Digital PDFs

DEC-08-LAS4A-A-D

December 2000

108 pages

Original

12MB

Document:	dec-08-las4a-a-d
Order Number:	DEC-08-LAS4A-A-D
Revision:
Pages:	108
Original Filename:	https://svn.so-much-stuff.com/svn/trunk/pdp8/src/dec/dec-08-las4a/dec-08-las4a-a-d.pdf

OCR Text

0 0

4K Mssemles
a

pal |l|
macro-8

§
O 0
*
O O
f
O O
f
O O
f
O O
f
O O
f
O O
f
O 0
§
0 O
f
O O
f
O O

digital equipment corporation

O G

DEC-O8éLAS4A-A-D

For

additional

from Software

Corporation,

ASSEMBLERS

PAL

III/MACRO-B

copies,

order

Distribution

Maynard,

No.

DEC-OB—LAS4A—A-D

Center,

Digital Equipment

Mass.

digital equipment corporation mdgnqrd. massachusetts
-

First

Printing, July

<:>1973 by Digital Equipment Corporation

The following are trademarks
Maynard, Massachusetts:

Digital Equipment Corporation,

CDP

DIGITAL

INDAC

PS/8

COMPUTER LAB

DNC

KA10

QUICKPOINT

COMTEX

EDGRIN

LAB-8

RAD-8

COMSYST

EDUSYSTEM

LAB—8/e

RSTS

DDT

FLIP

LAB-K

RSX

DEC

FOCAL

OMNIBUS

RTM

DECCOMM

GLC-8

OS/8

SABR

DECTAPE

IDAC

PDP

TYPESET

DIBOL

IDACS

PHA

UNIBUS

CHIP

1973

PREFACE

"HOW TO OBTAIN SOFTWARE INFORMATION"
this
various
document,
explains_ the
software users.
The

The

page,

located at

services

the

back

available to DIGITAL

this
postage prepaid "READER'S COMMENTS" form on the last page of
user's
critical
comments
All
evaluation.
requests. the

document

received
documents

ans
are

acknowledged
prepared.

and

will

considered

when

subsequent

software described in this document is furnished to the
purchaser
a license
for use on a single computer system and can be copied
in
such
use
(with inclusion of DIGITAL's copyright notice) only for
system, except as may otherwise be provided in writing by DIGITAL.
The

under

is
in this document is for information purposes only and
DIGITAL assumes no responsibility
change without notice.
and
for the use or reliability of software
equipment which is not
DIGITAL
no
assumes
it.
responsibility for any errors
supplied by

The

material

subject

which may

appear

this

document.

fun-‘- Ens-I. H.

con tents
1

Introduction to 4K Assemblers

.....

............

............
y

PAL III Programming

............................................

Character Set................;

Legal Characters
Illegal Characters
Numbers

..........

......

.....................................

..........

..............

...................

..........................................................

’

..........................

Format Effectors

Form Feed
Tabulations

.....

......................................

.......

...................................................

....................................................................

.............................
.

Statement Terminators

.......................................

......

.......

................

Statements

Labeis

.......................

.........................................

......................................

......................................
p

‘

Instructions

Operands
Comments

.........................

........................

................

...............................................

......

6
6

.....................

Coding Practices
Symbols

......

......................................

*7»

...................................................................

Internal Symbol

RepresentatiOn

Symbolic Addresses
Symbolic Instructions
Symbolic Operands
Symbol Table
Direct Assignment Statements

for PAL

III

.................................

.....

...........
g

‘

...........

.....

....................................................

.......

.....

...................................

.......

Expressions
Address

........

.............

................................

14
17

...........................

Current Address Indicator

...................

‘

.‘..:

11
11

.......

...................................

Assignments

...................

.....................

17
'

Indirect Addressing

Autoindexing

............................

‘

....................................

..................

.....

19‘

Instructions

....................................

..........................................................

Memory Reference Instructions
Microinstructions
Pseudo--Operators

Program

....................................

Preparation and Assembler Output
.

'MACRO- 8 Programming

....................

..........................

.....................

; .‘
.f I".
Operating Procedures
Summary of Diagnostic Messages for PAL III

Characters

........................................................................

.....

fl}.

............

........

....................................................

Expressions
Origin Setting

......................................................................
y

........................

'...;...‘.

...........

............

Link Generation
Literals

..................

..............................................................................

Field Pseudo—Op
Text Facility

.........................

Text Strings

.................................

..........

TextFacility

..................

.......

.....

Macro

.....................................

......................................................

.........

.......

.......................................................

............................................

-----------

................

Programming Hints
Dealing With a Limited Symbol Space

...........................................

.......

Summary of MACRO—8 Error Diagnostics
Pass 3

Output——Assembly Listing,

MACRO-8 Operating Procedures

AssemblerOutput

.......

................

..................

..............

..........

'47

.......

...............

53
53
57

58
58
61
61
62
62
66

.............................

.....

Symbol Table
Symbol Table Modification
Internal Symbol Representation for MACRO— 8
Memory ReferenceInstruction Recognition .7
Compatibility Between PAL III and MACRO-8 1
"

MACRO—81Pseudo-Operators

..............................................

...................

........

..........................................

'User Defined Macros

7 Defining

..........

........

...........

Single Character
Numbers

...................

.......................

.....

...........

68
69
59

Operating Instructions
MACRO—8 Switch Options
.....

Appendix A
Appendix B
Appendix C

..........................

........................

....................

'70
71

‘

.............................................................

.....

.............

.................................................................

vii

A-l

{Bi-1
C-l

4K assemblers

INTRODUCTION TO 4K

ASSEMBLERS

This manual contains descriptiOns of two PDP—8 4K Assemblers,
the first and most basic of which is PAL 111. It is assumed that

already familiar with the material presented in the
first five chapters of Introduction to Programming, as the PAL III
programming language is discussed in detail in that section.
In addition to PAL III, the MACRO-'8 Assembler is also dis—
cussed. MACRO-8 is similar to PAL III with the following addithe reader is

defined macrOs, double precision integers,
constants, arithmetic and Boolean operators, lit-

tional features:

floating—point
erals,

text

user

facilities,

and automatic

off-page. linkage generation.

MACRO—8 is recommended when any of these features is desired
and when a large symbol table is not required.

Appendix C contains a list of the permanent symbols for each
'

’

of these assemblers.
Several other System

Library Programs are useful in assembly
language programming. The Symbolic Tape Editor can be used to
change, correct, or create a program at the Teletype, and after
assembly, DDT or ODT are useful in debugging theprogram.
More

these and other useful programs can be found in Intro-

duction to Programming.
PAL III PROGRAMMING

PAL HI (an acronym for Program Assembly Language, version

111) is a two pass Assembler (with an optional third pass) designed
for the 4K PDP-8 family of computers. A program, written in the
l

PAL III

language, is translated by the Assembler into a

source

binary tape in two passes through the Assembler. The binary tape
is loaded

by the Binary Loader into the computer for execution.
During the first pass of the assembly, all user symbols are defined and placed in the Assembler’s symbol table. During the second pass, the binary equivalents of the input source language are
generated and punched. The Assembler’s third pass produces a
printed assembly listing (a listing of the program’s instructions
with the location, generated binary, «and source code side by side
on each line).
The Assembler requires a basic PDP-8 family computer with a
4K core memory, and a Teletype console. The Assembler can also
use either the high-speed reader, the high-speed punch, or both.
The user can change the Assembler’s permanent symbol table to
reflect his specific machine configuration, as explained under the
section on Altering the Permanent Symbol Table.
Character Set
‘

LEGAL CHARACTERS

The following characters are acceptableto PAL III:

1. The alphabetic characters:
2. The numeric characters:

‘L

Athrough Z
0 through 9*
:

3. The following special characters:
a.

Printing characters
+

‘
_

plus

;

semicolon

‘55

dollar sign

’
‘

‘

’

-—

minus
comma

equal sign

‘

1
‘

”

asterisk

Nonprinting keyboard characters:
'

SPACE

‘

TAB

RETURN
4.

Ignored characters:
FORM FEED

blank tape
RUBOUT

Leader Trailer (code 200)
LINE FEED

period
slash

ILLEGAL CHARACTERS
All other characters

are

illegal (excepthien used in a com-

ment) and cause the illegal character message:
IC

xxxx

nnnn

during pass 1, where xxxx is the Octal value of the offending
character and nnnn is the value of the current location counter
where it occurred. Illegal characters are ignored and assembly can
proceed. The current location counterlcontains the address in
which the next word‘of object code will be assembled. If the illegal
characteroccurs in the middle of a symbol, the symbol is termi—
nated at that point.
'

Numbers

Any sequence of digits delimited by punctuation
forms a number. For example:

characters

1.
'

35
..

4372.~

The pseudo—ops OCTAL and DECIMAL indicate to the Assem-

bler Which radix, or base, is to be .used in number interpretation.

initially set to octal (base 8) and remains octal unless it is changed. It can be changed to decimal (base 10) via the
DECIMAL pseudo-op, which indicates that all numbers which
follow are to be interpreted as decimal. This is then the case until
the occurrence of the OCTAL pseudo-op which converts the base
baCk to octal, and so on, (For an explanation of the internal repreThe radix is

sentation of numbers in the PDP-S,

see

Introduction

to Pro-

gramming.)
Format Efiectors

FORM FEED
The form feed code, if present in a PAL 111 program, will cause

the Assembler to output 12 blank lines during the pass 3 Assembly

listing.

This is useful «in

creating

page—by-page listing.

The

form feed is generated by typing a‘SHIFT/‘L on the Teletype.

TABULATIONS
Tabulations are used in the body of a source program to provide

listing. Tabs separate fields into columns (for
details see Chapter 7 in Introduction to Programming). For example, a line written:
a

neat

readable

Go,TAD

TOTAL/MAIN

LOOP

is much easier to read if tabs are inserted to form:
60:

TAD

TOTAL

/MAIN

LOOP

STATEMENT TERMINATORS
Either the semicolon (g) or the RETURN key may be used as .a
terminator. The semicolon‘is considered identical to a
carriage return/line feed except that it will not terminate a comstatement

ment. For example:

TAD

/THIS

COMMENT;

'TAD

The entire expression between the slash ( / ) and the carriage return
is

considered

comment.

TAD B.

Therefore the Assembler ignores the

If, for example, the user wishes to write a sequence of instructions to rotate the contents of the accumulator and link six places
A

to the right, it might look like

the following:

RTR

RTR
RTR»

However, the programmer can alternatively place all three instruc—
tions on a single line by separating them with the special character
semicolon (;) and terminating the line with a carriage return. The
above sequence of instructions can then be written:
RTR3RTR5RTR

These multi-statement lines

are

LIST,

particularly useful when setting
aside a section of data storage for use during processing. For example, .a 4-word cleared block could be reserved by specifying
either of the following formats:

LIST;

G
0
0
0

Either format may be used to input data words which may be in

the form of numbers,

symbols, or expressions. (Symbols and expressions will be explained later.) Each of the following lines gener—
ate one storage word in the object program:
DATA;

7777
A+C-B
S

123+B2

Statements
PAL HI

usually prepared on a Teletype,
with the aid of the Symbolic Tape Editor, as a sequence of statements. Each statement is written on a single line and is terminated
by typing the RETURN key. PAL HI statements are Virtually
format free; that is, elements of a statement are not placedin
numbered columns with rigidly controlled spacing, as in punched—
source

programs

are

card oriented, assemblers.

There are four types of elements in a PAL III statement which
are

identified by the order of their appearance in the statement, and

by the separating (or delimiting) character which follows or preI

cedes the element.
Statements are written in the general form:

label,

instruction

operand

/ comment

A statement must contain at least one of these elements and may
contain all four types. The Assembler interprets and processes the

binary instructions

statements, generating

one

words,

assembly process.

performing

data

LABELS

A label is the

symbolic name created by the programmer to
of a statement in the program. If present, the
the
location
identify
label is written first in a statement and is terminated by a comma.

There must be no intervening spaces between any of the characters
and the comma.

INSTRUCTIONS
An instruction may be one or more of the mnemonic machine
instructions (see Appendix C), or a pseudo-operation (pseudo-op)

which directs assembly processing.

(The assembly pseudo-ops
manual.) Instructions are terminated
with one or more spaces (or tabs if an operand follows) or with a
semicolon, slash, or carriage return.
are

described later in this

OPERANDS

Operands are usually the octal or symbolic addresses of the data
to be accessed when an instruction is executed, but they can be any
expression, or an argument of a pseudo-op. In each case, interpretation of operands in a statement depends on the statement instruc—
tion. Operands are terminated by a semicolon, slash, or carriage
return.

COMMENTS

Following a slash mark the programmer may add notes to a
statement. Such comments do not affect
assembly processing or
program execution, but areuseful in the program listing for later

analysis

debugging. The Assembler ignores everything from

the slash to the

carriage return. (For an example see the

section on Statement Terminators, preceding.)
It is possible to have nothing but) a carriage return on a line,
resulting in a space in the final listing. An error message is nOt
glven.

Coding Practices
A neat printout (or program listing, as it is usually called) makes
subsequent editing, debugging, and interpretation much easier
than if the coding were laid out in a haphaZard fashion. The coding
practices listed below are in
use, and will resultin a readable, orderly listing.

general

A title comment begins with a slash at the left hand margin.

Pseudo-ops may begin at the left margin; often, however,
they are indented one tab stopto line up with the executable
instructions.

Address labels begin at the left margin.
from succeeding fields by a tabulation.
6

They separated
are

Instructions, whether or not they are
by a label
field are indented one tab stop.
5. A comment is separated from the preceding field by one or
two tabs (as required) and a slash; if the comment Occupies
the whole line, it usually begins with a slash at the left
margin.
Symbols
A symbol is a string of letters and digits beginning with a letter
and delimited by a non- alphanumeric character. Although a symbol
may be any length, only the first Six characters are considered, and
any additional characters are ignored. Consequently symbols which

preceded

are

identical in their first six characters

are

considered identical.

Pseudo-opshave‘ fixed meanings, and cannot be redefined by
the programmer.

The Assembler has in its permanent symbol table definitions of
the

symbols for all FDR-8 pseudo—op codes, memOry reference,

operate and IOT (Input/ Output Transfer) instructions, which may
be used without

prior definition by the user. All other symbols
must be defined1n the source program. For example:
1.

Permanent symbols:

HLT

symbolic instruction whose value of 7402 is

taken by the Assembler from the permanent sym‘

bol table.

User defined symbols.
A

user

defined symbol. When used as a sym-

bolic address label, its value is the address of the

instruction it precedes. This value is
the Assembler.

The

user

assigned by

may asSign values to

symbols by using a direct assignment statement of
the form A
1234, which will be explained later.
=

INTERNAL SYMBOL REPRESENTATION FOR PAL III
Each permanent and user defined

symbol occupies four words

in the symbol table storage area, shown as follows:
Word 1

Word 2

Word 3

Word 4

‘

’

C2
C4
C6

Octal code or address

where C1,

C2,

C6 represent the first character, second
sixth character respectively. For a permanent sym—
.

character,
bol, word 4 contains the octal code of the symbol, while for a user
defined symbol, word 4 contains the address of the symbol. As an
example, the permanent symbol TAD is represented as follows:
.

Word 1 "—2 248 X1008 + 01 =1 24018 01' TA

Word 2

048 X 1008 + 00

04008 or D
‘

Word 3 = 0000
,

Word 4 2:: 1000 (octal code for TAD)
The

Assembler

distinguishes between pseudo—ops,
memory reference instructions, other permanent symbols, and user
defined symbols by their relative positiOn in the symbol table.
PAL

III

SYMBOLIC ADDRESSES

symbol Used as a label to specify a symbolic address must
appear as the first term in a statement and must be immediately
followed. by a comma. When used in this way, a symbol is assigned
a value equal to the current location counter and is said to be
defined. Permanent symbols (instructions, special characters, and
pseudo-ops) may not be used as symbolic addresses.
A defined symbol can be used as an operand, or as a reference
to an instruction. The user sets or resets the location counter by
typing an asterisk followed by the octal absolute address value, in
which the next program word is to be stored. If the origin is not
set by the user, PAL III begins assigning addresses at location 200.
A

/SET

*300
JMP
B;

DCA

LOCATION

COUNTER

300

(in the preceding example) is assigned a value
symbol B a value of 0302, and the symbol A a

The symbol TAG
of

0300,

the

value of 0303.
If

symbol is

,
.

defined

than

once

in this manner, the

Assembler will print the duplicate tag diagnostic:
8

DT‘

xxxx

nnnn

where xxxx is the

symbol, and nnnn is the value of the location
counter at the second occurrence of the symbol definition. The
symbol is not redefined. For example:
$300~
START,
CONTIN:

TAD

DCA

COUNTER

JMS

LEAVE

JMP

START

-74

COUNTER30
START,

The

CLA

CLL

symbol START would have

Value of 0300, the symbol
CON TIN would have a value of 0302, the symbol A would have
a

value of 0304, the symbolCOUNTER (considered COUNTE by

the Assembler) would have a value of 0305. When the Assembler

processed the next line itwould print (during pass 1):
DT

START

@306

Since the first pass of PAL III is usedto define all symbols in

symbol table, the Assembler will print a diagnostic if, at the
end of pass 1, there are any symbols remaining undefined. For
example:
the

*7170

A”

gig EMA

(The dollar sign mustterminate

HLT

all PDP-8 4K

(3MP

assembly pro-

grams.)

would produce the undefined address diagnostic:
UA

xxxx

nnnn

where xxxx is the symbol and nnnn is the location at which it was
first seen. The user’s entire symbol table is printed in alphabetical

order at the end of.pass 1. In the case of the

preceding example,

this would look as follows:
A

.717G/

7173

7174

The following are examples of legal symbolic addresses:

ADDR,
TOTAL,
SUNL
AL
The following are examples of illegal symbolic addresses:

AD)M,
7ABC, 1
LA BEL,‘
D+TAG,

(contains an illegal character)
(first character must be alphabetic)
(must not contain imbedded space)
(contains a legal but non—alphanumeric
character)
(must be terminated by a comma with no intervening spaces)
.-

‘

LABEL

SYMBOLIC INSTRUCTIONS

Symbols used as instructions must be predefined by the Assembler or by the
programmer. If a statement has no label the instruction may appear first in the statement and must be terminated by
a spaCe, tab, Semicolon, slash or carriage return. The following are
examples of legal operators:
TAD

PAGE
ZIP

(a mnemonic machine instruction operator)
.(an Assembler pseudo-op)
(legal only if defined by the user)

SYMBOLIC OPERANDS

Symbols used

operands normally have a value defined by

the user. The Assembler allows symbolic references to instructions
or

data defined elsewhere in the program. Operands may be num-

bers or expressions.

TOTAL,

TAD

AC1+TAG

The values of the two symbols AC1 and TAG, already defined
by the user, are combined by a two’s complement add. This value
is used as the address of the operand.
A

SYMBOL TABLE
The Assembler processes user defined symbols in source pro—
gram statements by adding them to its symbol table. The symbol

table cOntains all defined symbols along with the binary value

assigned to each Symbol.
Initially, the Assembler’s symbol table contains the mnemonic
op—codes of the machine instructions and the ASsembler pseudo—op
codes, as listed in Appendix C. As the source program is processed,
user defined symb01s are 'added’to the symbol table.
If, during pass 1, PAL III detects that the symbol table is full
(in other words, there is no more memory space to store symbols
and their associated values), the symbol table full diagnostic:
ST

‘

xxxx

nnnn

printed; xxxx, is the symbol that caused the overflow condition

and

nnnn

is the current location whenithe

overflow occurred.

The Assembler halts and may not be restarted.
More address arithmetic should be used to reduce the number

of symbols. It is also possible to segment a program and assemble

separately, taking care to generate proper links between the segments. (See the MACRO-8 section Dealing with a
Limited Symbol Space.) PAL III’s symbol capacity when using the
high-speed reader is 558 symbols. The permanent symbol table contains 80 symbols, leaving space for 478 possible user—defined symbols. When using the low—speed reader, PAL III’s symbol capacity
is 656 symbols, leaving space for 576 user—defined symbols.
the segments

DIRECT ASSIGNMENT

STATEMENTS

The programmer inserts new symbols with their assigned values

directly. into the symbol table by using a direct assignment statement of the form:

SYMBOL:

VALUE

VALUE may be a number or expression. No

space(s)

tab(s)

may appear between the symbol to the left of the equal sign and
the

equal sign. The following are examples of direct assignment

statements:

A=6

EXIT=JMP

1,6

C=A+B
.

All symbols to the

right of the equal sign must be already defined. The symbol to the left of the equal sign is subject to the same
restrictions asa symbolic address, and its assOciated value is stored
in the user’s symbol table. The use of the equal sign does not increment the'location counter. It is, rather, an instruction to the
’

Assembler itself.
A direct

assignment statement may also equate a new symbol
to the value assigned to a previously defined symbol. In this Case
the two symbols Share the same memory location.
I

BETA=17

GAMMA=BETA

The new symbol, GAMMA, is entered into the user’s symbol table

with the value 17.
The value assigned to a symbol may be changed as follows:
ALPHA=5

ALPHA=7
The second line of code shown

changes the value assigned to
ALPHA from 5 to 7. (This will generate an RD error, explained
below.)
SymbOls defined by use of the equal sign may be used in any
valid expression. For example:
*203
A=1®®

/DOES

NOT

UPDATE

CLC

B=4®®

/DOES

NOT

UPDATE

CLC

A+B

/THE

VALUE

saw

/THE

VALUE

110%

TAD

ASSEMBLED
ASSEMBLED

AT
AT

LOC.
LOC‘

20%
2G1-

symbol to the left of the equal sign has already been
defined, the redefinition diagnostic:
If the

will be
and

xxxx

nnnn

printed as a warning, where xxxx is the symbol name
is the value of the location counter at the point of

nnnn

‘

redefinition. The new value will be stored in the symbol table; for

example:
CLA=76®®

will cause the diagnostic:

CLA‘

sane

Whenever CLA is uSed after this point, it will haVe the value 7600.

Multiple assignments can be carried to two levels only. Where
X is some previously defined symbol or combination of symbols:

error

AZBZX

will assemble, but

A:B=C=X

will not assemble.

of this type will cause a “Pushdown Stack Overflow”

diagnostic of the form:
PO

xxxx

nnnn

This is a non—recoverable error condition which causes the assembly
to terminate. Continuation is not possible at
must be corrected and the

this point. The error
i

assembly restarted.

The

expression to the right of the rightmost equal sign must
composed completely of numbers and/0r previously defined
symbols.
The omission of the tag in a direct assignment statement will
cause various errors, depending on the placement of the statement.
If a statement such as:

before any symbol has been used in a program, PAL III
will generate an RD error message during pass 1, with a meaningoccurs

printout for the symbol being defined. If the statement :3
after a symbol has been used, the assembler assumes the
last symbol referenced is being redefined. For instance:
less

occurs

*BQG
CLA

will cause the diagnostic:
l3

AT.

(Nix

@201

during pass 1. In either case, PAL 111 may be restarted after pass
1, but attempting to continue to pass 2 will leave the Assembler in
a

state from which it can neither be continued for pass 3

nor re-

started for another pass 1.

Expressions
Symbols and numbers are combined by arithmetic and logical
operators to form expressions. There are three operators:
'

Signifies tWO’s cOmplement addition
Signifies‘ two’s complement subtraction
Space is interpreted in context

plus

minus

‘-—-

space

When a space occurs in an expression that does not contain a
,

memory reference

instruction, it‘means an inclusive OR is to be

performed. For example:
CLL

CLA

The symbol CLA has

advalue of 7200 and the symbol CLL has a

value of 7100; CLA CLL would produce 7300. User defined sym-

bols are treated as operate instructions. For example:

A:333
‘4222

CLA

Possible expressions andtheir Values using the symbols just defined
are

shown below. Notice that the Assembler reduces each expres—
'

‘

sion to one 4—digit (octal) word:
0333

A
.

—

0222

0555

’

A+B

“

‘

A—B
-——A

0111

7445

1—B

7557

B—l

0221

A B

0333

——71

7707

(an inclusive OR is performed)
'

etc.

‘An expression is terminated by either a comma, carriage return,
or

semicolon. If theeinformation generated is to be loaded, the cur-

rent location counter is incremented. For example:
A+43

B-75

A+B

words of information and the, current location
counter is incremented after each expression. The statement:

produces three
,

CLA

HALT=HLT

produces no information to be loaded (it produces an association
in the symbol table)
does
not increment the current locaand

hence

tion counter.

*4721
‘

TEMP,
TEM2,

.0.

The location counter is not incremented after the line TEMP; the
two

symbols TEMP and TEMZ are assigned the same value, in

this case 4721.

Since

PDP-8

instruction has an Operation code of three hits

as well as an indirect bit, a page bit, and seven address bits, the
Assembler must combine memory reference instructions in a manner

somewhat

from the

differently

Way in which

Combines

operate or IOT instructions. The Assembler differentiates between
the symbols in its permanent symbol table and user defined symbols.

tions:
-

The following symbols are used as memory reference instruc-

AND
TAD

Logical AND

0000

182

2000

DCA

3000

JMS

4000

,Two’ 3 complement addition
Increment and skip if zero
Deposit and clear accumulator
Jump to subroutine

J MP

5000

Jump

FADD

1000

Floating addition
Floating Subtraction
Floating multiply
Floating divide
Floating GET
Floating PUT

1000

FSUB

2000

FMPY

3000

FDIV

4000

FGET

5000

FPUT

6000
’

When the Assembler has prOcessed one of these
space folloWing' it acts

symbols,fithe

address field delimiter;

*4100
JMP

CLA

A has the value 4101, JMP' has the Value 5000, and the space acts
as a field

delimiter. These symbols are represented as follows:
A

100 001 000 001

JMP

101 000 000 000

The seven address bits of A are taken, ire:
000 001 000 001

The
zeros

remaining bits of the address are tested to see if they are
(page zero reference); if they are not, the current page bit is

set:

001..

0000117 000

The operation code is then ORed into the J MP expression to formi
l

101 011 000 0017
l

or, written more conciselyin octal:

5301
In addition to the above tests,‘ the‘page bits of the address field

compared With the page bits of'the current loCatiOn counter.
If the page bits of the address field are nonzero and do not equal
the page bits of the current location counter, an out-of-page reference is being attempted and the illegal reference diagnostic is
printed on pass 2 or pass 3. For example:
are

‘

*aioo
CLA

CLL

*7200
JMP

The

symbol in the address field ofthe JMP instruction has a

value of 4100 while the location coUnter
16

(the address where the
.»

instrhction is placed'in memory) has a valueoft7200. This instruction is

illegal because PAL III does not generate off-page references, and will be flagged during pass 2 or pass 3 by the illegal
reference diagnostic:
4100

Such

7200

NOTE
diagnostic would not be

generated
when using MACRO-8, which automatically
generates off—page references.
a

Address Assignments

The PAL III Assembler sets the origin,

starting address, of

the source program to absolute location (address) 0200 unless the

origin is otherwise specified by the programmer. As source stateare processed, PAL III assigns consecutive memory addresses to the instructions and data words of the object program.
This is done by automatically incrementing the current location
counter each time a memory location is assigned. A statement
which generates a single object program storage word increments
the location counter by one. Another'statement might generate six
storage words, incrementing the location counter by six.
Direct assignment statements and some Assembler pseudo—ops
ments

do not generate storage words and therefore do not affect the location counter.

CURRENT ADDRESS INDICATOR
The special character period (.)

always has a value equal to the
value of the current location counter. It may be used as any integer
or Symbol (except to the left of an equal sign), and must be preceded by a space when used as an operand, For example:
*200
JMP

.+2

is equivalent to J MP 0202. Also,
*300
9+2400

will produce in location 0300 the quantity 2700.

Consider:

*2200

CALLzLJMS

0027

The second line

( CALLZJ MS I .) does not increment the current
location counter, therefore, 0027 is placed in location 2200 and
CALL is placed in the user’s symbol table with an associated value
of 4600

(the octal equivalent of JMS I .)..

INDIRECT ADDRESSING
When the character 1 appears in a statement-between .a memory

reference instruction and an operand, the operand is interpreted
as the address (or location) containing the address of the operand
to be used in the current statement. Consider:

Tanpao
which is a direct address statement, where 40 is interpreted as the

address onpage zero containing the quantity to be added to the
accumulator. References to addresses on the current page and to
page zeromay be done directly. An- alternate way to note the page
zero reference is with the letter Z, as follows:
.

TAD

This is an optional notation, not differing in effect from the pre—
ViOUS example. Thus, if address 40 contains
0432, then 0432 is
added to the accumulator. Now consider.

TAD

which is an indirect address statement, where 40 is interpreted as
the address of the address containing the quantity to be added to
the accumulator.

Thus, if address 40 contains 0432, and address

432 contains 0456, then 456 is added to the accumulator.
18

NOTE
“Because the letter I is used to indicate indirect addressing, it is never used as a vari-

able. Likewise the letter Z, Which is

some-

times used to indicate a page zero reference,
is never used as a variable.
AUTOINDEXING

Interpage references are often necessary for obtaining operands
processing large amounts of data. The PDP-8 computers
have facilities to ease the addressing of this data. When one of the
absolute locations from 10 to 17 (octal) is indirectly addressed,

when

the contents of the location-is incremented before it is used as an

address and the incremented number is left in the location. This

allows the programmer to address consecutive memory locations

using a minimum of statements.
It must be remembered that initially these locations (10 to 17
on page 0) must be set to one less than the first desired address.
Because of their characteristics, these locations are called autoin—
dex registers. No"incrementation takes place when locations 10‘
to 17 are addressed directly. For example, if the instrirction to be
executed next'is in locati0n 300 and the data to be referenced'is
on the page starting at location 5000,
autoindex register 10 can be
to address the data as follows:
used
7

@276

1377

TAD

sala

DCA,1®~V
TAD
I
to

04777

/=Saca-1

@277

aaac

141a
.1

v/SET

‘/BEFORE

@377

4777

AUTO

/INCREMENT

USE

INDEX

5000
AN

ADDRESS

04777,4777

When the instruction in location 300 is executed, the contents

of location 10 will be incremented to 5000 and the contents of
location 5000 will be added to the contents of the accumulator.

When the instruction TAD I 10 is executed again, the contents of
location 5001 will be added to the accumulator, and so on.
Instructions
,

two basic groups of instructions: memory reference
and microinstructions. Memory reference instructions require an
There

are

operand; microinstructions do not require an operand.
19

MEMORY REFERENCE INSTRUCTIONS
In PDP- 8‘ computers

instructions

require a reference to
memory. They
appropriately designatedmemory reference
instructions, and take the following
some

are

format:

OPERATION
CODES (D— 5

MEMORY
pAGE

INDIRECT
ADDRESSING

‘

Bits 0

17 1

ADDRESS

Memory Reference Instruction Bit Assignments

through 2 contain the operatiOn che of the

be performed. Bit 3

instruction

tells the computer if the instructiOn is in-

direct, that is, if the address of the instruction specifies the loca—
tion of the address of the operand. Bit 4 tells the computer if the
instruction is
referencing the current page or page zero. This leaves
bits 5 thrOugh ll (7 bits) to specify an address. In these 7
bits,
200 octal or 128 decimal locations can be specified, the page bit
increases acCessible locations to 400 oCtal or 256 decimal. For a

list of the memory reference instructions and their
Appendix C.

codes, see

be followed

In PAL III a memory reference instruction must
by a space(s) or tab(s), an optional I or Z designation, and any
valid expression, and may be defined with the FIXMRI instruction,

explained under the section on Altering the Permanent Symbol
Table. Permanent symbols may be defined using the FIXTAB instruction, and may be used in address fieldsas shown below:
A=1234
FIXTAB
TAD

MICROINSTRUCTIONS

.
-

Microinstructions

put/ Output Transfer

are

divided into two groups: operateand In-

(IOT) microinstructions.
20

NOTE
If a programmer mistakenly makes an illegal

combination of microinstructions, the Assembler will perform an inclusive OR be—
tween them; for example;

interpreted as SPA
(7510)

CLL SKP is

(7100 7410)

Operate M icroinstructions

Within the operate group, there are two groups of microinstruc—

Group 1 microinstructions perform
clear, complement, rotate and increment operations, and are des—
ignated by the presence of a 0 in bit 3 of the machine instruction
word. (See Permanent Symbol Table list in Appendix C.)
tions which cannot be mixed.

OPERATION

‘I

CLA

CMA

CONTAINS
A 2) TO
SPECIFY
GROUPI

CLL‘

ROTATE
AND L
RIGHT

ROTATE 1
POSITION IF A0
ZPOSITIONS
IF

CML

IAC

ROTATE
AC AND L
LEFT

Group 1 Operate Microinstruction Bit Assignments

Group 2 microinstructions check the contents of the accumulator and link and, based on the check, continue to or skip the
next instruction. Group 2 microinstructions are identified by the
presence of a 1 in bit 3 and a 0‘ in bit 11 of the machine instruction

word (See Appendix C).

REVERSE
SKIP
OPERATION
CODE 7

CLA

SENSING OF
BITS 5,6,7

SZA

HLT

AFN/AA

01234567891011
V
CONTAINS A1
TO SPECIFY
GROUP 2

SMA

SNL

03R

CONTAINS A a
TO SPECIFY
GROUP 2

Group 2 Operate Microinstruction Bit Assignments
21

1 and

Group

Group 2 microinstructions cannot be combined

because bit 3 determines either one or the other.
Within Group 2, there are two groups of skip instructions. They
can be referred to as the

OR group and the AND group.
I

Group

‘
.

AND Group

SMA

SPA

SZA

SNA

SNL

SZL

The OR group is designated by a 01n bit 8, and the AND group

by a 1 in bit 8. OR and AND group instructions cannot be com-

bined because bit 8 determines either one or the other
If the
programmer does Combine legal skip instructions it is
important to note the conditions under which a skip may oCcur.
1.

OR Group———If these skips are combined in a statement, the
inclusive OR of the conditions determines the

skip.

For

‘

example:
SZA'SNL

The next statement is skipped if the accumulator contains
0000, or the link is a 1, or both conditions exist.
2.

AND Group—-—lf the skips are combined in a statement, the

logical AND
example:
SNA

of the, conditions determines the

skip.

For

SZL

The next statement1s skipped only if the
from 0000 and the link1s 0.

accumulator differs
*

Input/ Output Transfer Microinstructions
These microinstructions initiate

operation of peripheral equip-

ment and effect an information transfer between the central proces-

sorand the Input/ Output device(s).
The Permanent Symbol Table in Appendix. C contains the com,
monly used in IOTs for the disk, TTY, and high speed devices.
These and other IOTs

are

discussed in detail in the Small Com-

puter Handbook.
22

PSEUDO-OPERATORS‘

’
‘

The programmer usesp‘seudo-Dperators to direct the Assembler

perform certain tasks or to interpret subsequent coding in a
certain manner. Some pseudo-ops generate storage words in the
“object program, otherpseudo—ops direct the Assembler [as to how
to proceed with the assembly: Pseudo-ops are maintained in the
permanent symbol table; pseudo-ops should" not be used as variable
to

names within a program.

The function of each PAL III pseudo-op is described below.

Indirect Addressing

I‘M SymbOIic repreSentation for indirect addressing, must be
separated on eaCh side by at least one space.
For example:
'

DCA

ADD

The; value of the symbol ADDis’ used as the address of the address
in which the contents of the accumulator will be stored.

Optional method of denoting a

page

For example:
DCA

ADD

DCA

zero

reference.

ADD'

The two statements abovehave the same meaning and generate

the Same code where ADDis on page Zero.

Both Z and I can be present in the same

by at leastone space, as follows:
DQA z

‘

instruction, separated

ADD

which is the same as:

Don't

ADD

Radix Control

Numbers used in a source program are initiallyconsidered to be
23

octal numbers. However, if the programmer wishes to have certain

numbers

interpreted

decimal,

use

can

the pseudo-op

‘

VDECIMAL.
All following numbers are taken as. decimaluntil

DECIMAL
‘

OCTAL

Extended

the occurrence of the pseudo-op OCTAL.
Resets the radix to its original OCtal‘base.

Memory

When using more than one memory bank, the pseudo

op FIELD

instructs the Assembler to output a field setting. This field setting
is punched during pass 2 and is reCOgnized by the Binary Loader,
which in turn causes all subsequent information to be loaded into

the field specified by the expression.
FIELD

Where n. is an integer, apreviously defined symbol,
or

an expression within the range 0 < n < 7.

The FIELD pseudo-op causes a field setting

(binary word) of the

form:

‘

XXX

000

where.

to be output on the binary tape

000<xxx<1112

during pass 2 followed by an origin

setting of 200. This word18 read by the Loader, which then
loading information into the new field
The field setting is never remembered by the Assembler, and no
initial field setting is punched. A binary tape produced without
field settings may be loaded into any one field of core by appropriate manipulation of the Data Field switches when using the
Binary Loader. Asymbol in one field may be used to reference the

begins

location in any other field. The field to Which it refers is
determined by the use of the CDF and CIF instructions. (The pro—
same

grammer who is unfamiliar with the IOTs but'Wishes to use them
should experiment with several short test programs to satisfy him—

self as to their effect.) An example of
conservation might be.

this method of symbol space

*200

~‘

0200

@201

0005
0006

DATA:

0202

0007

0203

@004

@300

1200

‘DATA

@301

1203

DATA

*300
AC

/YIELDS

DATA

/YiELDs

DATA+4

/Y1ELDS

1200

IN
IN

1a»
1a
FIELD
@200

@006

0201

0007

0202

0003

0203

0002

@204

1200

TAD

@205

7200

CLA

0206

1204

TAD

NOTE
GDP

and CIF instructions must ‘be used

prior

to any

instruction referencing a loca—

tion outside of the

current field,

as
‘

in the following example:
*200

TAD

P301

VCDF 00

P301:

CIF

JMS

CIF

JMP

301'

FIELD

*200

NEXT;

TAD

P302

CDF

JMS ‘PRINT
HLT

P302;

302

PRINT:

TLS
TSF

JMP

--1

CLA
RDF

TAD

P6203

DCA

.+1

000

JMP
P6203:

6203

IMPRINT‘.
25

shown

When FIELD is used, the Assembler follows the new FIELD set-

ting with an origin at location 200. For this ‘reason, if the programmer wants to assemble code at location 400 in field 1 he must
write:

FIELD
*4GG

/CORRECT

EXAMPLE
‘

The following is incorrect and will not generate the desired codei

*Qflfl
FIELD

/INCORRECT
1

End of Tape

..
f

The pseudo—op PAUSE signals the Assembler to stop processing

the paper tape being read. The Current passiis not terminated, and

processing continues when the user depresses the CONTinue key.
When processing a segmented program, the programmer uses
the PAUSE pseudo-op as the last statement of each segment (tape)
to halt Assembler processing, giving him time to insert the next
segment of his program.
The PAUSE

pseudo-op should be used only at the physical

end of a tape or file and with two or more tapes of one program.
When a PAUSE statement is reached,
1.

The Assembler stops.

This is the physical end of the tape; the Assembler resets
A

the input buffer pointer.
3.

Operator intervention is required to put the next tape segment of the program in the reader and press the CONTinue

key.
If a PAUSE is encountered somewhere other than at the physical

end of a tape, some of the user code immediately after the PAUSE
will not be assembled. This occurs because PAL III has ‘an input

buffer to allow maximum use of reader speed. A tape is read in
until the buffer is filled or the physical end of the tape is reached.
26

The contents of the buffer are then processed. However, uponrec-

ognizing a PAUSE, PAL lll resets the buffer to empty and waits
for step 3 above.
End of Program

special symbol dollar sign ($) indicates the end of a prothe Assembler encounters the dollar sign, it termiWhen
gram.
nates the current pass. The Assembler must read a 58 after each
The

pass before it will correctly proceed with the assembly.

Altering the Permanent Symbol Table
PAL 111 contains a table of symbol definitions for the PDP 8
and its most common peripheral devices. These are symbols such
as TAD, DCA, and CLA, which are used in most PDP- 8 programs. This table is considered to be the permanent symbol table

1:3”

PAL Ill; all of the symbols it

If the user purchases

one

contains are listed in Appendix

more optional devices whose in—

struction set is not defined among the permanent symbols (for
example EAE or an A/D Converter), he would Want to add the
necessary. symbol definitions to the permanent symbol table in
every program he assembles. Conversely, the User who needs more
space for user defined symbols would probably want to delete all

definitions except the ones used in his program. For such purposes,
PAL III has three pseudo-ops that can be used to alter the perma-

nent symbol table. These pseudo-ops

are

recognized by the As3
are
pass they

sembler only during pass 1. During either pass 2 or

ignored and have
EXPUNGE

efiect.

the entire

Deletes

cept pseudo-ops.
FIXTAB
~

symbol table,

symbols

presently defined
permanent symbol table. All symbols

Appends

‘

permanent

all

before the occurrence of FIXTAB
of the

are

ex—

the

defined

made part

permanent symbol table until the Assem-

gbler is reloaded. For example, the PAL 111 EXtended Symbols
To

Tape ends with FIXTAB.

append the following RF08 disk lOTs to the Symbol table,

the programmer generates an ASCII tape of:
27

DCIM=6611
DIML=6615
DIMA=6616
DFSE=6621

DISK=6623
DCXA=6641
DXAL=6643

'Dxaczeeas
DMMT=6646
FIXTAB
PAUSE

The ASCII tape is then read into core ahead of the symbolic
program tape during pass 1. The PAUSE pseudo-op stops assem-

bly, and the Loader waits for the programmer to put the symbolic
program tape into the tape reader and press CONTinue.
Each time the Assembler is loaded, PAL III’s permanent symbol

table is restored to contain only the permanent symbols shownin
Appendix C.
After altering the

symbol table to fit his needs, the user might

want to keep PAL III in this state for future use. This can be done

by punching a binary of the section of core Occupied by PAL III
with its new symbol table.
To do this:
1.

Readin PAL III and modify

PAL III’s

symbol table

symbol table begins

desired.

location 2332 (octal).

Count all the symbols in the altered symbol table. Since each

symbol and its value require four words, multiplythis number by 4. Convert this number to octal and add it to 2332
(octal). This number is the upper limit ‘of PAL III. The
lower limit13 0001.
3.

Using the Binary

Punch Routine

(DEC- 08

-YXlA-PB),

which does a binary core

dump

the high-speed or Tele-

type punch, and the limits as stated in 2 above,
the PAL III Assembler.
4.

punch

out

The output of the Binary Punch Routine is the Assembler
with the modified symbol table and can be loaded with the

Binary Loader. This revised version of the Assembler can
thereafter be used instead ofthe original version.
The third pseudo-op used to alter the permanent symbol table
in PAL III

(and not present in MACRO-8) is FIXMRI which

may be used only after an EXPUNGE instruction:
28

‘

FIXMRI

Fix memory reference instruction. Memory reference

instructions are stored in the permanent symbol table

immediately following the pseudo-ops. The letters
FIXMRI must be followed by one space, the symbol
for the instruction to be defined, an equal Sign, and
the value of the symbol. The pseudo-op must be re—
peated for each memory reference instruction to be

‘

defined. All memory reference instructions must be
defined before the definition of. any other symbols.
For example:
EXPUNGE
FIXMRI

TAD=1QJ®®

FIXMRI

DCA=3$Q®

CLA=72®®
FIXTAB

PAUSE

When the preceding program segment is read into the Assembler

during pass 1, all symbol definitions are deleted and the three
symbols listed are added to the permanent symbol table. Notice
that CLA is not a memory reference instruction. This process is
often

performed to alter the Assembler’s symbol table so that it
contains only those symbols used at a given installation or by a
[given program. This may increase the Assembler’s capacity for
user defined symbols in the program.
Program Preparation and Assembler Output
In PAL 1H and MACRO—8, the source language or, symbolic tape is punched in ASCII code on 8—channel paper tape, using
‘

off—line Model LT-33

Teletype or the on-line Symbolic Editor.

prOgram should, begin with leader code, which may
be blank tape, code 200, or RUBOUTs.
Certain codes which the Assembler ignores may be used freely
to produce a more readable symbolic program listing.
These codes
are TAB and LINE FEED. The. Assembler also ignores extraneous
spaces, carriage return/line feed combinations, and blank tape.
In general,

When the Assembler encounters

form feed character, it causes

12 blank lines to be output on the listing

(in PAL III only).

The two programs below are identical and produce the same
binary code. The second, however, was generated using the TAB
29

‘

the‘Symbolie Editor and is easier to read. The first
program assembles faster only because there .is less paper tape to
functionof

be read into the computer.

PrOgram #1:
*GGG

/EXAMPLE

INPUT

/GENERAToR

G/START

BEGIN,

THE FORMAT

PRoGRAM
PROGRAM

KCC
FOR

KSF/wAIT

FLAG

.—I/FLAG

JMP

HRH/READ

SET

NOT

YET

CHARACTER

DCA

CHAR

TAD

CHAR

TAD

MSPACE/IS

SNA

CLA

SPACE?

HLT/YES
JMP

BEGIN+2/NO:INPUT AGAIN
STDRAGE'

CHAR G/TEMPCRARY

EQUIVALENT

MSPACE,-RAG/- ASCII
/END OF

EXAMPLE

Program #2:
*BOO
/EXAMPLE

/GENERATOR

INPUT

THE FORMAT

PROGRAM
'

/START

‘

BEGIN,

PROGRAM

’

‘

KCC

KSF

SJMP’--1‘
CHAR

TAD
TAD

MSPACE

YET

SNA

CLA

/IS

A.SPACE?
'

BEGIN+2

MSPACE,

-24O

/END

EXAMPLE

SET

CHARACTER

‘

CHAR

HLT

CHAR,

FLAG

FOR

/READ

DCA

JMP

/WAIT

/FLAG NDT

KRE

i/YES
/No:

INPUT

AGAIN

/TEMPoRARY STORAGE
/¥ASCII EQUIVALENT
;

The program consists of statements and pseudo—ops, is termi—
nated by the dollar sign (33.), and followed by some trailer code. If
the program is large, it can be segmented using the pseudo-op
PAUSE, which oftenfacilitates the editing of the source program

since each section‘will be physically smaller.
The Assembler

initially

sets

the current location counter to

0200. This counter is reset whenever the asterisk (*) is processed.

During pass 1, all illegal characters cause a diagnostic to be
printed. The tape should be corrected and reassembled.
The Assembler reads the source tape and defines all symbols
used. The user’s symbol table is printed (or punched) at the end
of pass 1. The symbol table is printed in alphabetical order. If
any symbols remain undefined, the undefined address diagnostic
is printed. If the program listed on the previous page were assembled, the pass 1 symbol table output would be:
‘

BEGIN

CHAR

MSPACE

cane
@213

,fl214

DUring pass 2, (the Assembler reads the source tape and generates the binary code“ using the symbol table equivalences defined during pass 1 The binary tape that is
punched may be
lOaded by the Binary Loader. This binary tape consists of
code, an origin setting, and data words. At the end
of pass 2, a checksum is punched on the binary tape, and trailer
code is generated.- During pass 2, the Assembler may diagnose an
illegal reference; When using the LT—33 punch, the diagnostic
is both printed and punched, and is preceded and followed by
RUBOUTS. The Binary Loader ignores
everything that has been
a
RUBOUTS
punched on tape between
During pass 3, the Assembler reads the sOurce tape and gen—
erates the code from the source statements. The assembly listing
is printed (Or punched). It consists of the current location counter,
the
generated code in octal, and thesource statement. The symbol
tableis printed at the end of the pass. If the sample program listed
above were assembled, the pass 3‘ output would be:
leader

*QOO

/EXAMPLE

/GENERATOR

a2aa

aaaa

@221

6232

KCC

@222

6931

KSF

BEGIN,

@223

5222

JMP

@224

6®36

KRB

.-1
.

DCA

CHAR

1213

TAD

CHAR

@207

1214

TAD

MSPACE

@210

7652

SNA

CLA

2211

7422

.HLT

52a2

(Rana

@214

75ma

MSPACE,

amp

PROGRAM

3213

anon

FORMAT

@265

C213

TO THE

/START

®2®6

@212

INPUT
PROGRAM

/wAIT

FOR

/FLAG

NOT

/READ

CHARACTER

SPACE?,

FLAG
SET

YET

/IS

/YES

BEGIN+2

/NO:

INPUT

AGAIN

/TEMPORARY STORAGE
/ASCII EQUIVALENT

-24a
‘

/END
BEGIN

22am

CHAR

2213'

MSPACE

EXAMPLE

'0214

Operating Procedures
The PAL III Assembler is

provided to DEC customers as a
binary tape, which is loaded into the PDP-8 memory by means of
the Binary Loader, using either the LT~33 reader or the highspeed reader. The Assembler also uses either the LT-33 reader
or the high—speed reader to read the source language tape, and it
uses either the LT.33
punch or the high—speed punch for output.
The selection of I/O devices is made when the Assembler is
started. The source language tape must be in. the proper reader,
with the reader and punch turned on.
table is printed
When using the high- speed punch, the
if
on the
bit 11 of the switch register is a O.
LT- 33 Teletype
if bit 11 of
The symbol table is punched on the high-speed
the switch register is a 1..
All diagnostics are printed on the LT— 33 except for the undee
fined address diagnostic when using the LT- 33 punch, or the
high—speed punch if it is included in the machine configuration and
turned on. The only diagnostic in pass ,2 will be illegal reference.
(Since this diagnostic is printed on the LT-33, it will also be
punched on the binary tape. It will, however, be ignored by the
Binary Loader.) The bit 11 switch option can also be used. during
.

symbol

punch

pass 3. If the machine is not equipped with a high-speed punch,
bit 11 must be set to 0.

binary tape of the PAL III Assembler, the
user is provided with an ASCII tape (PAL III Extended Symbols
Tape) containing symbol definitionsfor the instruction sets of the
available options to the PUP—8 (card readers, magnetic tapes, and
A/ D converters). A limited amount of space is available in a‘4K'
system; therefore, expanding the number of permanent symbols
that the Assembler recognizes will decrease the maximum number
of symbols the user has available
In addition to the

The

following is a description of

Assembler:
1.

steps

Load the ASsembler using either the

using the

III—33

PAL III

reader or the

high—speed reader (see Appendix A).

Set 0200 into the Switch Register; press ADDRess LOAD
Place the source language tape in the reader, turn on the
appropriate reader and the punch.

Set bits 0 and 1 of the Switch Register for the

These settings are:

proper mass

Pass 1 is

Bit 0

Bit I

pass 1
pass 2

pass 3

required so that the Assembler can initialize its

symbol table and define all user symbols. After pass 1 has
been made, either pass 2 or pass 3 can be made.
'

5.’ Bit ll switch options:
pass 1

Bit 11:1

Punch the'symbol table on the high—

speed punch if it is in the machine
configuration.
Print (and punch) the symbol table
on the LT- 33
(low- speed punch).
Punch binary tape on high—speed,
punch.
Punch binary tape on low— speed
punch.
'

Bit 1120

paSs 2

Bit 11:1
'

Bit 11:0

‘

paSs 3

Punch the assembly listing tape in
ASCII on the high- speed punch.
Print the
on
the

.Bit 11:1

Bulls-:0
}

—

Bit 10 switch options:
Bit 10:1
pass 3‘
~

tab stops

Bit 10:0;
.7

TAB

Output TAB as

(code

Output TAB (code 211)

assembly listing

LT-33

—

211

space

RUBOUT

and 377).

Bit 2 switch options:

passes 1 and 3

6.
‘

‘Bit‘ 2::
1

Suppress output of symbol table.

Bit 2:: 0

Output symbol table.

Press CLEAR and

CONTinue to begin pass 1 only Press
CONTinue to begin passes 2 Or 3. The Assembler halts at the
end of each pass. PrOceed from

step 3. If the AsSembler has
halted because of a PAUSE statement putthe next
into
the reader and press

CONTinue.

tape

SUMMARY OF DIAGNOSTIC MESSAGES FOR PAL 111
Pass I Diagnostics
all user symbols,
The Assembler reads the source tape,
and outputs the user symboltablein alphabetical order. Pass
V

defines
"

1 diagnostics are:

xxxx

nnnn

Illegal Character

Where XXXX is the value of the illegal character and nnnn is

the value of the Current 10cation
‘

was

procesSed

XXXX

counter

when the character
'
*

Redefinition

nnnn

symbol being redefined and nnnn is the
value of the current location 'counter at the point of redefinition. The symbol is redefined.

Where

XXXX

is the

AT nnnn
Duplicate Tag
An attempt is» being made to redefine a symbol using the
.ICOIIIIIIELXXXX“ is the symbol and nnnnis the value of the current location counter at the point of redefinition. The previous
value of the symbol is retained and the symbol is not re’

XXXX

defined.

Symbol Table Full
Where XXXX is the symbol causing the overflow and nnnn is
the value of the current location counter at thepoint of oven

XXXX

nnnn

flow. The Assemblerhalts and cannot be restarted.
PO

XXXX

i\Pushdown List Overflow

nnnn‘

’

attempt is being made to carry a multiple assignment to
more than two leVels; i.e., AszCE—tX.XXXX*is the Value of
the pushdown stack pointer (an internal address in PAL III).
An

The value of the current location counter When o’Verflow is

detected is

The Assembler halts at this

nnnn.

point without

reading more source tape. The CONTinue key has no effect
time; however, the Assembler can be restarted at location 200, as indicated in step 2 under Operating Procedures.
at this

XXXX

Undefined Address

nnnn

Where XXXX is the symbol that was used, but never defined,
and

nnnn

is the value of the current location counter when

symbol was first processed. This message is printed with
the symbol table at the end of pass 1. The symbol is assigned
a value equal to the highest address on the memory page

the

where it was first used.
Pass 2 Diagnostics
The Assembler reads the

tape, and, using the symbol
table defined during pass 1, generates and punches the binary
code. This binary tape can then be loaded by the Binary
source

Loader. The pass 2 diagnostic is:
IR

Illegal Reference
Where XXXX is the address being referenced and nnnn is the

XXXX

nnnn

value of the current location counter. The illegal address is
then treated

if it

were

the proper memory page. For

example:
*73E6
JMP

3(2)?

would produce:
IR

@3o7

73o6

and would generate 5307 to be loaded into location 7306.
’

Pass 3 Diagnostics

and, using the symbol
during pass 1, generates and prints the code

The Assembler reads the source tape
table defined

represented by the source statements. The current location
counter, the contents,- and the source statement are printed
side by side on one line. If bit 11 of the Switch Register is a 1
and the machine configuration includes the high—speed punch,
the assembly listing is punched in ASCII. The pass 3 diagnostic is Illegal Reference, as in pass 2.

MAQRD-S PROGRAMMING
Macro-8 is a 4K two-pass paper tape assembler similar to PAL

III, which contains several additional features which may be of
use

to more

advanced PDP—S'programmers. These features include

generation, literals, Boolean operations, double precision integer' input, floating point input, a text input facility, and user—
defined macros. The assembler is compatible in most respects with

link

PAL III; the areas of difference are noted later in this manual.
Characters
In addition

to those characters discussed under PAL III, the

following characters are used in MACRO—8.
Function

Symbol

Name

Ampersand

Exclamation Point

Combines symbols or numbers.

(Boolean AND)
Combines symbols or numbers

(Boolean OR)

”

Double Quote

()

Generates 8-bit ASCII code

Parentheses

Square Brackets
Angle Brackets

[1
<>

"NOTE: On an

Defines a literal on the current

page
Defines a page 0 literal
Defines a macro

tor—33':

is, generated by SHIFT / K-

[
]

is generated by SHIFT/ M

is generated by SHIFT/,

is generated by SHIFT/. (period)

Line Feed

(comma)
‘

(code 212) is ignored, as in PAL III, unless it immediately precedes a dollar sign. The sequence
line feed/ dollar sign is taken as unconditional end—

of—pass.
Expressions
All symbols

numbers (exclusive of pseudo-ops, macro
names, and double precision or floating point constants), may be
combined with certain arithmetic and logical operators to form exand

pressions. These operators are:
37

Name

Operator
7

Plus

,
_

Function

Signifies two’s complement ad-

dition*

Signifies two’s complement sub-

Minus
_traction*
Point
Signifies Boolean inclusive OR
Exclamation

—

;

(union)
Signifies Boolean AND
section) *
Interpreted in context; can signify an inClusiVe OR, or act

(inter-

Ampersand
Space

as a field

delimiter as in

‘

PAL 111

*As explained in Introduction to Programming.
NOTE
may not contain pseudo-ops
other than I and Z, macro names, double

Expressions

precision integers,

stants. To do

or,

so is an

code may result

being giVen.

floating point
error, and

without

con-

erroneous

error

message
’

Symbols and integers may be combined with any of the preced—
ingoperators. A symbolic expression is evaluated from left to right
with one exception—space, which serves as a delimiter. Grouping
of terms via the use of parentheses is not permitted in MACRO—8.
Consider the following examples:
A

-B

A+B

VALUE

0002

VALUE

0007

VALUE

0700.

0003

0005

0014

0007

0707

Since space is treated as a delimiter,
expression, the two expressions:

A!B+C
A

B+C

Ala-B

A113

A&B

7777

0003

‘0002

0002

0671

.0007

0005

0707

0000

separating two terms of an

are

not equivalent. In the first case the assembler interprets the

expression from left to right as follows:
(A!B)+C

In the second case, each term is considered separately and the re-

sults combined with an inclusive OR:
A

(3+0)

As in the following

example, the values of the two expres‘Sions will

differ:

'
-

314+5

7+5: 141..

3 4+5

3.1118 =13,

NOTE
‘

Expressions of the form A +B and A+ B
are

incorrect and may cause erroneous code

to be

generated Symbols and integers must

be combined with a

ing them

single operator separat-

Origin Setting

The origin is ordinarily set by use of the special character aster-

isk (*)

described in PAL III. All symbols to the right of the

asterisk must already have been defined. For example, if D has the
.

value 250 then:

*D+1®i
'WIll set the location counter to 0260.

programmer’s addressing problems, a convention
has been defined that divides memory into sections called pages.
Each page contains 2001: locations (128m) numbered 0 to 17711
To

ease

the

page. There are 408 or 32“, pages numbered 0 to 378. Some
examples of page numbers and the absolute and relative locations
on that

(addresses) are shown below. See Chapter 2 of. Introduction to
Programming for a Complete list of page numbers and their addresses It must be remembered, however, that there18 no physical
separation of pages in memory.
39

Absolute Address

Page
0

0——i77

0——177

zoo—~377

0——177

400——577

0——177

7400—57777

0-_177

7600——7777

0——177

To simplify page handling, the pseudo-op

PAGE

Relative Address,

PAGE

can be

used:

The PAGE pseudo-op resets the location counter to

the first address of page 11, where n is an integer,

previously defined symbol, or a symbol expression,
all of whose terms have been defined previously and
'

whose value is from 0 to 37, inclusive.
For example:

PAGE

PAGE ‘2

Sets the location counter to 0400

PAGE 6

Sets the location counter to 1400

When used without an argument, PAGE resets the
location counter to the first locatiOH on the next suc-

ceeding page. Thus, if a program is being assembled
into page 1 and the programmer wishes to begin the
page 2 he
pseudo-op PAGE, as follows:

segment

need only insert the

*206

JMP

-7

PAGE

CLA

In this case, the CLA will be assembled into location

400.

If, when the PAGE pseudo-op is

used without

current location counter is at the first

argUment, the
loCation of a page, it will not
an

be moved. In the following example, the code TAD B is assembled
into location 400:

*377
PAGE

If several consecutive PAGE

pseudo-ops are given,

the first

will cause the current location counter to be reset as specified. The
rest of the PAGE pseudo-ops in the sequence will be ignored.

NOTE
Since PAGE is a pseudo-op in. MACRO-8,
it is illegal

PAGE:

tag. Therefore the coding:

G
TAD

PAGE

may not be used:
Link Generation

symbolic addressing on the current
page of core memory, MACRO-8 automatically generates links for
off-page references. MACRO—8 compares the page bits of the adIn addition to handling the

dress field with the page bits of the location counter. If the page
bits of the address field are nonzero (not a page 0 reference) and
do not

equal the page bits of the location counter an off-page
referenceIS being attempted.

the

If reference is made to an address not on
page Where the
instruction is located, the assembler sets the indirect bit (bit 3),

and

indirect address linkage will be generated

the current

memory page If the off—page reference is already an indirect one,
the error diagnostic II (Illegal Indirect)
be generated during

passes 2 and 3.
When a link is

will

generated, the LG (Link Generated) message

will be printed on passes 2 and 3. In the case of several off-page
references to the

same

address, the link will be generated only

once, but the LG message will be printed each time.

*2117

‘

‘CLA

*2600

In the example above, the space preceding the user defined sym-

bol A acts as; an address field, delimiter. The Assembler will recognize that the register labelled A is not on the current page

(in this

case 2600 to 2777) and will generate a link to it as follows:

In location 2600 the Assembler will place the word 5777

which is equivalent to IMP I 2777.
2.

address 2777 (the last aVailable location on the current

page) the Assembler will place the word 2117 (the actual
address of A).

Although the Assembler will recognize and generate an indirect
address linkage when necessary, the programmer may indicate an
explicit indirect address by the pseudo—op I. This must be between
the instruction code and the address field, as it would be placed in
PAL III. TheAssembler cannot generate a link for an instruction

that is already specified as being an indirect reference and will print
message II

(Illegal Indirect). For example:

*2117
A,

CLA~

*2600
JMP

The

above coding will‘not work because A is not defined

page where J MP I A is attempted and
Literals

the

the indirect bit is already set.

Symbolic expressions appearing in the operand part of an in—
struction usually refer to locations containing the quantities being
operated upon. Therefore, the programmer must explicitly reserve
locations to hold his constants. The MACRO-8 language provides
a means (known as literals) for using a constant directly. Suppose,
42

for

example, that the programmer has an index which is to be

incremented by two. One way of coding this operation would be
as follows;

*2aa

CLA;.

TAD‘INDEX

C2,

"TAD

DCA

INDEX.

Using a literal, the same coding would be reWritten as:

CLA
TAD

INDEX

TAD

(2)

DCA

INDEX

The left parenthesis is a signal to the Assembler that the expression following is to be evaluated and assigned a word in the constants

table of the current page. This is the same table in which

the indirect address linkages

are

stored. In the previous example,

quantity 2 is stored in a word in the linkage and literals list
beginning at the top of the current memory page. The instruction
in which the literal appears is encoded with an address referring to
the address of the literal. A literal is assigned to storage the first
time it is encountered; subsequent reference to that literal from the
current page is made to the same register.
The use of literals in preference to the first method of handling
constants frees symbol storage space for variables.
If the programmer wishes to assign literals to page zero rather
than to the current page, he may use square brackets, [ and ], in
place of the parentheses. This enables him to reference a single
literal from any page of core. For example:
the

*200
TAD

[2]

TAD

[2]

*500

For the first and succeeding times the literal 2 is referenced, iden-

tical code is generated to a single location on page zero containing
the literal. In this case, the following code would be generated:

Location

Contents

0177

0002

0200

0500

1 177

Whether on page zero or the current page, the right
member is
are

(closing)

ignored and may be omitted. The following examples

acceptable:

TAD

(777

AND

[JMP

In the second

example, the instructionAND [JMP has the same
'

effect as:
AND

[5000

Literals can be nested.- For example:
*200
TAD

(TAD

(30

will generate the following:

Location

Contents

0200

1376

‘

0376~

1377

0377

0030
w

This type of nesting can be carried to two levelsas in the. preceding

example. Further nesting will cause BE (overlapping of internal
tables) and US (undefined symbol) error messages to be generated.
Literals are stored on each page starting at page address 177
(relative) and extending toward page address 0' (relative).’If a
literal is generated for a nonzero page and the origin is then set to
another page, the current page literal buffer is output (punched on

pass 2; printed on pass 3). This does not affect later execution. The
uSer may find that the literal buffer hasbeen dumped before the
end of a page to make room for more literals. The same‘literal Will
be generated if used after that point. If the origin is then reset to
the previously used page, the same literal will be generated if used

again, but it will not destroy previously used literals on that page.
Literals and links are stored only as far doWn as the highest in—
struction on the page. Further attempts to define literals will result
in a PE (Page Exceeded) or ZE (Page Zero Exceeded) error mes—
sage. Links are not stored in empty locations preceding the highest

location Used for instruction.
To

summarize, literals may take the following forms:

E"A

("A

where C is a constant, V is a variable, A is any Single ASCII charother than blank tape (code 000) or leader—trailer (code
200), I is an instruction, and E is an arithmetic expression, [ indiacter

page 0 reference and

( indicates a current page reference.
Arithmetic. expressions may consist of cOnstants, variables, and

cates

operators but must not include literals.
Single text characters may be combined with each other’in an
arithmetic expression but not with other constants or variables.
V

For example:
(VIA+I'B_UIC

is equivalent to
(son

An instruction may contain a literal; for example:
TAD

1 [son

(JMP

is valid;

I“

bowieveri

ran ca+<sa
Will not assemble since (A+ (50IS not a valid expression. Literals
may be used as the address part of a memory reference instruction.
TAD

(so

or in

place of an instruction:

(MSG14'
which causes ”the location addreSs of
sembled at the

.pointwhere

the literal (MSG'4A to beEaS—

it occurs in the program.

NOTE
If a large number of nested literals

par-

ticularly large numbers of literals are used,
the literal list may be output before the logical end of the page. This will not affect
'

later execution.

Field Pseudo- 0p
In addilon to punching the field setting and new origin setting
(*200) as in PAL III, the FIELD pseudo— op in MACRO- 8 first
causes all current page literals and links then all page 0 literals
and links to be output. For this
programming
reason the

following

techniques should be considered:
1.

Complete all coding

1n one

field

before moving to another.

If you return to a previous field, the literals and links will
not have been remembered.

exceedingly careful about referencing in one field a

defined in another. The field of a
variable is not stored (the assembler does not know what
variable which

was

field a variable is in). For example:
46

FIELD
~

1030a!

0340

10301

134a

334a

34a

FIELD

anaao

*3Da
TAD

(DOA

In this example, the codingDCA A in field 2 caused octal
.

3340 to be generated, but this may be taken as

refer-

ence to location 340 in field 2.

The argument to the FIELD pseUdo-op should be a value
0<N<7. If N is greater than 7, the bits to the left of the low order
~

‘

digit will be ignored. Thus field 8 Will be assembled as field 0 and
flagged as an illegal character if the radix is octal.
On pass 3, the octal address will be preceded by a single digit
denoting the current field. The first digit is merely a convenience for
the programmer and has no other effect on the assembly.
Text Facility

SINGLE CHARACTER TEXT FACILITY
If a single character is, preceded by a double quote (“), the 8—bit

value of ASCII code for the character is inserted instead of inter—
preting the letter as a symbol. For example;
CLA
TAD

("A

will place the constant 0301 in the accumulator.
The code“

will be assembled as 0256.

g
y

If a carriage return is desired as the text character, it must not
be the carriage return intended to end the line of code. To generate
0215 with this facility, it is necessary to generate:

“[CARRIAGE RETURN /LINE FEED]
[CARRIAGE RETURN /LINE4 FEED]

OI
:

“[CARRIAGE RETURN /LINE FEED]
OI'

“[CARRIAGE RETURN]
[CARRIAGE RETURN/LINE FEED]
.

~47

generated with the Symbolic
Editor, which outputs a line feed after each carriage
The latter cannot be

return. The Editor also cannot generate the sequence:

“[LINE FEED]
[CARRIAGE RETURN / LINE FEED]
which is necessary to generate the code 0212 with this facility. In
this ca'seitis best to use the octal 212 instead of “‘[LINE FEED].

On pass 3, certain characters, while causing the Correct octal

coding to be printed, will be interpreted as format-control characters in the source listing. A summary of special cases follows:
Source Code

Listing Appearance

'
“
'

‘

*200
“

00200

0215

00201

0215

“

’

‘7

"
’

“

°
‘

00201

0212

‘5

00203

0214

00204

0211

“

“(FORM FEED)

“(TAB)

TEXT STRINGS

(FORM FEED)
“(SPACES TO NEXT
TAB STOP)
‘

A string of text characters can be entered by giving the pseudoop TEXT followed by a single space (code 240), any delimiting
character (except blank tape
leader-trailer), a string of text, and

the

same

TEXT

delimiting character. For example:

ATEXTA

The‘character codes are stored two per word in ASCII code that
has been trimmed to the rightmost six bits. Following the last char—

acter, a 6-bit zero is inserted as a stop code. The above statement
would produce:
2405
3024

0000

The string in the following example:
TEXT

/.!./

would

produce:

5641
56@@

The TEXT

pseudo-op could also be used as part of a calling

sequence to a subroutine:

Example 1 :
JMS

MESS

TEXT

Example 2:
JMS

ADDMES,

MESS

NOWDS

/NO

ADDMES

/ADDRESS

WORDS

IN
OF

MESSAGE

TExr /MESSAGE/_
NOTE
While the TEXT pseudo-op
ters to be stored in

charac-

causes

trimmed code, the use

of the

single character control (“) causes
characters to be stored as a full 8-bit ASCII

code.

regard to delimiting characters, it should be noted that the
Symbolic Editor always generates carriage return ascarriage re—
turn/line feed and tab as tab/rubout or as multiple spaces. If car—
riage return (code 215) or tab (code 211) were chosen as
delimiter, line feed (code 212) or rubout (code 377) would become the first character of the text string.
With

NOTE
If no delimiter is seen, line feed/ dollar Sign

will terminate the pass. If no dollar sign is
found, an effective PAUSE (on high- speed

input) is executed, or 55 may be typed (on
low-speed input). (See HALTS at the end
of this manual.)

NUMBERS

The types of numbers allowed in MACRO-8 assemblies are in-

tegers, double precision integers, and double precision floating point
49

numbers. If the characters 8 and 9 are encountered in octal radix,

they are flagged as illegal characters and ignored. Thus 1283 in
octal radix becomes 123.

Double Precision Integers
Double precision integers may be positive or negative (stored as
two’s complement) according to their sign but may not be combined with operators in expressions. They are always taken as
decimal radix although the current radix of the program is not
disturbed. Each double precision integer is allotted two consecutive

words with the sign indicated by bit 0 of the first word, as shown
below:
'

Sign Bit

I WORD 1

‘

WORD 2
12

,
,

Double precision

Constants must be in the range:
—83‘88608<N<8388607

The double precision integer mode is entered throUgh the use of

pseudo-op DUBL. All numbers encountered after the occurrence of DUBL are- considered double precision integers (stored in
2 words) until an alphabetic character, *, or $ is encountered.
Each numberis terminated by a carriage return, semicolon (;) or
comment. For example:
the

*AQQ
DUBL

679467
‘

-3.
TAG,

CLA

Once the double precision mode has been entered, the programmer

*, or $ before proceeding with the program. With the exception of unique floating
point format characters (. and E), the same rules of character
must terminate it with an alphabetic character,

recognition apply for double precision integers as for floating point
constants.

The preceding section of code would produce:

Location

Contents
/

00400

0245

The numbers indicated under contents

00401

7053

are

the octal equivalent of the decimal

numbers

the

preceding example.

00402

0000

The CLA instruction is given the value

00403

0054

7200 as found in the permanent sym-

00404

7777

bOI table.

00405

7775

have a value of 0406.

00406

7200

The

'symbol‘ TAG would

Floating Point Constants
The

floating point input facility is designed. to assemble constants
for use by the PDP—8 Floating Point Package.
Double

precision floating point constants may be positive or
negative according to their sign but cannot be combined with
operators. Decimal radix is assumed but the current radix of the
program is not altered. Floating point iconstants are each assigned
three words and are stored in normalized form, as shown below:

23120th
P

116

Binary exponent:
Two’s complement
signed quantity

High order

412

Sign of

Mantissa

Low order

L
35

Mantissa

The exponent is a signed two’s complement quantity in one 12—

bit word. The signed two’s complement mantissa is stored in two
12-bit

words, maintaining 23 bits of significance, making a total

of three words for storage.
51

’

The double precision floating point mode is entered throughuse
of the pseudo-op FLTG. All numbers encountered after the use of

FLTG will be interpreted

double precision floating point con-

stants until the occurrence ofa
an *,

other than B. The

ber15:

general input

$ or an alphabetic character
format of a
point numa

floating

DDD .DDDEDD

where each D13 a decimal digit. While1n floating point mode,

character13 handled1n one of the following ways:

the format of

floating point

each

‘

If it is a legal part of

ber, it is handled as suchand is used to generate code.
If it is an asterisk, a dollar sign, or an alphabetic character
other than B, it terminates the current number and termi-

num-

nates floating point or double precision conversion.

If it is a format character

turn), it terminates the current number and is otherwise
ignored. Comments are ignored.
If it does not fall into any of the above categories, i.e‘.,
(%) an IC error message is generated. The number
is terminated, and the: character is ignored.

(i. e. ,,space slash, carriage re—_

In floating point format, the mantissa and the exponent may be

preceded by signs (+ or -—). The decimal point and the
exponent
may be omitted. For example:
*400
FLTG

+509.32E®2

-62.97E-4
1.®®E-2
TAG,

CLA

would produce upon executiOn:

Location.

Contents

00400

0003

00401

2427

00402

6675

00403

‘

7771

00404

4615

00405

2172
52

‘

Location

Contents

00406

7772

00407

2436

00410

5604
.

00411

7200

and the symbol TAG would be assigned a value‘of 0411.

Accuracy of floating point conversion compares favorably with
the PDP- 8 Floating Point Package. The most significant variation
is in the handling of large

negative exponents These are Calculated

by multiplying by an inverse rather than by dividing, due to space
limitations. No error checkingis donein the floating point routines;
users,

therefore, should stay within the range of 7 significant digit

mantissas and exponents in the range —600 to 600.

ilser DefinedMacros

program, it often happens that certain coding
sequences are used several times with diflerent arguments. If so, it
When writing

is convenient to generate the entire sequence with a single state—
ment. To do this, the coding sequence may be defined, as a

“macro,” using dummy arguments. A single statement referring to
the macro name, along with a list of real arguments, will generate
the correct sequence in line with the rest of the coding.

DEFINING A

MACRO

The macro name must be defined before it is used. The macro is
defined

pseudo-op DEFINE followed by the
macro name and a list of dummy arguments separated by spaces.
For example, a simple macro to move the contents of word A to
by

means

of the

word B and'leave the result in the accumulator could be coded
as follows:

DEFINE

MOVE

DUMMYl

DUMMY2

<CLA
TAD

DUMMYI

DCA

DUMMY2

TAD

DUMMY2>

The choice of symbols used as dummy arguments is arbitrary; how—

they may not be defined or referenced prior to the macro
definition. The actual coding of the macro is enclosed in angle
ever,

brackets.
53

The preceding definition of the macro MOVE can also be written
as follows:

DEFINE

MOVE

ARGIVARGB

<CLA3TAD ARGISDCA

AR623TAD

ARGB>

The definition of the macro is enclosed in angle brackets, as mentioned before and the semicolon characters indicate the termination

of a line of code, as in PAL 111..

When a macro name is processed by the Assembler, the real
arguments replace the dummy arguments. For example, assuming
that the macro MOVE has been defined as above:
*40@
A:

‘6

B:
7

the

MOVE AaB,

followmg codevis produced:
’

Location
}

Contents
0000

00400;
00401

7772

‘

00402

7200

00403

1200

00404
,

’

f3201

00405”,

1201

Notice that a macro definition has spaces separating the dummy
arguments and the macro call has commasseparating the macro

arguments.
A macro need not have any arguments. For example, a sequence
of

coding to rotate the

accumulator and link six

places to the

left might be coded as a macro by means of the following code:
DEFINE

ROTL

If; core space is tight and the sequence is to be used several times,
it would be better to code this as a subroutine:
54

ROTL:

‘R’I‘L,
RTL
RTL.

JMP

ROTL

The Subroutine occupies five locations and each call, I MS ROTL,

occupies 1. Each call to the macro ROTL occupies three locations.
The main advantage of the macro is the use of dummy arguments.
Its greatest disadvantage is the amount of symbol space required.
The entire macro definition is placed in the macro table, two
characters per word, with a dummy argument‘value replacing the
symbolic names. For example:
LOAD A

DEFINE
<CLA.~

TAD

stored,

in the macro

table, roughly follows:
as

[CLIA [TAID l7700l>00

,Where the vertical lines indicate successive 12-bit words. Comments
andline feeds are not stored. The macro definition can consist of
any valid coding, with the following restrictions:

‘1. Macros cannot be nested, iw.e., another macro name or
definition cannot appear in a macro definition and cannot
be brought in as an argument at the time a macro is refer-

enced.
TEXT (or
definition.

“

type) statements cannot appear in a macro

Arguments cannot be another macro name, a TEXT
pseudo-op or a character.
The symbols used as dummy arguments must not have
been
defined or referenced or subsequently
“

previously

redefined

‘

A macro cannot be redefined (the macro name may not be

used for any other purpose).

macro definition must

occur

in a comment

end with

>, which may not

(comments are ignored).

Dummy arguments may neither be pseudo-ops nor contain
pseudo—ops, and the real arguments are subject to the same
restriction.
55

The programmer who wishes to use pseudo-Ops in a macro
definition (not as arguments) should experiment with short programs using the intended macro to be sure that its expansion and
execution are legal and as. expected or desired.

Consider the following macro definition:

DEFINE

LOOP
DCA

<TAD

A
B
,

TAD.COUNT
ISZ

0‘2)

giving the macro name, a space and
the list of real arguments, separated by commas. There must be at
least as many arguments in the macro call as in the correspOnding
macro definition. When a macro is referenced, its definition is
found, expanded, and the real arguments replace the dummy arguments. The expanded macro is then prOcessed1n the normal
fashion
For example, the macro call:
A macro is referenced by

LOOP

x, Y2

in the

context of the program in which it appears, is'eq‘uivalent to:

TAD

DCA

TAD

COUNT

ISZ

JMP

.~2

The .macro table .sharesthe available

spacefwith. the symbol

(see Symbol Table). Thus the'prog'rammer must be aware

table

of the amount of room required by his macros and the fact that
each symbol occupies four words Of memory Also, the arguments
of

macro

the macro

call

are

temporarily stored in this
‘

is being expanded.

MACRO- 8

Pseudo-Operators

The

list

following

summarizes

MACRO- 8.
56

the

buffer space

while
‘

pseudo- ops available in

Description

Pseudo-op
DECIMAL

All following numbers interpreted as decimal

DEFINE

Used to define a macro

DUBL

Enter double precision integer mOde

EXPUNGE

Deletes permanent symbol table withexception
of pseudo—ops

dumped and a field

Causes the literals to be

FIELD

FLTG

setting to be output
Appends all presently defined symbols to the
permanent symbol table
Enter double precision floating point mode

Indirect addressing

OCTAL

Resets radix to its original octal base

PAGE

Resets the location counter

PAUSE

Signals assembler to stop processing paper tape
being read

FIXTAB
_

‘

TEXT

Allows a character string to be entered

Ignored (convenience to programmer only——
see PAL III)
End-of-pass (MACRO-8 recognizes line feed
followed by 33 as unconditional end-of—pass,
whether in a text string, macro definition, or
anywhere.)
,

Symbol Table
Because of the extra features of MACRO-8, there is less room

available for symbol storage than was found in PAL III. Programs

that were originally coded to be assembled by PAL III may have
many symbols to be assembled by MACRO—8v. If register
switches 9 and 10 are set to 1 during assembly, MACRO—8’s user

too

symbol table will be extended. These switches cause the macro
processor and double precision and floating point processor to be
deleted (see the operating instructions).
The high-speed reader buffer occupies 4008 locations which will
be used for symbol storage if low—speed input is requested. With
high-speed input, the symbol table capacity is 11510 symbols, each
of which requires four core locations. Using the
LT—33 as the
input device, the capacity is 1791., symbols.
57

SYMBOL TABLE MODIFICATION
Because of the small amount of core remaining to be used for
programmer symbols and the macro table, the following suggestions
are offered allowing a particular
individual to con—

installation or

serve symbol table space
By use of the pseudo——ops EXPUNGE and FIXTAB, unnecessary
‘

instruction mnemonics can be removed from the symbol table,
making more space available for programmer defined symbols and
macros. This also decreases assembly time as the unused instruction symbols are not involved in the Symbol table searches. The
mostoften used instruction mnemonics should be assembled first,
so that they“ will be in core next to the special characters and
pseudo——.ops This is desirable because the symbol search routine
starts with the pseudo-Ops at the top of the table (7577) and works

down.
an installation which does

have

Optional equipment
(RFO8 disk, TCOI DECtape, high-speed reader/ punch, etc.) avail—
able, the corresponding instruction sets can be removed. A sym—
bolic tape beginning with EXPUNGE, containing all necessary
instruction mnemonic definitions, and ending with FIXTAB and 35
could be assembled (only pass 1 is
necessary) by MACRO- 8 prior
to any other assemblies For

not

example:

EXPUNGE

AND=$®®®
TAD=1®®®
CLA=72®®

FIXTAB
_

The pseudo-op PAUSE could also be used in place of the dollar

sign "with the above tape, as the first tape of'a multiple tape assembly. The definitions will remain in the table until a subsequent
EXPUNGE or until the Assembler1s reloaded. See the list of per-

manent symbols1n Appendix C
INTERNAL SYMBOL

REPRESENTATION

FOR MACRO—8
Each permanent user

defined symbol occupies four words (loca-

tions) in the symbol table storage area, as follows:
58

012

l I
E

WORD 1
WORD 2.
WORD 3

C1 ><0'45(8)p 4432

45(8)

+C—4

C5 X 45(8)

+C6

C2 x

2 CHARACTERS
SECOND 2 CHARACTERS
FIRST

THIRD 2 CHARACTERS

OCTAL CODE OR ADDRESS

WORD 4

C1, C2,” C6 represent the first character, second character“
sixth character, respectively. (Symbols may consist Of
from one to six characters.) Bits 0 and 1 Of Word 1, and bit 0 of
word 2 are a three digit code denoting the type Of Symbol. Word 4
contains the octal code Of the symbol; if a user defined symbol,
where

Word 4 contains the address of the symbol. For example, the per—
manent

symbol TAD is represented as follows:

WORD l: 248 X 458 + 011213458 OR TA
WORD 2: 048 X 458 + 00: 2249 + 40009: 42248 OR D
WORD 3— 0000
’r

WORD 4

1000

(OCTAL CODE FOR TAD)

Note that the octal code for each character is always scaled by the
Assembler so that the Character is represented

using six bits of a

word. For example, ASCII code for T is 324, and was trimmed to

24; A is 301, trimmed to 01; etc. Digits 0
to the range 33

through 44.

through

are

scaled

MACRO— 8 recognizes eight categories Of symbols:
1.

Special Characters
These include single non- alphanumeric characters which
have special meaning to the Assembler, i.e., +
=1.
They are not affected by EXPUNGE or; FIXTAB and may
.

—-

redefined. The programmer may not define additional special characters. Special characters are recognized
not be

by the Assembler by their three digit code of 011.
Pseudo-ops
These are instructions to the Assembler, which, with the
exception Of FIELD, do not produce binary code. When
they are recognized by the Assembler, they are executed
immediately. They are not affected by EXPUNGE or
~

FIXTAB and may not be redefined. The programmer may
59

not define additional pseudo—ops.

Pseudo—ops may not oc(cOmma) nor may pseudo—

left of an
sign or
Z
in their proper context appear to
other
I
than
and
ops

cur tothe

the right of an 2 sign. Pseudo-ops are recognized by the

Assembler by their three digit code of 100.
Permanent Symbols

These

are

symbols which are part of the permanent as-

sembler symbol table or which were used1n the program

prior to the pseudo-op FIXTAB. They include such symbols as TAD, CLA, TLS, and any user-defined permanent
‘

symbols. They are designated as permanent symbols by
their position in the table and are not necessarily
defined.
For example, the coding sequence
TAD

FIXTAB

will

A to be entered as a permanent symbol with
no definition. It must later be defined, at whichtime it will
cause

be the same

any other permanent symbol. Permanent
be
redefined only if the new definition agrees
symbols may
as

with the old.

User-defined symbols
These include all

non-permanent, non—pseudo—op,

non—

special character symbols. They are recognized by the
Assembler by their place in the symbol table.
Defined symbols
These are permanent or user-defined symbols which
either are part of the assembler symbol table or have been
defined with
or a comma. Permanent symbols and those
defined with comma may only be redefined if the new defi—
nition agrees with the old. Any user- defined symbol may
be redefined with 2. They are recognized
the Assemby
bler by the three digit code 001.
Undefined symbols
These have never been defined with or a comma. They
are stored in the symbol table1n the order of their occur—
rence and are recognized by the Assembler by their three
digit code of 010.
‘

~—

Macro names

These are the names defined with the
60

pseudo——op DEFINE

the names of macros.

They are stored with the user—
defined and permanent symbols and are recognized by
the Assembler by their three digit code of 000.
Macro dummy arguments
as

‘

These are the arguments which follow the macro name in

the macro definition. These are stored with the permanent
and user-defined symbols and are recognized by the Assem—

bler by their three digit code of 101.
MEMORY REFERENCE INSTRUCTION RECOGNITION

Memory reference instructions are recognized by their use in an
expression, according to the following rule:
Given an expression SYMl SYMZ, SYMl is a memory reference instruction if and only if SYMl is a permanent symbol and
SYMZ is not a permanent symbol. In any other case SYMl and
SYMZ will be combined with an inclusive OR. For example:
Both permanent; will be combined with inclusive OR.

CLA CLL
A B

Both

user—defined; will be combined with in-

clusive OR.
B CLA

User-defined followed

by permanent; will be

combined with inclusive OR.
AND A

symbol followed by user—defined;
permanent symbol will be treated as a memory
reference instruction and user-defined symbol
Permanent

referenced address.

COMPATABILITY BETWEEN PAL 111 AND MACRO-8

MACRO-8 will assemble code produced for PAL HI with the

following exceptions:
1.

symbol table in MACRO-8 is considerably shorter
than the symbol table in PAL 111 due to the added features
of MACRO=8. See the section dealing with limited symbol
space for suggestions on breaking large tapes into smaller

The

ones.

pseudo-op. Refer to the
section on internal symbol representation for a discussion

MACRO—8

has

FlXMRl

of MACRO-8 memory reference instruction recognition.
Tags which are legal user-defined symbols in PAL III,
61

f such as

PAGE, DUBL, or FLTG, maybe pseudo-ops in

VMACRO— 8

While PAL III allows two levels of
A: B: O, MACRO— 8

definition with— i. e.,
level, A: O; B: O.
'—

only allows one

A special check wasinserted into PAL III for the conve—
,~

nience ofprogrammers who forgot the space in code in-

volving «7., (suchas IMP .~——l); no such check exists in
MACRO—8.

Programming Hints
1. Arrange coding in order Of ascending address to avoid
confusion.

possible.

Comment profusely wherever

,3.

All terms in any origin expression

(* or PAGE) need to

be defined prior to that expression so that all terms defined thereafter will be given the proper value on pass 1.
All terms to the
been

right of an equal sign (2)

must have

previously defined sothat the term on the left is

given the proper value on pass 1.
Since line feed/ dollar sign is recognized as unconditional

end-of—pass, it is wise to develop the habit of typing the
terminating 55 at the left hand margin so that it will be
caught if a TEXT or macro delimiter is missing. It is also
wise to keep several short tapes with just carriage return/
line feed, dollar sign, carriage return/ line feed in the tape
tray for assembling short subroutine tapes which end in
PAUSE. Often these Subroutines may be debugged separ—
ately prior to inclusionin a larger program.
If the link
generating facility is used it may be wise to
suppress the LG error message on pass 2. In MACRO- 8
(version DEC-O8—CMAB-PB)

location

1233

contains

4777 normally. To suppress the LG message reset loca-

tion 1233 to contain 7200 (CLA) on
reset to 4777 on pass 3

DEALING
1.

pass

2. It should be

WITH A LIMITED SYMBOL SPACE

Breaking long programs into smaller ones
Due to a“ shortage of symbol space it is often necessary to
break a. large program into several smallerones and assemble each one separately. Then the binaries may all be
62

loaded into c0re together; Record separately the map of.
where each program fits into core,- so that tWo segments
are not assembled and loaded into ‘the'Same area.
Make each segment as self-sufficient as possible. At the
beginning of each tape, insert comments indicating where

that segment expects other Segments to be and define tags
for locations which must be referenced; For

example, a

subroutine which calculates the sine of a number starts at
*400. It expects the double? precision multiplication sub’

routine to be at *2’00.‘

The beginning of the multiply subroutinet“
/DCUELE

PRECISION

/CALLED

WITH

ADDRESS

AND

/IN

LOCATION

FCR

MULTIPLY

ADDRESS

HIGH

SUBROUTINE

HIGH

ORDER

MULTIPLIER

MULTIPLICAND

JMS

FOLLOWING

EXAMPLE:

TAD

MULTI

JMS

MULT

MULT2

*2®@
MULT,

The beginning of the sine subroutine:
/DOUBLE PRECISION SINE SUBROUTINE
/EXPECTS DOUBLE PRECISION MULTIPLY
/TO

START

LOCATION

/CALLED WITH ADDRESS
/STORES

RESULT

SUBROUTINE

BOO
OF

ARGUMENT

LOCATIONS

BOO

AND

SOI

MULTP=2®G
*AGQ
SINE,

TAD

ARG

JMS

MULTI

ARG2

MULTI:

MULTP

Each subroutine

and assembled

is self—sufiicient and may be debugged

separately. It is only necessary to be sure
63

that each piece is. where it is expected to be. Comments in

the source listing are a great advantage here.

Reducing the number. of symbols in the program
Todecrease the number of symbols in a program, consider
thefollowing:
7a. Use of
instead of severaltags close together, i.e.,
.

JMP .+3 instead of J MP TAG. There is a reasonable

limit, around 108 instructions, beyond which a separate
tag is more feasible. A JMP .+17 can be trouble if a
line of code is removed later.

pointers and constants instead of scattering

b. Lists of

them through a page, for example:
LIST:

TAGI
TAGB

TAGB
TAG4

rather than:
START:

TAGI

END,

TAGB

Go.T

TAGS

These locations may then be referenced as. follows:
JMS

LIST+2

instead of:
JMS

Let the Link Generating Routine handle the reference:
JMS

TAGB

*d. If. links

are

not

wanted, make

Processor to free symbol space:
64

use

of the Literal

JMS

(TAG3

The effect of c and d is the same, with the exception
of

the

diagnostic message

(Link Generated).

(Many programmers avoid generating links so that if
is generated, it will be noted as an error.)
Use of Switch Options (see MACRO—8 Switch Options)
If a program does not use macros or floating point, delet—
ing those processors will provide mere symbol space. If
the program is very long and will only need to be assembled once, using the low speed reader instead of the high
speed reader gains 64- symbols worth of space.
Revising symbol table
If a program does not require all of the Assembler’s permanent symbols, the extra ones may be deleted by using
EXPUNGE, redefining the necessary ones, and using
FIXTAB. This gains one symbol space for each mnemonic
deleted. (See Appendix C for the permanent symbol
one

table.) If the EXPUNGE definition and FIXTAB are put
on a separate tape, this tape need only-be assembled on
pass 1.
Curtail the use of macros Wherever possible
Use of a subroutine in place of a macro will decrease the

‘

symbol space necessary since the macro is
stored in its entirety in the symbol table.
Punching a new assembler tape
if MACRO-8 is to be used frequentlywith a revised sym—
bol table, it may be wise to punch» a new tape of theAssembler with the revised table (and deleted processors, if any).
amount

To do so, it is necessary to obtain the source tapes for
the Binary Punch Routine (DEC—OB—YXlA—PB), edit it
to reset the

origin to 6000, and punch a new binary tape.
(The Binary Punch Routine currently resides in core in
the same locations as MACRO-8’s permanent symbol
table.) Load MACRQ-S, EXPUNGE, redefine, and modify as necessary. Then load the new Binary Punch Routine
and cause it to dump all core from locations 0 to 7577.
The resulting tape is the newAssembler. (Location 6000
is in the middle of the symbol table and. using it for the
origin should allow plenty of room for the Binary Punch
65

Routine. If there is

doubt, obtain the MACRO-8 source

listing for reference.)

SummaryofMACRO-S Error Diagnostics
*

The format of the error messages is:

ERROR CODE

ADDRESS

where ERROR CODE is a two character code which specifies the

type of error, and ADDRESS is either the absolute Octal address
where the error occurred Or the address of the error relative to the

last symbolic label (if there was one) on that page.

‘

'rAssembly can be continued after most errOrs. Onnonrecoverable
errors, assembly halts, and the program must be edited to correct
the error and assembly restarted.
Table 1

Error Code

MACRO-8 Error Messages
‘
'

Meaning

‘

'BE
‘

TWO MACRO-8 internal tables have overlapped.
This situation can usually be corrected by decreasing the number of current page literals used prior
to this point on the page. It is sometimes the result
of too many nested literals in an eXpression.

Illegal Character

IC.

A non-valid character was found other than in

a comment field or, a text field
A valid character was found under the wrong
conditions, i.e., 8 or 9 in octal radix

The illegal character is flagged and ignored, subject

.to conditions stated elsewhere in this chapter. (See
floating point constants, double precision integers,

etc.)
ID*

‘

Illegal redefintion of a symbol
An attempt was made to give a previously defined
symbol anew value. The symbol was not redefined (this is similar to the duplicate tag diagnostic
of PAL III).
-

7
'

Illegal Equals

‘

‘An equal Sign Was used in the wrong context. For

example:
TAD=®33
A+B=C
AZBT—Q

manna—s Error Messages

Table 1 (Can’t)
Error Code

Meaning

The

exPression to the left of the equal sign must
single symbol.

Only one level of definition is allowed in MACRO‘8 (unlike PAL 111 where two levels are allowed).

symbols may be redefined only if the
and old definitions match. (After EXPUNGE,

Permanent
new

only pseudo-Ops remain.) User-defined symbols
may be redefined with .—: at any time.
11

Illegal Indirect
An out—of— page reference was made, and a link
could not be generated because the indirect “bit was

already set. For example:
*2@@
TAD

PAGE
CMA

CLL

Illegal format in a Macro definition
The eXpression after the DEFINE pseudo-instruction does not comply with the. macro definition
position or structural rules. For example: a macro
name

is referenced before the macro definition.

Link Generated

A warning message;

a' link was generated for an
out-of-page. reference at this address. For example:

*BZQ

.
_
a

@02Q0

@177
g

@0377

_a4ee

means

‘714@

.,_
.

TAD

A.» /COMMENT

CLA

CLL

PAGE
,

RAP-

Missing Parameter in a macro call
An argument, called for by the macro definition, is
missing. For example:
‘

Table 1 (Can’t)

MACRO-8 Error Messages

Error Code

Meaning

DEFINE
‘

MAC

<TAD

CIA
DCA

.MAC

SUM

*2flfl

SUM=3®®

@GBQG

13®®

®®2®1

7$41

®®2®2

3600

Current, non—zero Page Exceeded

An attempt was made to override a literal with an

instruction, or override an instrUction with a literal.
This can be corrected by decreasing the number of
the page, or decreasing the number of
instructions on the page.

literals

Symbol Table Exceeded
symbol table and macro table overlap. Nonrecoverable error. The number of symbols or

The

macros

must be reduced.

Undefined Symbol
Asymbol has been processed during pass 2 that
was not defined by the end of pass 1.
Reference was made to a deleted processor through
of the following symbols: DUBL, FLTG,
DEFINE, <, >. Non-recoverable error. The As-

*one

sembler must be reloaded to handle the source
tape or the source tape must be edited to remove
the
ZE

illegal reference.

Page zero exceeded. Same as PE only with reference

to page 0.

Pass 3 Output—Assembly Listing

The output on pass 3 is a side—by—side listing of the source pro—
gram and its generated octal code. Depending on the setting of bit
11, this output is to either the Teletype or the high—speed punch.

The listing has the following characteristics:

The first 3 characters of the line are reserved for a two char—
acter error message, if necessary, and

single space. If there

is no error message for that line, the spaces are left blank. If
68

there is

more than one error message, the succeeding ones
will be added to the first with no intervening spaces and the
l

remainder of the line will be shifted to thelright.

‘

through 8th characters contain the addressinto
which coding on that line was assembled. If there was no.
assembled coding, these characters are spaces. The address is
a four digit octal address preceded by a one digit field number.
The 4th

The 9th and 10th characters are spaces.

through 14th Characters contain the assembled
code, if any. If there was none, these characters are also
The 11th

spaces.
The 15th and 16th characters are spaces.

The 17th through 72nd characters contain the source cod—
in g, including comments. Any tab characters (ASCII code 211)

appearing in the source code will be output as spaces to the
Tab stops occur every eight spaces in the source
code field. Since an LT—33 has, a 72 character line length,
there is only room for a total of 56 characters in the source
code and comment on one line. Any extra characters will be
typed over the last of the 56 characters. For neat listings On
next tab stop.

the

LT- 33 comments should be restricted

that the line

does not exceed 56 characters. If the listing is to be generated

high-speed punch for later printing on a wider device, this restriction can be modified without changing the
on

the

Assembler.
Since the

LT- 33 does not handle

form feed

(ASCII code

214) as a top--of form, MACRO— 8 will outputtthis character as six
blank lines, a form feed, and six blank lines, so that listings may be

conveniently divided into pages.
A sample of this format is given under the LG error message
in the previous section.
Dollar sign is not printed when it is the terminating pseudo- op.
(Itis printedin a text string or single character text.)
MACRO-8 Operating Procedures

ASSEMBLER OUTPUT
MACRO-8 is

two pass assembler with

optional third pass
which produces an octal/ symbolic assembly listing. During the first
pass, MACRO—8 processes the sourcetape and places all symbol
definitions and macro definitions in its symbol table and macro
a

table, respectively. During the second pass, MACRO-8 processes
the source tape and punches the binary format tape and symbol
table. This

punchedtable can be read by DDT. The third pass
provides a listing of the generated octal code and the original source
table.
language followed by a

printed symbol

Input Device

MACRO 8 can be used with either the

high-speed reader

or with

the low-speed (LT— 33 ) reader. The choice1s determined on first
paSs by-the location of thesource tape and is remembered for the

other two passes.

Output Device
On pass 1 the

only output is to the Teletype in the form of the
‘

error
8

messages which have been generated.

On pass 2 the device on which the binary tape is to be punched
is determined by bit 11 of the switch register. If bit 11 is set to 0,

the LT—33 'will be used. If bit 11. is" set to 1, the high-speed punch
will" be used. Thepunch should be turned'on before beginning the

pass. The symbol table will be punched On the same device (also
printed if the LT- 33 was specified.) Any error messages will ap—
LT-33
pear on the
On paSs 3, the same choices

will determine Whether

bit 11

the listing Will be produced on the Teletype

the high speed

punch.
OPERATING INSTRUCTIONS

Loader

(see Appendix A).

Leave the data field and in-

struction field set to this field for the remainder of the

assembly. '(On a-4K machine, load into field .0.)
2.
8

for pass I.
Prepare
to LINE
Turn

Teletype

OP-9.0".”

Place source in reader
*

Turn on reader

Set switch register to 0200

Press ADDRess LOAD
Set switch

and desired

options (see

MACRO- 8 Switch Options)

g. Press CLEAR and CONTinue
If there are multiple tapes each ending1n PAUSE when the
.

computer halts, remove the tape from the reader, insert the
70

next tape, and press CONTinue. Repeat until all tapes have

been read.

Passes2and3

_
7

_
.

Place source tapes in reader (if not the same reader as
the previous pass,
instructions from under pass

1).

‘

register for pass and desired option.
MACRO—8 Switch Options)
c... Turn on output device, if not already on.
(:1. Press CONTinue.
b. Set switch

(see

‘

If there are multiple tapes, followithe procedure for them as given
‘

under pass 1.

MACRO— 8

SWITCH OPTIONS

Switches should be set

noted to obtain

the desired option.

Switches not mentioned are ignored by the Assembler

PASS SETTING
Bit O

Bit 1

1‘

Pass .1

Option

Pass 1——to retain all

Pass 2

Pass 3

previously de—
fined symbols

Processor DeletiOn

Certain processors may be deleted to make room for more symbols:

Bit

Delete the floating point and double precision proces-

9 = 1

sors.

This increases the size of the

symbol table by

68“) symbols.
Delete the macro processor and the floating point and

Bit 10 = 1

double

precision processors.
space by 13110 symbols.

This increases

NOTE
Switches
1

9
is

and

are

sensed whenever

Once processors have
MACRO-8 must be reloaded

entered.

pass
been deleted,

handle

subsequent programs that use
double
macros,
precision integers, or float—
ing point numbers. Reference to a deleted
to

processor is a non-recoverable error.
71

symbol

Output Device
Bit 11 : O

Teletype printer (and punch, if turned on)

Bit 11 = 1

High-speed punch

Error messages Will be output to the Teletype, which must be
turned on-line.

‘

Halts
Locations mentioned in the following section refer to the version
of MACRO-8 released in March 1971

(DEC~08-CMAB).

The halt on $ is at location 4232. Next pass can be started from

this halt.
The halt on PAUSE is at location 1747.

added to the current pass from this halt.

Another tape may be

The halt on end—of—-tape in high-speed reader ‘( an effective pause)
is at location 340. This would occur if no 55 or PAUSE was seen.
Insert separate tape of $ and press CONTinue to end pass.
If the

LT-33

runs

out of

tape Without encOuntering al$

PAUSE, it stops reading and leaves the machine in an IOT loop.
(KSF; JMP .—1;). Type 33 on keyboard to end pass.
Other halts should be preceded by an error message.

’

appendx o

Ioodng procedures

Initializing the system
Before using the computer system, it is good practice to initialize
all units. To initialize the system, ensure that all switches and con'

trols are as specified below.
Main power cord is properly plugged in.

l-—-\

pmflewewe

Teletype is turned OFF.
Low—speed punch is OFF.
Low-speed reader is set to FREE.
Computer POWER key is ON.
PANEL LOCK is unlocked.

Console switches are set to O.
SING STEP is not set.

High—speed punch is OFF.
DECtape REMOTE lamps OFF.

The system is now initialized and ready for your use.

Loaders
READ—IN MODE (RIM) LOADER

When a computer in the PDP-8 series is first received, it is noth—

ing more than a piece of hardware; its core memory is completely
demagnetized. The computer “knows” absolutely nothing, not even
how to receive input. However, the programmer can manually
load data directly into core using the console switches.
The RIM Loader is the very first program loaded into the com—

puter, and it is loaded by the programmer using the console
A—l

switches. The RIM Loader instructs the computer to receive and

store, in core, data pUnched on paper tape in RIM coded format
(RIM Loader is used to load the BIN Loader described below)
There are two RIM loader programs: one is used when the in—
loWput is to be from
paper tape reader and the other

the

-speed

is used when input is to be from the high— speed paper tape reader.

The locations and corresponding instructions for both loaders are

listed in Table

A—1.

The procedure for loading (toggling) the RIM Loader into core
is illustrated in Figure A—l.

Table A-1.

RIM Loader Programs

InstrUction
Location

Low—Speed Reader

High-Speed Reader

7756

6032

6014

7757

6031

6011

7760

5357

7761

6036‘

6016

7762

7106

7763

7006

7764

7510

7765

5357

5374

7766

7006

7006
‘

7767

6031

6011

7770

5367

7771

6034

6016

7772

7420‘

7420

7773

3776

7774

3376

7775

5356

3376
r

5357
_

7776

0000

After RIM has been loaded, it is good programming practice to

verify that all instructions were stored properly. This can be done
by performing the steps illustrated in Figure A—2, which also
shows how to correct an incorrectly stored instruction.
When loaded, the RIM Loader occupies absolute locations 775 6
through 7776.
A—Z

) EZI‘IVIJJNI L
.LES AHVLOH
80.133138 HOUMSOJ. OW

‘

.LBS SBHOJJMS 8-9
0.1. 0381830
NOIiOflHiSNI «x-O'IBL-i

1.3 SEHOJJMS

“-6

OJ. (BL-H830
ViVO 0131;! *

EdVlOEG-X 88380 GWHOHS
GVO‘T INIH OiNl 073:] (D

SSBHd
.LX3 GVO‘T 800V

.138
01

QQLL

1‘
SSBHd
800V 0V0?

J38 as. =
.LSHL-J NOLLOHHLSNI

3838c] dBG

138 =88
~

.LX3N NOIJQHHLSNI

[

Ssaad d30

71V

SNOIiOflELLSNI
N

F
wna SI

emfitd 91-V

anvm

81119901 am NIH 199961

E‘V

(:g quouZE A;)
SET ROTARY
INDICATOR
SWITCH TO MD

SET SWITCHES
6*8 TO FIELD IN
WHICH RIM HAS
BEEN LOADED

I
,

PRESS
EXT ADDR LOAD

ITSET SR=7756#;]
,

PRESS
ADDR LOAD

[

PRESS
I

EXAM

L
J.

‘

MD=
CORRECT

INSTRU?CTION

[

SET SR=MA—1

ALL

VINSTRUCHONS

CHEgKED

PRESS
ADDR LOAD

GIM IS LOADED)

SET SR= CORRECT
INSTRUCTION

[

PRESS DEP

I—————————RJ

Figure A-Z.

Checking the RIM Loader

BINARY (BIN) LOADER—
The BIN Loader is a short utility program which, when in core,
instructs the computer to read binary-coded data punched on paper

tape and store it in core memOry. BIN is used primarily to load the
programs furnished in the software package (excluding the loaders
and certain subroutines) and the programmer’s binary tapes.
BIN is furnished to the programmer on punched paper tape in
RIM—coded format. Therefore, RIM must be in core before BIN
can be loaded. Figure A—3e illustrates the steps necessary to prop-

erly load BIN. And when loading, the input device (low- or highspeed reader) must be that which was selected when loading RIM.
A—4

LOAD Rm

_.._.

See

Figums CZ-I.C2-2

SET ROTARY
SELECTOR SWITCH
TO MD

SET SWITCHES
6-8 TO FIELD
WHICH CONTAINS
RIM

I
SET SWITCHES
9—11 TO FIELD IN
WHICH BIN IS
TO BE LOADED

PRESS

ADDR LOAD

EXT

'SET SR=7756

PRESS

ADDR LOAD'

WHICH

HIGH-SPEED

READER
?

READER

LOW-SPEED
READER

TURN HSR ON

PUT'enq
IN

TURN

LOAOER

PUT BIN LOADER
IN LSR

HSR

PRESS

CLEAR AND

TTY TO LINE

CONT

PUT LSR TO

STARTI

TAPE
READS IN
?

PRESS HALT

——————-——1
SET SWITCHES
6-8 To FIELD
BIN WAS LOADED
INTO

PRESS
EXT

ADDR LOAD

SET SR=7777

PRESS
‘

ADDRLOAD

PRESS EXAM

’YES
,'

BIN LOAOER

IS‘LQAQED

Figure

A-3

Loading the BIN Loader
A—S

When stored in core, BIN resides on the last page of core,

oc-

cupying absolute locations 7625 through 7752 and 7777.
BIN was purposely placed on the last page of core so that it
would always be available for use—the programs in DEC’s soft—
ware package do not use the last page of core (excluding the Disk
Monitor).

The programmer must be aware that if he writes a
which uses the last page of core, BIN will be wip-

program
ed out when that

happens,

the computer. When this
the programmer must load RIM and then BIN before
program

runs

he can load another binary tape.

Binary tapes to be loaded should be started on the leader-trailer
code (Code 200) otherwise zeros may be loaded into core, destroy—
ing previous instructions.
Figure A—4 lilustrates the procedure for loading binary tapes
,

into core.

SET ROTARY
SELECTOR SWITCH
TO

SET SWITCHES
6-8 TO FIELD IN
WHICH BIN IS
LOADED

I
SET SWITCHES 9'11
TO FIELD IN WHICH
PROGRAM IS TO BE
LOADED

RRESS

,EXTTQDPRHEQAD

SET SR To

7777‘
I

"PRESS

.599R LOAD

;

HIGH-SPEED READER

LOW-SPEED READER

w—

ONE

:—

gfiiliiliilliflifii

—
WI

CLEAR'

PRESS
AND

CONT
‘

NO_.

‘%

TAPE
READS IN
’?
I

YES

TING,

PRESS CONT

TAPE
STOPS AT
BEGHMWNG OF

,EADER/TRNLER
CODE
?

’

YES

VOTES

‘EIOBJECT TAPE-m.
‘

Figure

A4.

SIS LQQQEQM M!

Loading A Binary Tape mmg BEN
A57

ASCII-11 Character Set
Decimal

Character

Decimal

Equivalent
(A1 Format)

Character

8-Bit
Octal

6-Bit
Octal

Equivalent
(A1 Format)

301

241

—1952

302

160

”

242

——1888

303

224

243

304

03
04

42
43

288

244

——

E
F

352

245

——1696

246
247
250

-1568

-—1504
—1440

W
X
Y

305
306
307
310
311
312
313
314
315
316
317
320
321
322
323
324
325
326
327
330
331

332

260

928
992
1056
1120
1184
1248
1312
1376
1440
1504
1568
1632
1696
.——992

261

262
263

61
62

_928
1—864

Leader/Trailer

63
64

——800

Carriage RETURN

-——736
—-672

SPACE
RUBOUT

K
L

M
N

O
P

Q
R

S
T

U
V

264
265

416

480

544

608

672

736

14
15'
16
17
20

800

22
23
24

25
26
27

30
31

65
66

271

’

(
)
at

+
’

864

266
267
270

6
7

1 An

6-Bit
Octal

G
H

8—Bit
Octal

1—608
-—544
~-——480
-—416

/
g

;

<
Z

>
17
‘

@
[

\
]

¢(A)’-’3

<—

LINE FEED

—-1376

253

—1312

254

54
55

——1 184

256
257
272
273
274
275
276
277
300
333
334
335

336
337.
200

BELL
TAB

211

FORM

214

B—l

~1248

56
57
72

—1056

—~288

75‘
76
77

——1120
—352

~224

—160
~06
~32
32

1760

1824

1888

36
37

2016

~2016

abbreviation for American Standard Code for Information Interchange.

The character in parentheses is printed on some Teletypes.

—1632
’

212
215
240
377
000
207

Blank

1760

251
252

255

—-1824

1952

PAL 1H,

MACRO—8

following are the most commonly used elements of the
PDP—8 instruction set. For that reason they are foundlin the permanent symbol table within most assemblers. These instructions
are already defined within the computer. For additional information on these instructions and for a description of the symbols used
when programming other, optional, I/O devices, see the Small
Computer Handbook, available‘from the DEC Software DistribuThe

tion Center.
INSTRUCTION CODES
Mnemonic

Code

Time (lasecjl

Operation

Memory Reference Instructions
AND

0000

TAD

1000

182

2000

DCA

3000

JMS

4000

JMP

5000

Logical AND
Two’s complement add
Increment and skip it zero
Deposit and clear AC
Jump to subroutine
Jump

Floating-Point Instructions
FEXT
0000
Floating exit
FADD
1000
Floating add
FSUB
2000
Floating subtractiOn
3000
FMPY
Floating multiply
4000
FDIV
Floating divide
FGET
50001
Floating get
6000
FPUT
Floating put
FNOR
7000
Floating normalize

C—1

26
2.6

2.6
1.2

60
900
920
‘

1450
1480
‘

representative of the PDP-8/E.

2.6

1 Times are

2.6»

115
125
800

Mnemonic

Code

Operation

Sequence

Group 1 Operate Microinstructions (1 cycle-‘3)
No operation
NOP
7000

-———

IAC
RAL

7001

Increment AC

7004

Rotate AC and link left

one

RTL

7006

Rotate AC and link left two

RAR

7010

Rotate AC and link right

one

RTR

7012

Rotate AC and link right two

CML

7020

CMA

7040

Complemented link
Complement AC

CLL

7100

Clear link

CLA

7200

Clear AC

BSW

7002

3
J
_

Swap Bytes

in AC

Group 2 Operate Microinstructions (1 cycle)
HLT

7402

OSR

7404

SKP

7410

SNL

7420

SZL

7430

SZA

7440

SNA
SMA
SPA

Halts the computer
Inclusive OR SR with AC

3
1

Skip unconditionally
Skip on nonzero link
Skip on zero link
Skip on zero AC
Skip on nonzero AC
Skip on minus AC
Skip on positive AC (zero is positive)

7450
7500
7510

1
1

Combined Operate Microinstructions

CIA

7041

STL

7120

Complement and increment AC
Sent link to 1

GLK

7204

Get link (put link in AC, bit 11)

STA

7240

Set AC to

LAS

7604

Load AC with SR

——

MO Microinstructions
Load MQ from AC, then clear AC
7421
MQL
7501
Inclusive OR the MO with AC
MQA
CAM
7621
Clear AC and MO
SWP
7521
Swap AC and MO
ACL
7701
Load MQ into AC
Internal IOT Microinstructions

ION

6001

IOF

6002

Turn interrupt processor on
Disable interrupt processor

cycle is equal to 1.2 microseconds.
C-2

2, 3
1, 2
1, 4
2

2, 3

‘

Mnemonic

Code

Operation

Internal IOT Microinstructions (1 cycle)

SKON

6000

SRQ

6003

GTF

6004

RTF

6005

SGT

6006

CAF

6007

Skip if interrupt ON, and turn OFF
Skip on interrupt request
Get interrupt flags
Restore interrupt flags
Skip on greater than flag
Clear all flags

Keyboard/ Reader (1 cycle)
KSF

6031

Skip on keyboard/ reader flag

KCC

6032

Clear keyboard/ reader flag and AC;

KRS

Read

KRB

6034
6036

KCF

6030

KlE

6035

set reader run
_

keyboard/ reader butter (static)
Clear AC, read keyboard buffer
(dynamic), clear keyboard flags
Clear keyboard/ reader
AC 1 l to keyboard/ reader interrupt
enable FF.

Teleprinter/ Punch (1 cycle)
TSF
6041
Skip on teleprinter/ punch flag
6042
TCF
Clear teleprinter/ punch flag
6044
Load teleprinter/ punch and print
TPC
6046
Load teleprinter/ punch, print, and clear
TLS
teleprinter/ punch flag
6040
Set teleprinter/ punch flag
TFL
6045
TSK
Skip on printer or keyboard flag

High Speed Reader——Type PR8/ E ( 1 cycle)
6011
6012

Skip on reader flag

REC

6014

Clear flag and buffer and fetch

RPE

6010

Set interrupt enable for reader and

RCC

6016

Read reader buffer, clear flag and

PCB

6020

Clear

RSF
RRB

Read reader buffer and clear reader flag

Character

punch
buffer, and fetch character

interrupt enable for reader and
punch
C—3

Mnemonic

Code

Operation

Time (M860)

High Speed Punch—~Type PP8/ E (1 cycle)
PSF
6021
Skip on punch flag
PCF
6022
Clear flag and buffer
6024
Load punch buffer and punch character
PPC
Clear flag and buffer, load buffer and
PLS
6026
punch character
'

RPE

6010

Set interrupt enable for reader and

PCE

6020*

Clear interrupt enable for reader and

punch

Mnemonic

Code

RDF

6214

Read data field

RIF

6224

Read instruction field

RIB

6234

Read interrupt butter

RMF

6244

Restore memory field

CD1

62n3

Change to data field and instruction.

Operation
Memory Extension Control, Type MC8/ E (1 cycle)
62N1
CDF
Change to data field N
62N2
CIF
Change to instruction field N

field 11

C-4

PSEUDOEGPERATORS
The following is a list of the PAL III and MACRO—8 assembler

pseudo—ops.
(""3

x,
,

l (2/

“N_-,L:.</
1

”:M

”I

PAL III

MACRO—8

:0» V}

DECIMAL

M «ILL,

OCTAL

*1»
a” f»

FIELD

PAUSE

(SN

3%» KSQRERE’

5*& {)3

FRI?
I

EXPUNGE

EXPUNGE
FIXTAB

F IXTAB

PAGE
H

M’s/£53
9;T
ff: M
M

if 29,

‘v

,z/

4*
W

("a

ff
4-»?

>l<

1* IXMRI

DEFINE

DUBL

F LTG
TEXT

C—S

‘

VINDEX

Address

arithmetic

Address

assignments

Address

field delimiter

(PAL III),
(PAL III),

42
MACRO—8,
l6
PAL—ll,
Ampersand (&) operator
AND

code

ASCII—l

character

Double

(MACRO-8),

Assembler

(MACRO-8,
(PAL III),

Editor,

‘B—l

set,

End
End

symbolic tape,

program

Evaluation of

symbolic expression
38.
Exclamation point (I) operator
38
(MACRO-8),
Expressions

III),

27
(PAL III),
of tape (PAL III),
26
12
Equal sign (PAL III),
Error diagnostics
(MACRO—8), '66

(PAL

Autoindexing

50
(MACRO-8),
57
pseudo—op (MACRO—8),
53,
Dummy arguments (MACRO—8),
34
Duplicate tag (PAL III),

output

MACRO-8,
PAL—ll,

DUBL

(PAL III),

Assembly listing

(MACRO-8),
precision integers

constants

expressions
operators
(PAL III),

precision floating point

Double

III),

Arithmetic
Arithmetic
ASCII

precision constants
50
(MACRO’B),

skip instructions

group

(PAL

Double

(MACRO-8),

Binary punch routine
65
MACRO—8,
PAL III,
28
2,
Binary tape (PAL III),
BIN
A-4
(binary) loader,
A-5, A—7
loading,

MACRO-8,
PAL~III,

termination,
pseudo-op

MACRO-8,
PAL

B-l
codes,
set
(PAL III),
2,3
Characters (MACRO—8),
37
Checksum (PAL III),
31
6
Coding practices (PAL III),

tions
Comments

(PAL III),
(PAL III),

Compatibility
Constants

—

PAL

FIELD

instruc~

PAL
FIXMRI

and

table

address

FIXTAB

(MACRO-8),

17.
location counter

15,

DECIMAL

(MACRO-8),

pseudo-op

Delimiter

for

address

(PAL III),

(MACRO-8)

Halts

assignment

III,

(PAL

III),

field

(indirect address) pseudo—op
MACRO-8,
42, 57
PAL III,
23
34
Illegal characters (PAL III),
35
Illegal reference (PAL III),
14
Inclusive OR (PAL III),
I

messages,
34
III,

(PAL III),
Dollar sign ($)
pseudo-op,
57
MACRO—8,

53,

(MACROZB),
configuration

Hardware

Diagnostic

PAL

51
constants,
conversion accuracy,

III,
3, 24
Defined symbols (MACRO-8),

PAL

52
format,
FLTG pseudo—op
(MACRO-8), 52, 57
Format effectors
3
(PAL III),
Form feed
3
(PAL III),

PAL

Direct

57,

28,

pseudo—op'

MACRO-8,

DEFINE

24
(PAL III),
(PAL III),

pseudo-op
pseudo—op

number

DDT,

Floating point
(PAL III),

incrementation,

III,

MACRO-8,
PAL III,

indicator

(PAL III),
Current

24
(PAL III),
tape (PAL III),

pseudo—op
46, 57

Field setting

III

57,
27

MACRO-8,

15
6

MACRO-8,
Current

reference

memory

III,

Extended memory
Extended symbol

Character

Combining

EXPUNGE

summary

statements

(end of program)

Incrementing

current

location

(PAL III),
15, 17
Indirect addressing (PAL III), 18,23
counter

Indirect

address

linkage

(MACRO—8),

III),

PAL

Input device

transfer micro~
22
instructions (PAL III),
C—l
Instruction codes,'
Instructions (PAL III), '6, l9

Input/output

source
“

pseudo—op (MACRO—8),
39
Pages (MACRO-8),
1
PAL III programming,
Pass 1
(PAL III),
2, 31
PAGE

MACRO-8,
PAL III,

(PAL III),
(MACRO—8),
70,

'diagnostics,
Labels

5
(PAL III),
Leader code
(PAL III),
29, 31
62
Limited symbol space,
Link generation (MACRO—8),
41
42
Literals (MACRO-8),

RIM

(Read—In-Mode),

Loading procedures,
PAL III,
32
Location counter

Logical operators

Pass

40,

(PAL III),
output

III),_

diagnostics,

36
—

assembly listing,
68

(MACRO-8),

(binary),

Pass

(end of tape) pseudo—op
57
MACRO-8,
PAL III,
26
Period (.) character (PAL III),.17
Permanent symbols
60
(MACRO—8),
Permanent symbol table
(PAL III), 7
alterations,
27, 28
Permanent symbol tables,
C-1
Plus (+) operator
38
MACRO-8,
PAL III,
14

A—4‘

BIN

(PAL

diagnostics,

‘

Loaders

Pass

program

Output device

49
(MACRO—8),
symbol representation

Integers
Internal

39
Origin setting (MACRO—8),
Origin (starting address) of

(MACRO—8),

skip instructions

group

PAUSE

A—2

A—l,
Ael

(MACRO-8),‘
(PAL III),

Macro

(DEFINE)

definition

(MACRO-8),
restrictions,

53
55

MACRO—8

programming,

Macros,
Memory,

user

III,

(PAL

III),

PAL—III,

62
Programming hints,
Program preparation (PAL III),
C—5
Pseudo—operators,

MACRO—8,
PAL III,

Punching

.13

56,

Pushdown

new

assembler

list

tape

65
overflow

(PAL III),

44
(MACRO—8),
Nonprinting characters (PAL III),
Normalized form (MACRO-8),
51

Numbers

MACRO-8,
PAL III,

49
3

pseudo—op

MACRO-8,

PAL

3‘24

III,

(PAL III),
3, 23
34
(PAL III),
56
Referencing a macro (MACRO-8),
19
Registers, autoindex (PAL III),
RETURN key
4
(PAL III),
RIM (Read—In—Mode)
A—l
loader,
A—4
checking the loader,
‘

ROTL

MACRO-8,

Radix control

Redefinition

programs,

6
Operands (PAL III),
Operate microinstructions (PAL III),21
Operating instructions (MACRO—8), 70
Operating procedures
69

A—2

macro,
advantages and dis—
55
advantages (MACRO-B),

Semicolon

(7) as statement
terminator (PAL III),

Single

character

text

facility

PAL

III,

Operation
Operators,
MACRO-8,
PAL

literals

Nested

CDT,

57,

(MACRO—8),

OCTAL

techniques,

Multiple assignments

MACRO-8,

Memory reference instructions
61
MACRO-8,
PAL III,
15, 20
Microinstructions (PAL III),
Minus (—) operator
38
MACRO-8,
PAL

(MACRO—8),

Programming

(MACRO—8),"53
24
(PAL III),

extended

deletion

Processor

defined

III,

(MACRO-8),

20
(PAL III),
arithmetic and logical

code

Skip instructions (PAL III),
Source program preparation

(PAL III),

Tabulation (PAL III),
Tape editor, symbolic,

Space operator
38
MACRO-8,
PAL

III,

Terminating

59
Special characters (MACROcS),
Starting address (origin) of source

Terminators

program (PAL III),
Statements
(PAL III),

Text

Statement

terminators

PAL

(PAL III),

65, 71
option (MACRO-8),
11
Symbol capacity (PAL III),
59
Symbol categories (MACRO—8),
Symbol definitions, addition or
Switch

deletion

III),
of

expression
statements

III),

47
facility (MACRO—8),
48,
pseudo—op (MACRO-8),
48
strings (MACRO—8),

Trailer

code (PAL III),
31
Two-pass assembler (PAL III),

(PAL

addresses

capacity,

TEXT

Text

27
III),
8
(PAL III),
29
Editor (PAL III),
38
expression (MACRO—8),
instructions (PAL III),
IO
10
operands (PAL III),
29
tape (PAL III),
1
tape editor,
7
Symbols (PAL III),
62
Symbol space, limited,
57
Symbol table (MACRO—8),

Symbolic
Symbolic
Symbolic
Symbolic
Symbolic
Symbolic
Symbolic

(PAL

58
modification,
65
revision,
Symbol table (PAL III),

Undefined address

ll,

III),

MACRO-8,
PAL III,

Using

PAL

(page

assembler,

III

zero

MACRO-8,
PAL III,
7,

(PAL

Undefined symbols (MACRO-8),
6O
User—defined macros (MACRO-8),
53
Userwdefined symbols

reference)
57
23

pseudo—op

HOW

SOFTWARE

NEWSLETTERS,

OBTAIN

MAILING

SOFTWARE

INFORMATION

LIST

Communications Group,

Software

The

located

publishes software neWsletters for
Newsletters are published monthly, and keep
Maynard,

in
various DIGITAL products.
the user informed.about cussoftware products, documenta-

corporate headquarters

the

tomer software problems and solutions, new
tion corrections, as well as programming notes

There

two

are

similar

The

Software

The

Digital

techniques.

service;

levels

and

Dispatch
Software

News

Software Dispatch is part of
service applies to the following

The

Software Maintenance
software products:

the

Service.

This

PDPm9/15
RSX—llD

DOS/BATCH
RSTS~E

DECsystem—lo

Digital Software News for the PDP-ll and
PDP—S/lz are available to any customer
PDPe8/12 software.

Software News

for

purchased PDP-ll

Digital

the

who

has

collection of existing problems and solutions for a given software
A customer receives this publication
system is published periodically.
with his initial software kit with the delivery of his system.
This

collection would be either

Summary depending

formance

Software Dispatch Review
the system ordered.

Software Per—

mailing list of users who receive software newsletters is also mainUsers must sign—up for the news—
tained by Software Communications.
This can be done by either completing the form supletter they desire.
plied with the Review or Summary or by writing to:

Software

Communications

P.O.

Box

Maynard,
PROBLEMS

SOFTWARE

Questions
as

01754

Massachusetts

problems relating

DIGITAL's

software

should

reported

follows:

'North

Submitters:

South American

and

Upon completion of Software Performance Report
copy

remainder

send

and

(SPR)

form

remove

last

to:

Software

Communications

P.O.

Box

Maynard,

Massachusetts

01754

acknowledgement copy will be returned along with a blank SPR form
The acknowledgement will contain a DIGITAL assigned SPR
upon receipt.
The SPR number or the preprinted number should be referenced in
number.
Additional SPR forms may be obtained from
any future correspondence.

‘The

address.

the

above

All

International

Submitters:
last

Upon completion of the SPR form,

reserve

the

in the

nearest

DIGITAL

mainder
also

the

available

PROGRAMS

AND

SPR Center

from

our

SPR

copy

and

office.

send the
forms

SPR

re—
are

Centers.

MANUALS

Software and manuals should be ordered by title and order number.
United States, send orders to the nearest distribution center.

the

Digital Equipment Cerporation

Digital Equipment Corporation

Software Distribution Center
146 Main Street
Maynard, Massachusetts 01754

Software Distribution Center
1400 Terra Bella
Mountain View, California 94043

Outside of the United States, orders should be directed
Digital Field Sales Office or representative.

the nearest

SOCIETY

USERS

Society, maintains a user extechnical application information.
The Library contains approximately 1,900 programs for all
DIGITAL computer lines.
Executive routines, editors, debuggers, special
DECUS,

Digital Equipment Computers

change

center

functions,

for user—written

Users

programs

and

maintenance and various

games,

other

classes

of programs

are

available.

Program Library Catalogs

DECUS

of all

abstracts

and

Forms

FOCAL-8,
9, 15

PUP-8,

PDP-7/9,

are routinely updated and contain lists
according to computer line:

BASIC-8,

PDPan

RSTS-ll

PDP-ll,

PDPn6/10,
and

programs

information on acquiring and submitting programs
be obtained from the DECUS office,

the

DECUS

Library may
In

addition to the

DECUSCOPE

‘

’

PROCEEDINGS
THE

DIGITAL

EQUIPMENT USERS

catalogs,

DECUS

also

publishes

the

following:

Society's technical newsletter, published bi—monthly,
aimed at facilitating the interchange of technical in—
formation among users of DIGITAL computers and at disCirculaseminating news items concerning the Societyo
tion reached 19,000 in May, 1974.

-The

-Contains technical papers presented at DECUS
held twice a year in the United States, once
in Europe, Australia, and Canada.

Symposia
a

year

SOCIETY
MINUTES~OF

THE

DECsystem-lO

-A

sessions

report of the DECsystem—10

United

States

DECUS

held

the

two

Symposia.

SESSIONS

COPY-N-Mail

LUG/SIG

“A

monthly maiied communique

—Mailing
Group

Local User

(SIG)

Group

among

users

users.

and Special
providing closer
Specific product or

(LUG)

communique, aimed

communication among

DECsystem-lo

Interest

at
a

application.
information on the DECUS Library, publications, and other DECUS
activities is available from the DECUS offices listed below:

Further

DECUS

Digital Equipment Corporation
146

Main Street

Maynard,

Massachusetts

01754

DECUS

EUROPE

Digital Equipment Carp.
(Europe)
POO.

Box

1211

Geneva

340

Switzerland

International

4K Assemblers
8

Pal

III

Macro

DEC-08~LAS4A~A-D

READER'S

NOTE:

This form is for document comments only.
Problems
with software should be reported on a Software
Problem Report (SPR) form (see the HOW TO OBTAIN
SOFTWARE

Did

find

you

COMMENTS

errors

INFORMATION

page).

this manual?

so,

Did you find this manual understandable,
Please make suggestions for improvement.

specify by

usable,

and

page.

well—organized?

sufficient documentation on associated system programs
If not,
for use of the software described in this manual?
what material is missing and where should it be placed?
Is

there

required

Please

indicate

the

type of user/reader that you

Assembly language

DDD

Name

nearly represent.

programmer

Higher—level language

programmer

Occasional

(experienced)

User with

most

programmer

little programming experience

Student programmer

Non-programmer interested in computer concepts and capabilities
A

Date

Organization
Street

City

State

;

Zip

Code
or

Country
If

you

not

require

written reply,

please check

here,

Fold Here

Do Not Tear

Fold Here and Staple

FIRST CLASS

PERMIT NO. 33

MAYNARD, MASS.
BUSINESS REPLY MAIL
NO POSTAGE STAMP NECESSARY IF MAILED IN THE UNITED STATES

Postage will be paid by:

Software Communications
P.

Box

Maynard,

Massachusetts

01754