Digital PDFs

EK-FP11A-UG-001

2000

90 pages

Original

5.5MB

Document:	FP11-A Floating-Point Processor User's Guide
Order Number:	EK-FP11A-UG
Revision:	001
Pages:	90
Original Filename:	http://skn.noip.me/pdp11/resources/EK-FP11A-UG-001%20FP11-A%20Floating%20Point%20User's%20Manual.pdf

OCR Text

FP11-A
floating-point processor
user's guide

EK-FP11A-UG-001

FP11-A
floating-point processor
user's guide

digital equipment corporation • maynard, massachusetts

lst Edition, May 1978

Copyright © 1978 by Digital Equipment Corporation
The material in this manual is for informational
purposes and is subject to change without notice.
Digital Equipment Corporation assumes no responsibility for any errors which may appear in this
manual.
Printed in U.S.A.

This document was set on DIGITAL's DECset-8000
computerized typesetting system.

The following are trademarks of Digital Equipment
Corporation, Maynard, Massachusetts:
DEC
DECCOMM
DECsystem-IO
DECSYSTEM-20

DECtape
DECUS
DIGITAL
MASSBUS

5/82-14

PDP
RSTS
TYPESET-8
TYPESET-II
UNIBUS

CONTENTS
Page
CHAPTER 1

INTRODUCTION

1.1
1.2
1.2.1
1.2.2
1.3
1.4
1.5

GENERAL .......................................................................................................... 1-1
FEATURES ........................................................................................................ 1-1
Floating-Point Instruction Set Features ........................................................ 1-1
FPII-A Features .......................................................................................... 1-2
ARCHITECTURE .............................................................................................. 1-2
PHYSICAL DESCRIPTION .............................................................................. 1-2
RELATED DOCUMENTATION ...................................................................... 1-3

CHAPTER 2

INSTALLATION AND CHECKOUT

2.1
2.2
2.2.1
2.2.2
2.3
2.3.1
2.3.2

SCOPE ................................................................................................................ 2-1
FPII-A FLOATING-POINT PROCESSOR INSTALLATION ......................... 2-1
FPII-A Add-On Installation Procedure ........................................................ 2-1
BAII-L Box ................................................................................................. 2-4
FPII-AU UPGRADE KIT .................................................................................. 2-7
FPII-A U Power Components Installation .................................................... 2-7
FPI1-AU Logic Installation ......................................................................... 2-9

CHAPTER 3

REVIEW OF FLOATING-POINT NUMBERS

3.1
3.2
3.3
3.4
3.5
3.6

INTRODUCTION .............................................................................................. 3-1
INTEGERS ......................................................................................................... 3-1
FLOATING-POINT NUMBERS ....................................................................... 3-1
NORMALIZATION ........................................................................................... 3-2
FLOATING-POINT ADDITION AND SUBTRACTION ................................. 3-3
FLOATING-POINT MULTIPLICATION AND DIVISION ............................. 3-4

CHAPTER 4

DATA FORMATS

4.1
4.2
4.3
4.3.1
4.3.1.1
4.3.1.2
4.4
4.5

INTRODUCTION .............................................................................................. 4-1
FPII-A INTEGER FORMATS .......................................................................... 4-1
FPII-A FLOATING-POINT FORMATS .......................................................... .4-1
FPII-A Floating-Point Data Word ............................................................. .4-1
Floating-Point Fraction ........................................................................ 4-5
Floating-Point Exponent ...................................................................... 4-6
FPII-A PROGRAM STATUS REGISTER ....................................................... .4-7
PROCESSING OF FLOATING-POINT EXCEPTIONS .................................. .4-8
iii

CONTENTS (Cont)

Page
CHAPTER 5

FLOATING-POINT INSTRUCTIONS

5.1
5.2
5.3
5.3.1
5.3.2
5.3.3
5.3.4
5.3.5
5.3.6
5.3.7
5.3.8
5.3.9
5.3.10
5.3.11
5.3.12
5.3.13
5.3.14
5.3.15
5.3.16
5.3.17
5.3.18
5.3.19
5.3.20
5.3.21
5.3.22
5.4

FLOATING-POINT ACCUMULATORS .......................................................... 5-1
INSTRUCTION FORMATS .............................................................................. 5-1
INSTRUCTION SET .......................................................................................... 5-4
Arithmetic Instructions ............................................................................... 5-1 0
Floating-Modulo Instruction ...................................................................... 5-11
Load Instruction ......................................................................................... 5-11
Store Instruction ......................................................................................... 5-11
Load Convert (Double-to-Floating, Floating-to-Double) Instructions ........ 5-11
Store Convert (Double-to- Floating, Floating-to-Double) Instructions ........ 5-12
Clear Instruction ........................................................................................ 5-12
Test Instruction .......................................................................................... 5-12
Absolute Instruction ................................................................................... 5-12
Negate Instruction ...................................................................................... 5-13
Load Exponent Instruction ......................................................................... 5-13
Load Convert Integer-to-Floating Instruction ............................................. 5-14
Store Exponent Instruction ......................................................................... 5-15
Store Convert Floating-to-Integer Instruction ............................................. 5-16
Load FPll 's Program Status ...................................................................... 5-20
Store FPll 's Program Status ...................................................................... 5-20
Store FPll 's Status ..................................................................................... 5-20
Copy Floating Condition Codes ................................................................. 5-20
Set Floating Mode ...................................................................................... 5-20
Set Double Mode ........................................................................................ 5-20
Set Integer Mode ........................................................................................ 5-20
Set Long-Integer Mode ............................................................................... 5-20
FPI1-A PROGRAMMING EXAMPLES ......................................................... 5-21

CHAPTER 6

PROCESSOR ORGANIZA nON

6.1
6.2
6.2.1
6.2.2
6.2.3
6.2.4
6.2.5
6.2.6
6.2.6.1
6.2.6.2
6.2.7
6.3
6.4
6.5
6.6

INTRODUCTION .............................................................................................. 6-1
MICROPROCESSOR DESCRIPTION .............................................................. 6-2
Microprocessor Organization ....................................................................... 6-5
Arithmetic/Logical Operations .................................................................... 6-5
RAM ........................................................................................................... 6-8
Arithmetic Logic Unit (ALU) ....................................................................... 6-9
Q- Register .................................................................................................. 6-1 0
Source Operands and AL U Functions ........................................................ 6-1 0
Logical and Arithmetic Functions ....................................................... 6-1 0
Logical Functions for G, P, C n +4, and OVR ....................................... 6-10
Summary of Pin Definitions ........................................................................ 6-1 0
INSTRUCTION STATUS REGISTERS AND DECODE ................................ 6-1O
TRI-STA TE TRANSCEIVERS AND BUFFER ............................................... 6-10
BRANCH LOGIC AND TRI-STATE CONTROL ........................................... 6-10
CONSTANTS, BYTE AND SECTOR CONTROL, SHIFT CONTROL ........ 6-14
IV

CONTENTS (Cont)
Page
CHAPTER 7

MAINTENANCE

7.1
7.2
7.2.1
7.2.2
7.2.3
7.3
7.4
7.5

INTRODUCTION .............................................................................................. 7-1
FPII-A DIAGNOSTICS ..................................................................................... 7-1
MAINDEC DFFPAA ................................................................................. 7-1
MAINDEC DFFPBA .................................................................................. 7-1
MAINDEC DFFPCA .................................................................................. 7-2
KY11-LB PROGRAMMER's CONSOLE .......................................................... 7-2
FPII-A FLOW DIAGRAMS .............................................................................. 7-2
EXTENDER BOARD ......................................................................................... 7-3

APPENDIX A

OPTION POWER SPECIFICATIONS

FIGURES
Figure No.

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

Title

Page

KD 11-EA/FP11-A Signal Interface ..................................................................... 1-2
Maintenance Cable Installation ............................................................................ 2-5
Backplane Jumpers .............................................................................................. 2-6
Normalization ...................................................................................................... 3-3
Integer Formats ................................................................................................... 4-2
Floating-Point Data Formats ............................................................................... 4-2
Floating-Point Data Words .................................................................................. 4-3
Interpretation of Floating-Point Numbers ........................................................... .4-4
Unnormalized Floating-Point Fraction ............................................................... .4-5
FP11-A Status Register FormaL ......................................................................... .4-7
Floating-Point Accumulators ............................................................................... 5-1
Instruction Formats ............................................................................................. 5-2
Double-to-Single Precision Rounding ................................................................. 5-11
Single-to-Double Precision Appending ............................................................... 5-12
Integer Left-Shift Example ................................................................................. 5-14
Normalized Integer Example .............................................................................. 5-15
Store Exponent Example No.1 ............................................................................ 5-15
Store Exponent Example No. 2 ........................................................................... 5-16
Store Convert Integer Example ........................................................................... 5-17
KDII-EA/FPII-A Data Flow ............................................................................. 6-1
Simplified FPI1-A Block Diagram ....................................................................... 6-2
Microprocessor (AM2901) Block Diagram ........................................................... 6-3
RAM Register Usage ........................................................................................... 6-8
AM2901 Pin Connections ................................................................................... 6-14
Display Information ............................................................................................. 7-3
v

TABLES
Table No.

4-1
4-2

5-1
5-2
6-1
6-2
6-3
6-4
6-5
6-6
6-7
6-8
A-I
A-2

Title

Page

FPII-A Status Register ........................................................................................ 4-7
FPII-A Exception Codes ..................................................................................... 4-9
Format ofFPII-A Instructions ............................................................................ 5-3
FPII-A Instruction Set. ........................................................................................ 5-5
ALU Source Operand Contest .............................................................................. 6-5
ALU Function Control ........................................................................................ 6-6
ALU Destination Control .................................................................................... 6-7
Source Operand and ALU Function Matrix ......................................................... 6-9
ALU Logic Mode Functions .............................................................................. 6-11
AL V Arithmetic Mode Functions ...................................................................... 6-12
Logic Equations for ALU Functions .................................................................. 6-12
P, G, C N +4' OVR Functions .............................................................................. 6-13
PDP-II Family Models and Options Power Requirements .................................. A-I
PDP-II Family Options Power Requirements ..................................................... A-3

CHAPTER 1
INTRODUCTION

1.1 GENERAL
The FPII-A Floating-Point Processor is a hardware option that enables the PDP-II/34A central
processor to execute floating-point arithmetic operations. The FPII-A performs all floating-point
arithmetic operations and converts data between integer and floating-point formats. Floating-point
representation permits a greater range of number values than is possible with the conventional integer
mode. Thus, the FPII-A option provides a speedier alternative to the use of software floating-point
routines, and system speed is increased without complex arithmetic coding routines that consume
valuable CPU time. The FPII-A features both single- and double-precision (32- or 64-bit) capability
and floating-point modes.
The FPII-A is an integral part of the central processor. It operates using similar address modes, and
the same memory management facilities as the central processor. Floating-point processor instructions
can reference the floating-point accumulators, the central processor's general registers, or any location
in memory.

1.2 FEATURES
The following paragraphs summarize the features of the PDP-II /34A floating-point instruction set
and the FPII-A.
1.2.1

Floating-Point Instruction Set Features
• 32-bit (single-precision) and 64-bit (double-precision) data modes
• Addressing modes compatible with existing PDP-II addressing modes
• Special instructions that can improve input/output routines and mathematical subroutines
• Allows execution of in-line code (i.e., floating-point instructions and other instructions can
appear in any sequence desired)
• Multiple accumulators for ease of data handling
• Can convert 32- or 64-bit floating-point numbers to 16- or 32-bit integers during the Store class
of instructions
• Can convert 32-bit floating-point numbers to 64-bit floating-point numbers and vice-versa
during the Load or Store class of instructions.

I-I

1.2.2

FPII-A Features
• Performs medium-speed, floating-point operations on single- and double-precision data
• Has 17 (decimal) digit accuracy
• Contains its own microprogrammed control store
• Contains six 64-bit floating-point accumulators
• Contains error recovery aids

1.3 AR CHITECTURE
The FPII-A contains scratchpad registers, a floating exception address pointer (FEA), status and
error registers, and six general-purpose accumulators (ACO-AC5).

Each accumulator is interpreted to be 32 or 64 bits long depending on the instruction and the status of
the floating-point processor. For 32-bit instructions, only the left-most bits are used. The remaining
bits are unaffected.
The six general-purpose accumulators are used in numeric calculations and interaccumulator data
transfers. The first four registers (ACO-AC3) are also used for all data transfers between the FPII-A
and the central processor's general registers or memory.

1.4 PHYSICAL DESCRIPTION
The FPII-A consists of a single hex board [M8267 for the PDP-II /34A (KD ll-EA)) and modifications to the M7265 and M7266 boards used in the PDP-Il/34 central processor. (The modified boards
are designated M8265 and M8266, and the modified processor is designated as the KDII-EA). Figure
1-1 shows the basic signal paths between the central processor and the FPII-A. The bidirectional data
bus transfers instructions and data between the processors. An expanded control store in the KD IlEA accommodates floating-point requirements.

KD11-EA

PDP
11/34
CENTRAL
PROCESSOR
M8265
M8266
IKD11-EA)

INSTRUCTIONS/DATA

10 MICRO PROGRAM ADDRESS LINES

FP11-A
FLOATING
POINT
PROCCESSOR

CLOCK
M8267

INITIALIZE

11-5259

Figure I-I

KDlI-EA/FPII-A Signal Interface

1-2

1.S RELATED DOCUMENTATION
The following documents supplement this user's guide on the FPII-A Floating-Point Processor.
Manual

Document Number

BAII-K Mounting Box Manual
BAII-L Mounting Box Manual
DLlI-W Maintenance Manual
KDII-E Processor Manual (PDP-I 1/34)
M9301 Bootstrap Terminator Maintenance Manual
MMII-C/CPCore Memory Manual
MMII-D/DP Core Memory Manual
MSII-E-J MOS Memory Maintenance Manual
PDP-II Peripherals Handbook
PDP-l 1/04, 34,45,55 Processors Handbook
PDP-I 1/34 Processor Handbook
KDII-EA Processor Manual (PDP-II/34A)

EK-BAIIK-MM
EK-BAIIL-MM
EK-DLlIW-MM
EK-KDIIE-TM
EK-M9301-MM
EK-MMIIB-TM
EK-MMIID-TM
EK-MSIIE-MM
EP-PDPII-HB
EP-PDPll/04-HB
EP-11034-HB
EK-KDIEA-MM

1-3

CHAPTER 2
INSTALLATION AND CHECKOUT

2.1 SCOPE
This chapter provides the information necessary for unpacking, inspection, installation, and checkout
of the FPII-A and FPII-AU Floating-Point Processors.
2.2 FPll-A FLOATING-POINT PROCESSOR INSTALLATION
The FPII-A Floating-Point Processor option for the PDP-II /34A CPU consists of the following
parts:
I.
2.
3.
4.

M8267 - Floating-point module
H8821 - 20-pin over-the-top connector
54-12416 - IO-pin over-the-top connector
W9042 - Bus extender module

Prior to the installation of the FP II-A option, the + 5 Vdc current available to the PDP-II /34A CPU
backplane must be calculated. The following procedure is designed to help you calculate +5 Vdc
current drain and system configuration.

2.2.1

FPII-A Add-On Installation Procedure
l.

Verify system integrity by running the following diagnostics in the order given.
PDP-ll/34

CPU Test
Traps Test (at least Rev. C)
EIS Test
0-124K memory exerciser

DFKAA
DFKAB
FDKAC
DZQMC

2. Is CPU a PDP-ll/34A? (See serial name tag.)

Yes

No
An FPl1-A cannot be installed on a PDP-l 1/34. To upgrade a PDP-l 1/34 to
use an FPII-A, an FPII-AU kit must be used. Refer to Paragraph 2.3.

3. Is CPU box 26.7cm (10.5 in)?

Yes

No
Refer to Paragraph 2.2.2, BAll-L Box.

4. Calculate +5 Vdc current drain in the CPU backplane. Calculate +5 Vdc current drain for
all other backplanes in box (Figure 2-1).

2-1

Is the total current drain (all backplanes) greater than 57 A? (Does not include M8267
current.)
No

Yes

Is expander box available with room and current?
Yes

Refer to step 17.
Refer to step 6.
6.

Is the battery backup (BBU) option present?
No

J
7.

Yes

All jumpers must be out of all backplanes in box. Refer to Figure 2-2 and step
9.

Do backplane jumpers check as follows?
CPU backplane (DDll-PK).
+ 5 VB to +5 V jumper
In }
+ 15 VB to +15 V jumper
In
-15 VB to -15 V jumper
In

See Figure 2-2.

Refer to step 8.
8.

All other backplanes in box (DDII-DK, CK).

out}

+5 VB to +5 V jumper
+ 15 VB to + 15 V jumper
-15 VB to -15 V jumper

In
In

See Figure 2-2.

Refer to step 9.
9.

Is slot 3 open in CPU backplane?
Yes

Is cache (M 8268) in slot 3?
No

Yes

Cache is placed in slot 3 only if FP is not present. When FP is
added, cache is moved to slot 5. Slots 4A and B are reserved for
M9301/M9312 (Figure 2-1). Remove the over-the-top (OTT) connectors and move cache module (M8268) to slot 5. H8822 (OTT) is
necessary to complete the installation (Figure 2-1). Refer to step
10.
Refer to step 10.
10.

Is the MaS memory installed in any backplane other than the CPU backplane?
No

Yes

Add 0.5 A at +5 Vdc for each MaS board not installed in the CPU backplane
(CPU box only) to the current drain total for the CPU backplane calculated in
step 4.
Refer to step 11.

2-2

11. Is the CPU backplane current drain less than 25 A at +5 Vdc?
Yes

The devices must be moved from the CPU backplane to some other backplane
in the box in order to vacate slot 3 and to reduce the current drain at +5 Vdc to
25 A or less. This must be done without overloading the second +5 Vdc regulator in the box.
Is reconfiguring within the box possible?
Yes

Can the devices be moved to an expander box without overloading
the expander box?
No

Yes

Reconfigure the system until the current in the CPU
backplane is less than 25 A at +5 Vdc.

Refer to step 17.
Reconfigure within the box until the CPU backplane current drain at +5 Vdc is
less than 25 A.
Refer to step 12.
12. Install the FPI1-A module (M8267) in slot 3 of the CPU backplane.
13. Is KYII-LB (M7859) present?
No

Yes

Remove the two lO-pin maintenance cables, if necessary, from the CPU
(M8266) and install them in the FP module (M 8267) as shown in Figure 2-1.
Refer to step 14.

14. Install the two over-the-top (OTT) connectors as shown in Figure 2-1. Use H8822 if the
cache and FP are both present.
15. Turn power on and run the following diagnostics in the order given.
PDP-llj34
PDP-llj34
PDP-llj34
PDP-l1j34
PDP-llj34
PDP-llj34

CPU Test
Traps Test
EIS Test
FPP Diagnostic
FPP Diagnostic
FPP Diagnostic

At least Rev. C
Part 1
Part 2
Part 3

DFKAA
DFKAB
DFKAC
DFFPA
DFFPB
DFFPC

16. End
17. When it is impossible to reconfigure the box to accommodate the FPI1-A (M8267) without
overloading the +5 V regulator, one alternative is to move some devices to an expander box.
If an expander box is not present on the system, then there are two ways to proceed.
a.
b.

Remove some number of devices from the box to compensate for the 7 A at +5 Vdc
used by the FPII-A, and leave these devices out of the system.
Postpone installation until an expansion box can be added to the system.
Refer to step 16.
2-3

2.2.2

BAll-L Box
18.

Calculate the current drain at +5 Vdc for the backplane (DDII-PK) (Table 2-1 and Figure
2-1).

19. Is slot 3 open in the backplane?

Yes

No
Is the cache (M 8268) in slot 3?

Yes
Remove the over-the-top (OTT) connectors and move the cache
module (M8268) to slot 5. An H8822 OTT is necessary to complete
installation (Figure 2-1).
Refer to step 20.

Refer to step 20.
20.

Is the power supply an H777-AA, AB, BB (25 A version)?

Yes

No
The power supply must then be an H777-CA, CB, DA, DB (32 A version).
Is the total current drain at + 5 Vdc less than 25 A (without M8267)?

Yes

No
The devices must be moved from the CPU box to an expansion
box in order to vacate slot 3, if necessary, and to reduce the current
drain at +5 Vdc to 25 A or less. This must be done without overloading the expander box power supply.
Is the expander box available with enough room and power?

Yes

No
Refer to step 17.

Refer to step 12.
21.

Is the total current drain at +5 Vdc less than 18 A (without M8267)?

Yes

No
The devices must be moved from the CPU box to an expansion box in order to
vacate slot 3, if necessary, and to reduce the current drain at +5 Vdc to 18 A or
less. This must be done without overloading the expander box power supply.
Is the expander box available with enough room and power?

Yes

No
Refer to step 17.

Reconfigure the system until the CPU box current drain at +5 Vdc is less than
18 A.
Refer to step 12.
22.

Refer to step 12.

2-4

SLOT NO, 31M8267)

RED STRIPE
7011 411_10

2-5

RED STRIPE
7012214_20

SEE VIEW A

+5V JUMPER

o
[
o ]
0 GND

o 0
<§:]) ·15

PIN A01A1

o
<§:::]) +15
@:§) +15B
<§:::]) ·15B
<§:::]) +58
<§:::]) ·5

NOTES

<§:2)DCL
1. JUMPERS SHOWN ARE

<2::])AC LO

·15 TO ·15B

<2::]) LTC

+15 TO +15B

+!tTO +5B

<2::]) +20

2. USE #20 INSULATED
BUS WIRE FOR JUMPERS

(00) +5
~
VIEWA
11-5343

Figure 2-2

Backplane Jumpers

2-6

2.3 FPll-AU UPGRADE KIT
The FPII-AU upgrade kit contains power supply components necessary to increase the +5 Vdc current levels available from the 26.7 cm (l0.5 in) mounting box. The purpose of the upgrade kit is to
provide a method by which PDP-Il/34 CPUs can use the floating-point (FPII-A) option. Since the
FPII-A is an option for the PDP-ll/34A CPU, additional hardware is required to upgrade the PDP11/34 models to include the floating-point option. The following parts are required.
I.
2.
3.
4.
5.
6.
7.
8.

M8265 - Data path module
M8267 - FP module
M8266 - Control module
H7441 - Regulator module
H8821 - 20-pin over-the-top connector
54-12416 - lO-pin over-the-top connector
W9042 - Bus extender module
54-10834Y A - Power distribution board

The tools required are:
I.
2.
3.

2.3.1

Phillips screwdriver (medium and large)
Slot screwdriver (large)
90 degree offset Phillips screwdriver.
FPII-AU Power Components Installation
CAUTION
Turn off computer system and disconnect it from
power source before performing installation procedure.

1. Slide BA 11- K mounting box out of the cabinet assembly to the limits of the chassis slides.
2.

Release and remove mounting box top cover to gain access to H765 power supply assembly.

Loosen and remove cable clamp that secures the cables that are routed across the top of the
power supply.

Loosen and remove power supply cover.

Rotate the mounting box in such a manner that the bottom of the mounting box faces away
from the cabinet (box rotated 90 degrees).

Loosen and remove mounting box bottom cover to gain access to the power distribution
board located between the power supply and the backplane.
CAUTION
Do not remove the hinge screws (one on each side)
located at the junction of the power supply and the
module enclosure near the top side of the mounting
box.

Remove four flat-head screws (no washers) located approximately 10 cm (4 in) from the
bottom of the mounting box and at the junction of the power supply and the module enclosure assembly. The power supply can now be swiveled away from the module enclosure.

2-7

Locate the H744 +S Vdc regulator assembly. This regulator is the second module from the
right when viewing the bottom of the mounting box from the wire-wrap side of the backplane.

Locate and remove the two mounting screws and washers located just to the left of the H744
Mate-N-Lok connector. There will be a green safety wire secured by one of these screws.
NOTE
A 90 degree offset Phillips-head screwdriver is required to remove these screws and attached hardware.

10.

Locate and remove the last retaining screw and washer located on the back of the power
supply and to the left of the H744 decal.

11.

Release and remove the H744 Mate-N-Lok connector.

12.

Remove the H744 regulator by sliding it out through the top of the power supply assembly.
(Note that the mounting box may have to be rotated to accomplish removal.)

13.

Replace the H744 regulator with the H744I regulator (included in the upgrade kit).

14.

Replace mounting hardware removed in steps 9 and 10. Do not connect the H744I Mate-NLok connector at this time.

IS.

Release and remove the three Mate-N-Lok connectors connecting the remaining regulators
to the power distribution board assembly:

16. Locate and remove the black ground wire soldered to the power distribution board (located
near 116). Remove this ground wire from the power supply and the module enclosure assembly.
17.

Remove the +S V and ground fastons from the power distribution board (located near 114).

18.

Release and disconnect Mate-N-Lok connector from 18 on the power distribution board.

19.

Locate and remove four flat-head screws securing the power distribution board to the module enclosure assembly. These screws are located (two on each side) S em (2 in) from the
bottom of the mounting box and near the junction of the power supply and the module
enclosure.
NOTE
The removal of these four screws will allow the removal of the power distribution panel which in turn
will allow removal of all backplane Mate-N-Lok
connectors.

20.

Release and disconnect all backplane Mate-N-Lok connectors.

21.

Release and disconnect two Mate-N-Lok connectors connecting the power distribution
board to the power supply.

22.

Remove the power distribution board.

2-8

23.

Replace the power distribution board with the new 5410834- Y A power distribution board
(included in upgrade kit).

24.

Reverse procedure (steps 22-14 and steps 7-1) to install the new power distribution board
and to return system to normal.

2.3.2

FPll-AU Logic Installation
Refer to Paragraph 2.2.1 and calculate +5 V dc current drain and system configuration.

2-9

CHAPTER 3
REVIEW OF FLOATING-POINT NUMBERS

3.1 INTRODUCTION
This chapter briefly outlines some fundamentals of floating-point arithmetic. It provides useful background for more advanced topics in later chapters. The reader already familiar with floating-point
numbers and arithmetic may skip this chapter and continue to Chapter 4 for a discussion of FPII-A
data formats.
3.2 INTEGERS
All data within a computer system can be represented in integer form. The numbers that can be
represented in a 16-bit machine range in magnitude from OOOOOOg to 177777g (or from 010 to 65,53610)'
However, this presents problems with integer representation. A number between 1 and 2 (for example)
or numbers greater than 65,53610 can not be represented. Thus, integer representation imposes an
accuracy and a range limitation.
These limitations are imposed by the stationary position of the radix point (e.g., the decimal point in
base 10 notation or the binary point in base 2 notation). An integer's radix point is usually omitted in
integer representation because it always marks the integer's least significant place. That is, there are
never any digits to the right of an integer's radix point. For this reason, an integer is sometimes called a
fixed-point number.
Integer notation, however, can be modified to overcome the range and accuracy limitations imposed
by the fixed radix point. This is accomplished through the use of floating-point notation.
3.3 FLOATING-POINT NUMBERS
Floating-point numbers, unlike integers, have no position restrictions imposed on their radix points. A
popular type of floating-point representation is called scientific notation. With scientific notation, a
floating-point number is represented by some basic value multiplied by the radix raised to some power.

Example
Basic
value
/

/nent

1,000,00010 = l. X 106

~Radix
3-1

There are many ways to represent the same number in scientific notation, as shown in the example
below.
512. = 51200.
X 10-2
= 5120. X 10- 1
=
512.
X 100
= 51.2 X 10 1
=
5.12 X 102
=
.512X 103
The convention chosen for representing floating-point numbers with scientific notation in the FPII-A
requires the radix point to always be to the left of the most significant digit in the basic value (e.g., .512
X 10 3 in the above example). This modified basic value is called a fraction.
More examples of scientific notation are shown below.
Decimal
No.

Decimal
Scient. No.

Octal
Scient. No.

Binary
Scient. No.

64
33
1/2
1/16

0.64 X 102
0.33 X 102
0.5 X 100
0.625 X 10- 1

0.1 X 83
0.41 X 82
0.4 X 80
0.4 X 8- 1

X 27
0.1
0.100001 X 26
0.1
X 20
X 23
0.1

Note that in each of the examples above, only significant digits are retained in the final result and the
radix point is always (by convention) to the left of the most significant digit. Establishing the radix
point in a number whose basic value is greater than or equal to 1 is accomplished by shifting the
number to the right until the most significant digit is to the right of the radix point. Each right shift
causes the exponent to be incremented by 1. Similarly, establishing the radix point in a number whose
basic value is between 1 and 0 (i.e., a fraction) is accomplished by shifting the number to the left until
all leading Os are eliminated. Each left shift causes the exponent to be decremented by 1.
To summarize, the value of the number remains constant if its exponent is incremented for each right
shift of the basic value and decremented for each left shift. The representation for floating-point
fractions in the FPII-A is one in which all nonsignificant leading Os have been removed. The process
used to obtain this representation is called normalization, which is explained in more detail in Paragraph 3.4.
3.4 NORMALIZATION
In digital computers, the number of bits in a fraction is limited. Retention of nonsignificant leading Os
decreases accuracy by taking places that could be filled by significant digits. For this reason, a process
called normalization is used in the FPII-A. The normalization process consists of testing the fraction
for leading Os and left-shifting it until it is in the form 0.1 .... The exponent is accordingly decremented by the number of left shifts of the fraction. This ensures that the normalized number retains
equivalence with the original number. Since digits to the right of the binary point are weighted with
inverse powers of2 (i.e., 1/2, 1/4, 1/8 ... ), the smallest normalized fraction is 1/2 (0.10000 ... ). The
largest normalized fraction is 0.11111 .... Figure 3-1 shows an unnormalized fraction that must be
left-shifted six places to be normalized. The exponent is decremented by six to maintain equivalence
with the original number.

3-2

FRACTION

EXPONENT
UNNORMALIZED

100

011

NORMALI ZED

011

101

DECREASE

II
II

000

111

001

111

001

000

LEFT SHIFT FRACTION SIX PLACES

EXPONENT BY SIX

MA-0285

Figure 3-1

Normalization

Problem A - Represent the number 7510 as a binary normalized floating-point number.
I.

Integer conversion
7510 = 10010112

Convert to floating-point form
1001011.0 X 20 = 0.1001011 X 27
Fraction = 0.1001011
Exponent = III

Problem B - Represent the number 0.2510 as a binary normalized floating-point number.
I.

Integer conversion
0. 25 10 = 0.012

Convert to floating-point form
0.01 X 20 = 0.1 X 2-1
Fraction = 0.1
Exponent = -I

3.5 FLOA TING-POINT ADDITION AND SUBTRACTION
In order to perform floating-point addition or subtraction, the exponents of the two floating-point
numbers involved must be aligned or equal. If they are not aligned, the fraction with the smaller
exponent is shifted right until they are. Each shift to the right is accompanied by an incrementation of
the associated exponent. When the exponents are aligned or equal, the fractions can then be added or
subtracted. The exponent value indicates the number of places the binary point is to be moved to
obtain the integer representation of the number.
In the example below, the number 710 is added to the number 4010 using floating-point representation.
Note that the exponents are first aligned and then the fractions are added; the exponent value dictates
the final location of the binary points.
+0.101 000000000000 X 26 = 50 8 = 4010
+0.111 000 000 000 000 X 26 = 78 = 710

3-3

To align exponents, shift the fraction with the smaller exponent three places to the right and
increment the exponent by 3, and then add the two fractions.
+0.101 000000 000 000 X 26 = 508 = 4010
+0.000 111 000 000 000 X 26 = 78 = 710
+0.101 III 000 000 000 X 26 = 578 = 4710

To find the integer value of the answer, move the binary point six places to the right.
5

0.101 111.000000 000
~

3.6 FLOATING-POINT MULTIPLICATION AND DIVISION
In floating-point multiplication, the fractions are multiplied and the exponents are added. For floating-point division, the fractions are divided and the exponents are subtracted.
There is no requirement to align the binary point in floating-point multiplication or division.
Example:
Multiply 710 by 4010.
0.IIIOOOOX2 3 = 78= 710
XO.IOIOOOO X 26 = 508 = 4010

III
0000
11100
.10001100000000 X 29

(Result already in normalized form.)

Move the binary point nine places to the right.

--------

430
.100011000.00000 = 4308 = 28010
\

Example:
Divide 1510 by 510.
1.

.1111000 X 24
.1010000 X 2 3
1.010000) .1 II 1000 =
1.100000
10 I 0000) 1111000.000000
1010000
101000
101000

o
3-4

Exponent: 4 - 3 = 1

Result: 1.100000 X 2
Normalized ~ooooo X 22"
Normalized Fraction

Normalized Exponent

Move binary point two places to the right.
.11.00000 = 38 = 310

3-5

CHAPTER 4
DATA FORMATS

4.1

INTRODUCTION

The FPII-A requires its input data (operands) to be formatted. Formatting allows the FPII-A to
process operands in a meaningful way and produce correct results. There are four different formats for
operands input to the FPII-A: short-integer format (I), long-integer format (L), single-precision format (F), and double-precision format (D).
Output data from the FPII-A is also formatted. This output data is in the form of:
I.
2.

FPII-A status information and FPII-A exception information required by the CPU
Data sent to memory (via the CPU), which must be in I, L, F, or D format.

This chapter describes the FPII-A data formats. It is assumed that the reader is familiar with 2's
complement notation.

4.2

FPll-A INTEGER FORMATS

There are two integer formats, short (I) and long (L). The short-integer format is 16 bits long and the
long-integer format is 32 bits long. Data words (operands) in integer format are represented in 2's
complement notation. In both I and L formats, the most significant bit of the data word is the sign bit.
Figure 4-1 shows the integers 5 and -5 in both I and L formats.
Figure 4-2 illustrates the formats in which integers are arranged in memory. Integers sent to memory
must be in one of these formats. Integers received by the FPII-A are arranged and manipulated
according to the type of instruction being executed. Refer to Paragraphs 5.3.11 and 5.3.12 for descriptions of the ways in which incoming integers are manipulated during the load exponent and load
convert integer-to-floating instructions, respectively.

4.3

FPll-A FLOATING-POINT FORMATS

There are two floating-point formats, single-precision (F) and double-precision (D). The single-precision
format is 32 bits long and the double-precision format is 64 bits long. Figure 4-2 shows that the most
significant bit is the sign of the fraction (and the floating-point number being represented). The next 8
bits contain the value of the exponent, expressed in excess 200 notation (Paragraph 4.3.1.2). The
remaining bits (23 for single-precision, 55 for double-precision) contain the fraction. The fraction and
its associated sign bit are expressed in sign and magnitude notation (Paragraph 4.3.1.1).

4.3.1

FPll-A Floating-Point Data Word

Figure 4-3 illustrates the formats in which floating-point numbers are arranged in memory. Floatingpoint numbers sent to memory must be in one of these formats. Floating-point numbers received by
the FPII-A are arranged as illustrated in Figure 4-4.

4-1

INTEGER = 5
I.
SHORT

W OR D 1

--------I

1':1'01+1+1

INTEGER II)

SIGN BIT
I.
3130

LONGINTEGER(L)

I 0

-- -

I 0

WOR D 1

0 I 0 I 0 1 0 1

i'"---- WO R D 2 ------t
1514
0
1 0 1 0

I 0

1 0

5 I

SIGN B I T - - - ' N r E : G E R : - : ; - - - - - I-

WORD 1

SHORT INTEGER (I)

WORD 1

I·

LONG INTEGER (Ll

31 30

i'"---- W0 R D 2 --------I

15 14

SIGN BIT

Figure 4-1

MEMORY
WORD 1

~
31 30
SINGLE-PRECISION
FLOATING-POINT (F)
FORMAT IS

Integer Formats

2322
EXP

FRACTION
MEMORY
I-WORD 2-1

I
I
L--.I....-----1-----.I"1
tj

63 62
DOUBLE - PREC I SION
FLOATI NG- POINT (0)
FORMAT IS

5554

EXP

°1
0

MEMORY
WORD1

I·

MEMORY
WORD 2

-I I·

____________________

MEMORY
I-WORD 3 - i

MEMORY
I- WORD 4-i

t ~t ~

~~~

____________________--JJ

FRACTION
S = Sign
EXP = Exponent in excess 200 notation (refer to paragraph 3.3.1.2.).
Fraction 23 or 55 bit fraction in sign and magnitude
format.

MA-02BO

Figure 4-2

Floating-Point Data Formats

4-2

FRACTION
15
MEMORY

-- --EXP

IS ]

62-

INITIALLY LOADED
INTO FP11A

---;:>

--- ---

.-.:::--

.A.

I >< -::::- -

--::,.....,

-_--.----===---=::------!..
40 39

-I

..;-'-

...,....,S-'-J- - E X P - : = = = - t )

ZEROES

17~L.o

1
63

EXP

39 38
ZEROES

FRACTION
OVERflOW BIT
HIDDEN BIT _ _ _-'

(EXP;o'O. BIT 62=1)
11·5254

MEMORY

7
EXP!

~ -

6
I

Single Precision

°
F

- _

- - : [ 7 - __/

I
7

/- -

l7 ..... ./~---..... I °

/'/~

.r---L~-_/'/

56 /55
INITIALLY LOADED
INTO FPI1A

/,.?- -

-;>-?'

./'"
- .-

FRACTION

II I

Iii.,

\
\

/
/

/-/7_ _ _--,0

Ex_p_--,I

< - I_ _

7 6

FPll·AWORD
IN WORKING AREA

FRACTION

OVERflOW BIT
HIDDEN BIT - - - - '

(EXP;o'O. BIT 62=1)
11-5255

b.
Figure 4-3

Double Precision
Floating-Point Data Words

4-3

SIGN.

EXPONENT

14
0

MEMORY

FRACTION

.......

0
0

NUMBER 32 REPRESENTED
IN SIGN AND MAGNITUDE
FORMAT (NUMBER ASSUMED
NORMALIZED)

FPll
S

----0---,
ADDITIONAL
I

---~I--210

EXPONENT

J
I

FRACTION

HIDDEN
BIT

EXPONENT = 206 - 200 = 6 = 2"

OPERANDS

FRACTION = 1/2 (INSERTION OF HIDDEN BIT)

FLOATING POINT NUMBER = 2" X 1/2 = 32

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

EXPONENT

FRACTION

. . .. - - - -. . .

r - - - - - - - - ' ----------.,,,r-----~

MEMORY

FORMAT(NUMBER ASSUMED
NORMALIZED)

o
7
~

~~~~

1--~D~I~~7"-l

.......~........._'"-...r..............._ I -...................- - J _ 63

OPERANDS

~l- _.J
21 0

_ _ _ _- J

EXPONENT
EXPONENT = 177 - 200 = -1 = 2-1

HIDDEN
BIT

FRACTION
FRACTION = 1/2 + 1/4 + 1/8 = 7/8 (INSERTION OF HIDDEN BIT)

FLOATING POINT NUMBER = 2-1 X 7/8 = 7/16
MA-1448

Figure 4-4

Interpretation of Floating-Point Numbers

4-4

The sign bit, exponent bits, and fraction bits in the FPII-A data word have the same values as the data
word in memory. Note, however, that the FPII-A data word has more bits than its counterpart in
memory. This is because the FPII-A has provisions for generating an overflow bit and a "hidden" bit.
For purposes of discussion, the FPII-A data word can be thought of as being divided into two major
parts:
I.

A fraction, with its associated sign bit, hidden bit, and overflow bit.

An exponent.

4.3.1.1 Floating-Point Fraction - The fraction is expressed in sign and magnitude notation. The following simple example illustrates the idea behind sign and magnitude notation.

2's Complement
Notation

Sign and Magnitude Notation

000010

,....000010
Sign ~Magnitude

-2

111110

/100010
Sign ~ Magnitude

Only a change of sign bit is required to change the sign of a number in sign and magnitude notation.
Note that a positive n umber is the same in both notations.
Un normalized floating-point fractions have a range from approximately 0 through 2 as shown in
Figure 4-5. The FPII-A, however, normalizes all unnormalized fractions. That is, the fractions are
adjusted such that there is a 0 to the left of the binary point (bit 63) and a I to the right of the binary
point (bit 62). Thus, normalized fractions range in magnitude from 0.1000 ... to 0.1111 or from 1/2 to
approximately I.

~~~LZLEE~~NUMBER,--I0--'-.1_°. . .1_
. ~I_o....
1 ~: : 1 0 I 0 1 0 I' 1
0

LARGEST
NONZERO NUMBER

APPROXIMATELY

I
MA·1447

Figure 4-5

Unnormalized Floating-Point Fraction

4-5

The fraction overflow bit (bit 63) is set during certain arithmetic operations. For example, during
addition, certain sums will produce an overflow such as 0.1000 ... + 0.100 ... which yields 1.000 ....
This result must be normalized, so the FPII-A right-shifts the fraction one place and increases the
exponent by one.

Bit 62 is called the hidden bit. Recall that since fractions are normalized by the FPII-A, the bit
immediately to the right of the binary point (bit 62) is always a I. This bit is dropped when a fraction is
sent to memory and appended when a fraction is received from memory. This procedure allows one
extra bit of significance in floating-point fraction representation.

4.3.1.2 Floating-Point Exponent - The 8-bit floating-point exponent is expressed in excess 200 notation. The chart below illustrates the relationship between exponents in 2's complement notation and
exponents in excess 200 notation.

2's Complement

Excess 200

177 Most positive exponent

377 Most positive exponent

Positive
Exponents

Negttive
Exponents

Least positive exponent

200 Least positive exponent

377 Least negative exponent

177 Least negative exponent

Negative
Exponents

200 Most negative exponent

Most negative exponent

Note that an exponent in excess 200 notation is obtained by simply adding 200 to the exponent in 2's
complement notation. Thus, 8-bit exponents in excess 200 notation range from a to 377 (or from -200
to + 177). A number with an exponent of -200 is treated by the FPII-A as O.

For example, the number 0.12 is actually 0.1 X 20, and the exponent is represented as 10 000 000
because 2008 represents an exponent of zero. Figure 3-5 illustrates the range of floating-point numbers
that can be handled by the FPII-A. For simplicity, a fraction length of only three bits is shown.

4-6

4.4 FPll-A PROGRAM STATUS REGISTER
The FP ll-A contains a resident program status register that contains the floating-point condition
codes (carry, overflow, zero, and negative) that can be copied into the central processor. In other
words, FN, FZ, FV, and FC can be copied into the CPU's N, Z, V, and C condition codes, respectively. The program status register also contains 3 mode bits and additional bits to I!nable various
interrupt conditions. Figure 4-6 shows the layout of the program status register. Each bit shown in the
figure is described in Table 4-1.
NOTE
The FPll-A has no Unibus addresses. All FPll-A
registers are accessed by floating-point instructions
only.

INTERRUPT ENABLES
15

MODE BITS
7

CONDITION CODES
4

II I II I I I II I II I II I II I
FER

FlO

NOT
USED

I FI1UV

NOT
USED

FlV

FIU

FIC

NOT
USED

MA-1432

Figure 4-6

FPII-A Status Register Format

Table 4-1

FPll-A Status Register

Bit

Name

Function

FER

This bit indicates an error condition of the FPII-A.

FlO

Floating Interrupt Disable - All interrupts by the FPII-A are disabled
when this bit is on. Primarily for maintenance use. Normally clear.

Not Used

FIUV

Floating Interrupt on Undefined Variable - When this bit is set and a
-0 is obtained from memory, an interrupt occurs. If the bit is not set, -0
can be loaded and stored; however, any arithmetic operation treats it as
if it were a positive O.

FlU

Floating Interrupt on Underflow - When this bit is set, an underflow
condition causes a floating underflow interrupt. The result of the operation causing the interrupt is correct except for the exponent, which is
off by 400 8 • If the FlU is not set and underflow occurs, the result is set
to zero.

4-7

Table 4-1

FPll-A Status Register (Cont)

Bit

Name

Function

FIV

Floating Interrupt on Overflow - When this bit is set, floating overflow
causes an interrupt. The result of the operation causing the interrupt is
correct except for the exponent, which is off by 4008. If the FIV bit is
not set, the result of the operation is the same; the only difference is that
the interrupt does not occur.

FIC

Floating Interrupt on Integer Conversion Error - When this bit is set
and the store convert floating-to-integer instruction causes FC to be set
(indicating a conversion error), an interrupt occurs. When a conversion
occurs, the destination register is cleared and the source register is untouched. When FIC is reset, the result of the operation is the same;
however, no interrupt occurs.

Double-Precision Mode Bit - This bit, when set, specifies double-precision format and, when reset, specifies single-precision format.

Long-Precision Integer Mode Bit - This bit is employed during conversion between integer and floating-point format. If set, double-precision 2's complement integer format of 32 bits is specified; if reset,
single-precision 2's complement integer format of 16 bits is specified.

Truncate Bit - This bit, when set, causes the result of any floating-point
operation to be truncated rather than rounded.

Not Used

3-0

FN, FZ, FV,
and FC

These bits are the four floating-point condition codes, which can be
loaded in the CPU's N, Z, V, and C condition codes, respectively. This
is accomplished by the copy floating condition codes (CFCC) instruction. To determine how each instruction affects the condition codes,
refer to Table 5-1.

4.5 PROCESSING OF FLOATING-POINT EXCEPTIONS
Location 244 is the interrupt vector used to handle all floating-point interrupts. A total of six possible
interrupts can occur. These possible interrupt exceptions are encoded in the FPII-A exception code
(FEC) register. The interrupt exception codes represent an offset into a dispatch table, which routes
the program to the right error handling routine. The dispatch table is a function of the software. The
FEC for each exception is briefly described in Table 4-2.
Refer to the PDP-II /04,34,45,45 Processor Handbook for further details concerning FPII-A exceptions.
In addition to the FEC register, the CPU contains a 16-bit floating exception address (FEA) register,
which stores the address of the last floating-point instruction that caused a floating-point exception.

4-8

Table 4-2 FPll-A Exception Codes
FPll-A
Exception
Code

Definition

Floating Op Code Error - The FPll-A causes an interrupt
for an erroneous op code

Floating Divide by Zero - Division by zero causes an interrupt if FID is not set

Floating (or Double) Integer Conversion Error

Floating Overflow

Floating Underflow

Floating Undefined Variable
NOTE
The traps for exception codes 6, 10, 12, and 14, can
be enabled in the FPll-A program status register.
All traps are disabled if FID is set.

4-9

CHAPTER 5
FLOATING-POINT INSTRUCTIONS

5.1 FLOATING-POINT ACCUMULATORS
The FPII-A contains six general-purpose accumulators (ACO-AC5). These accumulators are 64-bit
read/write scratch pad memories with non-destructive readout.
Each accumulator is interpreted as being either 32 or 64 bits long, depending on the instruction and the
FPII-A status (Chapter 4). If an accumulator is interpreted as being 64 bits long, 64 bits of data
occupy the entire accumulator. If an accumulator is interpreted as being 32 bits long, 32 bits of data
occupy only the left-most 32 bits of an accumulator as shown in Figure 5-1.
The floating-point accumulators are used in numeric calculations and interaccumulator data transfers.
ACO-AC3 are used for all data transfers between the FPII-A and the CPU or memory.

64 BIT ACCUMULATOR

(

a
ACCUMULATORS

____________

______________

32 BIT ACCUMULATOR

r---------------r-------------~

3r---------------r-------------~

4
5

r---------------~------------~

MSB

LSB
MA-0277

Figure 5-1

Floating-Point Accum ulators

5.2 INSTRUCTION FORMATS
An FPII-A instruction must be in one of five formats. These formats are summarized in Figure 5-2.
The 2-bit AC field (bits 6 and 7) allows selection of scratchpad accumulators 0 through 3 only.
If address mode 0 is specified with formats Fl or F2, bits 2-0 are used to select a floating-point
accumulator. Only accumulators 5-0 can be specified in mode O. If 6 or 7 is specified in bits 2-0 in
mode 0, the FPII-A traps if floating-point interrupts are enabled (FID = 0). The FEC will indicate an
illegal op code error (exception code 2).

5-1

OC = 17

F51

8 7

FOC

FDST

SRC/DST

12 11
OC = 1 7

6 5

OC = 17

FSRC/FDST

65
AC

FOC

12 "

I
I

FOC

12 11

15
F41

A C

6 5

OC = 17

F31

FOC

12 "
OC = 17

F 21

12 "

FOC

I
11-3730

Figure 5-2

Instruction Formats

The fields of the various instruction formats (as summarized in Table 5-1) are interpreted as follows.
Mnemonic

Description

Operation Code - All floating-point instructions are designated by a 4-bit op
code of 178.

FOC

Floating Operating Code - The number of bits in this field varies with the
format; the code is used to specify the actual floating-point operation.

SRC

Source - A 6-bit source field identical to that in the PDP-II instruction.

DST

Destination - A 6-bit destination field identical to that in a PDP-ll instruction.

FSRC

Floating Source - A 6-bit field used only in format Fl. It is identical to SRC,
except in mode 0 when it references a floating-point accumulator rather than
a CPU general register.

FDST

Floating Destination - A 6-bit field used in formats F1 and F2. It is identical
to DST, except in mode 0 when it references a floating-point accumulator
instead of a CPU general register.

Accumulator - A 2-bit field used in formats F1 and F3 to specify FP11-A
scratchpad accumulators 0-3.

5-2

Table 5-1

Format of FPll-A Instructions

Instruction
Format

Instruction

Mnemonic

ABSOLUTE

ADD

CLEAR

COMPARE

F5
Fl

COPY FLOATING CONDITION CODES
DIVIDE

LOAD

LOAD CONVERT

LOAD CONVERT INTEGER

F3
F4
Fl

LOAD EXPONENT
LOAD FPll'S PROGRAM STATUS
MODULO

MULTIPLY

NEGATE

F5
F5
F5
F5
Fl

SET DOUBLE MODE
SET FLOATING MODE
SET INTEGER MODE
SET LONG INTEGER MODE
STORE

STORE CONVERT

STORE CONVERT
FLOA TING TO INTEGER

F3
F4
F4
Fl

STORE EXPONENT
STORE FPll 'S PROGRAM STATUS
STORE FPll 'S STATUS
SUBTRACT

TEST

ABSF FDST
ABSD FDST
ADDF FSRC, AC
ADD FSRC,AC
CLRFFDST
CLRD FDST
CMPF FSRC, AC
CMPD FSRC, AC
CFCC
DIVF FSRC, AC
DIVD FSRC, AC
LDFFSRC,AC
LDDFSRC,AC
LDCFD FSRC, AC
FDCDF FSRC, AC
LDCIF SRC, AC
LDCID SRC, AC
LDCLF SRC, AC
LDCLD SRC, AC
LDEXPSRC,AC
LDFPS SRC
MODF FSRC, AC
MOOD FSRC, AC
MULF FSRC, AC
MULD FSRC, AC
NEGFFDST
NEGDFDST
SETD
SETF
SETI
SETL
STF AC, FDST
STDAC, FDST
STCFD AC, FDST
STCDF AC, FDST
STCFI AC, DST
STCFL AC, DST
STCDI AC, DST
STCDL AC, DST
STEXP AC, DST
STFPS DST
STST DST
SUBF FSRC, AC
SUBD FSRC, AC
TSTF FDST
TSTD FDST

5-3

5.3 INSTRUCTION SET
Table 5-2 contains the instruction set of the FPII-A. Some of the symbology may not be familiar.
Therefore, a brief description follows.
I.

A floating-point flip-flop, designated FD, determines whether single- or double-precision
floating-point format is specified. If the flip-flop is cleared, single-precision is specified and
is designated by F. If the flip-flop is set, double-precision is specified and is designated by D.
Examples are NEGF, NEGD, and SUBD.
NOTE
Only the assembler or compiler differentiates between NEGF and NEGD or LDCID or LDCLD instructions. The Floating-point does not differentiate
between the instructions but depends upon the value
of FD and FL as usually controlled by SETD,
SETF, SETC, and SETI instructions (i.e., LDCID
-+ SETI -+ SETD -+ LDCLD).

An integer flip-flop, designated FL, determines whether short-integer or long-integer format
is specified. If the flip-flop is cleared, short-integer format is specified and is designated by I.
If the flip-flop is set, long-integer format is specified and is designated by L. Examples are
SET! and SETL.

Several convert-type instructions use the symbology defined below.
CIL,FD - Convert integer to floating
CFD,IL - Convert floating to integer
C F D or CD F - Convert single-floating to double-floating or convert double-floating to
single-floati'ng

UPUM is defined as the largest possible number that can be represented in floating-point
format. This number has an exponent of 377 (excess 200 notation) and a fraction of allIs.
Note the UPUM is dependent on the format specified. LOUM is defined as the smallest
possible number that is not identically O. This number has an exponent of 001 and a fraction
of all Os except for the hidden bit.

The following conventions are used when referring to address locations.
(xxxx) = the contents of the location specified by xxxx
ABS (address) = absolute value of (address)
EXP (address) = exponent of (address) in excess 200 notation

Some of the octal codes listed in Table 5-2 are in the form of mathematical expressions.
These octal codes can be calculated as shown in the following examples.
Example 1: LD FPS Instruction
Mode 3, register 7 specified (F instruction format)
170100 + SRC
SRC field is equal to 37
Basic op code is 170 I 00
SRC and basic op code are added to yield 170137.

5-4

Example 2: LDF Instruction
AC2, mode 2, and register 6 specified (Fl instruction format).
172400 + C * 100 + FSRC
AC = 2
2 * 100 = 200
172400 + 200 = 172600
FSRC is equal to 26
172600 + 26 + 172626
7.

AC v 1 means that the accumulator field (bits 6 and 7 in formats Fl and F3) is logically
ORed with 01.
Example:
Accumulator field = bits 6 and 7 = AC2 = 10. AC v 1 = 11.

The information in Table 5-2 is expressed in symbolic notation to provide the reader with a quick
reference to the function of each instruction. The following paragraphs supplement the information in
Table 5-2.

Table 5-2

FPII-A Instruction Set

Mnemonic

Instruction Description

Octal Code

ABSF FDST
ABSD FDST

Absolute
FDST +- minus (FDST) if FDST ~ 0; otherwise FDST +- (FDST)
FC +- 0
FV +- 0
FZ +- 1 if exp (FDST) = 0; otherwise FZ +- 0
FN +- 0

170600+ FDST
F2 Format

ADDF FSRC, AC
ADDD FSRC, AC

Floating Add
AC +- (AC) + (FSRC) if IAC I + (FSRC)
~ LOLIM; otherwise AC +- 0
FC +- 0
FV +- 1 if IAC I~ UPLIM; otherwise FV +- 0
FZ +- if (AC) = 0; otherwise FZ +- 0
FN +- 1 if (AC) < 0; otherwise FN +- 0

172000+AC* 100+ FSRC
FI Format

CLRF FDST
CLRD FDST

Clear
FDST +- 0
FC +- 0
FV +- 0
FZ +- 1
FN +- 0

170400+ FDST
F2 Format

5-5

Table 5-2

FPll-A Instruction (Cont)

Mnemonic

Instruction Description

Octal Code

CMPF FSRC, AC
CMPD FSRC, AC

Floating Compare
FC +- 0
FV +- 0
FZ +- 1 if (FSRC) - (AC) = 0; otherwise
FZ +- 0
FN +- 1 if (FSRC) - (AC) < 0; otherwise
FN +- 0

173400+ AC* 100+ FSRC
Fl Format

CFCC

Copy Floating Condition Codes
C +- FC
V+- FV
Z +- FZ
N +- FN

170000
F5 Format

DIVF FSRC, AC
DIVD FSRC, AC

Floating Divide
AC +- (AC)/(FSRC) if I(AC)/(FSRC) I
~ LOLIM; otherwise AC +- 0
FC +- 0
FV +- 1 if IAC I~ UPLIM; otherwise FV +- 0
FZ +- 1 if EXP (AC) = 0; otherwise FZ +- 0
FN +- 1 if (AC) < 0; otherwise FN +- 0

174400+ AC* 100+ FSRC
Fl Format

LDF FSRC, AC
or
LDD FSRC, AC

Floating Load
AC +- (FSRC)
FC +- 0
FV +- 0
FZ +- 1 if (AC) = 0; otherwise FZ +- 0
FN +- 1 if (AC) < 0; otherwise FN +- 0

172400+ AC* 100+ FSRC
Fl Format

LDCDF FSRC, AC
LDCFD FSRC, AC

Load Convert Double-to-Floating or
Floating-to-Double
AC +- CF,D or CD,F (FSRC)
FC +- 0
FV +- 1 if IAC I ~ UPLIM; otherwise
FV +- 0
FZ +- 1 if (AC) = 0; otherwise FZ +- 0
FN +- 1 if (A C) < 0; otherwise FN +- 0

177400+ AC* 100+ FSRC
Fl Format
F, D-single-precision to
double-precision floating
D, F -double-precision to
single-precision floating

If the current format is single-precision float-

ing-point (FD = 0), the source is assumed to
be a double-precision number and is converted to single-precision. If the floating-truncate bit is set, the number is truncated;
otherwise, it is rounded. If the current format
is double-precision (FD = 1), the source is assumed to be a single-precision number and
loaded left-justified in the AC. The lower half
of the AC is cleared.

5-6

Table 5-2

FPll-A Instruction (Cont)

Mnemonic

Instruction Description

Octal Code

LDCIF SRC, AC
LDCID SRC, AC
LDCLF SRC, AD
LDCLD SRC, AC

Load and Convert from Integer to Floating
AC +- CIL,FD (SRC)
FC +- 0
FV +- 0
FZ +- 1 if (AC) = 0; otherwise FZ +- 0
FN +- 1 if (AC) < 0; otherwise FN +- 0
CIL,FD specifies conversion from a 2's complement integer with precision I or L to a
floating-point number of precision F or D. If
integer flip-flop IL = 0, a 16-bit integer (I) is
double specified, and if IL = 1, a 32-bit integer (L) is specified. If floating-point flip-flop
FD = 0, a 32-bit floating-point number (F) is
specified, and if FD = 1, a 64-bit floatingpoint number (D) is specified. If a 32-bit integer is specified and addressing mode 0 or
immediate mode is used, the 16 bits of the
source register are left justified, and the remaining 16 bits are zeroed before the conversion.

177000+AC*100+SRC
F3 Format

LDEXPSRC,AC

Load Exponent
AC SIGN +- (AC SIGN)
AC EXP +- (SRC) + 200 only if ABS (SRC)
< 177
AC FRACTION +- (AC FRACTION)
FC +- 0
FV +- I if (SRC) > 177; otherwise FV +- 0
FZ +- 1 if EXP (AC) = 0; otherwise FZ +- 0
FN +- 1 if (AC) < 0; otherwise FN +- 0

176400+AC*100+SRC
F3 Format

LDFPS SRC

Load FPII-A's Program Status Word
FPS +- (SRC)

170100+SRC
F4 Format

MODF FSRC, AC
MODD FSRC, AC

Floating Modulo
AC v I +- integer part of (AC)*(FSRC)
AC +- fractional part of (AC)*(FSRC)
- (AC v 1) if I(AC)*(FSRC) I
~ LOLIM or FlU = I; otherwise AC +- 0
FC +- 0
FV +-1 if IAC I~ UPLIM; otherwise FV +- 0
FZ +- 1 if (AC) = 0; otherwise FZ +- 0
FN +- I if (AC) < 0; otherwise FN +- 0

171400+AC*100+FSRC
Fl Format

LDCIF = Single Integer
to Single Float
LDCID = Single Integer
to Double Float
LDCLF = Long Integer
to Single Float
LDCLD = Long Integer
to Double Float

The product of AC and FSRC is 48 bits in
single-precision floating-point format or 59
bits in double-precision floating-point format.
The integer part of the product
[(AC)*(FSRC)] is found and stored in AC v 1.
The fractional part is then obtained and
stored in AC. Note that multiplication by 10
can be done with zero error, allowing decimal
digits to be stripped off with no loss in precision.

5-7

Table 5-2

FPll-A Instruction (Cont)

Mnemonic

Instruction Description

Octal Code

MULF FSRC, AC
MULD FSRC, AC

Floating Multiply
AC +- (AC)*(FSRC) if I(AC)*(FSRC) I
~ LOLIM; otherwise AC +- 0
FC +- 0
FV +-1 if IA<;:: I~ UPLIM; otherwise FV +- 0
FZ +- 1 if (AC) = 0; otherwise FZ +- 0
FN +- 1 if (AC) < 0; otherwise FN +- 0

171000+AC*IOOFSRC
FI Format

NEGF FDST
NEGD FDST

Negate
FDST +- minus (FDST) if EXP (FDST) ~ 0;
otherwise FDST +- 0
Fe +- 0
FV +- 0
FZ +- 1 if if EXP (FDST) = 0; otherwise FZ
+-0
FN +- 1 if (FDST) < 0; otherwise FN +- 0

170700+ FDST
F2 Format

SETD

Set Floating Double Mode
FD +- 1

170011
F5 Format

SETF

Set Floating Mode
FD +-0

170001
F5 Format

SETI

Set Integer Mode
FL +- 0

170002
F5 Format

SETL

Set Long-Integer Mode
FL +- 1

170012
F5 Format

STF AC, FDST
STD AC, FDST

Floating Store
FDST +- (AC)
FC +- FC
FV +- FV
FZ +- FZ
FN +- FN

174000+AC*I00+FDST
FI Format

5-8

Table 5-2

FPll-A Instruction (Cont)

Mnemonic

Instruction Description

Octal Code

STCFD AC, FDST
STCDF AC, FDST

Store Convert from Floating-to-Double or
Double-to-Floating
FDST +- CF,D or CD,F (AC)
FC +- 0
FV +- 1 if IAC I~ UPLIM; otherwise FV +- 0
FZ +- 1 if (AC) = 0; otherwise FZ +- 0
FN +- 1 if (AC) < 0; otherwise FN +- 0

176000+ AC* 100+ FDST
Fl Format
F, D-single-precision to
double-precision floating
D, F -double-precision to
single-precision floatmg

The STCFD instruction is the opposite of the
LDCDF instruction; thus, if the current format is single-precision floating-point (FD =
0), the source is assumed to be a single-precision number and is converted to double-precision. If the floating truncate bit is set, the
number is truncated; otherwise, it is rounded.
If the current format is double-precision (FD
= I), the source is assumed to be double-precision number and loaded left-justified in the
AC. The lower half of the AC is cleared.
STCFI AC, DST
STCFL AC, DST
STCDI AC, DST
STCDL AC, DST

Store Convert from Floating-to-Integer
Destination receives converted AC if the resuiting integer number can be represented in
16 bits (short integer) or 32 bits (long integer).
Otherwise, destination is zeroed and C-bit is
set.

STCFI = Single Float to
Single Integer
STCFL = Single Float to
Long Integer
STCDI = Double Float
to Single Integer
STCDL = Double Float
to Long Integer

FV +- 0
FZ +- I if (DST) = 0; otherwise FZ +- 0
FN +- 1 if (DST) < 0; otherwise FN +- 0
C +- FC
V+- FV
Z +- FZ
N +- FN
When the conversion is to long integer (32
bits) and address mode 0 or immediate mode
is specified, only the most significant 16 bits
are stored in the destination register.

5-9

175400+ A C* 100+ DST
F3 Format

Table 5-2

FPll-A Instruction (Cont)

Mnemonic

Instruction Description

Octal Code

STEXP AC, DST

Store Exponent
DST +- AC EXPONENT - 2008
FC +- a
FV +- 0
FZ +- I if (DST) = 0; otherwise FZ +- 0
FN +- I if (DST) < 0; otherwise FN +- 0
C +- FC
V+- FV
Z +- FZ
N +- FN

175000+C*100+DST
F3 Format

STFPS DST

Store FPII-A's Program Status Word
DST +- (FPS)

170200+DST
F4 Format

STST DST

Store FPII-A's Status
DST +- (FEC)
DST + 2 +- (FEA) if not mode 0 or not immediate mode

170300+DST
F4 Format

SUBF FSRC, AC
SUBD FSRC, AC

Floating Subtract
AC +- (AC) - (FSRC) if I(AC) - (FSRC) I
~ LOLIM; otherwise AC +- 0
FC +- 0
FV +- I if AC UPLIM; otherwise FV +- 0
FZ +- I if (AC) = 0; otherwise FZ +- 0
FN +- I if (AC) < 0; otherwise FN +- 0

173000+ A C* 100+ FSRC
Fl Format

TSTF FDST
TSTD FDST

Test
Floating
FC +- 0
FV +- 0
FZ +- I if EXP (FDST) = 0; otherwise FZ +- 0
FN +- I if (FDST) < 0; otherwise FN +- 0

170500+ FDST
F2 Format

5.3.1 Arithmetic Instructions
The arithmetic instructions (Add, Subtract, MUltiply, Divide) require one operand in a source (a
floating-point accumulator in mode 0, a memory location otherwise) and one operand in a destination
accumulator. The instruction is executed by the FPII-A and the result is stored in the destination
accumulator.
The Compare instruction also requires one operand in a source and one operand in a destination
accumulator. However, the two operands remain in their respective locations after the instruction is
executed by the FPII-A, and there is no transfer of the result.

5-10

5.3.2 Floating-Modulo Instruction
The Floating-Modulo (MOD) instruction causes the FPI1-A to multiply two floating-point operands,
separate the product into integer and fractional parts, and store one or both parts as floating-point
numbers. The whole-number portion goes into an odd-numbered accumulator and the fraction goes
into an even-numbered accumulator.
The whole-number portion of the number, when expressed as a floating-point number, contains an
exponent greater than 201 in excess 200 notation, which means that the whole number has a decimal
value of some number greater than one and less than UPUM, where UPUM is the greatest possible
number that can be represented by the FPII-A.
The fractional portion of the number, when expressed as a floating-point number, contains an exponent less than or equal to 201 in excess 200 notation. This means that the fraction has a value less than
one and greater than LOUM, where LOUM is the smallest possible number that can be represented
by the FPII-A.

5.3.3 Load Instruction
The Load instruction causes the FPII-A to take an operand from a source and copy it into a destination accumulator. The source is a floating-point accumulator in mode 0 and a memory location otherwise.
5.3.4 Store Instruction
The Store instruction causes the FPII-A to take an operand from a source accumulator and transfer it
to a destination. This destination is a floating-point accumulator in mode 0 and a memory location
otherwise.
5.3.5 Load Convert (Double-to-Floating, Floating-to-Double) Instructions
The Load Convert Double-to-Floating (LDCDF) instruction causes the FPII-A to assume that the
source specifies a double-precision floating-point number. The FPII-A then converts that number to
single-precision, and places this result in the destination accumulator. If the floating-truncate (FT)
status bit is set, the number is truncated. If the FT bit is not set, the number is rounded by adding a I to
the single-precision segment if the MSB of the double-precision segment is a 1 depending on the prior
conditions set up by the FD bit (Figure 5-3). If the MSB of the double-precision segment is 0, the
single-precision word remains unchanged after rounding.

,
DOUBLE PRECISION
SEGMENT

SINGLE PRECISION
SEGMENT

MA-0288

Figure 5-3

Double-to-Single Precision Rounding

The Load Convert Floating-to-Double (LDCFD) instruction causes the FPII-A to assume that the
source specifies a single-precision number. The FPII-A then converts that number to double-precision
by appending 32 zeros to the single-precision word, and places this result in the destination accumulator.
Note that for both Load Convert instructions, the number to be converted is originally in the source (a
floating-point accumulator in mode 0, a memory location otherwise) and is transferred to the destination accumulator after conversion.

5-11

5.3.6 Store Convert (Double-to-Floating, Floating-to-Double) Instructions
The Store Convert Double-to-Floating (STCDF) instruction causes the FPII-A to convert a doubleprecision number located in the source accumulator to a single-precision number. The FPII-A then
transfers this result to the specified destination. If the floating-truncate (FT) bit is set, the floatingpoint n umber is truncated. If the FT bit is not set, the number is rounded. If the MSB (bit 31) of the
double-precision segment of the word is ai, 1 is added to the single-precision segment of the word,
depending on the prior conditions set up by the FD bit (Figure 5-3); otherwise, the single-precision
segment remains unchanged.
The Store Convert Floating-to-Double (STCFD) instruction causes the FP11-A to convert a singleprecision number located in the source accumulator to a double-precision number. The FPII-A then
transfers this result to the specified destination. The single-to-double precision is obtained by appending zeros equivalent to the double-precision segment of the word (Figure 5-4).
Note that for both Store Convert instructions, the number to be converted is originally in the source
accumulator and is transferred to the destination (a floating-point accumulator in mode 0, a memory
location otherwise) after conversion.

L -________~~_ _------~

SINGLE PRECISION
SEGMENT

DOUBLE PRECISION
SEGMENT
11- 3728

Figure 5-4

Single-to-Double Precision Appending

5.3.7 Clear Instruction
The Clear instruction causes the FPII-A to clear a floating-point number by setting all its bits to O.
5.3.8 Test Instruction
The Test instruction causes the FPII-A to test the sign and exponent of a floating-point number and
update the FPII-A status accordingly. The number tested is obtained from the destination (a floatingpoint accumulator in mode 0, a memory location otherwise). The FC and FV bits are cleared. The FN
bit is set only if the destination is negative. The FZ bit is set only if the exponent of the destination is
zero. If the FIUV status bit is set, a trap occurs (after the test instruction is executed) if a minus zero is
encountered.
5.3.9 Absolute Instruction
The Absolute instruction causes the FPII-A to take the absolute value of a floating-point number by
forcing its sign bit to O. If mode 0 is specified, the sign of the number in the floating-point destination
accumulator is forced to O. The exponent of the number is tested, and if it is 0, zeros are written into
the accumulator. If the exponent is non-zero, the accumulator is unaffected.
If mode 0 is not specified, the sign bit of the specified data word in memory is zeroed. The exponent of
this word is tested, and if it is 0, the entire data word in memory is zeroed. If the exponent is non-zero,
the integer exponent is restored to memory.
Absolute and Negate instructions are the only instructions that can read and write a memory location.

5-12

5.3.10 Negate Instruction
The Negate instruction causes the CPU (or the FPII-A, in mode 0) to complement the sign of an
operand. If mode 0 is specified, the sign of the number in the floating-point destination accumulator is
complemented. The exponent of the number is tested, and ifit is 0, zeros are written into the accumulator. If the exponent is non-zero, the accumulator is unaffected.
If mode 0 is not specified, the sign bit of the specified data word in memory is complemented. This
word is then transferred from memory to a floating-point accumulator. The exponent of this word is
tested, and if it is 0, the entire data word is zeroed and transferred back to memory. If the exponent is
non-zero, the original fraction and exponent are restored to memory.

5.3.11 Load Exponent Instruction
The Load Exponent instruction causes the floating-point processor (FPP) to load an exponent from
the source (a floating-point accumulator in mode 0, a memory location otherwise) into the exponent
field of the destination accumulator. In order to do this, the 16-bit, 2's complement exponent from the
source must be converted to an 8-bit number in excess 200 notation. This process is described further
below.
Assume that the 16-bit, 2's complement exponent is coming from memory. The possible legal range of
16-bit numbers in memory is from 000000 to 177777 8• On the other hand, the possible legal range of
exponents in the FPl1-A falls into two classes.
1.
2.

Positive exponents (0 through 177) - When 200 is added to any of these numbers, the sum
stays within the legal 8-bit exponent field (i.e., from 200 through 377).
Negative exponents (177601 through 177777) - When 200 is added to any of these numbers,
the sum stays within the legal 8-bit exponent field (i.e., from 1 through 177).

Notice that all legal positive exponents coming from memory have something in common: their 9 highorder bits are all Os. Similarly, all legal negative exponents from memory have their 9 high-order bits
equal to 1. Therefore, to detect a legal exponent, only the 9 high-order bits need be examined for all 1s
or all Os.
Any number from memory outside these ranges is illegal and will result in either an overflow or an
underflow trap condition.
Example 1: LDEXP 000034
Exponentof34
200

00000000

00011100
10000000
10011100
234

The upper 9 bits all equal 0, so this is a legal positive exponent. The number 234 is sent to the 8-bit
exponent field of the specified accumulator.
Example 2: LDEXP 201
Exponentof201
200

201
,----+-----'-,. -'---00000000
10000001
+
0
10000000
_ _ _ _ L---~

00000001

Overflow
This is an illegal positive exponent. Notice that when 200 is added to the exponent, an overflow
occurs.
5-13

Example 3: LDEXP 100200

2
Exponent of 100200

1000000

200

~L...-_---l

0 0
1 00000
10000000
00000000

Underflow
This is an illegal negative exponent. Notice that when 200 is added to the exponent, a result is
produced that is more negative than can be expressed by the 8-bit exponent field. Thus, an underflow occurs.

Example 4: Special Case - Exponent of 0: LDEXP 177600
Exponent of 177600

11111111

10000000
10000000

00000000

This is the one case where the 9 high-order bits are all equal, but the exponent is illegal. This is
because 177600 represents an exponent of O. This exponent causes an underflow condition to
exist; that is, it is treated as an illegal negative exponent.

5.3.12 Load Convert Integer-to-Floating Instruction
The Load Convert Integer instruction takes a 2's complement integer from memory and converts it to
a floating-point number in sign and magnitude format. If short-integer mode is specified, the number
from memory is 16 bits and is converted to a 24-bit fraction (single-precision) or a 56-bit fraction
(double-precision), depending on whether floating or double mode is specified. If long-integer mode is
specified, the number from memory is 32 bits and is converted to a single- or double-precision number,
depending on whether floating or double mode is specified. The integer is loaded into bits 55-40 if
short integer is specified or into bits 55-24 if long integer is specified. It is then left-shifted eight places
so that bit 55 is transferred to bit 63 (Figure 5-5).

63 62 6160 59 58 57 56 55 54 53 52 51 50 49 48 47 46 45 44 43 42 41 40
10101010101010101'1'1'111110101'11111'11101010111

11·5307

Figure 5-5

Integer Left-Shift Example

The integer is then assigned an exponent of 2178 short integer. This is the result of adding 2008 (since
the exponent is expressed in excess 200 notation) to 178 which represents 15\0 shifts. This number of
shifts is the maximum number required to normalize a number. If long-integer mode is specified, the
integer is assigned an exponent of 2378 which represents 31 \0 shifts.

5-14

The 2's complement integer is tested by examination of bit 63 to see if it is a positive or negative
number. The number is then normalized by left-shifting until bit 63 becomes a 1. If bit 63 is 1 (negative
number), the integer is negative, the sign bit is set, the number is 2's complemented, and then normalized.
To normalize a number, bit 63 (MSB) of the fraction must be equal to a and bit 62 must be made equal
to 1. To do this, the integer is shifted the required number of places to the left and the exponent value is
decreased by the number of places shifted (Figure 5-6).
Shift integer 15 places to the left to normalize.
Bit 59 = 0, bit 58 = 1
Decrease exponent by 1510 which is 178.

EXP= 2178
- 17 8
200 8

When loading a long integer with an FD = 0, if the long integer contains more than 24 significant
digits, then less significant digits will be truncated with some loss of accuracy.

115308

Figure 5-6

Normalized Integer Example

5.3.13 Store Exponent Instruction
The Store Exponent (STEXP) instruction causes the CPU to access a floating-point number in the
FPII-A, extract the 8-bit exponent field from this number, and subtract a constant of 200 (since the
exponent is expressed in excess 200 notation). The exponent is then stored in the destination as a 16bit, 2's complement, right-justified number with the sign of the exponent (bit 07) extended through the
8 high-order bits.
The legal range of exponents is from a to 377, expressed in excess 200 notation. This means that the
number stored ranges from -200 to 177 after the constant of 200 has been subtracted. The subtraction
of 200 is accomplished by taking the 2's complement of 200 and adding it to the exponent field (Figures
5-7 and 5-8).

II
15

FLOATING POINT

NUMBER IN '.11-'

EXPONENT (8 BITS)
0
0
0

FRACTION

SIGN EXTENSION
EXPONENT
TRANSFERRED
TO MEMORY

0
0
0
0
0
0
0
00
\
0
0
0
0
__________________________
________________________

(ORACCUMULATOR)~

~~,

7 \

~ BIT 7 IS EXTENDED TO
THE 8 HIGH ORDER BITS.
MA-1433

Figure 5-7

Store Exponent Example No. I

5-15

FLOATING.POINT
NUMBER IN FP11-A

EXPONENT (8 BITS)
0
0
0

FRACTION

SIGN EXTENSION

EXPONENT
TRANSFERRED
TO MEMORY
(OR ACCUMULATOR)

1~1
15

8~6

BIT 7 IS EXTENDED TO
THE 8 HIGH ORDER BITS.
MA-1430

Figure 5-8

Store Exponent Example No.2

Two examples that illustrate the process follow: one using an exponent greater than 200 and the next
using an exponent less than 200.
Example 1: Exponent = 207
Exponentof207
2's Complement of 200

10000111
+ 10000000

Result = 7

/00000!.!l
Sign Bit

Example 2: Exponent = 42
00100010
+10000000

Exponentof42
2's Complement of 200
Result = -42

/1~0~
Sign Bit

5.3.14 Store Convert Floating-to-Integer Instruction
The Store Convert Floating-to-Integer instruction causes the FPP to take a floating-point number and
convert it to an integer for transfer to a destination.

The four classes of this instruction are as follows.
1.
2.
3.
4.

STCFI - Convert single-precision, 24-bit fraction to a 16-bit integer (short-integer mode).
STCFL - Convert single-precision, 24-bit fraction to a 32-bit integer (long-integer mode).
STCDI - Convert double-precision, 56-bit fraction to a 16-bit integer (short-integer mode).
STCDL - Convert double-precision, 56-bit fraction to a 32-bit integer (long-integer mode).

5-16

The (normalized) floating-point number to be converted is transferred to the FPP. The FPP works
with the sign bit and one of the following.
1.
2.
3.

The 15 MSBs of the fraction for Floating-to-Integer and Double-to-Floating conversion
The 31 MSBs of the fraction for Double-to-Long conversion
The entire fraction for Floating-to-Long conversion.

The FPP subtracts 201 from the exponent to determine if the floating-point number is a fraction. If the
result of the subtraction is negative, the exponent is less than 201, and the absolute value of the
floating-point number is less than 1. When converted to an integer, the value of this number is 0; a
conversion error occurs, the FZ bit is set, and Os are sent to the destination. If the result of the
subtraction is positive ( or zero), it indicates that the exponent is greater than (or equal to) 201, and the
floating-point number can be converted to a non-zero integer (Figure 4-9).

I I

I 0 I 0 I 0 I 0 I0 I I

BEFORE I
SHIFTING

1 0 0I 0 0 0

AFTER
SHIFTINGI
PLACES

t0 I 1 1 01 0 1 0 01 0 1 0

o 1 o 1 o 1 o 11

10 10

"----..,----J
4

MSB
MA-0287

Figure 5-9 Store Convert Integer Example

A second test is made by the FPP to determine if the floating-point number to be converted is within
the range of numbers that can be represented by a 16-bit integer (I-format) or 32-bit integer (Lformat).
Consider the range of integers that can be represented in I and L formats and their floating-point
equivalents.
I-Format
(16 bits)

Floating-Point
Equivalent

L-Format
(32 bits)

Floating-Point
Equivalent

Most Positive
Integer

077777

+.IIII...X215

17777777777

+ .1111... X 231

Least Positive
Integer

000001

+.IOO ... X 21

00000000001

+.IOO ... X 2'

Least Negative
Integer

177777

-.IIII...X216

37777777777

-.llll...X 232

Most Negative
Integer

100000

-.IOOO ... X 2 16

20000000000

-.IOO ... X 232

NOTE
MSB of integer = sign of integer.
5-17

Thus, the exponent of a positive floating-point number to be converted must be less than 1610 (220 in
excess 200 notation) to convert to I-format or 3210 (240 in excess 200 notation) to convert to L-format.
The exponent of a negative number to be converted must be less than or equal to 1610 or 3210 to
convert to 1- or L-formats, respectively.
The FPP tests whether the floating-point number to be converted is within the range of integers that
can be represented in I-or L-format by subtracting a constant of 208 (for short integers) or 40s (for long
integers) from the result of the first test (result of first test = biased exponent - 2018 = unbiased
exponent - 1). If the result of the subtraction is positive or 0, it indicates that the floating-point number
is too large to be represented as an integer. In that case, a conversion error occurs and Os are sent to the
destination. If the result of the subtraction is a negative number other than -I, the floating-point
number can be represented as an integer without causing an overflow condition. If the result of the
subtraction is -1, the exponent of the floating-point number is either 220 (short) or 240 (long), and
converson proceeds. However, the floating-point number is within range only ifits sign is negative and
its fraction is .100 ... (i.e., ifit is the most negative integer; see table above). If, in this case, the number
is not the most negative integer, it will be detected by a third conversion error test (see below) after
conversIOn.
To convert the fraction to an integer, the FPP shifts it right a number of places as specified by the
following algorithms.
Short integer:

No. of right shifts = 208 + 2018 - biased exponent - 1

Long integer:

No. ofright shifts = 40 8 + 2018 - biased exponen t - 1

Regardless of the condition of the FT bit, the fractional part of the number is always truncated during
this shifting process.
If the floating-point number is positive, the integer conversion is complete after shifting, and the
number is transferred to the appropriate destination. If, however, the floating-point number is negative, the integer must be 2's complemented before being sent to its destination.
After conversion, the FPP performs a third test for a conversion error by comparing the MSB of the
(converted) integer with the sign bit of the original (uncon verted) n urn ber. I f these signs are not eq ual,
there has been a conversion error and the FPP traps if the FIe bit is set. This test is performed to detect
a floating-point number with an exponent of 220 (short) or 240 (long) that has not been converted to
the most negative integer.

Example 1: Store Convert Floating-to-Integer (STCFI)
Exponent = 203
Sign = 0
Fraction (24 bits) = .100000000000000000000000
15 MS Bs 0 f fraction = .100000000000000
203 (excess 200) = 2
Fraction = 1/2
Integer to be stored = 1/2 X 2 = 4

Test 1: Is the number to be converted a fraction?
Exponent:
No

203 8
-201

Since this result is positive, the given floating-point
number is not a fraction and conversion may proceed without error.
5-18

Test 2: Is the floating-point number to be converted within range? (We are working with a
positive short integer.)
Result of Test 1:

2
-20

Yes

-16

Indicates that the number to be converted is within
range and can be represented as a 16-bit integer.
No conversion error occurs.

How many right shifts? Use algorithm:
20 8 + 2018 - 2038 - 1 = 208 - 38 = 158 = 1310

= 13 right shifts
This example involves a positive number, so conversion is complete after 13 right shifts. If
the number had been negative, the integer would have been 2's complemented.
3.

Test 3: The MSB of the converted integer and the sign bit of the original floating-point
number are compared. Since they are equal, no conversion error occurs.

Example 2: Store Convert Floating-to-Integer (STCDL)
Exponent = 2408
Sign = 0
31 MSBs of fraction = .1000000000000000000000000000000
1.

Test 1: Is the number to be converted a fraction?
Exponent:

240 8
-201

Since this result is positive, the given floating-point
number is not a fraction, and conversion may proceed (i.e., no conversion error occurs).

37 8

Test 2: Is the floating-point number to be converted within range? (We are working with a
positive long integer.)
Result of Test 1:

37
-40
-1

We know the number is out of range by examining
the sign bit (in fact, this number is one greater than
the most positive integer that can be represented).
However, the FPP does not know this yet, and conversion proceeds without error at this point.

How many right shifts? Use algorithm:
40 8 + 2018 - 2408 - 1 = 0

= No right shifts
Converted 32-bit integer = 200000000008
Since the number is positive, conversion is now complete (i.e., no need for 2's complementing).
5-19

Test 3: The most significant bit of the converted integer (which is 1) and the sign bit of the
original floating-point number (which is 0) are compared. Since they are not equal, a conversion error occurs, which we predicted in Step 2.

5.3.15 Load FPll's Program Status
This instruction causes the FPP to transfer 16 bits from the location specified by the source to the
floating-point status (FPS) register. These 16 bits contain status information for use by the FPII-A in
order to enable and disable interrupts, set and clear mode bits, and set condition codes (Paragraph
4.4).
5.3.16 Store FPII 's Program Status
This instruction causes the FPP to transfer the 16 bits of the FPS register to the specified destination.
5.3.17 Store FPll's Status
The Store FPll 's Status (STST) instruction causes the FPP to read the contents of the floating exception code (FEC) and floating exception address (FEA) registers when a floating-point exception (error)
occurs.
If mode 0 addressing is enabled, only the FEC is sent to the destination accumulator. If mode 0
addressing is not enabled, the FEC is stored in memory followed by the FEA. In memory, the FEA
data occupies all 16 bits of its memory location, while the FEC data occupies only the lower 4 bits of its
location.
When an error occurs and the interrupt trap in the CPU is enabled, the CPU traps to interrupt vector
244 and issues the STST instruction to determine the type of error.
NOTE
The STST instruction should be used only after an
error has occurred, since in all other cases the instruction contains irrelevant data or contains the
conditions that occurred after the last error.
5.3.18 Copy Floating Condition Codes
The Copy Floating Condition Codes (CFCC) instruction causes the FPP to copy the four floating
condition codes (FC, FZ, FY, FN) into the CPU condition codes (C, Z, Y, N).
5.3.19 Set Floating Mode
The Set Floating Mode (SETF) instruction causes the FPP to clear the FD bit (bit 07 of the FPS
register) and indicate single-precision operation.
5.3.20 Set Double Mode
The Set Double Mode (SETD) instruction causes the FPP to set the FD bit (bit 07 of the FPS register)
and indicate double-precision operation.
5.3.21 Set Integer Mode
The Set Integer Mode (SETI) instruction causes the FPP to clear the I L bit (bit 06 of the FPS) and
indicate that short-integer mode (16 bits) is specified.
5.3.22 Set Long-Integer Mode
The Set Long-Integer Mode (SETL) instruction causes the FPP to set the IL bit (bit 06 of the FPS) and
indicate that long-integer mode (32 bits) is specified.

5-20

5.4 FPll-A PROGRAMMING EXAMPLES
This paragraph contains two prcgramming examples using the FPII-A instruction set. In example 1,
A is added to B, D is subtracted from C, the quantity (A + B) is mUltiplied by (C - D), the product of
this multiplication is divided by X, and the result is stored. Example 2 calculates DX3 + CX2 + BX +
A, which involves a 3-pass loop.
Example 1: [(A + B) * (C - Dj]/X
SET F
LDF
ADDF
LDF
SUBF
MULF
DIVF
STF

A,ACO
B,ACO
C,ACI
D,ACI
AC1,ACO
X,ACO
ACO,Y

;LOAD ACO FROM A
;ACO HAS (A + B)
;LOAD AC 1 FROM C
;ACI HAS (C - D)
;ACO HAS (A + D)*(C - D)
;ACO HAS (A + D)*(C - D)/X
;STORE (A + D)*(C - D)/X IN Y

Example 2: DX3 + CX2 + BX + A
Loop 2
r , - - - -A '---~~

ACO = [(D * X + C) * X + B] * X + A
Loop I
~,-----------y,---------~I

Loop 3
ACO = [DX2 + CX + B] * X + A
ACO = DX3 + CX2 + BX + A

LOOP;

SETF
MOV #3,%0
MOV#D+4,%1
LDF (6)+,ACl
CLRF ACO
ADDF -(4),ACO
MULF ACI,ACO
SOB%O,LOOP
ADDF -(4),ACO
STF ACO,-(6)

5-21

;SET UP LOOP COUNTER
;SET UP POINTER TO COEFFICIENTS
;POP X FROM STACK
;CLEAR OUT ACO
;ADD NEXT COEFFICIENT
;TO PARTIAL RESULT
;MULTIPLY PARTIAL RESULT BY X
;DO LOOP 3 TIMES
;ADD X TO GET RESULT
;PUSH RESULT ON STACK

CHAPTER 6
PROCESSOR ORGANIZATION

6.1 INTRODUCTION
The FPII-A Floating-Point Processor connects to the KDII-EA central processor via a tri-state bus
(Figure 6-1). This interface allows addressing of floating-point memory utilizing the memory management option.

SERV BR PFAIL

BR PF PEND

'MPC (10)

TRI STATE AMUX
KD11-EA
M8265
M8266

FLOATING
POINT
MODULE
MB267

LOAD IR

PROC CLOCK

PRoe INIT

(16)

AMUXO-AMUX15

TRISTATE BUS

UNIBUS

11-5250

Figure 6-1

KD1I-EA/FPII-A Data Flow

The CPU software must initiate floating-point operation and originate addresses and data since control of the FPII-A resides in the CPU.

6-1

The FPII-A depends on the CPU to fetch instructions and data from memory in order to initiate
floating-point operations. If the instruction is not a floating-point instruction, it is ignored by the
FPII-A. If the decoded instruction is a floating-point instruction (i.e., contains an op code of
17XXXX), the FPII-A causes the CPU to branch to the FPII-A ROM (read-only memory) microstates associated with floating-point instructions.
The simplified block diagram illustrated in Figure 6-2 shows the major functions of the FPII-A.

FP11-A

KD11-EA

FP11-A
AM2901

BRANCH
(BUT)
FP11-A
CONTROL
- BRANCH
-CONSTANTS
- BYTE
- SECTOR
- SHIFT
- CLOCKS
-A.BPORTMUX

TRISTATE

16-BIT
TRISTATE BUFFER

11-5248

Figure 6-2

Simplified FPII-A Block Diagram

6.2 MICROPROCESSOR DESCRIPTION
The principal data manipulation element in the floating-point processor is the AM2901 microprocessor (Figure 6-3). The basic microprocessor is 4 bits wide and 16 of these units are cascaded to
make up a 16 by 64-bit word for the FPII-A. A general discussion of the microprocessor followed by a
description of its integration into the overall floating-point processor is given in the following paragraphs.

6-2

2i-

RAM 0'>.
LOIR 1

RAM3

" ROlLI

-~.
MUX

AWORD
ADORES
S

c>::~

MUX

16 BIT BY 4 BIT 2-PORT RAM

RAM
EN

Ao Al A2 A3

Bo Bl B2 B3

r-<

B)
--<
--< B,

B WORD

ADDRESS

MUX

TTl
III
CLOCK
CP

E A LATCH
Ao Al Az A3

E B LATCH

8081 8z 83

D,.I

'--

I.
I.

.
Cnr

ALU
FUNCTION
DECODE

2-IN
MUX

L....-.

THREE
STATE
CONTRO L

2-IN
MUX

A
y,

3-IN
MUX

2-IN
MUX

3-IN
MUX

I
I

I
I
I

7,;'1
MUX

MUX

I
II

~"O

IO/CI

3-IN
MUX

MUX

I::

1""-

ALU
SOURCE
OPERAND
DECODE

ROlLI

1;;.-'
MUX

ALU
DESTINATION
DECODE

Q EN

S
I.

"
I.

~IO

~ G

I---< P

ARITHMETIC LOGIC UNIT (ALU)

Q REGISTER

O,z/

I
I

C'(
2-IN
MUX

I'
DIRECT
DATA
S
INPUT

3i-i,i, Fj ,i, Fj ~i" Fj f

Qo
"LO/RI

'---< Bl

l
I

I "

~ Cn +4

I---< OVR

YF'

y.
11-5267

Figure 6-3

Microprocessor (AM2901)
Block Diagram

6-3

6.2.1 Microprocessor Organization
As shown in Figure 6-3, the major components of the microprocessor are the RAM, the arithmetic
logic unit (ALU), and the Q-register.
Information contained in any of the 16 64-bit words of the RAM may be read from the A-port as
controlled by the 4-bit A-word address (Ao-A3) field. Similarly, data in any of the 16 words of the
RAM as defined by the B-address (Bo-B3) field input may be simultaneously read from the B-port of
the RAM. It is also possible to apply the same address code to both the A and B select fields, in which
case the identical file data will appear at both the RAM A-port and B-port simultaneously.
New data is always written into the file word specified by the B-address field of the RAM. The RAM
data input field is driven by a 3-input mUltiplexer which permits shifting of the ALU output. The 3input multiplexer allows data to be shifted right 1 bit position, left 1 bit position, or not shifted in
either direction.

6.2.2 Arithmetic/Logical Operations
The arithmetic logic unit (ALU) is capable of performing three arithmetic and five logical operations
on the two 4-bit input words Ro-R3 and So-S3. The R-input field to the ALU receives its input from a
2-input multiplexer while S receives its signals from a 3-input multiplexer. The 2- and 3-input multiplexers both have an inhibit capability. This is the equivalent of an "0" source operand.
If the five data inputs to the ALU are combined into pairs, 10 combinations of registers are possible,
i.e., AB, DA, AQ, OA, OB, BQ, BD, DO, DQ, and OQ, as illustrated in Table 6-1. The microprocessor
uses eight of these operand pairs. Selection of the ALU source operand pairs is accomplished by the
microinstruction inputs 10, 11, and 12.

Table 6-1

ALU Source Operand Contest
ALU Source
Operands

Microcode

L
L
L
L
H
H
H
H

L
L
H
H
L
L
H
H

L
H
L
H
L
H
L
H

Octal
Code

a
I

-'"l
3
4
5
6
7

A
A
0
0
0
0
0
0

Q
B
Q
B
A
A
Q

The direct-data (D) source-operand input port is used to insert all data into the working registers
inside the 2901 microprocessor. The D-input can also be used by the ALU to modify any of the
internal data files of the RAM via the F outputs of the ALU.
The Q-register is a separate 4-bit register intended primarily for multiplication and division routines.
This register can also be used as an accumulator or buffer register for certain applications.

6-5

The ALU performs three arithmetic and five logical functions as directed by the three control bits 13,
14, and 15 (Table 6-2).

Table 6-2 ALU Function Control
Microcode

1.1

L
L
L
L
H
H
H
H

L
L
H
H
L
L
H
H

L
H
L
H
L
H
L
H

Octal
Code
0
I

3
4
5
6
7

ALU
Function

Symbol

R Plus S
S Minus R
R Minus S
R OR S
-RAND S
RANDS
R EX OR S
R EX NOR S

R + S
S - R
R - S
RVS
R/\S
R/\S
RVS
RVS

ALU output data may be routed to one of eight possible destinations as defined by control bits 16, 17,
and 18. ALU output data may be a data output from the device or it may be stored in the RAM or the
Q- register.
The data output of the microprocessor uses a 2: I multiplexer whose inputs are the A-port of the RAM
or the ALU outputs (F). Selection of these outputs is controlled by bits 16, 17, and 18 of the microinstruction control input (Table 6-3). Note that the left- and right-shift functions in Table 6-3 are reversed for the F P ll-A application.
The FPII-A uses 16 AM2901 units connected in cascade with 3 levels of look-ahead carry logic. This
configuration results in a 64-bit word. Carry generate (G), and carry propagate (P), are unit outputs
for use with a look-ahead carry generator. Carry out (C n +4) is also generated by the microprocessor
and is available as an output carry flag in a status register. C n and C n +4 are both active high.
Three additional outputs are generated by the ALU. These are F3, F = 0, and overflow (OVR). F3
represents the most significant bit (sign) of the ALU and can be used to determine positive or negative
results without enabling the 3-state outputs or while enabling the A-port to output. F3 is non-inverted
with respect to the sign bit output Y3. F = 0 output is used for zero detect and is an open-collector
output that can be wire ORed between microprocessor slices. F = 0 is high when all F outputs are low.
OVR is a flag indicating an arithmetic operation exceeds the available range and is high when the
overflow condition exists, i.e., when C n and C n +4 are not of the same polarity.
Inputs to the RAM are via a 3-input mUltiplexer. The multiplexer allows input data to be entered into
the RAM in three modes:
• Shifted left one place
• Shifted right one place
• Unshifted.

6-6

The shifting is accomplished by two ports: RAM-LOjRI and RAM-ROjLI. Both ports consist of a
buffer driver with a tri-state output and an input to the multiplexer. In the shift-up (X2) mode, the RO
buffer is enabled and the RI multiplexer input is enabled. In the shift-down (-2) mode, the LO buffer
and LI input are enabled. In the no-shift mode, both the LO and RO buffers are in the high-impedance
state and the multiplexer inputs are not selected. The microinstruction control bits 16 , 17, and 18 operate the shifter as shown in Table 6-3.

Table 6-3
RAM
Function

Microcode

ALU Destination Control
Q-Register
Function
y

RAM
Shifter

Q
Shifter

RAMo RAM3
LO/RI L1/RO

Qo
Q3
LO/RI L1/RO

Octal
Code

Shift

Load

Shift

Load

Output

None

ALU
(F)

None

ALU
(F j )

None

ALU
(F j )

Left
(Down)

ALU
(F j + I )

Left
(Down)

Q-Reg

IN]

IN~

(Oi+ I )

Left
(Down)

ALU
(F i + I )

IN3

Right
(Up)

ALU
(F i _ l )

Right
(Up)

Q-Reg

INo

Right
(Up)

ALU
(F j _ , )

INo

(OJ--l)

X = Don't care_ Electrically. the shift pin is a TTL input internally connected to a three state output which
is in the high impedence state_

The Q-register is also driven from a 3-input multiplexer. In the no-shift mode, the multiplexer enters
ALU data into the Q-register. Operation for the shift-up or shift-down modes is the same as for the
RAM as indicated in Table 6-3.

6-7

The RAM, Q-register, and the A and B data latches are controlled by the clock input. When enabled,
data latches are also controlled by the clock input and data is clocked into the Q-register on the
positive-going transition of the clock. When the clock input is high, the A and B data latches are open
and will pass any data that is present at the RAM outputs. When the clock is low, the latches are closed
and will retain the last data entered. If the RAM-EN is enabled, new data is entered into the RAM file
(word) defined by the B-address field when the clock input is low.
6.2.3 RAM
The FPll-A RAM register usage is shown in Figure 6-4. This unit, located in the microprocessor, is
the scratch pad area where the results of arithmetic and logical operations are temporarily stored. The
contents of the RAM are read into the AL U under control of the FPll-A microcode. It consists of 16
64-bit words (each of the 16 microprocessors (AM2901) contains a 16- X 4-bit RAM).

FPS

17
16

FCCR

FEC

ZERO

EFSRC

ZERO

EAC

FSRC

AC5

AC4

AC3

AC2

AC1

ACO

7
6

SECTOR 3
BYTE 6

SECTOR 2
BYTE 5

SECTOR 1

BYTE 4

BYTE 3

BYTE 2

F-REG

SECTOR 0

SECTOR 3

BYTE 7

BYTE 1

BYTE 0

E-REG

X-REG

11-5300

Figure 6-4

RAM Register Usage

6-8

Six of the 64-bit registers are allocated for the accumulators and are accessible to the programmer via
the FPII-A instruction register. Registers 6 and 7 are unused while registers 10-17 are set aside for
special functions. Registers 10-17 are accessed only by the control ROM. Registers 10-14 constitute a
working storage area for the FPII-A microcode. Other functions included are the floating-point status
register, condition codes, and exception codes.

6.2.4 Arithmetic Logic Unit (ALU)
The ALU is the data path component that actually performs the arithmetic/logical operation under
command of the microcode (Table 6-4). R-inputs are fed in via a 2-input mUltiplexer whose inputs are
the direct data (0) inputs and the output of the A-port of the RAM. The S-inputs include the A- and
B-ports of the RAM and the Q-register outputs.

Table 6-4 Source Operand and ALU Function Matrix
0

A,Q

A,B

O,Q

O,B

O,A

D,A

O,Q

0,0

C'n=L
R Plus S
Cn=H

A+Q

A+B

O+A

D+Q

A+Q+I

A+B+I

Q+I

B+I

A+I

D+A+I

D+Q+I

D+I

Cn=L
S Minus R
C'n=H

Q-A-I

B-A-I

Q-I

B I

A-D-I

Q-D-I

-D-I

Q-A

B-A

A-D

Q-D

-D

Cn=L
R Minus S
Cn=H

A-Q-I

A-B-l

-Q-I

-B-1

-A-l

D-A-l

D-Q-l

D-l

A-Q

A-B

-Q

-A

D-A

D-Q

R OR S

AVQ

AVB

D-A

DVQ

RAND S

A!\Q

A!\B

D!\A

D!\Q

RAND S

A!\Q

A!\B

D!\A

D!\Q

REX-OR S

AVQ

AV'B

DVA

DVQ

REX-NOR S

AVQ

DVQ

12 I 0 Octal
ALU
Source
Is 4 3 Octal
ALU
Function

AVB

+ = Plus: - = Minus: V= OR,!\ = AND: V = EX OR

6-9

-DVA

ALU output data (F) may be routed to the Q-register or RAM, or may be multiplexed with the A-port
output data from the RAM as Yo-Y 3. The ALU function decode determines the arithmetic or logical
function to be performed, while the ALU destination decode determines which of the indicated registers the data is routed to or whether it will be a data output of the device itself.
The ALU source operand decode performs the actual register selection. All three of these functions are
controlled by bits 10-18 of the control word.
6.2.5 Q-Register
The Q-register is used primarily during mUltiply and divide operations to store multiplier or product
operators. Its contents may be shifted left or right or remain unshifted and the register may route data
to the ALU or receive input from that device.
6.2.6 Source Operands and ALU Functions
This paragraph summarizes the arithmetic and logic functions performed by the ALU and presents
ALU logic and arithmetic functions in separate tabulations.
6.2.6.1 Logical and Arithmetic Functions - The ALU performs five logical and three arithmetic functions on eight source operand pairs. ALU logic functions and appropriate control bit values (10-15) are
shown in Table 6-5. The carry input (C n ), has no effect in logic mode but does affect operations in
arithmetic mode (Table 6-6). Both carry-in LOW (C n = 0) and carry-in HIGH (C n = I) are defined.
6.2.6.2 Logical Functions for G, P, Cn +4 , and OVR - The four signals, G, P, C n +4, and OVR, as
described in Paragraph 6.2 are designed to indicate carry and overflow conditions when the microprocessor is in the add or subtract mode. Table 6-7 indicates the logic equations for these four signals
for each of the eight ALU functions. The R- and S-inputs are the two inputs selected according to
Table 6-8.
6.2.7 Summary of Pin Definitions
The AM2901 pin definitions are summarized in Table 6-7. Pin assignments for the AM2901-40-pin
dual in-line package are shown in Figure 6-5.

6.3 INSTRUCTION STATUS REGISTERS AND DECODE
The FPII-A contains a 12-bit instruction register and two flip-flop's that are bits of the status registers
(FD;FL). The possible FPII-A instruction formats are presented in Chapter 2. One bit of the status
register (FD) specifies double- or single-precision format and the other (FL) designates single- or
double-precision integer format. The outputs of these registers are fed to the floating-point instruction
decode register which consists of two 512- X 4-bit ROMS. The ROM outputs then generate microprocessor control (MPC) outputs to control the microprogram.
6.4 TRI-STATE TRANSCEIVERS AND BUFFER
The 8097 tri-state status gales are arranged as tri-state transceivers to communicate between the
AMUX bus (KDII-EA) and the T-bus (FPII-A), which are 16-bit tri-state buses.
74S173s, which are tri-state flip-flops, are used to buffer Unibus data being passed to the AM2901s.
6.5 BRANCH LOGIC AND TRI-ST ATE CONTROL
This logic is controlled by condition code and T-bus branch ROMs in the FPII-A. Branch (BUT) bits
are routed to the ROMs whose outputs condition a group of gates. These gates are enabled by the
various branch conditions that may arise during floating-point operations and direct (point) the microprogram counter to the appropriate code in the microprogram to service the branch function. The
BUT conditions include the condition codes, exponent negative, exponent zero, carry, overflow, and
T-bus bits 0, 1,5,6,8,9, to, II, and 14. These functions include items such as fraction Z-bit, fraction
negative, bus request, and the like.

6-tO

Table 6-5

ALU Logic Mode Functions

Octal
1543,1210

40
41
45
46
30
31
3 ::;
36
60
61
65
66
70
71
75
76
72
73
74
77
62
63
64
67
32
33
34
37
42
43
44
47
50
51
55
56

Group

Function

AI\Q
AI\B
AND

DI\A
DI\Q
AVQ
AVB
OVA
DVQ

EXOR

EX NOR

AVQ
AVB
DVA
DVQ
AVQ
AVB
DVA
DVQ

Q
Invert

Pass

B
A
D
Q
B
A

Pass

Q
B
A

0
0
"Zero"

0
0
0
AI\Q
~!\B

Mask

6-11

D!\A
D!\Q

Table 6-6 ALU Arithmetic Mode Functions

Cn = 0 (Low)

Octal
Is 43, 12 I 0

Group

ADD

05
06
02
03
04
07
I 2
I 3
I4

PASS

Decrement

-'1 !-

23
24
I 7
I0
I I
I 5
I 6

I 's Compo

Subtract
(I 's Comp.)

20
2I
25
26

Cn = 1 (High)

Function
A+Q
A+B
D+A
D+Q
Q
B
A
0

Group

Function

ADD plus
one

A+Q+I
A+B+I
D+A+I
D+Q+I

Increment

Q-I
B-1
A-I
0-1
-Q-I
-B-1
-A-I
-0-1
Q-A-I
B-A-I
A-D-I
Q-D-I
A-Q-I
A-B-I
D-A-I
D-Q-I

Pass

2's Compo
(Negate)

Subtract
(2's Comp.)

Table 6-7 Logic Equations for ALU Functions
Definitions (+ =OR)
Go = RoSo
G I = R)S)
G 2 = R2 S2

Po = Ro + So
PI = RI + SI
P 2 = R2 + S2
P 3 = R3 + S3

G 3 = R 3 S3

6-12

Q+I
B+I
A+I
D+I
Q
B
A
0
-Q
-B
-A
-0
Q-A
B-A
A-D
Q-D
A-Q
A-B
D-A
D-Q

Table 6-8 P, G, C N +4' OVR Functions

Is 4 3

01
I

Cn + 4

OVR

P1 P2 P ,P O

G 1 + P 1G 2 + P1 P 2 G, + P 1 P 2 P,G O

C 1 C4

Function

R+S

S - R

Same as R + S equations, but substitute R, for R, in definitions

R - S

SaIlle as R + S equations, but substitute S, for S, in definitions

•

RV S

Low

P1 P 2 P,P O

P 3 P 2 P,P O +C n

P 1 P 2 P,P O +C n

R /\ S

Low

G 1 +G 2 +G,+G O

G 3 +G 2 +G, +G o +C n

G J +G 2 +G, +G o +C n

R AS

Low

R \f S

RV S

+ = OR

G J +G 2 +G,+G O

Same as RA S equations, but substitute R, for R, in lkfinitions

•

Same as R \f S, but substitute R, for R, in definitions

•

p}G J +P}P 2 G 2 +P 3 P 2 P,G, +p}P 2 P,P O

P 3 G 3 +P1 P 2 G 2 +P1 P 2 P,G,
+P1 P 2 P S ,P O (Go +C n )

Compll'Illent of
C n +4 at Il'ft

c5E

OVR

<;'+4

GND

LO/R1

Am2901

•
2

RAM3 RAMo

VCC

ROlLI LOIRI

ROlLI

NOTE: PIN 1 IS MARHa FOR ORIENTATION.
11-5247

Figure 6-5

AM2901 Pin Connections

6.6 CONST ANTS, BYTE AND SECTOR CONTROL, SHIFT CONTROL
The constant ROMs contain the fixed-value numbers required for certain floating-point functions.
The magnitude of some of the constants depends on whether the floating-point numbers are single- or
double-precisionend short or long integer. Thus, FD and FL are used as ROM-gating signals for
proper constant selection.
The BYTE control lines enable AM2901 outputs onto the T-bus by 8-bit bytes. Any high-byte/lowbyte combination may be enabled.
Sector control is used to independently select one of four 16-bit sectors of the AM2901. Each sector
clock clocks four AM2901 slices (16 bits) which are used internally to load the RAM or Q-register.
Shift and rotate signals are generated to operate the input multiplexers to the RAM and Q-registers of
the AM290 1. These registers may be left and right shifted and controls are provided for injection of 1s
or Os into the bit stream as shifts are carried out.

6-14

CHAPTER 7
MAINTENANCE

7.1 INTRODUCTION
This chapter describes some of the maintenance tools and techniques available for maintenance of the
FPII-A floating-point option. Descriptions of the diagnostics, programmer's console, display features, and documentation aids are also included.
7.2 FPII-A DIAGNOSTICS
Three diagnostics are available to validate and diagnose the FPII-A. However, since the KDII-EA
data path is used extensively on floating-point instructions, CPU tests should be run prior to running
floating-point diagnostics if there is any doubt about the CPU. Successful running of CPU tests does
not rule out the possibility that a KDII-EA failure may cause only floating-point instructions to fail.
The three FPII-A diagnostics are listed below with a short description of each. The diagnostics should
be run in the same order as they are listed because succeeding diagnostics have been run successfully.
Otherwise, faulty diagnosis of the failed micro-step and where the problem is located may result.
7.2.1 MAINDEC DFFPAA
This diagnostic tests the following floating-point instructions.
LDFPS
STFPS
CFCC
SETF, SETD, SETI, and SETL
STST
LDF and LDD (all source modes)
STD (mode 0 and I)
ADDF, ADDD, and SUBD (most conditions)
7.2.2 MAINDEC DFFPBA
This diagnostic tests the following floating-point instructions.
ADDF, ADDD, and SUBD (all conditions not listed in DFFPAA)
CMPD and CMPF
DIVD and DIVF
MULD and MULF
MODD and MODF
This diagnostic also makes use of a special testing module (M8267-TA), which allows the diagnostic to
check the ability of the floating-point to abort an ADD, SUB, MVC, DIV, or MOD instruction if an
interrupt request occurs during the initial portion of one of these instructions. The extra hardware
tested using the special test module is minimal and it is expected to be used only during manufacturing
for more complete testing. The diagnostic automatically checks for the test module, and only if present, performs the special instruction abort test. A message at the beginning of the program indicates
the presence of the test module and its use by the diagnostic. If the module is not present, no message is
generated.
7-1

7.2.3 MAINDEC DFFPCA
This diagnostic tests the following floating-point instructions.
STF and STD (all modes)
STCFD and STCDF
CLRD and CLRF
NEGF and NEGD
ABSF and ABSD
TSTF and TSTD
NEGF, ABSF, and TSTF (all source modes)
LDFBS (all source modes)
LDCIF, LDCLF, LDCID, and LDCLD
LDEXP
STFPS (all destination modes)
STCFL, STCFI, STCDL, and STCDI
STEXP
STST

7.3 KYll-LB PROGRAMMER'S CONSOLE
Normal console and maintenance features provided by the KYII-LB programmer's console to debug
and diagnose the KD Il-EA processor are directly extendable in use to the FPII-A floating-point
option. These features include the normal console functions of examining and depositing into memory
and general registers, single-instruction stepping, the console maintenance features of single microinstruction stepping, and displaying MPC lines, Unibus data, and Unibus address lines.
The KYII-LB displays the additional MPC line (MPC 09 L) if the proper cable connections between
the KYII-LB and FPII-A modules are made. Thus, single micro-stepping the machine through floating-point micro-code is possible.
A change in the KD II-EA processor from the KD ll-E processor enables the AMUX lines onto the
Unibus data lines in the manual clock mode. (K YII-LB maintenance cables are attached, the console
is in MAINT mode, and the HLT ISS key has been depressed.) The AMUX to Unibus drivers are not
enabled, however, if the current micro-step is a DATI, at which time some other device (memory, I/O)
will be driving the Unibus data lines. Since the console can display the Unibus data lines (EXAM key
in MAINT mode), the AMUX lines are being indirectly displayed most of the time. This new feature is
directly extendable to the FPII-A in that the AMUX lines are the data path link between the KDllEA and the FPII-A. At any micro-step, the AMUX lines may be displayed and while running floatingpoint micro-code, the T-bus lines of the FPII-A are defaulted onto the AMUX lines. This means that
if the AMUX lines are not specifically being used in a floating-point micro-instruction, the T -bus will
be enabled onto the AMUX, allowing the T-bus to be displayed. Also, whenever the T-bus is not being
explicitly used, 2 bytes of the 64-bit data path are enabled onto the T -bus. The actual source of the data
on the AMUX lines at any micro-step may be determined from the FPII-A flow diagrams. Refer to
the K Yll-LB Programmer's Console Maintenance Manual for more information on the use and operation of the KYII-LB for maintenance. Refer to the FPII-A print set for information regarding the
proper installation of the FPII-A and KYII-B.

7.4 FPII-A FLOW DIAGRAMS
Each micro-step in the FPII-A flow diagrams denotes what will be displayed on the Unibus data lines
when the manual clock is enabled. This information is given just below the dotted line in each block.
The information may be a constant (such as 100000) or may be defined in a general way such as
Q(B7:BO), which indicates that bytes 7 and 0 of the Q-register will be displayed. Refer to Figure 7-1.

7-2

1457

8-L

F12 ~ SRl (F12)
E12 ~ ZERO
-

JUMP!8-M

D ~ ZERO: F12 (B6)

DISPLAY INFORMATION
11-5641

Figure 7-1

7.5

Display Information

EXTENDER BOARD

A special extender board (W9042) and two extender cables are included with the FPII-A module on a
hex extender module. The FPII-A print set shows the correct methods of using the W9042 extender
board and the included cables.

7-3

APPENDIX A
OPTION POWER SPECIFICATIONS

A-I

Table A·I

PDp·ll Family Models and Options Power Requirements

Model/Option

Description

H765
(115/230 Vac)

Power Supply

Regulator Units

I 5 V Regulator
(5411086)

Power line monitor

11/05-S

KDII-B
MMII-U
3 SPC
2 M930s
Total Amperes

8.0
5.4
6.0
2.5
16.6

KDII-A
KEII-F
KEII-E
KJI I-A (optional)
KTII-D
KWII-L
SPC
M981
MFII-U (16K)
M930
Total Amperes

10.5
2.0
3.0
0.5
2.5
0.5
2.0

11/35-S

+5 V(CPU)

Current Needed (Amperes)
+5 V (Options) -IS V +20 V

-5 V

+15 V

AC Line Current
(Amperes)

21.

0.25

0.05
4.4

0.51

4.4

0.51

1.25
6.1
1.25
8.6

4.4

0.5\

6.0

0.25

0.05

5.0

MFII-U/MMII-U*
(Active)
(Standby)

16K sense
core memory
(double SU)

6.1
5.4

4.4
0.56

0.51
0.41

2.2
0.8

MFI I-UP/MMI I-UP
(Active)
(Standby)

16K sense
core with parity
(double SU)

7.3
5.4

4.4
0.56

0.5\
0.41

2.3
0.8

MFII-L (MMII-L)
(Active)
(Standby)

8K core
memory
(double SU)

3.4
1.7

6.0
0.5

1.8
0.3

MFII-LP (MMII-LP)
(Active)
(Standby)

8K parity
core memory
(double SU)

4.9
1.7

6.0
0.5

2.
0.3

*Noninterleaved.
·*Refer to appropriate appendix for regulator unit output current.

A-2

Table A-I

Model/Option

Description

MMII-S

Same as MMII-L
except in SU
configuration (1 SU)

PDP-I 1/04

KDII-D
M9301
M9302

PDP-ll Family Models and Options Power Requirements (Cont)

+5 V (CPU)

Current Needed (Amperes)
+5 V (Options) -15 V +20V

-5 V

+15 V

AC Line Current
(Amperes)

Same as
MFII-L

5.0
2.0
1.2

7.0

Memory:
See individual
memory listings.

o LlI-W (optional)
M7850 (optional)
KYII-LA
KYII-LB (optional)

2.0
1.0
0.1
3.0

0.15

0.05

0.06

PDP-ll/34A
PDP-ll/34A

9.0
KDII-EA
M9301
M9302

11.5
2.0
1.2

9.0

Memory:
See individual
memory listings.
DLlI-W (optional)
M7850
KYII-LA
KYII-LB (optional)

2.0
1.0
0.1
3.0

0.15

0.05

0.06

FPIIA

M8267

7.0

MMII-CP

8K core
memory

3.0

3.5

0.2

16K core
memory

3.0

4.0

0.5

MMII-WP
(Active)
(Standby)

32K parity
core memory
(double SU)

6.1
5.5

3.4
0.6

0.74
0.64

2.1
0.8

MMII-YP
(Active)
(Standby)

32K parity
core memory

5.0
5.0

3.5
0.6

0.4
0.4

2.0
0.8

MMII-DP

A-3

Table A-I

PDP-ll Family Models and Options Power Requirements (Cont)

+5 V (CPU)

Current Needed (Amperes)
+5 V (Options) -IS V +20 V

Model/Option

Description

MSII-EP

4KMOS
MUD memory

1.5 (+5)
0.5 (+5B)***

0.1

0.34

MSII-FP

8KMOS
MUD memory

1.5 (+5)
0.5 (+5B)***

0.1

0.36

MSII-IP

16K MOS
MUD memory

1.5 (+5)
0.5 (+5B)***

0.1

0.4

M7850

Parity control for
MUD memories

1.0

-5 V

+15 V

AC Line Current
(Amperes)

***Current from +5 Vb rail if Battery Backup Option is used. If there is no Battery Backup Option. then 2.0 A is drawn from +5 V.

Table A-2 PDP-ll Family Options Power Requirements
Current Needed (Amperes)*

AC Line
Current
(Amperes)

Option

Mounting Code

Description

Power
Harness

AAII-D

I SU

D/A converter subsystem

7009562

3.0

0.3

AR-II

SPC

ADC and DACs

N/A

5.0

0.5

BA614

(AAII-D)

D/A converter

3.0

0.3

BM792-Y

SPC

Bootstrap loader

0.3

CDII-A/B

ISU

1000 cpm, 80-co\.
card reader controller

7010117

2.5

0.25

CDII-E

I SU

1200 cpm, 80-co\.
card reader controller

7010117

2.5

0.25

CMII

SPC

200 cpm, 80-co\.
card reader controller

1.5

0.15

+5 V

*+20 V not used in this configuration.

A-4

-IS V

-5 V

+15 V

Table A-2

PDP-ll Family Options Power Requirements (Cont)

Power
Harness

Current Needed (Amperes) *

AC Line
Current
(Ampere!

Option

Mounting Code

Description

CRII

SPC

300 cpm, 80-col.
card reader controller

1.5

0.15

DAII-DB

ISU

Unibus link

4.0

0.4

DAII-F

I SU

Unibus window

7010117

5.0

0.5

DBII-At

ISU

Bus repeater

7009562

3.2

0.31

DCII-A

I SU

Dual clock and system unit

7010117

0.2

0.02

DCII-DA

(DC I I-A)

Full duplex module set

DDII-B

I SU

Peripheral mounting panel

7010117

DHII-AA

DLB SU

Prog. async 16-line multiplexer

DHII-AD

DLB SU

Dlll-A

+5 V

-15 V

-5 V

+15 V

2.0

0.2

7010118

8.4

0.42

Modern control

7010118

10.8

0.665

0.4

1.33

I SU

Async 16-linc MUX

7010117

4.7

0.25

0.6

Dlll-AC

ISU

Async 16-line MUX

DLlI

SPC

Async interface

DMII-B

(DHII)

16-line modem control

(DHIl)

2.4

0.24

DNII-A

I SU

Auto calling system unit

7009562

2.6

2.5

DPII-D

I SU

Half/full duplex sync interface

7009562

2.56

DPII-C

(DPII-D)

Data/sync register extender

0.77

0.08

DPII-K

(DPII-D)

Internal DPII clock

0.18

0.02

0.2

0.9

1.0
1.8

0.25

.15

.016

o.v...

0.07

DQII-D
DQII-D

0.2

0.21

0.28

0.62
ISU

Full/half duplex sync interface

7010117

6.0

0.07

0.04

0.62

* +20 V not used m this configuration.
tWhen installing a DBll-A bus repeater in a BAII-K 10.5 Inch Mounting Box, the AC LO and DC LO wires must be removed from the harnesses ofal! the options
(located in the same box) after the DBII-A.

A-5

Table A-2

PDP-ll Family Options Power Requirements (Cont)

+15 V

AC line
Current
(Amperes)

0.07

0.04

0.62

0.02

0.05

Current Needed (Amperes)*

Power
Harness

+5 V

-15 V

7010117

6.0
0.4

Option

Mounting Code

DQII-E

1 SU

DFCII-A

(DU/DP CLOCK)

Level converter clock recovery

DQII-K

(DQII-D/A)

Crystal clock

DRII-B

SPC

General purpose DMA

DRII-C

1 SU

General purpose digital interface

DRII-K

SPC

Digital I/O

N/A

0.15

DUII-D

SPC

Full/half duplex

2.2

2.5

0.05

0.27

DUII-EA

SPC

Sync prog. interface

2.6

0.20

0.07

0.33

DVII

DBL SU

Sync MUX

13.5

.083

0.435

0.5

KGII-A

SPC

Comm. arith unit

1.2

0.12

KWII-L

(CPU)

Line clock

0),

0.08

KWII-P

SPC

Prog. line clock

1.0

0.1

LCII-A

SPC

LA30 control

1.5

0.15

LPII-R

SPC

1200 LPM printer

1.0

0.1

LPII-S

SPC

900 LPM printer

1.0

0.1

LPII-W

SPC

240 LPM printer

1.5

0.15

LPII-V

SPC

300 LPM printer

1.5

0.15

LSII-A

SPC

60 LPM printer

1.5

0.15

LVII-B

SPC

Electrostatic printer, 500 LPM

1.5

0.15

MRII-DB

2 SPC

Bootstrap

Description

I Full/half duplex sync interface

-5 V

0.05
7009562

0.012

3.3

0.32

1.5

0.15

*+20 V not used in this configuration.

A-6

0.6

Table A-2

PDP-ll Family Options Power Requirements (Cont)

Power
Harness

Current Needed (Amperes)*

Option

Mounting Code

Description

PCII

SPC

Papertape

PRII

SPC

Papertape (reader)

RHII

DBLSU

RKII-D

Disk and control

TAII-A

SPC

Dual cassette interface

VTII

Graphic processor

6.5

VRII-A

SPC

Pushbutton box

7010115

+5 V

-15 V

-5 V

+15 V

AC Line
Current
(Amperes)

1.5

0.15

1.9

0.19

8.0

0.8

100.

0.8
0.4

*+20Y not used in this configuration.

A-7

FPll-A FLOATING-POINT PROCESSOR
USER'S GUIDE
EK-FPIIA-UG-OOI

Reader's Comments

Your comments and suggestions will help us in our continuous effort to improve the quality and usefulness of our
publications.

What is your general reaction to this manual? In your judgment is it complete, accurate, well organized, well
wr~ten, etcJ Is ~ easy to use? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

What faults or errors have you found in the manual? ~~~~~~~~_ _ _ _ _ _~_~_ _ __

Does this manual satisfy the need you think it was intended to satisfy? ~~_ _ _ _ _ _~~_~~_
Does it satisfy your needs? ~~_~~_ _~~~~~__ Why? _______________________

Please send me the current copy of the Technical Documentation Catalog, which contains information on
the remainder of DIGITAL's technical documentation.

Name ~~~_ _ _ _ _ _ _ _ _ _ _ _ __

Street _ _ _ _ _ _~_ _~_~~~~_ _ __
City ~_~~_~~_ _ _ _ _ _ _ _ _ __
Title
State/Country _ _ _ _ _ _ _ _ _ _ _ _ _ __
Company~~~_~_~_ _ _ _~_ ____
Department ~_ _ _ _ _ _ _~~_~_~ Zip

Additional copies of this document are available from:
Digital Equipment Corporation
444 Whitney Street
Northboro, Ma 01532
Attention: Communications Services (NR2/M 15)
Customer Services Section
Order No. ____E_K
__
-F_P_l_l_A_-U_G_-O_O_I_ _ _ __

-----------~~-----------

-- -- -- -

Do Not Tear-Fold Here and Staple

FIRST CLASS
PERMIT NO. 33
MA YNARD, MASS.
BUSINESS REPLY MAIL
NO POSTAGE STAMP NECESSARY IF MAILED IN THE UNITED STATES
Postage will be paid by:

Digital Equipment Corporation
Technical Documentation Department
Maynard, Massachusetts 01754

digital equipment corporation

Printed in U .S.A .