHA+l=O .

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

Logical AND operator
Logical inclusive OR operator

For example, the conditional assembly statement:
.IF OF

SYMl & SYM2

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

6-42

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 (0) in the assembly listing.

6.10.2

Subconditional Assembly Block Directives:

Subconditional directives may be placed
blocks to indicate:

conditional

assembly

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

condition

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

The unconditional assembly
conditional assembly block.

within

.IFF, .IFT, .IFTF

body

code

within

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

6-43

GENERAL ASSEMBLER DIRECTIVES
Table 6-6
Subconditional Assembly Block Directives
Subconditional
Directive

Function

.IFF

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

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.

.IFT

;TESTS TRUE. SYM IS DEFINED. ASSEM;BLE REMAINDER OF CONDITIONAL ASSEM;BLY BLOCK •

. ENDC

6-44

GENERAL ASSEMBLER DIRECTIVES
EXAMPLE 2:

Assume that symbol X is defined and
defined.

• Il" DF

• IF DF
.!PF

that symbol Y is not

;TESTS TRUE, SYMBOL X IS DEFINED •
;TESTS FALSE, SYMBOL Y IS NOT DEFINED •
;TESTS TRUE, SYMBOL Y IS NOT DEFINED,
;ASSEMBLE THE FOLLOWING CODE.
;TESTS FALSE, SYMBOL Y IS NOT DEFINED.
;DO NOT ASSEMBLE THE FOLLOWING CODE •

• ENDC
.ENDC
EXAMPLE 3:

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

. IF DF

MOV

A,Rl

;TESTS TRUE. A IS DEFINED .
;ASSEMBLE THE FOLLOWING CODE.

;TESTS FALSE. A IS DEFINED. DO NOT
;ASSEMBLE THE FOLLOWING CODE.

• HF

MOV

Rl,RO

.IF NDF B

;NESTED CONDITIONAL DIRECTIVE IS NOT
;EVALUATED •

• ENDC
.ENDC
EXAMPLE 4:

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

• IF DF

.IF DF

;TESTS FALSE. SYMBOL X IS NOT DEFINED •
;DO NOT ASSEMBLE THE FOLLOWING CODE.
;NESTED CONDITIONAL DIRECTIVE IS NOT
;EVALUATED.

.HF

;NESTED SUBCONDITIONAL DIRECTIVE IS
;NOT EVALUATED.

.!FT

;NESTED SUBCONDITIONAL DIRECTIVE IS
;NOT EVALUATED .

. ENDC
.ENDC

6-45

GENERAL ASSEMBLER DIRECTIVES
6.10.3

Immediate Conditional Assembly Directive:

.!IF

An immediate conditional assembly directive provides a means for
writing a 1-line conditional assembly block.
In using this directive,
no terminating .ENDC statement is required, and the condition to be
tested
is completely expressed within the line containing the
directive.
Immediate conditional assembly directives are of the form:
.IIF
where:

cond,arg,statement
represents one of the legal condition
tests
def in~d for conditional assembly blocks in Table
6-5.

cond

represents any
and/or tab).
arg

legal

separator

(comma,

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

space;

with
the
i.e.,
an
macro-type

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

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

For example, the immediate conditional statement:
.ITF

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 6-5, an illegal argument, or a null argument specified in an
.IIF directive results in an error code (A) in the assembly listing.

6.10.4

PAL-llR Conditional Assembly Directives

In order to maintain compatibility with programs developed under
PAL-llR, the following conditionals remain permissible under MACR0-11.
It is advisable, however, to develop future programs using the format
for MACR0-11 conditional assembly directives.
Directive
.IFZ or
.IFNZ or
.!FL or
.IFG or
.IFLE
.IFDF
.IFNDF

.IFEQ
.IFNE
.IFLT
.IFGT

Arguments

Assemble Block if

expression
expression
expression
expression
expression
symbolic argument
symbolic argument

expression=O
expression not equal 0
expression<O
expression>O
expression is < or =O
symbol is defined
symbol is undefined

The rules governing these directives are the same as for the
conditional assembly directives previously described.
6-46

MACR0-11

CHAPTER 7
MACRO DIRECTIVES

7.1

DEFINING MACROS

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

7.1.1

.MACRO Directive

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

label:
where:

directive.

name, dummy argument list

label

represents an optional statement label.

name

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

dummy
argument
list

legal

separator

(comma,

space,

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

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

ABS

A,a

;DEFINES MACRO ABS WITH TWO ARGUMENTS.

7-1

MACRO DIRECTIVES
NOTE
to

appear

definitions,
a
constructed with the concatenation character
cause the macro directive to be ignored. This may
result in improper termination of the
macro
definition.
This NOTE also applied to .!RP,
.IRPC, and .REPT.

7.1.2

.ENDM Directive

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

every

definition

must

.ENDM

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

name

.ENDM
.ENDM

macro

ABS

;TERMINATES THE CURRENT
;MACRO DEFINITION.
;TERMINATES THE CURRENT
;MACRO DEFINITION NAMED ABS.

If specified, the symbolic name in the .ENDM statement must match the
name specified in the corresponding .MACRO directive. Otherwise, the
statement is flagged with an error code (A) in the assembly listing
(see Appendix D).
In either case, the current macro definition is
terminated. Specifying the macYo name in the .ENDM statement thus
permits
MACR0-11
to
detect
missing
.ENDM
statements
or
impropexly-nested macro definitions.
The .ENDM directive may be followed by a comment field, but
contain a label, as shown below:
.MACRO
JSR
.WORD
• ENDM

TYPMSG MESSGE
RS,TYPMSG
MESSGE

must

not

;TYPE A MESSAGE.
;END OF TYPMSG MACRO •

An .ENDM statement encountered by MACR0-11 outside a macro definition
is flagged with an error code (0) in the assembly listing (see
Appendix D) •
NOTES
1.

Labels on .ENDM directives are ignored.

Illegal labels will cause the
to be bypassed.

7-2

directive

MACRO DIRECTIVES
7.1.3

.MEXIT Directive

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

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

7.1.4

flagged

MACRO Definition Formatting

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

7.2

CALLING MACROS

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

name

real arguments

label

represents an optional statement label.

name

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

7-3

MACRO DIRECTIVES
real
arguments

represent
symbolic
arguments
which
replace
the dummy arguments specified in the
.MACRO
directive. When multiple arguments are specified,
they are separated by any
legal
separator.
Arguments
to the macro call are treated as
character strings whose usage is determined by the
macro definition. Note that MACR0-11 accepts the
ASCII value of lower-case alphabetic characters
when .ENABL LC has been specified.

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

7.3

MOV

(RO) ,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

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

REN A,B,C

REN

ALPHA,BETA,<Cl,C2>

associated

macro

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

<MOV

X,Y>,#44,WEV

causes the entire expression
MOV

X,Y

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

7-4

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

~/<MOV

X,Y>/,#44,WEV

causing the entire character string
argument.

passed

The following macro call:
REN

#44,WEV-/MOV X,Y/

however, contains only two arguments (#44 and WEV-/MOV X,Y/), because
the up-arrow is. a unary operator (see Section 3.1.3) and it is not
pr~ceded b~. a~ argument separator.
As snown in the examples above, spaces can be used within bracketed
argument constructions to increase the legibility of such-expressions.

7.3.1

Macro Nesting

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

LEVELl DUM1,D.UM2
<DUMl>
<DUM2>

.MACRO
DUM3
ADD
MOV
.ENDM

LEVEL2 DUM3
#10, RO
RO, (Rl) +

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

<MOV

X,RO>,<MOV

R2,RO>

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

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

7-5

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

LVl A,B

.MACRO

LV2 C

.ENDM
.ENDM

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

7.3.2

Special Characters in Macro Arguments

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

MOV

PUSH ARG
ARG,-(SP)

.ENDM

PUSH

X+3(%2)

causes the following code to be generated:
MOV

7.3.3

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

Passing Numeric Arguments as Symbols

When macro arguments are passed, an absolute symbol value can be
passed which is treated by the macro as a numeric string. An argume~t
preceded by the unary operator backslash (\) is treated 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

. ENDM
.MACRO
.WORD
. ENDM

CON A,B
4

INC

x,c

B=B+l
A'B:

C=O

;B IS TREATED AS A NUMBER IN CURRENT
;PROGRAM RADIX .
;A'B IS DESCRIBED IN SECTION 7.3.6 .

7-6

MACRO DIRECTIVES
The above macro call

XO:

.WORD

(INC) would thus expand to:

Note in this expanded code thaf the label XO:
is the result of the
concatenation of two real arguments.
The single quote (') character
in the label A•B:
causes the real arguments X and
0 to be
concatenated as they are passed during the expansion of the macro.
This type of argument construction is described in further detail
in
Section 7.3.6.

A subsequent call to the same macro would generate the following code:
Xl:

.WORD

and so on, for later calls.
The two macro definitions are necessary
because the symbol associated with dummy argument B (i.e., C) cannot
be updated in the CON macro definition, because its numeric value has
already been substituted for its symbolic name, i.e., the character 0
has replaced C in the argument string.
In the CON macro definition,
the number passed is treated as a string argument.
(Where the value
of the real argument is_O, only a single O character is passed to the
macro expansion.)
~
Passing numeric values in this manner is useful in identifying source
listings.
For
example,
versions of pro~rams ~reated
through
conditional assemblies ot a single source program can be
identified
through such codi'ng_ as tfiat shown-be!ow.
Assume, for example, that
the symbol ID in the macro earl (IDT) has been equated elsewhere in
the source program to the value 6.
.MACRO
. !DENT
.ENDM

IDT SYM
/V05A'SYM/

IDT

\ID

;ASSUME THAT THE SYMBOL ID TAKES
;ON A UNIQUE 2-DIGIT VALUE .
;WHERE V05A IS THE UPDATE
;VERSION OF THE PROGRAM.

The above macro call would then expand to:
. !DENT

/V05A6/

where 6 is the numeric value of the symbol ID.

7.3.4

Number of Arguments in Macro Calls

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

7.3.5

Creating Local Symbols Automatically

A label is often required in an expanded macro.
macro

facilities

In the conventional
thus far described, such a label must be explicitly
7-7

MACRO DIRECTIVES
specified as an argument with each macro call.
Be careful in issuing
subsequent calls to the same macro, to avoid specifying a duplicate
label as a real argument.
This concern can be eliminated through a
feature of MACR0-11 which 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$, where n is a decimal integer within the range
64 through 127, inclusive.
Such local symbols are created by MACR0-11
in numerical order, as shown below:
64$
65$

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

ALPHA, A,?B

TST
BEQ
ADD

A
B
#5,A

;CONTAINS DUMMY ARGUMENT B PRECEDED BY
;QUESTION MARK.

B:
.ENDM
A local symbol is generated automatically by MACR0-11 only when a real
argument of the macro call
is either null or missing, as shown in
Example 1 below, which reflects the expansion of the ALPHA macro
defined above.
If the real argument is specified in the macro call, however, MACR0-11
inhibits the generation of a
local symbol and normal argument
replacement occurs, as shown in Example 2 below.
EXAMPLE 1:

Generate a Local Symbol for the Missing Argument:

ALPHA
TST
BEQ
ADD

Rl
Rl
64$
#5,Rl

;SECOND ARGUMENT IS MISSING.
;LOCAL SYMBOL IS GENERATED.

64$:
EXAMPLE 2:

Do Not Generate a Local Symbol:

ALPHA
TST
BEQ
ADD

R2,XYZ
R2
XYZ
#5,R2

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

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

restricted

the

first

Note that automatically-created local symbols resulting local symbols
from the expansion of a macro, as described above, do not in any way
influence local symbol block boundaries.
In other words,
such
automatically-created local symbols do not establish a local symbol
block in their own right.
7-8

MACRO DIRECTIVES
However, when a macro has several arguments earmarked for automatic
local symbol generation, subsfituEiri<i- a specific labei- for one such
argument introduces a risk that assembly errors will result. This is
because MACR0-11 constructs its argument substitution list at the
point of macro invocation. Therefore, the appearance of any label,
thE~
.ENABL LSB .directive, or the .PSECT directive, in the macro
expansion will create a new local symbol block.
This could leave
local
symbol references in the previous block and the symbol
definitions in the new one, resulting in error codes in the assembly
listing (see Appendix D). Furthermore, a subsequent macro expansion
that generates local symbols in the new block may duplicate one of the
symbols in question, resulting in an additional error code (P) in the
assembly listing.

7.3.6

-~~Y~o~d

Arguments

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

arguments.

name==str ing

name

represents the dummy argument,

string

represents the real symbolic argument.

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

separators

When a keyword argument appears in the dummy argument list of a macro
definition, the specified string becomes the default real argument at
mac:ro call.
Whe!n a keyword argument appears in the real argument list of a macro
call, the specified string becomes the real argument for the dummy
argument that exactly matches the specified name, whether or not the
dummy argument was defined with a keyword. If a match fails, the
entire argument specification is treated as the next positional real
argument.
A keyword argument may be specified anywhere in the dummy
argument list of a macro definition and is part of the positional
ordering of argument.
On the other hand, a keyword argument may be
specified anywhere in the real argument list of a macro call but does
not affect the positional correspondence of the remaining arguments.
1
2
3
4
5
6

7
8
9

.LIST

DEFINE A MACRO HAVING KEYWORDS IN DUMMY ARGUMENT LIST
.MACRO
.WORD
.WORD
.WORD
.ENDM

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

12
13
14
15
16

NOW INVOKE SEVERAL TIMES

7-9

MACRO DIRECTIVES
17 000000
000000
000002
000004
18
19 000006
000006
000010
000012
20
21 000014
000014
000016
000020
22
23 000022
000022
000024
000026
24
25 000030
000030
000032
000034
26
27 000036
000036
000040
000042
28
29
30

7.3.7

OOOOOOG
OOOOOOG
OOOOOOG

TEST
.WORD
.WORD
.WORD

A,B,C
A
B

000040
000030
000020

TEST
.WORD
.WORD
.WORD

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

000001
000005

TEST
.WORD
.WORD

OOOOOOG

~WORD

BLOCK=5
1
5
TEMP

000005
000000

TEST
.WORD
.WORD
.WORD

OOOOOOG
000001
000000

OOOOOOG

TEST
.WORD
.WORD
.WORD

ooooooc

TEST
.WORD
.WORD
.WORD

000001

.END

000001
000000

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_ s_ingle 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/
''A 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/
'X, 'Y

In expanding the first line, the scan for the first argument
terminates upon finding the first ' character.
Since A is a dummy
argument, the ' is removed. The sean 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.
·
7-10

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

7.4

MACRO ATTRIBUTE DIRECTIVES:

.NARG, .NCHR, AND .NTYPE

Three directives are available in MACR0-11 which allow the user to
determine certain attributes of macro arguments. The use of these
directives permits selective modifications of a macro expansion,
depending on the nature of the arguments being passed.
These
directives are described separately below.

7.4.1

.NARG Directive

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

.NARG

symbol

label

represents an optional statement label.

symbol

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

7-11

MACRO DIRECTIVES
An example of the .NARG directive follows:

, T ITLe' NARG

2
3

,MAC~O

NOPP,~IJM

, f\JARt;

SVM

,IF

E,,,~V~

6
1

, lFF

• ME)( IT

,s::IEPT

8
q

NUM

NOi'

EN('}M
,ENOt:
,ENOM

10
\t

13
\4
1-;

NOPP

0'71000~

,NARG
,tF
• MEX IT

sv~

E'Q,SVM

.IFF

,RE:PT

NOP

,e:N"M
,E:NDC

NOPI'
, t..tARG

,JF

SVM
EO,SVM

,ME)(tT

,TFF
.~EPT

NOP

,ENl)M
NOP

000000

0t'Je!240

001"0t'J2

t'1Cll0240

NOP

00000a

00!021.l"

N(lP

00000~

"'"'"'240

NOP
NC'IP

000010
0001(1112

000ttaC!I
OIOl02Ut'!

NOP
.E~Oe

000001

7.4.2

ENI')

.NCHR Directive

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, which is useful in calculating the length of
macro arguments, takes the following form:
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
(see
Appendix D) .
7-12

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

legal

separator

(comma,

space,

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

An example of the .NCHR directive follows:

,TITLF.

NCH~

.~ACRO

, NCMr:t
, WORI')

,ASCII

7
8

,EVEN
,ENOM

C1o.UR,MESS
SVM,MF.:SS
SVM

/MESS/

1l'I

~0Qlt'l~QI

C~A~

<~ELLC'I>

Ql~Ci)Q\f.'1'5

, ~iC HR

00001?10

i;,~0Ql05

0000(:11?

11 ~

, wnRD
,ASCII

SVM,MELLO
SVM

0~\?10P.3

MSG 11

'"'5

/HELL.01

\1 a

0~0Jei~u
0t"~!Z'Ol5

t1U

"'"'0ti!0&

117

,EVEN

12
13

1 ll
7.4.3

(:Jt~f)l('IC.Jl

,EN~

.NTYPE Directive

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

.NTYPE

symbol,aexp

where:

label

represents an optional statement label.

symbol

represents any legal symbol.
This symbol
is
equated to the 6-bit addressing mode of the
following argument.
If a symbol is not specified,
the .NTYPE directive is flagged with an error code
(A) in the assembly listing.
represents any
and/or tab) .
7-13

legal

separator

(comma,

space,

MACRO DIRECTIVES
·aexp

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

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

, T tTLE

3
4

.~At:RO

SAVf,UG

.~TYPE

,IF

SVM,ARG

5
6

MOV

,IFF
MnV
,ENOC

10
t1
12
13
1a
15

NTVJtE

F.~ 1 SYM&7P'
A~G,•(SPl

r~EGISTE~

iURG, .. CSP)

tNON•~E'GISTF.'.?

MOOE
~10,,

,!~Jl'~M

~00000

Ql00Cl!00t

TEMPI

ei0ei~OJ2

0001'10,
0~00r2!2

, WC'}R"

SAVE
, NTYPF.'
,TF

~1
sv~,Rt

Mnv

PtP.'ttib

,!FF
MOV

F.Q, SVt-4"'70
l:?1 1 •($P)

r~F::GISTF.R

•Rt,•CSP)

rNON•REGISTER

,Et-.1nc

"10"E"
~C'IDE

01?10C'J0lJ
00Ql~b7

SAVE'

TE~I'

t.JTVPE
,IF

SYM,TEMP

'4CW

TFMP,•(SP)

tRE't.HSTER

•TEt.itP,•(SP)

r llJON•REGISTE'~ M(lOE

EQ,SY~lf.70

,TFF
0~00QllJ

~1274~

MtW

"'"et"m"'

MOOF

,ENF')C

20
21
~(llQ112J0 t

,FND

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

7.5

addressing

modes,

refer

.ERROR AND .PRINT DIRECTIVES

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

.ERROR

expr

;text

7-14

MACRO DIRECTIVES
where:

label

represents an optional statement label.

expr

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

text

represents the specified message
the .ERROR directive.

Upon encountering an .ERROR directive anywhere in
MACR0-11 outputs a single line containing:

associated
source

An error code (P)

The sequence number of the .ERROR directive statement

The value of the current location counter

The value of the expression, if one is specified

The source line containing the .ERROR directive.

with

program,

For example, the following directive:
.ERROR

;INVALID MACRO ARGUMENT

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

Seq. Loe.
No. No.

Exp.
Value

512 005642

000076

Text
.ERROR

;INVALID MACRO ARGUMENT

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

7.6

INDEFINITE REPEAT BLOCK DIRECTIVES:

.IRP AND .IRPC

An indefinite repeat block is a structure that is similar to a macro
definition;
essentially a macro definition that has only one dummy
argument. At each expansion of the indefinite repeat range, this
dummy argument is replaced with successive elements of a specified
real argument list. An indefinite repeat block directive and its
associated repeat range are coded in-line within the source program.
This type of macro definition and expansion does not require calling
the macro by name, as required in the expansion of conventional macros
previously described in this section.
An indefinite repeat block can appear either within or outside another
macro definition, indefinite repeat block, or repeat block (see
Section 7./).
The rules for specifying indefinite repeat block
arguments are the same as for specifying macro arguments (see Section
7. 3) •

7.6.l

.IRP Directive

The .IRP directive is used to replace a dummy argument with successive
real arguments specified in an argument string. This replacement
process occurs during the expansion of an indefinite repeat block
range. This directive takes the following form:
7-15

MACRO DIRECTIVES
label:

.IRP

sym,<argument list>

(range of indefinite repeat block)

.ENDM
where:

label

represents an optional statement label.

sym

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

legal

separator

(comma,

space,

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

range

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

.ENDM

indicates the end of the indefinite
range.

repeat

block

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

7.6.2

.IRPC Directive

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

.IRPC

sym,<string>

(range of indefinite repeat block)

.ENDM
where:

label

represents an optional statement label.

7-16

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

sym

represents any
and/or tab).

legal

separator

(comma,

space,

represents a list of 9haracters enclosed within
angle brackets to be used in the expansion of the
indefinite repeat range.
Although the
angle
brackets
are
required only when the string
contains separating characters, their use
is
recommended for legibility.

range

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

.ENDM

indicates the
range.

end of the indefinite repeat

block

An example -0f the use of the .IRPC directive is shown in Figure 7-1.
1
2

,TITLE
,LIST

IRPTST
ME

X,<AA,BB,cc,oo,EE,FF>
X, CR0) +

0'110000G

,IRP
MOV
,ENOM
MOV
MOV
MOV
MOV
MDV
MOV

0Pl011l00G
000000G
000000G
000000G
000000G
000000G

,IRPC
MOVB
1 £NOM
MOVB
MOVB
MOVB
MOVB
MOVB
MOVB

4
5
6
7
8

0001'!0111
0tlHiHIJ04
"100~10

000014
000020
000024

016720!
016720
016720
016120
01672'11
016720

l'llll0000G
000"00G
000000G
C!J'110000G
0P10000G

AA,(R0)+
BA 1 CR0)+
CC,(R0)+
DO,(R0)•
EE, CR0H
FF,(R0)•

10
11
12
lJ

0Pl01l!J0
000034
000040
000044
0011l050
000054
14
15
16

112741
112741
112741
112741
112741
112741

7.7

1 •(Rl)
#B 1 •CR1)
#C, .. (R1)
#0,•(FH)
#f, .. (FH)
#F, '"' ( Rl)

,END

000001 1

Figure 7-1

X,<ABCDEF">
#tX,•CRl)

Example of .!RP and .IRPC Directives

REPEAT BLOCK DIRECTIVE:

.REPT, .ENDR

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

7-17

MACRO DIRECTIVES
label:

.REPT

exp

(range of repeat block)

.ENDM
where:

7.8

label

represents an optional statement label.

exp

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

range

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

.ENDM
or
• ENDR

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

MACRO LIBRARY DIRECTIVE:

.MCALL

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

argl,arg2, ••• argn

argl,
arg2, .•.
argn

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

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

both

The /ML switch under RSX-11 and the /LIBRARY qualifier under IAS,
specified in connection with an input file specification, indicate to
MACR0-11 that the file is a macro library.
When a macro call is
encountered in the source proqram, MACR0-11 first searches the user
macro library for the named macro definitions, and, if necessary,
continues the search with the system macro library.

7-18

MACRO DIRECTIVES
Any number of such user-supplied macro files may be designated.
In
cases of multiple library files, the search for the named macros
begins with the last such file specified.
The search continues in
reverse
order until the required macro definitions are found,
terminating again, if necessary, with a search of the system macro
library.
If any named macro is not found upon completion of the search, i.e.,
if the macro is not defined, the .MCALL statement is flagged with an
error code (U) in the assembly listing.
Furthermore, a statement
elsewhere in the source program which attempts to expand such an
undefined macro is flagged with an error code (0) in the assembly
listing.
The RSX-11 and IAS command strings to MACR0-11, through which a file
specification is supplied, are described in detail in Sections 8.1.2
and 8.2.2, respectively.

7-19

MACRO DIRECTIVES

PART IV
OPERATING PROCEDURES

CHAPTER 8
OPERATING PROCEDURES

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 listing, the assembly listing, and the symbol table
listing. An optional cross-reference listing of symbols and macros is
available.
A sample assembly listing is provided in Appendix I.

8.1

RSX-llD AND RSX-llM OPERATING PROCEDURES

The following sections describe those MACR0-11 operating procedures
that apply exclusively to the RSX-llD and RSX-llM system environments.

8.1.1

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

Following the entry of CTRL/C (AC) from an operator's console,
the
Monitor Console Routine
(MCR)
indicates its readiness to accept a
command by prompting with the following sequence:
MCR>
MCR then waits for input.
One of five methods can be employed to initiate MACR0-11, as described
below.

8.1 .. 1.1 Method 1 - Direct MACR0-11 Call - The terminal
method 1 is:

sequence

for

MCR>MAC
MAC>macll-cmd-string
The monitor console routine
(MCR)
accepts MAC as input, causing
MACR0-11 to be activated.
Since an assembly 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 (see
Section 8.1.2 below). 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 a CTRL/Z
(AZ]I is entered.

8-1

OPERATING PROCEPURES
8.1.1.2 Method 2 - Using RUN
method 2 is:

Facility - The

terminal

sequence

for

MCR>RUN ..• MAC/UIC=[g,m]
MAC>macll-cmd-string
Method 2 is identical to method 1, except that the MCR RUN command is
used, which requires the entire task name, including the 3-dot prefix.
In addition, the default UIC is changed for one execution.
As in
method 1 above, MACR0-11 again solicits command string input (see
Section 8.1.2) after the assembly of the indicated files is completed.
The /UIC is optional.

8.1.1.3
3 is:

Method 3 - Single Assembly - The terminal sequence for method
MCR>MAC macll-cmd-string

In method 3, no prompting from MACR0-11 occurs, since the command
string input is included in the MCR command line. As in methods 1 and
2, the expression macll-cmd-strng is any legal, syntactically correct
MACR0-11 command string of the form described in Section 8.1.2.
MACR0-11 then assembles source files under control of the command
string and exits when finished.

8.1.1.4 Method 4 - Install, Run Immediately, and Remove On Exit - The
terminal sequence for method 4 is:
MCR>RUN $MAC/UIC=[g,m]
MAC>macll-cmd-string
This method is used when the MACR0-11 assembler is not permanently
installed in the system.
MCR installs MAC from the system program
directory and requests it under the specified UIC. As in method 1,
MACR0-11 solicits command string input
(see Section 8.1.2). When
MACR0-11 exits, it is automatically removed from the system.
NOTE
MACR0-11 can be terminated by entering a
CTRL/Z
(~Z)
at any time a request for
command string input is present.

8.1.1.5 Method 5 - Using Indirect Filename Facility - Any one of the
following sequences may be used in initiating RSX-ll's indirect file
facility for command string input:
MCR>MAC
MAC>@filespec
or
MCR>RUN ... MAC/UIC=[g,m]
MAC>@filespec
or
8-2

OPERATING PROCEDURES
MCR>MAC @f ilespec
or
MAC>RUN $MAC/UIC=[g,m]
MAC>@f ilespec
These forms use the indirect file facility of
RSX-11,
which
effectively accomplishes the substitution of "@filespec" for the
"macll-cmd-string" input employed in methods 1 through 4.
The file
specified as "@filespec" contains MACR0-11 command strings. After
this file is opened, command lines are read from the file until the
end-of-file is detected. Only three nested levels of indirect files
are permitted in MACR0-11.

8.1.2

RSX-11 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.

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;
however, the
entire command string must fit on an 80-byte command line.
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 ~ondition,
resulting in the error message "MAC -- ILLEGAL FILENAME" on the
command output device (see Section 8.4). 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 file specified.

8-3

OPERATING PROCEDURES

Each file specification contains
accordance
with
the
standard
specifications):

the following information
RSX-11
conventions
for

(in
file

filespec /switch:value •••
where:

f ilespec

is the standard RSX-11 file
described in Section 8.3 below.

/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:
/SW
/NOSW
/-SW

specification

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 point (.).
Any numeric value preceded by a number sign (#) 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
sign (+) or a minus (-) sign.
The positive
specification is the default assumption.
If an
explicit octal declaration is specified (#), the
sign indicator, if included, must precede the
number sign.
All switch values must always
colon ( : ) •

preceded

The number of permissible switch specifications,
accompanying switch values, and interpretations
thereof are program-dependent, i.e., the switch
specifications are interpreted in the context of
the program to which they apply.
The switch
options applicable to MACR0-11 are described in
Table 8-2 below.
A syntactical error detected in the command string causes MACR0-11 to
output the following error message to the command output device (see
Section 8.4):
MAC -- COMMAND SYNTAX ERROR
followed by a copy of the entire command string.
Table 8-1 lists the default values for each file specification.

8-4

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

8.1.3

Device
Directory

Filename

Type

Object
File

System
device.

Current.

None

.OBJ

Listing
File

Device used
for object
file.

Directory
used in
Object file.

None

.LST

Source 1
File

System
device.

Current.

None

.MAC

Source 2
to
Source n
File

Device used
for source 1 or
last source file
specified.

Directory
used for
source 1 or
last source
file specified.

None

.MAC

User
Macro
Library

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

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

.MLB

System
Macro
Library

System
device.

[ 1' 1]

RSXMAC

.SML

Indirect
Command
File

System
device.

Current.

None

.CMD

RSX-11 File Specification Switches

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
(see Section 8.1.2).
The available switches for use in
MACR0-11 file specifications under RSX-11 are listed in Table 8-2.

8-5

OPERATING PROCEDURES
Table 8-2
MACR0-11 File Specification Switches for RSX-11
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).
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, beginning with
the last user macro file specified,
continuing in reverse order with each
such file specified, and terminating, if
necessary, with a search of the system
macro library file.
If a required macro
definition is not found upon completion
of the search, an error code (U) results
in the assembly listing
(see Appendix
D).
This means that a user macro
library file must be specified in the
command line prior to the source file{s)
that 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
use~
macro
library
file
must be
specified in the command line prior to
the source files which require it, in
order to resolve macro definitions.

8-6

OPERATING PROCEDURES

Table 8-2 (Cont.)
MACR0-11 File Specification Switches for RSX-11
Switch
/PA: 1 (see Note)
/PA: 2 (see Note)

Function
Assemble the associated
assembly pass 1 only.
Assemble the associated
assembly pass 2 only.

file

during

file

during

/SP (see Note)

Spool listing output (default value).

/NOSP (see Note)

Do not spool output.

/CR: [arg]

Produce a cross-reference listing.
Section 8.1.4.

See

NOTE
The /ML, /PA and /SP switches do not
interact
with
or override MACR0-11
directives.
Rather, they have meaning
only in the command line itself.
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, /PA, /EN, and
/DS;
the options /ML and /PA apply only to the file
immediately
preceding the option so specified, whereas the /EN arid /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.
For
example, in the following command string element:
/LI:SRC/LI:MEB
the· switch specification /LI:MEB will override
that
specified
previously as /LI:SRC.
If both switch values are desired, they can be
specified in the syntactically correct form shown below:
/LI:SRC:MEB
Examples:
1.

MAC>OBJFIL,LSTFIL/NL:BEX:COM/LI:ME=SRCFIL
This command string suppresses the listing
of
binary
extensions and the source comments, and lists the macro
expansions.
Furthermore, it causes all listing directives in
the source program having the arguments BEX, COM, and ME to
be overridden.
In this example, the object output is sent to
the file named OBJFIL.OBJ, and the listing and symbol table
output is sent to the file named LSTFIL.LST.

8-7

OPERATING PROCEDURES
2.

MAC>OBJFIL,LISTM/NL:TOC=SRCFIL
This command string causes the assembly listing's table of
contents to be suppressed along with all other listi~g output
(except the symbol table), when the general no-list mode
prevails for
listing files,
e.g., when the /NL switch is
present in the file specification without an argument.

8.1.4

Cross-Reference Processor

(CREF)

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

User-defined symbols

Macro symbols

Permanent symbols

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

SYM
MAC
REG
MT

where:
SYM

specifies user-defined symbols

MAC

specifies macro symbols

REG

specifies register symbols

PST

specifies permanent symbols.

(default)

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

and

macro
that

symbols
type

/CR:SYM
produces a cross-reference listing of user-defined symbols only.
No
listing of macro symbols is generated.
Thus, to produce all four
types of cross-reference listings, you must specify all four arguments
(the order
in which they are specified is not significant).
Use a
colon to separate arguments.
For example:
/CR:REG:SYM:MAC:PST
The CREF processor is more fully described in the Utilities
Manual supplied with your system.
Figure 8-1 illustrates a comp1ete cross-reference listing.
8-8

Reference

OPERATING PROCEDURES

P!RMAN!NT SVMBDL TABLE c~oss REFERENCE
SYMBOL

R!FERENC!S

ADO

2•227
2•241
2•271

BCC
BCS

8EQ
BISB

BITB

BLO
B~f

BR
CL.R

2•203

2•237
2•235
2•250
2•232
2•225
2•282
2•207

CLRB

2•30U

C"4P

2•2et2
2•252
2•285
2•205
2•2St

INC
JMP
JS~

MOV

2•?0LI

2•245
2• 2M~

MOVB
RTS
SEC:
TST
TSTB
1 SLKB
,BLl<W

,BVTE

,ENO

,ENOC

,IOENT
,IF

2•31~
2•23~

2•21?
2•2Q2
2•2Llb
2•2qq
2•151
2•77
2•122
2•152
2-u,q
2•333

1•1 .,q

, LIST

,NARG

,NLIST
,PSECT

,RAD50
, TITLE
,WORO

2•25&

2•313

2•23&

2•215

2•2qS

2•251

2•273

2•302

2•2&4

2•2b7

2•280

z.zqe

2•224
2•2'.53
2•331
2•2(1'8
2•111
2•20b
2•21.17

2•231

2•212

2•3211

2•314

2•210

2•21q

2•220

2•222

2•20q
2•248
2•277
2•31e
2•307

2•221
2•24Q
2•278
2•32q

2•221
2•255
2•27q

2•22ft
2•2&0

2•A0
2• 124
2•1bLI
2•170

2•81
2•125
2•1&5

2•82
2•12b

2• u,115

2•127
2•tU

2•16&

2•tb7

2•1b7

2•tU

2• 1 oU
2•1'70

2•16'5
2•170
2•210
2•222
2•243

2•tb5
2•i?05
2•Z10
2•222
2•241

2•1&&
2•205
l•210

2•300

2•JH

2•305
2•274

2•27f>

2•312
2•23C1
2•320

2•284

2•330

2•318
2•18
2•123
2•1bLI
2•tbq

2•83

2• 170

2•220
2•240
2•281
2•164

2•281
2•1U

2•258
2•281
2•165

2•281

2•258
2•311

l•tbb

2•108

2•311
2•110

1•75

1•83

t•CJ8

1• 123

1•t45

2•208

2•210

2•221.'1

2•222

2•240

2•88
2•14q
2•1
2•138

2-11 e
2• t bLI

2•12q
2•tb5

2•1b6

2•161

2•Uq

2• 140

2•141

2•142

2-2se
,~ACRO

2•1b5
2•2
2•1Et4
2•toq
2•2"!
2•220
2•243

1•3
1•17LI
2•1bq
2·~08

eIFF

C8EI'

2-2se

2•208
2•222
2•243

2-ne

2•2'22

2•243
2• 270

1•\qq

t•4o
2•1.&0

2• 205
2•311
1•1
2•15
z.1q

1•2
2•135
Figure 8-1

Sample CREF Listing

8-9

OPERATING PROCEDURES
8 .. 2

IAS MACR0-11 OPERATING PROCEDURES

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

8.2.1

p~ocedures

Initiating MACR0-11 Under IAS

The MACRO command is used under
IAS to begin MACR0-11 assembler
operations. The command 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. 2. 2, below).
The command can be issued whenever the IAS Program Development System
(PDS)
is at command level in interactive mode.
This condition is
signified by the appearance of the prompting sequence:
PDS>
at the user's terminal.
The MACRO command can be input either
directly from the terminal
(interactive mode) or from a batch file
(batch mode). When the specified assembly has completed, MACR0-11
terminates operations and returns control to PDS.
(Refer to the IAS
User's Guide for further information about interactive and batch mode
operations.)

8.2.2

IAS Command String Format

A MACR0-11 command string can be specified using any one of the three
formats shown below. The first two formats apply to interactive mode
operation, while the last format applies to batch mode operation.
Note that,
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 following the prompting message:
FILES?
then waits for the user to specify the input file(s).
In batch mode, the command and its arguments must appear on
line unless the PDS line continuation symbol (-) is used.
The command formats for each operating mode are as follows:
Interactive Mode

input
[
filespec /LIBRARY+ ...

PDS> MACRO [qualifiers]
or
PDS> MACRo[qualifiers]

input
[
FILES? filespec /LIBRARY+ ...

8-10

the

same

OPERATING PROCEDURES

Batch Mode
$MACRO [qualifiers]

input
[
filespec /LIBRARY+ .•.

where:
input
f ilespec

is the specification of an input file
(see Section 8.3) 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.

/LIBRARY

specifies that an input file is a macro
library file. As noted in Section 7.8,
any macro that is defined externally
must
be
identified
in the .MCALL
directive before it can be retrieved
from a macro library file and assembled
with the user program.
In locating
macro definitions, MACR0-11 initiates a
fixed search algorithm, beginning with
the last user macro file specified,
continuing in reverse order with each
such file specified, and terminating, if
necessary, with a search of the system
macro library file.
If a required macro
definition is not found upon completion
of the search, an error code (U) results
in the assembly listing (see Appendix
D) .
This means that a user macro
library file must be specified in the
command line prior to the source file(s)
that uses any macros defined in the
library file.
If more than one library
file is specified, the libraries will be
searched in right-to-left order.

qualifiers

specifies one or more of the following:

output
/OBJECT[:filespec]

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

/NOOBJECT

Do not produce an object file.

output
/LIST[:filespec]

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

/NOLI ST

Do not produce a listing file.
The
default in interactive mode is /NOLIST
and in batch mode is /LIST.

8-11

OPERATING PROCEDURES
NOTE
When no listing file is specified, any
errors encountered in the source program
are displayed at the terminal from which
MACR0-11 was initiated.
/CROSSREFERENCE[:argl ... arg4]
Produce a cross-reference listing. Argl
through arg4 are as described in Section
8.1.4.
This
qualifier
may
be
abbreviated to /C.
PDS accepts the MACRO or $MACRO command as input and initializes the
MACR0-11 assembler, which in turn processes the specified files
according to the options indicated in the command string.
When the
operation is complete, MACR0-11 returns control to PDS to obtain the
next command line either from the terminal or from the batch stream.

8.2.3

IAS Indirect Command Files

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 is specified in the form:
@f ilespec
where:
@

specifies that the name that
file.

follows

indirect

filespec

is the file specification of a file (see Section 8.3)
that contains a command string. The default extension
for the file name is .CMD.

An indirect command file reference must always be the rightmost
in the command (see Section 8.2.4 for examples).

8 .. 2. 4

entry

IAS 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.

Where the indirect command file TEST.CMD contains the command
string:
MACRO/OBJECT:MYFILE
8-12

A+B

OPERATING PROCEDURES
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:FOO

@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 FOO.LST is placed
on disk unit 1.
4.

$MACRO/LIST:DKO: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 O.

8.3

IAS/RSX-11 FILE SPECIFICATION FORMAT

The general form for a file specification in IAS/RSX-11 systems is
shown below.
Detailed inforciation is provided in the applicable
system user's guide or operating procedures manual (see Section 0.3 in
the Preface).
dev: [g,m]name.ext;ver
where:
dev:

is the name of the physical device where the desired
file resides. A device name consists of two characters
followed by a 1- or 2-digit device unit number
(octal)
and a colon (e.g., DPl:, DKO:, DT3:). The default
device under RSX-llD and RSX-llM is as 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 (UFO) code.
This code
consists of a group number {octal), a comma {,), and an
owner {member) number {octal) all enclosed in brackets
{[]). An example of a UFO code is:
[200,30].
The
default
Identification
!AS, this can
command.

name

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

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

8-13

OPERATING PROCEDURES
.ext

is a 1- to 3-alphanumeric character filename extension
or type that is preceded by a period (.). An extension
is normally used to identify the nature of the file.
Default values depend on the context of the file
specification and are as follows:
.CMD = Indirect command (input) file
.LST
A listing (print format) file
.MAC = MACR0-11 source module (input file)
.OBJ = MACR0-11 object module (output file)
Intermediate CREF input file created
.CRF
by MACR0-11.

;ver

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

the

highest

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

8.4

MACR0-11 ERROR MESSAGES

MACR0-11 outputs an appropriate error message to the command output
device when one of the error conditions described below is detected.
These error messages reflect operational problems and should not be
confused with the diagnostic error messages (see Appendix D) produced
by MACR0-11 during assembly.
All the error messages listed below, with the exception of the "MAC -COMMAND I/O ERROR" message, result in the termination of the current
assembly; 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.
Error Message

Meaning

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"
for meaning.

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.

8-14

OPERATING PROCEDURES

Meaning

Error Message
MAC -- COMMAND SYNTAX ERROR

An error was detected ~n the
syntax
of
the
MACR0-11
command line.

MAC -- ILLEGAL FILENAME

Neither the device name nor
th~
filename was present in
the input file specification
(i.e.,
the
input
file
specification is null), or a
"wild
card"
convention
(asterisk) was employed in an
input
or
output
file
specification.
"Wild card"
options {*) are not permitted
in
MACR0-11
file
specifications.

MAC -- ILLEGAL SWITCH

An
illegal
switch
was
specified
for a file,
an
illegal value was specified
with a switch, or an invalid
use of a switch was detected
by MACR0-11.

MAC -- INDIRECT COMMAND SYNTAX ERROR

The name of the
indirect
command
file
(@filename)
specified in 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.
(Only 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
Section F.3.

MAC -- INVALID FORMAT IN MACRO LIBRARY

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

MAC -- I/0 ERROR ON INPUT FILE

In reading a record from a
source input file or macro
library file,
an error was
detected by the file system,
e.g., a line containing more
than 132(10)
characters is
encountered.
This
message
may
also indicate that a
device problem exists or that
either a source file or a
macro library file has been
corrupted
with
incorrect
data.
8-15

OPERATING PROCEDURES
Error Message

Meaning

MAC -- I/O ERROR ON MACRO LIBRARY FILE

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

MAC -- I/O ERROR ON OUTPUT FILE

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

MAC -- I/O ERROR ON WORK FILE

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

MAC -- OPEN FAILURE ON INPUT FILE

1.
2.
3.
4.
5.
6.

MAC -- OPEN FAILURE ON OUTPUT FILE

1.
2.
3.
4.
5.
6.

8-16

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

OPERATING PROCEDURES
Error Message

Meaning

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 (on the order
of 13,000 to 14,000), it has
run out of space.
This means
either the source program is
too large to start with, or
it contains a condition that
leads to excessive size, such
as a macro expansion that
recursively
calls
itself
without
a
terminating
condition.

The MACRO Assembler uses its stack for the following purposes:
1.

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

Control Section Information.
.ASECT, or .CSECT.

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

Work space (for code conversion, etc.).

Five

words

for

each

PSECT,

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

8-17

approximately

APPENDIX A
MACR0-11 CHARACTER SETS

A.l

ASCII CHARACTER SET
EVEN
PARITY
BIT

7-BIT
OCTAL
CODE

CHARACTER

0
1

000
001

NUL
SOH

002

STX

003

ETX

004

EOT

005

ENQ

1
1

006
007
010

ACK
BEL
BS

0
0

Oll

012

HT
LF

1
0

014

VT
FF

015

016

017

1
0

020
021

DLE
DCl

023

DC3

024

DC4

013

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

0
1
1

CAN
EM
SUB
ESC
FS
GS
RS

1
1
0

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

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
0

llO
lll

H
I

1
0
0

1
1
0

0
1
0
1
1
0
0

1
1
0
1
0
0

1
0
1
1
0

1
0

0
1
1
0
0

1
0
1
1
0

1
0
0

1
0

us
SP

REMARKS
also

file

#
$
%
&
I

Accent acute or apostrophe.

(
)

I
0

1
2
3
4
5
6
7

8
9

>
?
@

A
B

c
D
E

F
G

A-2

ERR,

(SYNC) .

MACR0-11 CHARACTER SETS
EVEN
PARITY
BIT

--1
0
1
0
0
1
0
1
1
0
1
0
0

0
0

112
113
114115
116
117
120
121
122
123
124
125
126
127
130
131
132
133
134
135
136
137
140

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

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

177

1
0
0
1

0
1
1

* "

7-BIT
OCTAL
CODE

'CHARACTER

REMARKS

K
L
M

N
0
p
Q

x
y

z
[

]

shift/k.
shift/l.
shift/m.

*
**Accent grave.

a
b

a
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.

Appears as # or t on some machines.

** ·-Appears as

on some machines.
A-3

MACR0-11 CHARACTER SETS
A.2

RADIX-50 CHARACTER SET
Character

ASCII Octal Equivalent

Radix-50 Equivalent

space

A-Z

101-132

1-32

unused

0-9

60-71

36-47

The maximum Radix-50 value is, thus,
47*50**2+47*50+47=174777
The following table provides a convenient means of translating between
the ASCII character set and its Radix-50 equivalents.
For example,
given the ASCII string X2B, the Radix-SO equivalent is (arithmetic is
performed in octal):
X=ll3000
2=002400
B=000002
X2B=ll5402

Single Char.
or
First Char.
Space
A
B

c
D
E

F
G
H

I
J
K
L
M
N

0
p
Q

u
v
w
x
y

z
$

000000
003100
006200
011300
014400
017500
022600
025700
031000
034100
037200
042300
04540ff
050500
053600
056700
062000
065100
070200
073300
076400
101500
104600
107700
113000
116100
121200
124300

Second
Character

Third
Character

Space
A
B

E
F
G
H

I
J
K
L
M

0
p
Q

u
v
w
x
y

z
$

000000
000050
000120
000170
000240
000310
000360
000430
000500
000550
000620
000670
000740
001010
001060
001130
001200
001250
001320
001370
001440
001510
001560
001630
001700
001750
002020
002070
A-4

E
F
G
H

I
J
K
L
M
N

0
p
Q

u
v
w

x
y

z
$

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

MACR0-11 CHARACTER SETS
SINGLE CHAR.
OR
FIRST CHAR.
127400

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

135600
140700
144000
147100
152200
155300
160400
163500
166600
171700

SECOND
CHARACTER

Unused
0
1
2
3
4
5
6
7
8
9

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

A-5

THIRD
CHARACTER

Unused
0
1
2
3
4
5
6
7
8
9

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

APPENDIX B
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

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

B-1

MACR0-11 ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

Format

Address
Mode
Name

Address
Mode
Number

@R or
(ER)

(ER)+

Autoincrement

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

@(ER)+

Autoincrement
Deferred

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

- (ER)

Autodecrement

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

@- (ER)

Autodecrement
Deferred

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

E (ER)

Index

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

@E(ER)

Index Deferred

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

Immediate

The expression
operand itself.

the

@#E

Absolute

The expression
E
is
address of the operand.

the

Relative

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

Relative
Deferred

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

B.3

ASSEMBLER DIRECTIVES

The MACR0-11 assembler directives are
table.
For a detailed description
B-2

Meaning
R

contains

the
the ad-

summarized in the following
of each directive, the table

MACR0-11 ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES
contains references to the appropriate sections in
manual.

Form

Section
Reference

the

body

the

Operation

6.3.3
7.3.6

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

6.3.3

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

""Bn

6.4.1.2

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

"Ce:icpr

6.4.2.2

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

6.4.1.2

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

"Fn

6.4.2.2

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

"On

6.4.1.2

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

"Rccc

6.3.7

Convert ccc to Radix-50 form.

.ASCII /string/

6.3.4

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

B-3

MACR0-11 ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

Form

Section
Reference

Operation

.ASCIZ /string/

6.3.5

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

.ASECT

6.8.2

Begin or
resume
program section.

.BLKB exp

6. 5. 3

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

.BLKW exp

6.5.3

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

.BYTE expl,exp2, .•

6.3.l

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

.CSECT [name]

6.8.2

Begin or resume named or unnamed
relocatable program section. This
directive
is
provided
for
compatibility
with other PDP-11
assemblers.

.:OSABL arg

6.2

Disables the function specified
the argument.

.ENABL arg

6.2

Enables (invokes)
the
function
specified by the argument.

• :END [exp]

6. 6 .1

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

.ENDC

6.10.1

Indicates the end of a
assembly block •

• ENDM [name]

7 .1. 2

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

.ENDR

7.7

Indicates the end of the current
repeat block.
This directive is
provided for compatibility
with
other PDP-11 assemblers.

.EOT

6.6.2

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

the

absolute

conditional

MACR0-11 ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

Form

Section
Reference

Operation

.ERROR exp1text

7.5

User-invoked
error
directive1
causes output to the listing file
or the
command
output
device
containing the optional expression
and the statement containing the
directive.

.EVEN

6.5.1

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

.FLT2 argl,arg2,.~.

6.4.2.1

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

.FLT4 argl,arg2, .••

6.4.2.1

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

.GLOBL syml,sym2, .•.

6.9

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

.IDENT /string/

6 .1. 5

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

.IF cond,argl

6.10.1

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

• IF'F

6.10.2

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

• IE'T

6.10.2

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

. H'TF

6.10.2

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

.IIF cond,arg,
statement

6.10.3

Acts
as
a 1-line
conditional
assembly block where the condition
is
tested
for
the
argument
specified.
The
statement
is
assembled only if the condition
tests true.
B-5

MACR0-11 ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

Form

Section
Reference

Operation
.IRP sym,
7.6.1
Indicates
the
beginnin~
of an
indefinite repeat block in which
the symbol specified is replaced
with successive elements of the
real argument list enclosed within
angle brackets •

<argl,arg2, ••• >

• IRPC sym,<string>

7.6.2

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

• :LIMIT

6.7

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

.LIST [arg]

6 .1.1

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

• MACRO name,argl,
arg 2, •••

7 .1.1

Indicates
definition
name
and
arguments.

.MCALL argl,arg2, •..

7.8

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

• MEX IT.

7 .1. 3

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

.NARG symbol

7.4.1

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

• NCHR symbol,<string>

7.4.2

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

.NLIST [arg]

6 .1.1

Without an argument, the .NLIST
directive decrements the listing
level count
by
1.
With
an
argument, this directive suppresses
that
portion
of
the
listing
specified by the argument.
B-6

the start of a macro
having the
specified
the
following dummy

MACR0-11 ASSEMBLY LANGUAGE AND ASSEMBLER DIRECTIVES

Form

Section
Reference

Operation

.NTYPE symbol,aexp

7. 4. 3

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

.ODD

6.5.2

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

.PAGE

6 .1. 6

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

.PRINT exp;text

7.5

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

.PSECT name,attl, ...
attn

6. 8. 1

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

.RADIX n

6.4.1.1

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

. RAD50 /string/

6.3.6

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

.REPT exp

7.7

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

. SBTTL string

6 .1. 4

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

. TITLE string

6 .1. 3

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

. WORD expl,exp2, ..

6.3.2

Generates successive words of data;
each word contains the value of the
corresponding specified expression.

B-7

APPENDIX C
PERMANENT SYMBOL TABLE (PST)

The permanent symbol table (PST)
contains those symbols which are
automatically recognized by MACR0-11. These symbols consist of both
op codes and assembler directives.
The op codes
(i.e.,
the
instruction set)
are listed first, followed by the directives which
cause specific actions during assembly.
For a detailed description of the instruction set, see the appropriate
PDP-11 Processor Handbook.

C.1

OP CODES

MNEMONIC
ADC
ADCB
ADD
ASH
ASHC
ASL
ASLB
ASR
ASRB
BCC
BCS
BEQ
BGE
BGT
BHI
BHIS
BIC
BICB
BIS
BISB
BI'r
BI'rB

BLE
BLO
BLOS
BL'r

OCTAL
VALUE

005500
105500
060000
072000
073000
006300
106300
006200
106200
103000
103400
001400
002000
003000
101000
103000
040000
140000
050000
150000
030000
130000
003400
103400
101400
002400

FUNCTIONAL NAME
Add Carry
Add Carry (Byte)
Add Source To Destination
Shift Arithmetically
Arithmetic Shift Combined
Arithmetic Shift Left
Arithmetic Shift Left (Byte)
Arithmetic Shift Right
Arithmetic Shift Right (Byte)
Branch If Carry Is Clear
Branch If Carry Is Set
Branch If Equal
Branch If Greater Than Or Equal
Branch If Greater Than
Branch If Higher
Branch If Higher Or Same
Bit Clear
Bit Clear (Byte)
Bit Set
Bit Set (Byte)
Bit Test
Bit Test (Byte)
Branch If Less Than Or Equal
Branch If Lower
Branch If Lower Or Same
Branch If Less Than

C-1

PERMANENT SYMBOL TABLE (PST)

MNEMONIC

OCTAL
VALUE

BMI
BNE
BPL
BPT
BR
BVC
BVS
CALL
CCC
CLC
CLN
CLR
CLRB
CLV
CLZ
CMP

100400
001000
100000
000003
000400
102000
102400
004700
000257
000241
000250
005000
105000
000242
000244
020000

CMPB

120000

COM
COMB

005100
105100

DEC
DECB

005300
105300

DIV
EMT
FADD
FDIV
FMUL
FSUB
HALT
INC
INCB

071000
104000
075000
075030
075020
075010
000000
005200
105200

IOT
JMP
JSR
MARK
MFPI

000004
000100
004000
006400
006500

MFPS

106700

MOV
MOVB

010000
110000

MTPI

006600

MTPS

106400

MUL
NEG
NEGB
NOP
RESET
RETURN
ROL
ROLB
ROR

070000
005400
105400
000240
000005
000207
006100
106100
006000

FUNCTIONAL NAME
Branch If Minus
Branch If Not Equal
Branch If Plus
Breakpoint Trap
Branch Unconditional
Branch If Overflow Is Clear
Branch If Overflow Is Set
Jump To Subroutine (JSR PC,xxx)
Clear All Condition Codes
Clear C Condition Code Bit
Clear N Condition Code Bit
Clear Destination
Clear Destination (Byte)
Clear V Condition Code Bit
Clear z Condition Code Bit
Compare Source To
Destination
Compare Source To
Destination (Byte)
Complement Destination
Complement Destination
(Byte)
Decrement Destination
Decrement Destination
(Byte)
Divide
Emulator Trap
Floating Add
Floating Divide
Floating Multiply
Floating Subtract
Halt
Increment Destination
Increment Destination
(Byte)
Input/Output Trap
Jump
Jump To Subroutine
Mark
Move From Previous
Instruction Space
Move from PS
(LSI-11)
Move Source To Destination
Move Source To Destination
(Byte)
Move To Previous
Instruction Space
Move to PS
(LSI-11)
Multiply
Negate Destination
Negate Destination (Byte)
No Operation
Reset External Bus
Return From Subroutine (RTS PC)
Rotate Left
Rotate Left (Byte)
Rotate Right

C-2

PERMANENT SYMBOL TABLE (PST)

MNEMONIC

OCTAL
VALUE

RORB
RTI

106000
000002

RTS
RTT

000200
000006

SBC
SBCB
SEC
SEN
SEV
SEZ
SOB
SUB

005600
105600
000277
000261
000270
000262
000264
077000
160000

SWAB
SXT
TRAP
TST
TSTB
WAIT
XOR

000300
006700
104400
005700
105700
000001
074000

sec

FUNCTIONAL NAME
Rotate Right (Byte)
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)
Wait For Interrupt
Exclusive OR

OP CODES FLOATING POINT PROCESSOR ONLY
MNEMONIC

OCTAL
VALUE

ABSD
ABSF
ADDD
ADDF
CFCC

170600
170600
1 72000
1 72000
170000

CLRD
CLRF
CMPD
CMPF
DIVD
DIVF
L_DCDF

170400
170400
173400
173400
174400
174400
177400

LDC FD

177400

LDC ID

177000

LDC IF

177000

LDCLD

177000

LDCLF

177000

LDD
LD:EXP

172400
176400

FUNCTIONAL NAME
Make Absolute Double
Make Absolute Floating
Add Double
Add Floating
Copy Floating Condition
Codes
Clear Double
Clear Floating
Compare Double
Compare Floating
Divide Double
Divide Floating
Load And Convert From
Double To Floating
Load And Convert From
Floating To Double
Load And Convert Integer To
Double
Load And Convert Integer To
Floating
Load And Convert Long
integer To Double
Load And Convert Long
Integer To Floating
Load Double
Load Exponent

C-3

PERMANENT SYMBOL TABLE (PST)

C.2

MNEMONIC

OCTAL
VALUE

LDF
LDFPS
MFPD

172400
170100
106500

MOOD

171400

MODF

171400

MTPD
MULD
MULF
NEGD
NEGF
SETO
SETF
SETI
SETL
SPL
STCDF

106600
171000
171000
170700
170700
1 700ll
170001
170002
170012
000230
176000

STCDI

175400

STCDL

175400

STCFD

176000

STCFI

175400

STCFL

175400

STD
STEXP
STF
STFPS
STST
SUBD
SUBF
TSTD
TSTF

174000
175000
174000
170200
170300
173000
173000
170500
170500

FUNCTIONAL NAME
Load Floating
Load FPPs Program Status
Move From Previous Data
Space
Multiply And Integerize
Double
Multiply And Integerize
Floating
Move To Previous Data Space
Multiply Double
Multiply Floating
Negate Double
Negate Floating
Set Double Mode
Set Floating Mode
Set Integer Mode
Set Long Integer Mode
Set Priority Level
Store And Convert From
Double To Floating
Store And Convert From
Double To Integer
Store And Convert From
Double To Long Integer
Store And Convert From
Floating To Double
Store And Convert From
Floating To Integer
Store And Convert From
Floating To Long Integer
Store Double
Store Exponent
Store Floating
Store FPPs Program Status
Store FPPs Status
Subtract Double
Subtract Floating
Test Double
Test Floating

MACR0-11 DIRECTIVES
DIRECTIVE
. ASCII
.ASCIZ
.ASECT
.BLKB
.BLKW
.BYTE
.CSECT

FUNCTIONAL SIGNIFICANCE
Translates character string to ASCII equivalents .
Translates character string to ASCII equivalents;
inserts zero byte as last character.
Begins absolute program section
{provided for
compatibility with other PDP-11 assembliers).
Reserves byte block in accordance with value of
specified argument.
Reserves word block in accordance with value of
specified argument.
Generates successive byte data in accordance with
specified arguments.
Begins relocatable program section
{provided for
compatibility with other PDP-11 assemblers).

C-4

PERMANENT SYMBOL TABLE (PST)
DIRECTIVE
• DSABL
. ENABL
. END
. ENDC
.ENDM
.ENDR
. EOT
.ERROR
. EVEN
.FLT2
.FLT4
. GLOBL
.!DENT
. IF
.IFF
.IFT
.IFTF
.IIF
.IRP
.IRPC
.LIMIT
.LIST
.MCALL
.MEXIT
.NARG
.NCHR
.NLIST
.NTYPE
. ODD
. PAGE
. PRINT
.PSECT
. RADIX
.RAD50
.REPT

FUNCTIONAL SIGNIFICANCE
Disables specified function .
Enables specified function •
Defines logical end of source program •
Defines end of conditional assembly block •
Defines end of macro definition, repeat block, or
indefinite repeat block.
Defines end of current repeat block (provided for
compatibility with other 'PDP-11 assemblers).
Define End of Tape condition (ignored) .
Outputs diagnostic message to listing file or
command output device.
Word-aligns the current location counter •
Causes two words of storage to be generated for
each floating-point argument.
Causes four words of storage to be generated for
each floating-point argument.
Declares global attribute for specified symbol(s) •
Labels object module with
specified
program
version number.
Begins conditional assembly block •
Begins
subconditional
assembly
block
(if
conditional assembly block test is false).
Begins
subconditional
assembly
block
(if
conditional assembly block test is true).
Begins subconditional assembly block
(whether
conditional assembly block test is true or false).
Assembles immediate conditional assembly statement
(if specified condition is satisfied).
Begins
indefinite
repeat
block;
replaces
specified symbol with specified successive real
arguments.
Begins
indefinite
repeat
block;
replaces
specified
symbol
with
value
of successive
characters in specified string.
Reserves two words of storage for high and low
addresses of task image.
Controls listing level count and
format
of
assembly listing.
.MACRO Denotes start of macro
definition.
Identifies required
macro
defini~ion(s)
for
assembly.
Exit from current macro definition or indefinite
repeat block.
Equates specified symbol to
the
number
of
arguments in the macro expansion.
Equates specified symbol to
the
number
of
characters in the specified character string.
Controls listing level count
and
suppresses
specified portions of the assembly listing.
Equates specified symbols to the addressing mode
of the specified argument.
Byte-aligns the current location counter •
Advances form to top of next page .
Prints specified message on command output device .
Begins specified program section having specified
attributes.
Changes current program radix to specified radix .
Generates data block having Radix-50 equivalents
of specified character string.
Begins repeat block and replicates it according to
the value of the specified expression.
C-5

PERMANENT SYMBOL TABLE (PST)

DIRECTIVE
.SBTTL
.TITLE
.WORD

FUNCTIONAL SIGNIFICANCE
Prints specified subtitle text as the second line
of the assembly listing page header.
Prints specified title text as object module name
in the first line of the assembly listing page
header.
Generates successive word data in accordance with
specified arguments.

The MACR0-11 directives listed above are summarized in greater
in Appendix B.

C-6

detail

APPENDIX D
DIAGNOSTIC ERROR MESSAGE SUMMARY

D. 1

MACR0-11 ERROR CODES

A diagnostic error code is printed as the first character in a
source
line which contains an error detected by MACR0-11.
This error code
identifies a syntactical problem or some other type of error condition
detected during the processing of a source line.
An example of such a
source line is shown below:
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 types of
error conditions produce this diagnostic message,
all the possible directives which may yield a
general assembly error have been categorized below
to reflect specific classes of error conditions:
CATEGORY 1:

ILLEGAL ARGUMENT SPECIFIED .

. RADIX -- A value other than
specified as a new radix .

2, 8,

• LIST/.NLIST -- Other than a
legally defined
argument
(see Table 6-1) is specified with the
directive .
. ENABL/.DSABL -- Other than a
legally defined
argument
(see Table 6-2) is specified with the
directive .
. PSECT -- Other than a legally-defined argument
(see
Table
6-3)
is
specified with the
directive .
. IF/.IIF -- Other
than
a
legally
defined
conditional test (see Table 6-5) or an illegal
argument expression value is specified with the
directive .
. MACRO -- An illegal or duplicate symbol
in dummy argument list.

D-1

found

DIAGNOSTIC ERROR MESSAGE SUMMARY
Meaning

Error Code
A
(Cont'd)

CATEGORY 2:

NULL ARGUMENT OR SYMBOL SPECIFIED.

.TITLE -- Program name is not specified in the
directive,
or
first
non-blank
character
following
the directive
is a
non-Radix-50
character .
. IRP/.IRPC -- No dummy argument is specified in
the directive .
. NARG/.NCHAR/.NTYPE -- No symbol
in the directive.
.IF/.IIF -- No
conditional
specified in the directive.

specified

argument

CATEGORY 3:
UNMATCHED DELIMITER/ILLEGAL
CONSTRUCTION •

ARGUMENT

• ASCII/.ASCIZ/.RAD50/.IDENT -- Character string
or argument string delimiters do not match, or
an illegal character is used as a delimiter, or
an illegal argument construction is used in the
directive .
. NCHAR -- Character string delimiters do not
match,
or
an
illegal character is used as a
delimiter in the directive.
CATEGORY 4:

GENERAL ADDRESSING ERRORS.

This type of error results from one of
possible conditions:

several

Permissible range of a branch
instruction,
i.e.,
from -128(10) to +127(10) words, has
been exceeded.

A statement makes
invalid
use 0£
the
current
location
counter,
e.g.,
a
".=expression" statement attempts to
force
the
current location counter
to cross
program section (.PSECT) bounda~ies.

A statement contains an
invalid address
expression.
In cases where an absolute
address expression is required,
specifying
a global symbol, a relocatable value, or a
complex relocatable value (see Section 3.9)
results in an invalid address expression.
Similarly, in cases where a
relocatable
address expression is required, either a
relocatable
or
absolute
value
is
permissible,
but a global symbol or a
complex relocatable value in the
statement
likewise results
in an
invalid address
expression.
Specific cases of this type of
error are those which follow:

D-2

DIAGNOSTIC ERROR MESSAGE SUMMARY
M~aning

Error Code

.BLKB/.BLKW/.REPT -- Other than an absolute
value or an expression which reduces to an
absolute value has been specified with the
directive.
4.

Multiple expressions are not separated by a
comma.
This condition causes the next
symbol to be evaluated as part of the
current expression.

CATEGORY 5:

ILLEGAL FORWARD REFERENCE.

This type of error results from either
possible conditions:
1.

A
global
assignment
(symbol==expression)
contains
reference to another symbol.

two

statement
forward

An expression defining
the value of the
current location counter contains a forward
reference.

Bounding error.
Instructions or word data are
being assembled at an odd address.
The locatio~
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 pass 1, and
proceeds with assembly pass 2.

Illegal character detected.
Illegal characters
which are also non-printable are replaced by a
question mark (?) on the listing.
The character
is then ignored.

Input line is greater than 132(10)
characters in
length.
Currently, this error condition is caused
only through excessive substitution
of
real
arguments for dummy arguments during the expansion
of a macro.

Multiple definition of a
label.
A label was
encountered which was equivalent (in the.first six
characters) to a label previously encount~red.

A number contains a digit that
is not in the
current program radix.
The number is evaluated as
a decimal value.

Opcode
error.
Directive
out
of
context.
Permissible nesting level depth for conditional
assemblies has been exceeded.
Attempt to expand a
macro which was unidentified after .MCALL search.

D-3

DIAGNOSTIC ERROR MESSAGE SUMMARY
Error Code

Meaning

Phase error. A label's definition of value varies
from one assembly pass to another or a multiple
definition of a local symbol has occurred within a
local symbol block. Also, when in a local symbol
block defined by the .ENABL LSB directive, an
attempt has occurred to define a local symbol in a
program section other than that which was in
effect when the block was entered. A P error code
also appears if an .ERROR directive is assembLed.

Questionable syntax. Arguments are missing, too
many arguments are specified, or the instruction
scan was not completed.

Register-type error.
An invalid use
of
or
reference to a register has been made, or an
attempt has been made to redefine a standard
register symbol without first issuing the .DSABL
REG directive.

Truncation error. A number generated more than 16
bits in a word, or an expression generated more
than 8 significant bits during the use of the
.BYTE directive or trap (EMT or TRAP) instruction.

Undefined symbol.
An
undefined
symbol
was
encountered
during
the
evaluation
of
an
expression; such·an undefined symbol is assigned
a value of zero. Other possible conditions which
result in this error code include unsatisfied
macro names in the list of .MCALL arguments and a
direct assigment (symbol=expression)
statement
which contains a forward reference to a symbol
whose
definition
also
contains
a
forward
reference;
also, a local symbol may have been
referenced that does not exist in the current
local symbol block.

Instruction error. The instruction so flagged is
not compatible among all members of the PDP-11
family. See Section 5.3 for details.

D-4

APPENDIX E
SAMPLE CODING STANDARD

E.l

INTRODUCTION

Standards

eliminate variability and the requirement to make
a
Much of the difficulty in establishing standards stems from
the notion that they should be optimal.
However, to be successfully
applied,
standards must represent an agreement on certain aspects of
the programming process.

de~cision.

This Appendix contains DIGITAL's PDP-11 Program Coding Standard.
It
i~> suggested that this be used as a model to assist users in preparing
standards for their own installations.

E .. 2

LINE FORMAT

All source lines shall consist of from one to a maximum of eighty
characters
(not including the audit trail added by the SLIPR (SLP in
RSX-llM) editor. This program is described in the applicable RSX-llM
or RSX-llD utilities manual or in the IAS Editing Utilities Reference
Manual (see Section 0.3 in the Preface).
Assembly language code lines shall have the following format:
1.

Label Field - if present, the label shall start at tab stop 0
(column 1) .

Operation field - the operation field shall start at tab stop
1 (column 9).

Operand field - the operand field shall start at tab
(column 17).

Comments field - the comments field shall start at tab stop 4
(column 33) and may continue to column 80.

stop

Comment lines that are included in the code body shall be delimited by
a line containing only a leading semicolon.
The comment itself
contains a leading semicolon and starts in column 3.
Indents shall be
1 tab.
If the operand field extends beyond tab stop 4
(column 33)
simply
a space and start the comment.
Comments which apply to an
instruction but require continuation should always line up with the
character position which started the comment.
le~ave

E-1

SAMPLE CODING STANDARD
E.3

COMMENTS

Comment all coding to convey the global role of an instruction, rather
than simply a literal translation of the instruction into English. In
general this will consist of a comment per line of code.
If a
particularly difficult, obscure, or elegant instruction sequence is
used, a paragraph of comments must immediately precede that section of
code.
Preface text, which describes formats, algorithms, program-local
variables, etc., will be delimited by the character sequence ;+at the
start of the text and ;- at the end;
these delimiters facilitate
automated extraction of narrative commentary. The comment itself will
start in column 3.
For example:

;+
THE INVERT ROUTINE ACCEPTS
; A LIST OF RANDOM NUMBERS AND
; APPLIES THE KOLMOGOROV ALGORITHM
; TO ALPHABETIZE THEM.
;-

E.4

E.4.1

NAMING STANDARDS

E.4.1.1 General Purpose Registers - Only
permitted as register names;
and may
purpose:
R0=%0
Rl=%1
R2=%2
R3=%3
R4=%4
R5=%5
SP=%6
PC=%7

the
not

following names are
be used for any other

;REG 0
;REG 1
;REG 2
;REG 3
;REG 4
;REG 5
;STACK POINTER (REG 6)
;PROGRAM COUNTER (REG 7)

E.4.1.2 Hardware
Registers - These
registers
must
be
named
identically to the hardware definition. For example, PS and SWR.

E.4.1.3 Device Registers - These are symbolically named identically
to the hardware notation. For example, the control status register
for the RK disk is RKCS. Only this symbolic name may be used to refer
to this register.

E-2

SAMPLE CODING STANDARD
E.4.2

Processor Priority

Testing or altering the processor priority is done using the symbols
PRO, PRl, PR2, ..•••• PR7
which are equated to their corresponding priority bit pattern.

E.4.3

Other Symbols

Frequently-used bit patterns such as CR and
conventional symbolics on an as-needed basis.

E.4.4

will

made

Using the Standard Symbolics

The register standards will be defined within the assembler.
All
other standard symbols will appear in a file and will be linked prior
to program execution.

E.4.5

Symbols*

E.4.5.1 Global Symbols - Global symbols should be easily recognized
by their format. The following standards apply and completely define
symbol standards for PDP-11 Medium/Large software products.

symbol

pos-1

pos-2

pos-3

pos-4

pos-5

pos-6

length

ncm-qlbl··sym

letter

a-num/
null

>=-1

gl.bl-sym

$/.

a-num/
null

>=-1

glbl-of fset

letter

$/.

a-num

a-num/
null

>•3

qlbl-bi t··ptrn

letter

a-num

$/.

a-num/

a-num/
null

>=4

le>cal-sym

number

>•2

* Symbols that are branch targets are also called labels, but we

will

always use the term "symbol".

** Number is in the range O<number<65535.
*** The use of $ or .
software.

for global names is reserved

E-3

for

DEC-supplied

SAMPLE CODING STANDARD
where:
a-num
non-glbl-sym
local-sym
glbl-sym
glbl-offset
glbl-bit-ptrn

is an alphanumeric character.
are non-global symbols.
local symbols, as defined by
MACR0-11.
are global symbols (addresses) •
are global offsets (absolute
quantities).
are global bit patterns.

A program never contains a .GLOBL statement without showing cause.

E.4.5.2

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
Global Bit Pattern Symbols
Al$20
B3.6

JI.M
Local Symbols
37$
271$

E.4.5.3 Program-Local Symbols - Self-relative
address
arithmetic
(.+n)
is absolutely forbidden in branch instructions~ its use in
other contexts must be avoided if at all possible and practical.

E-4

SAMPLE CODING STANDARD
Target symbols for branches that exist solely for positional reference
local symbols of the form

~Luse

<num>$:
Use of non-local symbols is restricted, within reason, to those cases
where
reference
to
the
code
occurs external to the code.
Local-symbols are formatted such that the numbers proceed sequentially
down the page and from page to page.

E.4.5.4 Macro Names - The last two characters (with
the
last
cha:racter possibly being null) have special significance. The next to
last character is a $, the last, a character specifying the mode of
the: macro.
For example, in the three macro forms in-line, stack, and p-section,
the in-line form has no suffix, the stack has an <S>, and the
p-section a <C>. Thus the Queue I/O macro can be written as any of
QIO$
QIO$S
QIO$C
depending on the form required. These are not reserved letters.
the form of the name is standard.

E. 51

E.5.1

Only

PROGRAM MODULES

General Comments on Programs

In our software, a program provides a single distinct function.
No
limits exist on size, but the~single function limitation should make
modules larger than lK a rarity./ Since any software may eventually
exploit the virtual memory capacity of the 11/40 and 11/45, programs
should make every attempt to maintain a dense reference locus
(don't
promiscuously branch over page boundaries or over a large absolute
address distance).
All code is read-only. Code and data areas are distinct and each
contains explanatory text. Read-only data should be segregated from
read-write data.

E.5.2

The Module Preface

Each program module in the system shall exist as a separate file. The
filename will reflect the name of the module and the file extension
shall be of the form 'NNN'. The 'NNN' signifies the edit number or
the version number. The version number shall be changed only when a
new base level is created. Furthermore, if no corrections are made to
a file from one base level to the next, the version number will not be
changed. Crhe availability of ~le <;.ontrol ~rvices and File Control
Primitives will greatly simplify version number maintenanc~. Program
modules adhere to a strict format.
This format adds to
the
readability and understandability of the module.
The following
sections are included in each module:
E-5

SAMPLE CODING STANDARD
-tor the Code Section:
1.

~.TITLE

a moati.l.e
used.

statement that specifies the name of the module.
If
contains more than one routine, subtitles may be

2. · An . IDENT statement specifying the version number.
PDP-11 version number standard appears in section E.10.
3.

A .PSECT statement that defines the program section in
the module resides.

A copyright statement, and the disclaimer.

The
which

COPYRIGHT (C) 1976
DIGITAL EQUIPMENT CORPORATION, MAYNARD, MASS.
THIS SOFTWARE IS FURNISHED UNDER A LICENSE FOR USE
ONLY ON A SINGLE COMPUTER SYSTEM AND MAY BE COPIED
ONLY WITH THE INCLUSION OF THE ABOVE COPYRIGHT
NOTICE.
THIS SOFTWARE, OR ANY OTHER COPIES THEREOF,
MAY NOT BE PROVIDED OR OTHERWISE MADE AVAILABLE TO
ANY OTHER PERSON EXCEPT FOR USE ON SUCH SYSTEM AND TO
ONE WHO AGREES TO THESE LICENSE TERMS.
TITLE TO AND
OWNERSHIP OF THE SOFTWARE SHALL AT ALL TIMES REMAIN
IN DEC.
THE INFORMATION IN THIS DOCUMENT IS SUBJECT TO CHANGE
WITHOUT NOTICE AND SHOULD NOT BE CONSTRUED AS A
COMMITMENT BY DIGITAL EQUIPMENT CORPORATION.
DEC ASSUMES NO RESPONSIBILITY FOR THE
USE
OR
RELIABLILITY OF ITS SOFTWARE ON EQUIPMENT WHICH IS
NOT SUPPLIED BY DEC.

The version number of the file.
The PDP-11 version number standard is
E.10.

described

section

The name of the principal author and the date
module was first created.

The name of each modifying author and
the
date
of
modification.
Names and modification dates appear one per
line and in chronological order.

A brief statement of the function of the module.
Note:

which

the

Items 1-8 should appear on the same page.

A list of the definitions of all equated local symbols used
in the module.
These definitions appear one per line and in
alphabetical order.

10.

All local macro definitions, preferably in alphabetical order
by name.

11.

All local data.
a.
b.
c.

The data should indicate

Description of each element (type, size, etc.)
Organization (functional, alpha, adjacent, etc.)
Adjacency requirements

E-6

SAMPLE CODING STANDARD
12.

A more detailed definition of the function of the module.

13.

A list of the inputs expected by the module.
the
calling
sequence
if non-standard,
settings, and global data settings.

This includes
condition code

14.

A list of the outputs produced as a result of

15.

A list of all effects (including side effects) produced as a
result of entering this module.
Effects include alterations
in the state of the system not explicitly expected in the
calling sequence, or those not visible to the caller.

16.

The module code.

E.5.3

entering this
module.
These include delivered results, condition code
settings, but not side effects.
(All these outputs are
visible to the caller.)

Formatting the Module Preface

Rules:
1.

The first eight items appear on the same page and will not
have explicit headings.
Item 3 may be omitted if the blank
p-section is being used.

Headings start at the left
indented 1 tab position.

Items 7-14 will have headings which start at the left margin,
preceded and followed by lines containing only a leading <;>.
Items which do not apply may be omitted.

margin*;

descriptive

text

A template for the module preface follows.
FILE-EXAMPL.SOl
.TITLE
.IDENT
.PSECT

EXAMPLE

/01/
KERNEL

COPYRIGHT (C) 1976
DIGITAL EQUIPMENT COPORATION, MAYNARD, MASS.
THIS SOFTWARE IS FURNISHED UNDER A LICENSE FOR USE ONLY ON A
SINGLE COMPUTER SYSTEM AND MAY BE COPIED ONLY WITH THE
INCLUSION OF THE ABOVE COPYRIGHT NOTICE.
THIS SOFTWARE, OR
ANY OTHER COPIES THEREOF, MAY NOT BE PROVIDED OR OTHERWISE
MADE AVAILABLE TO ANY OTHER PERSON EXCEPT FOR USE ON SUCH
SYSTEM AND TO ONE WHO AGREES TO THESE LICENSE TERMS.
TITLE
TO AND OWNERSHIP OF THE SOFTWARE SHALL AT ALL TIMES REMAIN
IN DEC.
THE INFORMATION IN THIS DOCUMENT IS SUBJECT TO CHANGE WITHOUT
NOTICE AND SHOULD NOT BE CONSTRUED AS A COMMITMENT BY DIGITAL
EQUIPMENT CORPORATION.

*ThE~ left margin consists of a <;> a <space> then the heading, so
text of the heading begins in column 3.

E-7

the

SAMPLE CODING STANDARD
DEC ASSUMES NO RESPONSIBILITY FOR THE USE OR RELIABILITY OF
ITS SOFTWARE ON EQUIPMENT WHICH IS NOT SUPPLIED BY DEC.
VERSION 01
JOE PASCUSNIK l-JAN-72
MODIFIED BY:
RICHARD DOE 21-JAN-73
SPENCER THOMAS 12-JUN-73
Brief statement of the module's function
EQUATED SYMBOLS
List equated symbols
LOCAL MACROS
Local Macros
LOCAL DATA
Local data

;+
Module function-details
INPUTS:
Description of inputs
OUTPUTS:
Description of outputs
EFFECTS:
Description of effects
Begin Module Code

E.5.4

Modularity

No other characteristic has more impact on the ultimate engineering
success of a system than does modularity. Modularity for PDP-11
Software Engineering's products consists of the application of the
single-function philosophy described in section E.5.1, and adherence
to a set of calling and return conventions.

E.5.4.1 Calling Conventions
conventions must be observed.

(Inter-Module) - The

E-8

following

calling

SAMPLE CODING STANDARD
Transfer of Control
Macros will exist for call and return. The actual transfer will
be via a JSR PC instruction.
For register save routines, a
JSR Rn,SAVE will be permitted.
The CALL macro is:
CALL

subr-name

The RETURN macro is:
RETURN
Register Conventions
On entry, a subroutine minimally saves all registers it intends
to alter except result registers.
On exit it restores these
registers.
(State preservation is assumed across calls.)
Argument Passing
Any registers may be used, but their use should follow a coherent
pattern.
For example, if passing three arguments, pass them in
RO, Rl and R2 rather than RO, R2, RS.
Saving and restoring
occurs in one place.

E.5.4.2
macro.

Exiting -

All subroutine exits occur through a single RETURN

E.5.4.3 Intra-Module Calling Conventions Designer optional, but
consistency favors a calling sequence identical to that of the
inter-module sequence.

E.5.4.4 Success/Failure Indication - The C bit will be used to
return the success/failure indicator, where success equals O, and
failure equals 1. The argument registers can be used to return values
or additional success/failure data.

E.5 . 4.5 Module Checking Routines Modules are responsible for
verifying the validity of arguments passed to them. The design of a
module's c~lling sequence should aim at minimizing the validity checks
by minimizing invalid combinations. Programmers may add test code to
perform additional checks during checkout. All code should aim at
discovering an error as close (in terms of instruction executions) to
its occurrence as possible.

E.6

E.6.1

FORMATTING STANDARDS

Program Flow

Progrrams will be organized on the listing such that they flow down the
page!, even at the cost of an extra branch or jump.
E-9

SAMPLE CODING STANDARD
For example:

[:ocess
BBB

AAA

COMMON

shall appear on the listing as:
TST
BNE

BBB

CMN

AAA:

BBB:

CMN:

Rather than:
TST
BNE

BBB

CMN

AAA:

CMN:

BBB:

E-10

SAMPLE CODING STANDARD
E.6.2

Common Exits

A common exit appears as the last code_sequence on the listing.
the flow·chart:

[

EXIT

will appear on the listing as:
PRl:
BR

EXIT

PR2:

PR3:

PR4:

EXI r:
1

And not as:
PRl:

EXI'r:

PR2:
BR

EXIT

PR3:

PR4:

E-11

Thus

SAMPLE CODING STANDARD
E.6.3

Code with Interrupts Inhibited

Code that is executed with inter!upts inhibited, shall be flagged by a
three semicolon (;;;) comment delimiter.
For example:
;ENABLE BY RETURNING
;BY SYSTEM SUBROUTINES,

.• ERTZ:
BIS
BIT
BEQ
RTT

#PR7,PS
#PR7,+2(SP)
10$

10$:

E.7

; ; ;
; i ;
; ; ;
; ; ;
; ; i
; ; ;
; ; ;
; ; ;
; ; ;

INHIBIT INTERRUPTS

M
M
E
N
T

PROGRAM SOURCE FILES

Source creation and maintenance shall be done in base levels.
A base
level is defined as a point at which the program source files have
been frozen.
From the freeze point to the next base
level,
corrections will not be made directly to the base level itself.
Rather a file of corrections shall be accumulated for each file in the
base
level.
Whenever an updated source file
is desired,
the
correction file will be applied to the base file.
The accumulation of corrections shall proceed until a logical breaking
point has occurred
(i.e.
a milestone or significant implementation
point has been reached). At this time all accumulated corrections
shall be applied to the previous base level to create a new base
level.
Correction files will then be started for the new base level.

E.8

FORBIDDEN INSTRUCTION USAGE
1.

The use of instructions or index words
previous instruction.
For example:
MOV

@PC,Register

BIC

Src,Dst

literals

the

uses the bit clear instruction as a literal.
This may seem
to be a very "neat" way to save a word but what about
maintaining a program using this trick?
To compound the
problem, it will not execute properly if I/D space is enabled
on the 11/ 45.
In this case @PC is a D bank reference.
2.

MOV

#ALPHA,PC

t~ansfers 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 0£ control

E-12

SAMPLg CODING STANDARD
take place. What if one would like to get a jump trace of
the execution of a program (a move trace is unheard of)? As
a more general issue, perhaps even other operations such as
ADD and SUB from PC should be discouraged. Possibly one or
=two words can be saved by d§liig these upeta:tIOhs but how many
such occurrences are there?
3.

'I'he 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 s~quence:
CMP

-(Rl), (-Rl)

CMP

- ( Rl) , - ( Rl)

The intent of this instruction sequence is to subtract 8 from
register Rl (not to set condition codes).
This can be
accomplished in approximately 1/3 the time via a
SUB
instruction (9.4 vs. 3.8 microseconds) at no additional cost
in memory space. Another question~ here is also, what if Rl
is odd?
SUB always wins since it will always execute
properly and is always faster!

E.9

E.9.1

RECOMMENDED CODING PRACTICE

Conditional Branches

When using the PDP-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
BLOS

(BCC)
(BCS)

A common pitfall is to use a signed branch (e.g. Jf3GT)\ when comparing
~o
memory addresses.
1
oes well until the two addresses have
o~ld§l Le £>igns;
thaf..is, one of them goes across
"5'?5trrre!.
This type of coding error usually shows itself as a result of
re-·linking at different addresses and/or a change in - size of the
program.

E .10

PDP·-11 VERSION NUMBER STANDARD

The PDP-11 Version Number Standard applies to all modules, parameter
files, complete programs, and libraries which are written or caused to
be written, as part of the PDP-11 Software Development effort. It is
used to provide unique identification of all released, pre-released,
and in-house software.
It is limited in Fhat, as currently specified~nly six characters of
19 re
....i.dJi•ntificatiop
ysed.
Future implemen adons of the Macro
Assembler, Task Builder, and Librarian should provide for at least
E-13

SAMPLE CODING STANDARD

Version Identifier
<form>

= <form> <version> <edit> <patch>
Used tg identify a particular form of a modul~ or
where applicable, as in the case of
LINK-11. One alphabetic character, if used, and
null (i.e.-;-a binary 0) if not used.

prog[~m,

U sed to identify the release, or generation, of a
program.
Tn!o decimal digits, starting at O~, and
incremented at the discretion of the project in
order to reflect what, in their opinion, is a
major change.

<edit>

Used to identify the level to which a particular
release, or generation, of a program or module has
been edited.
An edit is defined to be
an
alteration
to the source form.
Twp gecime.l.
di2its, beginning at 01, and incremented with each
edit; null if no 'eaTrS.

<patch>

Used to identify the level to which a particular
release, or generation, of a·program or module has
been patched. A patch is defined as an alteration
to a binary form.
One alphahet,i.c character,
startiR~ at...B, and running sequentially toward
z,
each time a set of patches is released; null if
no patches.

These fields are interrelated.
When <version> is changed, then
<patch> and <edit> must be reset to nulls. It is intended that when
<edit> is incremented, then <patch> will be re-set to null, because
the various bugs have been fixed.

E.10.l

Displaying the Version Identifier

The visible output of the version identifier should appear as:
Key <letter> <form> <version> - <edit> <patch>,
where the following Key Letters have been identified:
V
X
Y

released or frozen version
in-house experimental version
field test, pre-release, or in-house release version

Note that 'X' corresponds roughly to individual support, 'Y' to
support, and 'V' to company support.

group

The dash which separates <version> from <edit> is used only if <edit>
and/or <patch> is not null. When a version identifier is displayed as
part of program identification, then the format is:
Program
<space><key-lette~><form><version>-<edit><patch>

Name

E-14

SAMPLE CODING STANDARD
Examples:
PIP X03
LINK VB04-C
MACRO YOS-01

E.10.2

Use of the Version Number in the Program

All sources must contain the version number in an .IDENT directive.
For programs (or libraries) which consist of more than one module,
each individual module will follow this version number standard.
The
version number of the program or library is not necessarily related to
the version numbers of the constituent modules;
it is perfectly
reasonable, for example, that the first version of a new FORTRAN
library, VOO, contain an existing SIN routine, say VOS-01.
Parameter files are also required to contain the version number in an
.!DENT 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, e.g.,
the! FORTRAN Library, will have an identification module in the first
position.
An identification module exists
solely
to
provide
identification, and normally consists of something like:
;OTS IDENTIFICATION
.TITLE FTNLIB
.!DENT /003010/
.END

E-15

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, particularly those with the BK subset
version of MACR0-11, should become thoroughly familiar with the
conventions discussed herein. In this regard, Appendix F addresses
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.
In addition to the obvious advantages
accruing from small, distinct, highly-functional bodies of code, one
can usually avoid the problem of insufficient dynamic memory during
assembly by practicing such a policy. Other suggestions as ; to how
available memory can be best utilized are discussed in the following
SE~ctions.

F. 1

GEN:ERAL 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 - four words.

Program sections - six words.

Macro names - four 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.

am1c s orage
The usage of
scope of local symbol blocks through
~IACR0-11 directives {see Sections 3 n 5 and

.ENABL

t~e
i~!rz:ir---,,_..._.._

the
LSB

_________

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. IfJ>ossible, 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
a prefix file wh'
efines all possibly-use
s mbols for
assem
,
are encourage
m ols 1n o og1ca
ciasses. Bach class so grouped can then become a shortened pre~ix
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 appropriate example of this idea is seen in the definition of
standard symbols.
The 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
IOERR$ - I/O return status codes
FILIO$ - File-related I/O function codes
SPCIO$ - Special I/O function codes

F.2

MACRO DEFINITIONS AND EXPANSIONS

By far, 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. An RSX-11 utility program {called
SQZ for RSX-llD only) and a Librarian function switch {/SZ) are
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-llD utilities manual (see
Section 0.3 in the Preface).
It often seems expedient to append a macro definition prefix file to
each
assembly to provide commonly-used macros.
This practice,
however, may produce the undesirable allocation of valuable dynamic
F-2

ALLOCATING VIRTUAL MEMORY
storage for unnecessary macros. This side effect can be avoided by
specifying that the prefix file containing the
macros
is
a
user-supplied macro library file (see Table 8-1). This action imposes
the stipulation that the names of all desired macros must be listed as
arguments in the .MCALL directive (see Section 7.8).
Storage for macro text can be re-used effectively by redefining
certain types of macros to null after they have been invoked. This
practice releases their dynamic memory for the storage of later macro
te~:t
and also eliminates the overhead and the need for dynamic memory
which would otherwise be required during the subsequent invocation and
expansion of such non-redefined macros. The practice of redefining
macros to null applies mainly to those that only define symbolic
assignments, as shown in the example below. The redefinition process
may be accomplished as follows:
SY Ml
SYM2

.MACRO
VALl
VAL2

DEFIN

OFE'l
OFE'2
OFE'3

SYMBOL
OFFl+SIZl
OFP2+SIZ 2

OFE'N

OFFM+SIZM

;DEFINE SYMBOLIC ASSIGNMENTS.

;DEFINE SYMBOLIC OFFSETS.

. MACRO
.ENDM

DEFIN

.ENDM

DE FIN

;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. So doing ensures more efficient
use of dynamic memory.
The! following system macros have
property after once being invoked:

the

automatic

null

DRERR$ - Directive return status codes
IOERR$ - I/O return status codes
FILIO$ - File-related I/O function codes
SPCIO$ - Special I/O function codes
CSI$

- Command String Interpreter codes and offsets

GCMLD$ - Get Command Line codes and off sets
BDOFF$ - FCS buffer descriptor off sets
FCSBT$

- FCS bit value codes

FDOFF$

- FCS file descriptor block off sets

FSROF$ - FCS file storage region (FSR) off sets
NBOFF$ - FCS filename block off sets
F-3~

redefinition

ALLOCATING VIRTUAL 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 improve performance.
The first measure involves shifting the burden of symbol definition
from MACR0-11 to the Task Builder. In most cases, the definition of
system I/O and 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 (see Section 3.9 and
Section D.l, category 4 of error code A). The Task Builder attempts
to resolve all global refe~ences from user-specified default libraries
and/or the system object library (SYSLIB.OLB).
Furthermore, by
applying the selective search option for object modules consisting
only of global symbol definitions, the actual additional burden to the
Task Builder is minimal.
A second way of making more dynamic memory available is to produce
only one output file (either object or listing), as opposed to two.
The additional file descriptor block (FOB) and file storage region
(FSR) required to support the second output file are allocated from
available dynamic memory at the start of each assembly.
Furthermore,
the size of the file storage region allocated is the minimum required
for the second (listing) output file. For disk files, this is 264(10)
words, and for direct line printer output, it is 74(10) words.
The final way of increasing available dynamic memory is related only
to the operating environment. Under RSX-llM, MACR0-11 allocates all
storage between its highest address and the end of its partition as
dynamic memory.
Consequently, the amount of working storage can be
increased by installing and running MACR0-11 in a larger partition.
In IAS and RSX-llD, the assembler's dynamic memory is fixed at link
time.
If a larger assembler is not available, you may build one by
increasing the size of the task's stack.
This is accomplished by
altering the STACK= option in the command file to build MACR0-11.

F-4

APPENDIX G

FEATURES/FUNCTIONS NOT SUPPORTED BY THE RSX-llM SK ASSEMBLER

ITEM

REFERENCE

Search of PST in operand expressions

3.2.2

ftF unary operator

3.7 and
6.4.2.2

Z error code

5.3

.ENABL, .DSABL Symbolic Arguments:
ABS
CDR
FPT
LC
PNC

6.2
6.2
6.2
6.2
6.2

Floating-point Storage Directives:
.FLT2
.FLT4

6.4.2.1

PAL-llR Conditional Assembly Directives

6.11.3

. EOT' directive

6.6.2

More than one level of indirect command files

8.1.1.5

No Floating Point Processor op-codes

C.l

No CREF support

8 .1. 4

G-1

6.4.2~1

APPENDIX H
WRITING POSITION INDEPENDENT CODE

H.l

INTRODUCTION TO POSITION INDEPENDENT CODE

The output of a MACR0-11 assembly is a relocatable object module.
The
Task Builder binds one or more modules together to create an
executable task image. Once built, a task can generally be loaded and
executed only at the virtual address specified by the Task Builder at
link time.
This is because the Task Builder has had 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
(i.e.,
dependent on the virtual addresses to which it was bound).
All PDP-11 processors offer addressing modes that make it possible to
write instructions that are not dependent on the virtual addresses to
which they are
bound.
A
body
of
such
code
is
termed
position-independent and can be loaded and executed 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-llD and RSX-llM,
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.
Code that is not position-independent 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 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
deferred register mode
autoincrement mode
deferred autoincrement mode
autodecrement mode
deferred autodecrement mode

When
using
these
addressing
position-independence, providing
been supplied such that they are
virtual memory location.

modes,
you
are
guaranteed
the contents of the registers have
not dependent upon a particular

H-1

WRITING POSITION INDEPENDENT CODE
when
a
The relative addressing modes are position-independent
relocatable address is referenced from a relocatable instruction.
These modes are as follows:
A

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
(i.e.,
@#A}
may be
employed 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 (e.g., a control
the reference is position-independent.
For example:
MOV

2(SP) ,RO

;POSITION-INDEPENDENT

MOV

N{SP) ,RO

;POSITION-INDEPENDENT

block

offset},

N=4
If, however,
X is a relocatable
position-dependent.
For example:
CLR

ADDR {Rl)

address,

the

reference

;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.
E'or example:
MOV

H.2

@#$DSW,RO

;RETRIEVE DIRECTIVE STATUS

EXAMPLES

The RSX-11 library routine, PWRUP, is a FORTRAN callable subroutine to
establish or remove a user power failure AST entry point address.
Imbedded within the routine is the actual AST entry point which 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 from this routine.
The first
example has been modified to illustrate position-dependent references,
(see
Figure
H-1).
The
second example,
Figure H-2,
is the
position-independent version.
H-2

WRITING POSITION INDEPENDENT CODE
PWRUP::
CL.R
CALL

- (SP)
.X.PAA

.WORD
MOV
MOV
BNE
CLR
BR

1., $DSW
$0TSV,R4
(SP)+,R2
10$
- (SP)
20$

MOV
MOV

R2,F.PF(R4)
#BA,-(SP)

;SET AST ENTRY POINT
;PUSH AST SERVICE ADDRESS

CALL
. BYTE

. X.EXT
109. ,2 .

;ISSUE DIRECTIVE, EXIT •

MOV
MOV
MOV

RO I - (SP)
Rl,-(SP)
R2,-(SP)

;PUSH (SAVE) RO
;PUSH (SAVE) Rl
;PUSH (SAVE) R2

;ASSUME SUCCESS
;PUSH (SAVE) ARGUMENT ADDRESSES ONTO
;STACK
;CLEAR DSW, AND SET Rl=R2=SP
;GET OTS IMPURE AREA POINTER
;GET AST ENTRY POINT ADDRESS
;IF NONE SPECIFIED, SPECIFY NO POWER
;RECOVERY AST SERVICE

10$ ::

;

20$ ::

BA:

;

Figure H-1

Position-Dependent Code

PWRUP::
CLR
CALL
. WORD
MOV
MOV
BNE
CLR
BR

- (SP)
.X.PAA
1. I $DSW
@#$0TSV,R4
(SP)+,R2
10$
- (SP)
20$

;ASSUME SUCCESS
;PUSH ARGUMENT ADDRESSES ONTO STACK
;CLEAR DSW, AND SET Rl=R2=SP .
;GET OTS IMPURE AREA POINTER
;GET AST ENTRY POINT ADDRESS
;IF NONE SPECIFIED, SPECIFY NO POWER
;RECOVERY AST SERVICE

MOV
MOV
ADD

R2,F.PF(R4)
PC I - (SP)
#BA-. I (SP)

;SET AST ENTRY POINT
;PUSH CURRENT LOCATION
;COMPUTE ACTUAL LOCATION OF AST

CALL
. BYTE

. X.EXT
109. ,2 .

;ISSUE DIRECTIVE, EXIT .

10$:

;

20$:

ACTUAL AST SERVICE ROUTINE:
1)
2)
3)
4)
BA:

MOV
MOV
MOV

SAVE REGISTERS
EFFECT A CALL TO SPECIFIED SUBROUTINE
RESTORE REGISTERS
ISSUE AST EXIT DIRECTIVE
RO,-(SP)
Rl,-(SP)
R2,-(SP)
Figure H-2

;PUSH (SAVE) RO
;PUSH (SAVE) Rl
;PUSH (SAVE) R2
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.

H-3

WRITING POSITION INDEPENDENT CODE
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 task's virtual
address space.
The
MACR0-11
Assembler
provides
a
way
of
checking
the
position-independence of code.
In an assembly listing, MACR0-11
inserts a ' character following the contents of any word which
requires the Task Builder to perform a relocation operation. In some
cases this character indicates a position-dependent instruction;
in
other cases, it merely draws the user's attention to the use of a
symbol which may or may not be position-independent. The cases which
cause a ' character to be inserted in the assembly listing are as
follows:
1.

Absolute mode references are flagged with a ' character when
the reference is relocatable.
References are not flagged
when they are position-independent (i.e., absolute).
For
example:
MOV

ADDR(Rl) ,RS
@ADDR(Rl) ,RS

;NON-PIC IF ADDR IS RELOCATABLE.
;NON-PIC IF ADDR IS RELOCATABLE.

Relative and relative deferred mode references are flagged
with a ' character when the address specified is relocatable
with respect to another program section. For example:
MOV
MOV

;PIC ONLY IF ADDR IS ABSOLUTE.

Index and index deferred mode references are flagged with a '
character when the offset is relocatable. For example:
MOV
MOV

@#ADDR,Rl

ADDRl,Rl
@ADDRl,Rl

;NON-PIC WHEN ADDRl IS BOUND
;TO ANOTHER PROGRAM SECTION

Immediate mode references to relocatable addresses are always
flagged with a ' character.
MOV
MOV

#3,RO
#ADDR,Rl

;ALWAYS POSITION-INDEPENDENT.
;NON-PIC WHEN ADDR IS RELOCATABLE.

There is one case in which the MACR0-11 assembler 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 (i.e., MOV $0TSV,R4 in Figure F-1).
Those references requiring more than simple Task Builder relocation
are also indicated in the assembly listing. Simple global references
are flagged with the letter G. Those which contain multiple global
references or complex relocation, are flagged with the letter C (see
Section 3.9 and Chapter 4).
In such cases, it is difficult to
positively state which are or are not position-independent. However,
in general, it is safe to apply the guidelines discussed earlier in
this Appendix to the resulting address value produced by the Task
Builder.

H-4

APPENDIX I
SAMPLE ASSEMBLY AND CROSS REFERENCE LISTING

,TITLE
,ID!NT

1
2
3
4

5
6

7
8
q

10
11
12
13
14

1'7
18
iq

20
21
22

25
26

27
28

30
31
32
33

P3PRE
/05/

MASS,

THIS SOFTWARE IS FURNISHED TO PURCHASER UNDER A LICENSE FOR USE
ON A SINGLE COMPUTER SYSTEM AND CAN BE COPIED (WITH INCLUSION
OF DEC'S COPYRIGHT NOTICE) ONLY FOR USE IN SUCH SYSTEM, EXCEPT
AS MAY OTHERWISE BE PROVIDED IN WRITING BY DEC,
THE INFORMATION IN THIS DOCUMENT IS SUBJECT TO CHANGE WITHOUT
NOTICE AND SHOULD NOT BE CONSTRUED AS A COMMITMENT av DIGITAL
EQUIPMENT CORPORATION.
DEC ASSUMES NO RESPONSIBILITY
OF ITS SOFTWARE ON fQUIPMENT
VERSION 05
C, MONIA 05•FEB•7b
PERFORM PRELIMINARY SYMBOL TABLE PROCESSING
LOCAL MACROS
DEFINE CANNED SECTION TABLE ENTRIES
SECTN N~ME,FLAGS,SEG
WHERE I
NAME•SECTION NAME
oFFSET•OFFSET IN SEGMENT DESCRIPTOR To RECEIVE SECTION ADDRESS,
BLANK IF NONE,
FLAGS•SECTION FLAGS BYTE CONTENTS
SEG•'R00T' IF SECTION IS DEFINED IN ROOT SEGMENT ONLY

35
36
37

3q
40

41
42
43
44
45
46
47
48
4q

50
51
52

53
54

55
56

FOR THE USE OR RELIABILITY
WHICH IS NOT SUPPLIED BY DEC,

.•.•2

SS0•,

,WORD
,WORD

0
0

,RAD50

/NAME/

,BYTE
,IF B
,BYTE
,IFF
,BYTE
,ENOC
,IF NB
,WORD
,IFF

FLAGS
SEG

SSS•,

····"
, •SSS

•l

OFFSET
OFFS!T

I-1

.WORD
.ENDC
SCTLGH•.•SS0
,WORD
.ENDM

58
5q
&0
&1
&2
&3
&4
&5
&6
&7
&8

11!

LOCAL DATA
COMMON STORAGE SHAPED WITH MODULE 'PRCLM'

*** NOTE ***

THE CONTENTS OF THE SECTION DEFINED BELOW MUST MATCH
THE SECTION CONTENTS SPECIFIED IN MODULE 'PRCLM•.

72
73

I
N

74
75 00000111
7&
77 00111000
78 000024
7q 0001112&
00111030
80 00'11032
81 00111034
82 e.10'11054
83 1110005&
84 1110006111
85 001110&2
86 11111100&4
87
88 000000
8q
90
qi

93
q4

95
q6
q7
98

100
Ult

102
trill

104

105

106
107
108

uq
110

.PSECT

125257
01113340

CRSCTI eBLKW
CRVSCt • BLt<W
LOADNM1 .RAD50
MOONt
LCLSYM1
SRCPF1
SRVSG1
SVDEFt
SVENTt
SVFLG1

,BLl<W
,BLKW
• BLKW
.BLKW
,BLKW
,BLKW
,BLl<W

PRCLSO,D,OVR
1111 •
1
/SLOAD/

e•
1
1
1
1
1

113

FIRST HALF OF MODULE NAME
LOCAL SYMBOL STORAGE
SEARCH PATH FLAG
r VIRTUAL ADDRESS OF SEARCH SEGMENT
SYMBOL DEFINITION FLAG
SYMBOL ENTRY ADDRESS
CURRENT SYM80L FLAGS

,PSECT
GSD DISPATCH VECTOR TABLE
THIS TABLE CONTAINS THE ADDRESS OF SUBROUTINES WITHIN MODULE
'PRCLM' THAT PROCESS SPECIFIED GSD RECORD TYPES. PRIOR TO
ISSUING A SUBROUTINE CALL TO ANY OF THESE ENTRY POINTS, THE
FOLLOWING MUST BE SETUP1
INPUTS I
R0•REAL ADDRESS OF SEGMENT DESCRIPTOR
Rt•ADDPESS OF LOCAL SYMBOL DESCRIPTOR CLCLSYM)
R4•AODRESS OF DUMMY ELEMENT DECRIPTOR
RS•REAL ~DDRESS OF SEGMENT DESCRIPTOR
CRSCT•CURRENT SECTION DATA (SETUP BY PSTNME OR SCTNME)
MODN•NAME OF SEGMENT BEING PROCESSED
SCRELM•AOORESS OF DUMMy ELEMENT DESCRIPTOR
SCRSEG•REAL AODRSS OF SEGMENT DESCRIPTOR
SCRVEL•0 (VIRTUAL ADDRESS OF CURRENT ELEMENT DESCRIPTOR)
SCRVSG•VIRTUAL AOD~ESS OF SEGMENT DESCRIPTOR

111

112

CURRENT SECTION DAU
VIRTUAL ADDRESS OF CURRENT SEGMENT
NAME OF MANUAL LOAD ROUTINE

*** NOTE ***

THE CONTENTS OF THIS TABLE ARE ESTABLISHED BV MODULE 'PRCLM'•
ANY CHANGE IN THAT MODULE MUST BE REFLECTED HEREIN.

114
115
11'
117

.PSECT

118 109001!1

119
12111 811191111110
i21 i60i1Jfj2
122 1119005114
123 B00H6
124 0Bfl111UJ
125 01111GH2
126 10111014
i 27 101!10 u

r MODULE NAME

SCTr.i;.;E' • BLi<i'i

SECTION r.iA!oiE

tNTSVMi
XI' RA DR i
SYMDCLi
PSTNMEI
VIDENTi

.PSECT

129 0111111000
130

131

DUMMY ELEMENT DESCRIPTOR

132
133
134
135 00000111

DUMMYI

136

139 0011111110

14111 1011111112
141 01111014
142 00011116
14-3 011111111120
1144
145
14&
147
148
149 1110011122
150

0121111000
00011100
111001111110
1111110000
001110111111
r/10001110

1
1
111

12!

r.
0
111
0

LINK TO NEXT
ADDRESS OF CMT TABLE
HIGHEST SECTION NUMBER
ELEMENT SWITCH WORD
MODULE IDENTIFICATION
I

MODULE NAME
VIRTUAL ADDRESS OF CMT TABLE

SYMBOL ENTRy FOR OVERLAY CONTROL ROUTINE
.RADS0
,BLl<W
.BLKB

125311

151
152 080021

,WORD
BLl<W
• 8Ll<W
,WORD
.wORD
•WORD
.WORD
1 WORO
,wORO
1

137
H

INTERNAL SYMBOL
TRANSFER ADDRESS
r SYMBOL DECLARATION
PROGRAM SECTION NAME
VERSION IDENTIFICATION
VIRTUAL ARRAY NAME

.BLl<W
• BLl<lll
.BLKW
.BLKW
.BLl<W

VSTN"lEi .Bl.KW

128

138 09011106

GSDVSD,D,OVR

"IODNMEI aBLl<W

,BYTE

000

153

/SMA/
1
1

r FIRST HALF OF NAME

SECOND HALF OF NAME
FLAGS
UNUSED

154

SECTION ENTRy LIST

155
156
157
158

THE SECTIONS DESIGNATED AS 'ROOT' ARE INCLUDED IN THE MAIN
ROOT SEGMENT ONLY. ALL OTHERS ARE INCLUDED IN EACH SEGMENT OF
THE TASl<

159

1b0

1&1
1b2 900030
1b3
164 0H032
165 01110042
1&6 0011111152
167 000062
168 1111110072
169 1111110102
170 0111111112

e.1111111000

SCTNMI

RTSPC I

.wORD
SEC TN
SECTN
SEC TN
SEC TN
SEC TN
SEC TN
SECTN

: DUMMY ZERO AT START OF TABLE
SSALVC,cCSSRELICS$TYP>,sSGAUT+2 : AUTOLDAD VECTOR SPACE
SSRGOS,<CSSRELICSSTYP>,S!GREG+2,ROOT J REGION DESCRIPTOR
SSRTS,cCSSALO!CSSGBLICSSREL>
: GLOBAL SEGMENT RETURN POINT
!SSGD0 1 <CSSAL01CSSREL1CSSTYP>,,R00T t DELIMITING SECTION FOR SEGMENT TABLES
SSSG01,cCSSREL1CSSTYP>,SSGSEG+2,ROOT r SEuMENT DESCRIPTORS
SSSGD2,cCSSALO!CSSREL!CSSTYP>,,ROOT r DELIMITING SECTION FOR SEGMENT DESCRIPTORS
SSWNDS,cCSSRELICSSTYP>,SSGWN0+2 1 RDOT , W[NOOW DESCRIPTORS

293 llHHl542
294 8011.1546
295 800550
296 00111552
297 899554
298 1!1095'56
299 900560
3H 111'1121562
31111 800564
31/J2 8005'72
31113 11JH574
304 11Je10514
305 e1110576
306 000604
307 H00P.10
308 eH612
309 0210612
318 ee0&14
H

I
~

311 01i"IJ620
312 000024
313 01110626
314 0ee63111
315 0ee632
316 eee634
·'311 0000411.1
318 909640
319 000042
320 800642
321
322
323
324
325
326

012700
00111e13o'
01232111
Hl 434
1111232lll
012320
005010
Ul5740
11101404
026767
000000G
0 01!10 0111 G
001022
11i!15010
032767
11.101ll0210G
177316
ell.I 1402
112710
11.11110000G
010500
11.1127!i'11
080034'

#LCLSYM+2,RllJ

MOV

CR3)+,(R0)+
30$
(R])+, (Rlt'I)+
CR3)+, (Rill)+
(R0)
•(R0)
10$
SCRVSG, SIHSEG

COPY SECTION NAME
IF EQ AT ENO OF LIST
COPY SECOND HALF OF NAME
COPY FLAGS
CLEAR LENGTH
ROOT SEGMENT ONLY?
J IF EQ NO
r AT ROOT?

BEQ
MOV
MOV
CLR
TSTB
BEQ
CMP

POINT TO LOCAL DESCRIPTOR

)ii

t0SI

BNE

205

CLRB
BIT

(R0)
#SliilSNH,TSKSW

IF NE NO
CLEAR HIGH FLAGS BYTE
BUILDING RESIDENT LIBRARY?

BEQ
MOVB
15SI

20s1
30!1

15S
#CSSRES, CR111)

t
t

IF fQ NO
FLAG SECTION AS TASK·RESIDENT

330 000650
331 "80652
332
333

1-tJ
t""
CJ:]

)ii
CJ:]

3:
tJ:1
t""

MOV
·MOY

R5 1 R0

CALL
MOV
ADD
Clo!P
BEQ
MOV

fPSTNME
R5,R0
(R3) 1 R0
R0 1 R5
20!
CRVSC, CRiil)

TST

#LCLSV~ 1 R1

(R]) +

COPY REAL ADDRESS OF SEGMENT
GET ADDRESS OF LOCAL DESCRIPTOR
r ENTER SECTION IN TABLES
GET ADDRESS OF SEGMENT OESCRIPTO~
OFFSET TO PROPER SLOT IN DESCRIPTOR
NO OFFSET?
IF EQ NO OFFSET TO BE SETUP
t STORE SECTION ADDRESS
CLEAR c, ADVANCE TO NEXT ENTRY

)ii

(')

00
00
~

CJ:]
t'Jj
CJ:]

CJ:]

z
(')

RETURN

CJ:]

INSERT A GLOBAL SYMBOL IN SEG"!ENT.

t""

LOCAL SyMBOL DESCRIPTOR AND CURRENT SECTION MUST BE SETUP

t-:3

z
G)

327

328 ee06u
329 000644

00
00

01051110
06130111
11.12Hlll5
001402
016710
11100021.l"
11.105723

MOV

012701
000034.
"1050111
000001

INSYMI

MOV

*LCLSyM,Rt

GET ADDRESS OF LOCAL DESCRIPTOR

MOV
CALLR

R5,R0
•SY MDCL

COPY ADDRESS OF SEGMENT DESCRIPTOR
ENTER SYMBOL IN TABLE

.ENO

SYMBOL CROSS REP'ERENCE

SYMBOL

REFERE~JCES

CR
CRSCT
CRVSC

• 000015
2100000 A
000024 R

#2•77

··~

• ****** GX
CSSGBL • ****** GX
CSSREL • ****** GX

CSU LO

CSSRU
CSSTYP

• **"'*** GX
• ****** GX

DUMMY
ESLCMT
ESLNUM
ES LS WT

llJ0001110 R

INSCT
INSYM
INTSYM
LCLSVM
L.F

I
Ul

CREF

......
••1.uc.

LOADNH
HARKS
MOON
MODN"4E
PSTNME
PlPRE
RTSPC
RSS11M
SCTLGH
SCTNH
SCTNME
SGSGBL
SPA
SRCPF
SRVSG
SWSNH
SY DEF
SV!NT
SYFLG
SVHDCL
SVSD!F
SVSGBL
SVSREI.
SYS WK
SSGAUT
SSGNME
SS GREG
UGS!G

• ******
• ******
• ******

•• 00H14
000011

GX
GX
GX

0011J540 R
ee06U R
00011J04 R
000034 R
• 000012
000026 R
000el22 R
9'!10032 R
900000 R

000012 R
000174 R
00H52 R
A0001'l10
• 000010
•
e0011J30 R
91/J0002 R

• ******

• 0001/J40
11100054 R
000056 R

• ******

GX
00CIJ0&0 R
llJ00062 R

000064 R
0000U R

• ••••••
•• ******
******

a ******
• ******
• ******
• ******
• ******
SSGSTB • ******
SSGSTS • ******

GX
GX
GX
GX
GX

V01

#1•197
•2•78
2•166
2•166
2•164
2•1b9
2•307
2•U4

2•316
2•161

2•16q

2•165
2•170

2•166

2•167

2•168
00

2•165

2•167

2•263

2•284

2•168

2•U9

2•170

#2•135
•2•264
•2•265
2•20&
*1•197
*1•197
2•210
2•281
#2•122
#2•81
*1•197
#2•79
#2•149
#2•80
#2e120
#2•125
2•209
#2•166
#1•34
*2•164
#2•169
#2•U2
#2•121
2•274
#1•197
#2•82
#2•83
2•235
#2•84
#2•85
•2•86
#2•124
2•279
2•230
2•2'79
2•237
2•164
2•261
2•165
2•168
2•227
2•2'74

l'tJ
t'I
tEl

00
00

tEl

ttl
t'I
i<

#2•291
*2•328
2·216

2•2Q3

2•31111

2•32q

2•228
2•260

2•245

(')
~

(I)

2•311
#2•218
2•272

00
~

tEl
tsj

*2•165
#2•170
2•268

*2•U16
2•272

*2•167

•2•16e

tEl
~

tEl

z
(')
tEl

t'I
H

1-3

2•305

Ci)

~-:n1

2•279
2•250
2•262
2•2511

2•277

2•278

R!GISTER

SYMBOL CROSS REl'ERENCE

SYMBOL

REFERENCES

•2•205
•2•222
•2•270
2•2H
•2•246
•2•255
•2•276
•2•284
2•298
•2•312
•2•284
2•22C)
•2•329
•2•230
•2•248
•2•268
2•313
*2•263
•2•219
2•262
2•312

R1
R2
R3
R4
R5
H

I
O'\

CR!F

V01

•2•208
•2•240
•2•281
•2•209
•2•247
*2•256
•2•277
•2-2q3
•2•299
•2•313
•2•207
2•233

•2•2U
•2•243
•2•311
•2•226
•2•248
•2•260
*2•278
•2•294
2•304
2•314
•2•221
2•23q

*2•212
•2•257
•2•320
•2•227
2-21.1q
•2•261
•2•27C)
•2•296
2•3'!17
2•316
2•224
•2•242

2•244
2•250
2•262
2•280
•2•2CJ7
•2•30q
•2•33111
•2•228
•2•31121

•2•237
2•24q
2•272
•2•318
2•264
•2•223
2·274
2•314

2•239

*2•245

•2•247

•2•294

•2•2q6

•2•2''7

2•265
2•226
2·277
2•330

2•266
2•2'52
2·278

2•261
2.30q

•2•220
•2•258

Cll

)ii

tlj

tot

ti:l
)ii

Cll
Cll
ti:l

"'t<tot
)ii

zt j

()

Cll
Cll
~

ti:l
tzj

ti:l

()

ti:l

tot

Cll

G')

MACRO CROSS R!l'ER!NCE
MACRO NA.ME

REFERENCES

CALL

#1•46
2•222
2•270

CALLA
DEi'

!RR ORS
JOL
RETURN
SAVRG
SAVVR

SEC TN

SOB
,DSECT

I
......J

#1•75

#1•8J
#1•123

CREF

2•205
2•240
2•281
#2•285
#1•14n

2•208
2•2U3
2•311
#2•331
#1•1q7

#2•212
2•21q

#2•320

#2•164
•2-1n

#2•165
#2•170

#1•145

#1•187
#1•156
#1•166
#2•48

#2•168
#1•115
u-qe

2•210
2•257

V01

2•220
2•258

PERMANENT SVMBDL TABLE CROSS REHRENCE
SYMBOL

RE,ERENCES

ADD

2•227
2•241
2•271
2•203
2•315
2•237
2•235
2•250
2•232
2•225
2•282
2-2rn
2•304
2•202
2•252
2•285
2•205
2-2a0_
2•311
2•204
2•228
2•247
2•262
2•277
2•29&
2•329
2•230
2•212
2-·2q2
2•24&
z.2qq
2•151
2•77
2·84
2•123
2•137
2•152
2•166
2•169
2•333
1•179
2•169
1•3
1•174
2•166
2•169
2•205
2•208
2•22111
2•222
2•240
2•243

ace

BCS
BEQ

BISB
BIT
BITB
BLO
ISNE
BR
CLR
CLRB
CMP
INC
JMP
JSR
MDV
H

I
00

MOVB

SEC
TST
TSTB
.BLl<B
,BLl<W

,BYTE
,END
,ENDC
,IDENT
,Ill'

2•256

2•313

2•236

2•275

CREF

2•2q5

2•300

V01

2•306

2•305
2•274

tll

):ii

ttt

2•251

2•273

2•302

2•264

2•267

2•280

2-2qe

2•224
2•253
2•331
2•208
2•243

2•231

2•272

2•301

2•2Ul
2•257

2•219
2•258

2•220
2•270

2•222
2•281

2•206
2•22q
2•248
2•263
2•278
2•297
2•330
2•23q
2•320

2•20q
2•233
2•24q
2•265
2-21q
2•30q

2•221
2•2ll2
2•255
2•266
2-2ea
2•'310

2•223
2•24ll
2•260
2•2&@
2•zq3
2•312

2•226
2•2ll5
2•261
2•276
2•29ll
2•31&

t:"'
t%J

):ii

2•314

tll
tll
t%J

2•307

°'i<
t:"'

):ii

(')
~

tll
tll
~

t%J
t:l:j

2•318

t%J

2•78
2•85
2•12"
2•150
2•1&4
2•167
2•170

2•80
2·8o
2•125
2• 17"
2•1&4
2•167
2•170

2•81
2•120
2•12&

2•82
2•121
2•127

2•83
2-122
2•13&

2•1&5
2•168

2•16'5
2•168

2•106
2•1&q

2•165
2•169
2•2
2•16'.I
2•167
2•1711.1
2•205
2•210
2•220
2•222
2•2ll0
2•243

2•1&6
2• 1'70

2•10'7

2•16'7

2•168

2•16ll
2•167
2•1'70
2•208
2•210
2•2211!
2•222
2•24111
2•257

2•165
2•168
2•205
2•?08
2•210
2•220
2•222
2•243
2•257

2•16'5
2•168
2•21115
2•208
2•210
2•220
2•2ll0
2•243
2•2'5'7

2•166
2•169
2•205
2•208
2•210
2•222
2•240
2•243
2•257

(')

t%J

t:"'
H

tll

INDEX

Absolute addresses, 6-14
Absolute binary output, 6-14
Absolute expression, 3-16
Absolute mode, 5-5, 5-7
Absolute module, 6-34
Absolute program section, 6-37
Address boundaries, 6-31
Address mode syntax, B-1
Address modes, 5-1
Addressing forms, summary, 5-7
Allocating byte data, 6-17
Allocating dynamic memory, F-1
Allocating word data, 6-18
Allocation requirements, 6-35
Alternate radix, 6-25
Ampersand, 3-1
Angle brackets, 3-3, 3-15,
6-4, 6-25, 6-28, 7-4 to
7-5, 7-16 to 7-17
Apostrophe, 7-10
Argument substitution, 7-16
Arithmetic addition operator
or autoincrement indicator
3-1
Arithmetic division operator
3-1
Arithmetic multiplication
operator, 3-1
Arithmetic subtraction
operator or autodecrement
indicator, 3-1
ASCII character set, A-1
ASCII conversion, 3-14
ASCII conversion characters
6-19
.ASCII directive, 6-20
.ASCIZ directive, 6-21
.ASECT directive, 6-38
Assembler directives7 6-1,
B-1, B-2
Assembler version, 6-8
Assembly language, B-1
Assembly listing, 2-6
Assembly pass 1, 1-1
Asterisk, 3-1
At sign, 3-1
Attribute of the current
location counter, 3-12
Autodecrement deferred
mode, 5-3, 5-7
Autodecrement mode, 5-7
Autoincrement deferred
mode, 5-3, 5-7
Autoincrement mode, 5-2, 5-7

"B operator, 6-25
Backslash, 3-1
Binary operators, 3-15
Blank lines, 2-2
Blocks of storage, reserving,
6-30
.BLKB directive, 6-30
.BLKW directive, 6-30
Branch instruction addressing,
5-8
.BYTE directive, 6-17

~c operator, 6-27
Calling conventions, E-8
Calling macros, 7-3
Changing default radix, 3-13
Changing value of location
counter, 3-12
Character set, 3-1
Character substitution, 7-16
Code and data separation, 6-38
Code or data sharing, 6-38
Coding standard, E-1
Colon, 3-1
Comma, 3-1
Comment, 6-14, E-2
Comment field, 2-5
Comment field indicator, 3-1
Complementing an
argument, 6-27
Complex relocatable
expression, 3-16
Complex relocation, 4-1
Concatenated, 6-35
Concatenation of macro
arguments, 7-10
Conditional assembly block,
6-41
Conditional assembly
directive, 6-41, 6-42
Conditional branches, E-13
Continuation lines, 2-2
Creating local symbols
automatically, 7-7
Creating program sections,
6-36
Cross-reference listing
(CREF) , 8-7, 8-8
Cross-reference processor, 8-8
.CSECT directive, 6-17
Current location counter, 2-2,
3-11, 3-14, 5-6, 6-29

Index-1

INDEX (CONT.)

D operator, 6-25
Data storage directives, 6-17
Date, 6-8
Default object module name,
6-11
Default register
definitions, 6-15
Deferred addressing
indicator, 3-1
Defining macros, 7-1
Device registers, E-2
Diagnostic, 7-14
Diagnostic error message
summary, D-1
Direct assignment operator,
3-1
Direct assignment statements,
3-7
Directives, 2-5, 5-9, 6-1
Double ASCII character
indicator, 3-1
Double colon, 3-1, 3-7
Double equal sign, 3-1, 3-7
Double quote, 3-1, 3-14, 6-19
.DSABL directive, 3-7, 3-9
6-13 to 6-15, 6-27
Duplication of code, 7-17

EMT, 5-8
.ENABL directive, 5-7, 6-13,
6-15, 6-27
.END directive, 6-31
.ENDC directive, 6-41
.ENDM directive, 7-2
End of the source input, 6-31
.ENDR directive, 7-17
Entry-point instructions, 6-33
.EOT directive, 6-31
Equal sign, 3-1
Error codes, D-1
.ERROR directive, 7-14
Error messages, 8-14·
Evaluation of expressions,
3-15
.EVEN, 6-29
Exclamation point, 3-1
Executable task image, 4-1
Exiting, E-9
Expressions, 3-14, 3-15
External expression, 3-15,
3-16
External symbols, 6-40
Externally-defined macro, 7-18

~F operator, 3-14, 6-27
File specification default
values, 8-5

File specification format,
8-13
Finding address mode of macro
arguments, 7-13
Finding number or characters
in strings, 7-12
Floating-point
data, 6-26
number, 6-28
number specification, 6-27
rounding, 6-15, 6-27
storage directives, 6-27
truncation, 6-14, 6-27
.FLT2 directive, 6-27
.FLT4 directive 6-27
Forbidden instruction
usage, E-12
Form-feed, 6-13, 7-3
Format control, 2-6
Formatting standards, E-9
Forward referencing, 3-8
Function control switches, 8-6
Function directives, 6-13

General purpose registers, E-2
General registers, 3-9
Global
label, 6-40
references, 6-15
symbol, 2-3, 6-40
symbol directory, 1-2
.GLOBL directive, 3-7, 6-39
GSD, 1-2

Hardware registers, E-2
Horizontal formatting, 2-6

IAS command string format,
8-10
.IF directive, 6-41
.IFF directive, 6-43, 6-44
.IFT directive, 6-43
.IFTF directive, 6-43
.IIF directive, 6-46
Illegal characters, 3-3
Immediate conditional
assembly, 6-46
Immediate expression
indicator, 3-1
Immediate mode, 5-4, 5-7
Immediate mode deferred, 5-5
Implicit .WORD directive, 2-5,
6-18
Indefinite repeat block
directives, 7-15

Index-2

INDEX

Index deferred mode, 5-4, 5-7
Index mode, 5-4, 5-7
Indirect command files, 8-12
Initial argument or expression
indicator, 3-1
Initial register indicator,
3-1
Initiating MACR0-11 under
!AS, 8-10
Initiating MACR0-11 under
RSX-llM/RSX-llD, 8-1
Instruction set, C-1
.!RP directive, 7-15
.IRPC directive, 7-15, 7-16
Item or field terminator, 3-1

Keyword arguments, 7-4, 7-9

Label field, 2-2
Label terminator, 3-1
Left angle bracket, 3-1
Left parenthesis, 3-1
.LIMIT directive, 6-31
Line format, E-1
Linking, 4-1, 4-39
.LIST directive, 6-1
Listing conditional
assemblies, 6-4
Listing control
directives, 6-1
Listing control switches,
8-6
Listing level count, 6-2
Listing of binary extensions,
6-3
Listing of comments, 6-4
Listing of generated
binary code, 6-3
Listing of macro clas~, 6-4
Listing of macro definitions,
6-4
Listing of macro expansion
binary code, 6-4
Listing of repeat range
expansions, 6-4
Listing of source lines, 6-3
Listing of the current
location counter, 6-3
Listing of the symbol table,
6-4
Listing output spool, 8-7
Local symbol block, 6-14
Local symbol block delimiters,
3-10
Local symbols, 3-6, 3-10, 3-11

(CONT . )

Location counter, 6-36
Location counter control
directives, 6-29
Logical AND operator, 3-1,
6-::42
Logical inclusive OR
operator, 3-1, 6-42
Lower-case ASCII, 6-14
Macro arguments, 7-6
Macro attritube directives,
7-11
Macro call, 2-5, 7-3, 7-5
Macro call arguments, 7-4
Macro call numeric argument
indicator, 3-1
Macro definition, 7-1, 7-15
Macro definition arguments,
7-4
Macro definition formattin'g,
7-3
Macro definition termination,
7-2
Macro definitions and
expansions, F-2
Macro directives, 7-1
Macro expansion termination,
7-3
Macro library, 8-6
Macro library directive, 7-18
Macro name, 7-1, 7-4
Macro names, E-5
Macro nesting, 7-5
Macro qualifiers, 8-10
Macro symbol table, 3-5
MACR0-11 character sets, A-1
MACR0-11 directives, 5-9, C-4
MACR0-11 symbols, 3-5
.MCALL directive, 7-18
Memory allocation, 6-32, 6-33,
6-38
Memory allocation and mapping,
6-32
.MEXIT directive, 7-18
Minus sign, 3-1
Modularity, E-8
Module checking routines, E-9
Module preface, E-5
Multi-defined label, 2-4
Multiple definitions of local
symbols, 3-11
Multiple labels, 2-4

Naming standards, E-2
.NARG directive, 7-11
.NCHR directive, 7-11, 7-12

Index-3

INDEX (CONT.)

Negative numbers, 3-14
Nested conditional directives,
6-43
Nested macros, 7-3, 7-5
.NLIST directive, 6-1, 6-11
.NTYPE directive, 7-11, 7-13
Number of arguments in
macro calls, 7-7, 7-11
Number sign, 3-1, 5-4
Numbers, 3-13
Numeric control, 6-24
Numeric control operators,
6-26, 6-27
Numeric directives, 6-26

0 operator, 6-25
Object module, 4-1
Object module name, 6-11
Octal radix, 3-13
.ODD directive, 6-29
Op codes, 2-4, C-1
Operand field, 2-4
Operand field separator, 3-1
Operating procedures, 8-1
Operator field, 2-4
Order of symbol table
search, 3-6
Other symbols, E-3
Overlaid, 6-35
Overlays, 6-33

.PAGE directive, 6-13
Page eject, 7-3
Page ejection, 6-13
Page formatting, 2-6
Page headings, 6-8
Page number, 6-8
PAL-llR conditional assembly,
6-46
Passing numeric arguments
as symbols, 6-46
Percent sign, 3-1
Permanent symbol table, 3-5,
C-1
Plus sign, 3-1
Position independent code, H-1
.PRINT directive, 7-14
Processor priority, E~3
Program boundaries directive,
6-31
Program counter, 3-9, 5-1
Program modules, E-5
Program section access, 6-33
Program section name, 6-33
Program sections, 3-12, 6-32

Program source files, E-12
Program-local symbols, E-4
Programming standards and
conventions, 2-1
.PSECT directive, 3-12, 6-32,
6-35

operator, 6-23
.RAD50 directive, 3-13, 6-24
Radix control, 6-24
Radix control operators, 6-25
Radix-SO character set, A-4
Radix-50 control operator,
6-23
Radix-50 data, 6-22
.RADIX directive, 3-13, 6-24
Read-only access, 6-33
Read/write access, 6-33
Register deferred mode, 5-2
Register expression, 5-1
Register, mode, 5-1, 5-7
Register standards, E-2
Register symbols, 3-9
Register term indicator, 3-1
Relative addresses, 6-14
Relative addressing mode, 5-6
Relative deferred mode, 5-6,
5-7
Relative mode, 5-5, 5-7
Relocatability, ~-34
Relocatable expressions, 3-16,
4-1
Relocatable module, 6-34
Relocatable program sections,
6-37
Relocation, 4-1
Relocation bias, 2-2, 6-34
Repeat block di~ective, 7-17
.REPT directive, 7-17
Reserving storage, 6-30
Reserving storage space, 3-13,
6-29
Right parenthesis, 3-1
RSX-11 command string format,
8-3
RSX-11 file specification
switches, 8-5
RSX-llM BK assembler, G-1

.SBTTL directive, 6-8, 6-11
Scope of the program section,
6-33
Semicolon, 3-1
Sending messages to listing
file, 7-14

Index-4

INDEX

Separating and delimiting
characters, 3-2
Single ASCII character
indicator, 3-1
Single quote, 3-1, 3-14, 6-19,
7-10
Slash, 3-1
Source line sequence numpers,
6-3
Space, 3-1
Special characters, B-1
Special characters in macro
arguments, 7-6
Spool listing output, 8-7
Stack pointer, 3-9
Statement format, 2-1
Storing Radix-50 data, 6-23
Subconditional assembly, 6-43
subtitle, 6-8
Success/failure indication,
E-9
Symbol control directive, 6-39
Symbol examples, E-4
Symbol table listing, 1-2
Symbolic arguments of listing
control directives, 6-3,
6-4
Symbols, E-3
Symbols and expressions, 3-1
System macro libraries, 7-18

Tab, 3-1
Tab character, 2-2
Table of contents, 6-4, 6-11
Task builder, 2-2, 4-1
Teleprinter mode, 6-4
Terminal argument or
expression indicator, 3-1
Terminal register indicator,
3-1

(CONT • )

Terminating directives, 6-31
Terms, 3-14
Time-of-day, 6-8
.TITLE directive, 6-11
Title of the object module,
6-8
Transrating to ASCII,
6-20, 6-21
Translating to Radix-50,
6-22
Trap instructions, 5-8

Unary and binary operators,
3-4
Unary control, 6-24
Unary operator ordering, 6-28
Unary operators, 3-15
Unconditional assembly, 6-43
Undefined symbols, 3-7, 3-14
Universal unary operator
or argument indicator, 3-1
Up arrow or circumflex, 3-1
Up-arrow, 3-3
Up-arrow (~) construction, 7-5
User symbol table, 3-5
User-defined and macro
symbols, 3-5
User-defined macro libraries,
7-18
Using the standard symbolic~,
E-3

Version number, 6-12
Version number standard, E-13

.WORD directive, 3-11, 6-18

Index-5

IAS/RSX-11 MACR0-11
Reference Manual
DEC-11-0IMRA-B-D
READER'S COMMENTS
NOTE:

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

Did you find errors in this manual?

If so, specify by page.

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

Is therE~ sufficient documentation on associated system programs
required for use of the software described in this manual? If not,
what material is missing and where should it be placed?

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

Assembly language programmer
Higher-level language programmer
(] Occasional programmer (experienced)
[] User with little programming experience

Student programmer
[] Non-programmer interested in computer concepts and capabilities
Name
Date------------Organization _________________________________
Street-·---------------------------~---------

City

----~--------State

_________ Zip Code _________
or
Country

Do Not Tear - Fold Here and Tape -

No Postage
Necessary
if Mai led in the
United States

1111

BUSINESS REPLY MAIL
FIRST CLASS PERMIT N0.33 MAYNARD MASS.
POSTAGE WILL BE PAID BY ADDRESSEE

RT/C SOFTWARE PUBLICATIONS

TW/A14

DIGITAL EQUIPMENT CORPORATION
1925 ANDOVER STREET

Do Not Tear - Fold Here

TEWKSBURY, MASSACHUSETTS

01876

-I

I
I