Digital PDFs

AA-2124D-TC

October 1977

150 pages

Original

6.8MB

view download

OCR Version

6.3MB

view download

Document:	ftnExtns Oct77
Order Number:	AA-2124D-TC
Revision:	0
Pages:	150
Original Filename:	AA-2124D-TC_ftnExtns_Oct77.pdf

OCR Text

FORTRAN/RT-11
Extensions Manual
Order No. AA-2124D-TC

October 1977

This document describes the purpose and function

of the FORTRAN/RT-11 language extensions for
real-time support, and FDT, the FORTRAN debugging technique.

FORTRAN/RT-11
Extensions Manual
Order No. AA-2124D-TC

SUPERSESSION/UPDATE INFORMATION:

This document completely supersedes
the document of the same name,

Order No. DEC-11-LRTEA-C-D,
published September 1976.
OPERATING SYSTEM AND VERSION:

SOFTWARE VERSIONM:

RT-11 VO3

FORTRAN/RT-11 Extensions V02
FDT V02

To order additional copies of this document, contact the Software Distribution

Center, Digital Equipment Corporation, Maynard, Massachusetts 01754.

digital equipment corporation - maynard. massachusetts

First Printing,
Reviged:

March 1975

June
September
October

1975
1976
1977

The information in this document is subject to change without notice
and should not be construed as a commitment by Digital Equipment
Digital Equipment Corporation assumes no responsibility
Corporation.
for any errors that may appear in this document.

The software described in this document is furnished under a license
and may be used or copied only in accordance with the terms of such
license.

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

1975, 1976, 1977 by Digital Equipment Corporation

The postage prepaid READER'S COMMENTS
document requests the user's critical
paring future documentation.

The

following are

trademarks

form on the last page of this
evaluation to assist us in pre-

of Digital Equipment Corporation:

DIGITAL

DECsystem-10

MASSBUS

DEC

DECtape

OMNIBUS

PDP

DIBOL

0Ss/8

DECUS

EDUSYSTEM

PHA

UNIBUS
COMPUTER LABS

FLIP CHIP
FOCAL

RSTS
RSX

COMTEX

INDAC

TYPESET-8

DDT

LAB-8

TYPESET-10

DECCOMM

DECSYSTEM-20

TYPESET-11

CONTENTS

PREFACE

RT-11

1.1

WHAT

FORTRAN

ARE

THE

EXTENSTIONS
FORTRAN

Control

Digital

and

THINGS

1.2

EXTENSIONS?

Time-Interval
Input

and

Sampling

Output

Analog-to-Digital
Sampling
Digital-to-Analog
Graph Plotting

(A/D)

Conversion and

(D/A)

Conversion

KNOW ABOUT

RT-~11 AND

THE

FORTRAN

EXTENSIONS

Completion Routines
Considerations in Using
Routines

Software

Action

[]

W
W

Service

Routine

CTRL/C

Requirements

Acquisition

for

High-Volume

CONVENTIONS
Syntax Conventions
Notation Conventions
DESCRIPTION

Programming
Data

Completion

Clock

SYSLIB Library
The RT-11 User

.
.

NN
N B RS

e
L

1.2.1

THE

FORTRAN

EXTENSTONS

CLRD (Clear Display) Routine
CVSWG (Convert Switched Gain

ROUTINES

Value)

W
b
U

DRS (Digital Read-in Sampling) Routine
DXY (Display X-Y Data Pairs) Routine
FLTl6 (lo-bit Flcating—-Point Conversion)
Routine
F5H (Flash) Routine
FXY (Flash X-Y Data Pairs) Routine
HIST (Time Interval Sampling Techniqgue)

DIS

[]

I
N S

L)
.

TSNS

Routine

(Display

Routine
=

CHAPTER

IADC

Data)

Routine

(Single Analog-to-Digital

Conversion)

Routine

1.4.11

IDIR

(Digital

Reading
Memory

1.4.12

Input

Digital

Reading)

Input

Routine

Bit and Byte Manipulation of a
Reading the Software Clock
IDOR (Digital Output Register)

Loading
Memory

Reading

Location

Digital

Output

Single

Word

Routine
Register or

Location

and

Resetting

iii

the

Software

Clock

CONTENTS

(CONT.)

.
L]

(Stop

SETR

(Set

1-59

Sampling)

Routine

the

Programmable

A REAL-TIME SUPPORT
System Configuration
Standard

Changing

the

Revised

Specifications
Configuration

The Real-Time Support Library
Generating the Library
Accessing the Library
MESSAGES

DEBUGGING

INTRODUCTION
USING
FDT

TECHNIQUE

(FDT)

FDT

COMMAND

TYPES

Control

Commands

Information

Transfer

FDT

Control

Commands

FDT

CONVENTIONS

AND

Commands
TERMINOLOGY

Syntax

Conventions
Current Procedure

The

Location

Offset
Named

Specification

Location
Location

[~ V8]

L)
*

Locations
Subscripted Name Locations
Mode Codes
FDT

Pause

Vb
~N oy

THE

CONTINUE
DIMENSTION
ERASE

GOTO
a

Label

MACRO

Implicit

Macro

L]
[]
[

STEP

STOP

START

[]

PAUSE

NAME

RESET

TYPE

WATCH

FDT

COMMANDS

.
.

.
[
L]

wur o

Defining

Definition

DESCRIPTION

o
*

U ab

o
e«
*

U1 U
O
U
Gt

NNDNNMNDNNDNDND

Relative

BB WWWNNN

ERROR

L1BRARY

System

Configuration

Assembling

Clock)

HENOO U UL

.
.
L] L]
L]
| CW S
O Sy Sy Wy
L]
L}
L]
[0
W N

BUILDING
A

1-55
1-56

Routine

Display)

Rate

.«

Routine

Control)

Routine

Program

NRONONND
NN

Routine

1
NNNNNNNNN!}JNNNNNNNN
|

L]
-

RTS

U LUt
Ut

=
) Y Iy SR

(Wait)

SDIS

[

(LED

Routine

[F9)

(Repeated

(Relay

]

REL

LWATIT

LED

Binary

to BCD)
BCD to Binary)
Display) Routine

w N

(Convert

FORTRAN

(Convert

CHAPTER

KB2BCD
KBCD2B

Page

TECHNIQUES

Named

AND

ADVANCED

CAUTIONS
LY

WHAT

(CONT.)

CONTENTS

How

PITFALLS

Common

FDT

Octal

Generates

Addresses

Offsets

Names

Relative

Addressing

Subscript
Gl
> W

Addressing

Format

Conversion

Routines

On-line

Debugging

Technique

Execution

APPENDIX

ARGUMENT

APPENDIX

OBJECT

APPENDIX

ERRCR

APPENDIX

FDT

COMMAND

APPENDIX

FDT

LOCATION

APPENDIX

FDT

MODES

APPENDIX

FDT

ERROR

AND

(ODT)

Speed

NN
.
L]

N
Ll
N
L]
>
L]

Indirect

DEFAULT

SOURCE

SUMMARY

TABLE

FILES

TABLE

SUMMARY
SPECTFICATION

FORMATS

MESSAGES

INDEX

Index~1

and

Circular

Array

Arrays

Pointers

TABLES

TABLE

2-1

FDT

Mode

Codes

[}

Linear

Circular

FIGURE

FIGURES

PREFACE

The

RT-11

real-time
RT-11

FORTRAN

support

documented

the

and

Manual

describes

FDT,

FORTRAN

following

RT-11

the

Introduction

Pocket
System

Guide (DEC-11-ORRCB-A-D)
User's Guide
(DEC-11-ORGDA-A-D)

RT-11

Advanced

RT-11

Documentation

Programmer's

System

Message

System

Generation

RT-11

System

Release

RT-11/RSTS/E
PDP-11

Manual

the

FORTRAN

(DEC=-11-ORMEB-A-D)

Manual

Notes

(DEC-11-ORGMB-A-D)

(DEC-11-ORNRB-A-D)

following

User's

Language

(DEC-11-OKAPA-A-D)

(DEC-11-ORDDB-A-D)

manuals:

Guide

Reference

sets

laboratory

(DEC-11-LRRUB-A-D)

Manual

DEC~11-LFLRA-C-DNI)
FORTRAN IV Installation Guide

RT~11
three

Guide

Directory

RT-11l

documented

technique.

(DEC-11~-ORITA-A-D)

RT-11

FORTRAN-callable

debugging

manuals:

RT-11
RT-11

FORTRAN

The

Extensions

routines,

(DEC-11-LFLRA-C-D

and

(DEC-11-LRSIA-A-D)

peripherals

supported

the

FORTRAN

extensions are described in the following sets of manuals.
You should
be familiar with the function of
the
peripherals
hardware
on
your
system,

1.,

and

with

ADl1-K

the

terms

Analog

used

Digital

these

manuals.

Converter

User

Manual

Multiplexer

User's

Manual

and

Control

(EK-AD11K-0P-001)

AM11-K

Multiple

Gain

(EK-AMI1IK-TM-002)
AAll-K 4-channel D/A

Display

User's

Manual

(EK-AAL1K-TM-001)
DR11-K

Interface

User's

Guide

and

Real

Time

Maintenance

Manual

(EK-DRIIRK-MM-001)

KWll-K

Dual Programmable
(EK-KW11K-OP-001)
'

AR1l

User's

LPS-11 Laboratory
(EK-LPS1I-0P-0023)
LPS11-S

Guide

Laboratory
(EK-LPS11-MM-002)

Clock

User

Manual

(DEC-11-HARUG-B-D)

Peripheral
Peripheral

vii

System
System

User's

Guide

Maintenance

Manual

CHAPTER

RT-11

1.1

WHAT

ARE

THE

FORTRAN

EXTENSIONS

EXTENSIONS?

The RT-11 FORTRAN Extensions are a library of routines
data
acquisition
and
processing.
The
extensions
subprograms written in MACRO
assembler
language
and
FORTRAN.

counter

The

and

extensions

Specifically, the
following sets of

allow

you

laboratory peripherals

control

converter,

programmable

clock

FORTRAN program.

FORTRAN real-time
extensions
peripherals hardware:

the K-series
hardware
AM11-K
multiple
gain
clock

from a

for
real-time
themselves
are
callable
from

support

any

the

(ADll-K
analog-to-digital
converter,
multiplexer,
AAll-K digital-to-analog

DR11-K digital

interface,

and

KWll-K

programmable

counter)

the

ARIll

real-time

the

LPS1l

analog

system

laboratory peripheral

See

Section

1.5

for

The

FORTRAN

extensions

discussion
can

system

combined

configuration.
to

provide

five

kinds

operations:
time
interval
control
and sampling, digital input and
output, analog-to-digital conversion and
sampling,
digital-to-analog
conversion,
and
graph plotting.
The following section describes the
functions available and lists the routines you use
to
perform
those
functions.
Detailed
instructions
for
wusing each routine appear in
Sections

1.4.1

Time-Interval
Timed

through

Control

1.4.20.

and

Sampling

program delay.
Use SETR in single-interval

the
delay
interval.
Schedule

event

interval,

the

and

mode

LWAIT

completion

specify

time

mode to
counter

the

1length

end

the

interval.

Use SETR to request a timed program delay,
a completion routine to execute at the end
Repeated time intervals.
Use SETR in repeated-interval
of
the intervals.
The clock
at the end of each interval.

the

detect

and to
of the

specify
the
generates an

designate
delay.

duration
interrupt

FORTRAN EXTENSIONS

RT-11

Schedule

periodic

events.

Use
SETR
to
request
repeated
time
1intervals,
and
designate
a
completion
routine
that
executes
at
completion of each interval.

Time

the

stamping.

Use IDIR or IDOR to read the current value of
the
software
the software
describes
which
1.2.2,
Section
(see
clock

clock).
during
binary

Measuring

clock
software
the
read
You can also use DRS to
Use FLT16 to convert the unsigned
digital sampling.

integer

the

time

value

intervals

to a

real

number.

between events.

in
SETR
to
run
the
clock
external event
an
Whenever
last
1intervening
since
the

Use HIST to read the time and
mode.
timing
external-event
occurs, HIST reads
the
time

<cannot do this with an ARl1l because
(You
event.
external
its clock does not have the external event timing mode.)

Measuring the latencies of a series
initial

events

relative

event.

Use HIST to read the clock value at the time of
the
event,
and SETR to run the clock in external-event timing from zero
AR1ll
mode.
(You cannot do this with an
AR1ll
because
the
clock does not have the required timing mode.)

Digital

Input

Read

and Output

digital

input.

Use IDIR to read the value of the
IDIR

accepts

data

binary

format.

Generate

digital

in either

digital

binary or

BCD

input

format

and

stores

output.

the
Use IDOR
to
1load
a
digital
output
register.
Use
conversion
routine
KB2BCD
to generate BCD format data for
output.

Digital

sampling

regular

time

intervals.

Use DRS to sample from a digital input register, and SETR in
at
repeated-interval
mode
to
generate
clock
interrupts
regular intervals (thus triggering the DRS sampling).

Continuous digital sampling at regqular time intervals.
Request digital sampling at regular time intervals using DRS
and
SETR, and designate a completion routine to process the
data
Digital

call

the

sampling

Use DRS

to SETR.

external

sample

events.

from a digital

input register.

Continuous digital sampling on external events.
Request digital sampling on external events

designate
call

Analog-to-Digital
Read

one

completion

using

DRS,

routine to process the data

and

in the

DRS.

(A/D)

value

from

Conversion and Sampling
an

Use IADC to take a
CVSWG
to
convert

analog

integer data

format

switched

was

gain

input.

and
single sample from an
A/D
channel,
the
digitized
value from a nonstandard

used).

real

number

(if

autogain

RT-11

Read

FORTRAN

EXTENSIONS

a series of values from one or more analog input channels.
Use SETR
to
set
the
sampling
rate
with
the
clock
in
repeated-interval mode, and RTS to collect a single sweep of
A/D
samples.
Use
LWAIT
to
detect
the
completion
of

sampling,

nonstandard

and

CVSWG

integer

data

convert

format

the digitized values
real

from

numbers.

Read values continuously from one or more analog
and store the values on a mass storage device.
Use SETR
to
set
the
sampling
rate
with

input

channels

the
clock
in
repeated-interval
mode,
and
RTS
to
sample
A/D
data
continuously.
Designate a completion routine to handle
the
data throughput to disk in the call to RTS, and use CVSWG to
convert the digitized values from nonstandard
1integer
data
format

real

Digital-to-Analog

(D/A)

Generate

analog

Use

numbers.

Conversion
voltage.

IDOR to generate

digital

output

value

for

D/A

converter.

Generate

series

analog

voltages.

Use SETR in repeated-interval mode to control
and
to
designate
a completion routine.
1In
routine, you use IDOR to generate the digital
for a D/A converter.
‘
Graph

the time base,
the completion
output
value

Plotting

Data

scaling.
Use CLRD to multiply the data in an array
values are in the required range.

Single display of Y-axis data.
Use FSH to display a data

Continuous

display

Y-axis

array

data

that

the

data

once.

(refresh

scopes).

(a)

Use SETR in repeated-interval mode to
control
the
refresh
interval,
and
to
designate
a completion routine.
1In the
completion routine, you call FSH to display a data array.

(b)

Use DIS

to display

designate
you

a data

array

a completion routine.

use SDIS

stop

the

constantly,

display.

and

SETR

1In the completion routine,

Continuous display of Y-axis data (storage scopes).
Use FSH to display a data array once, and
1IDOR
to
control
the intensify bit and erase bit of the scope controller.
Continuous

(a)

display

X-Y

data pairs

(refresh

scopes).

Use SETR in repeated—-interval mode to
control
the
refresh
interval
and
to
designate
a
completion routine.
1In the
completion routine, you call FXY to display the X and Y data
arrays.

(b)

Use DXY to display the X and Y data arrays
SETR
to
designate a completion routine.
routine, you use SDIS to stop the display.

constantly,
and
1In the completion

RT-11 FORTRAN EXTENSIONS

Continuous display of X-Y data pairs (storage scopes).
Use FXY to display the X and Y data arrays once, and

control

the

intensify

bit

and

erase

bit

IDOR to
the scope

controller.

1.2

THINGS TO KNOW ABOUT RT-11 AND THE FORTRAN EXTENSIONS

There are three variants of the RT-11 monitor:
RT-11 Single Job (S8J),
RT-11
Foreground/Background
(FB),
and
RT-1ll
Extended Memory (XM).
Consult the RT-11 documentation for descriptions of
the
features
of

the

monitor variants.
FB.

and RT-11

The FORTRAN extensions run under both RT-11 SJ

CAUTION

Do not use
RT=-11 XM.

The

FORTRAN

extensions

the

FORTRAN

run with

extensions

FORTRAN

under

IV version

which

includes

virtual-array support.
(See the FORTRAN Language Reference Manual for
description of the VIRTUAL array
declarator.)
However,
the
FORTRAN
extensions
1impose
<certain restrictions on the use of FORTRAN virtual
arrays.
The extensions routines cannot refer to virtual arrays.
That
is,
the
array
arguments passed to the extensions routines cannot be
virtual-array names.
If your FORTRAN program passes a
virtual
array
to
an
extensions routine, the results are invalid and unpredictable.
There is no error message.
Your
FORTRAN
main
program
can
contain
virtual
arrays
and
you can copy the conventional arrays used by the
extensions routines into virtual arrays.

NOTE

Virtual
or

1.2.1

arrays

EQUIVALENCE

cannot

appear

COMMON

statements.

Completion Routines

In the FORTRAN extensions package, completion routines are subroutines
called
from
within
an
interrupt
service
routine.
The
FORTRAN
extensions library provides standard
interrupt
service
routines
to
handle
all
interrupts
from
the
<clock
and laboratory peripherals.
Essentially, you write your own
completion
power of interrupt service routines.

routines

extend

the

Several of the FORTRAN extensions routines allow you
to
designate
a
completion routine name, and the conditions under which you want it to
execute.
The idea behind completion routines is that
you
will
want
them
to
execute
wupon
completion
of
some operation that generates
interrupts--for example, at the end of data sampling.
The
condition
that you specify is always in terms of some number of interrupts.
For
example, you can specify that a completion routine is to execute after
a
certain number of clock interrupts.
Depending on the mode of clock
operation,
defined

event

the

number

routine
of

interrupts.

data

executes
samples,

after
or

after

defined
a

defined

1interval,
number

after

external

RT-11

The

RT-11

monitor

into

subroutine

whenever

the

satisfied,
If

the

interrupt

completion

EXTENSIONS

conditions

<call

appropriate

the

other

puts

FORTRAN

1in

the

interrupt

service

routine

and

has

completion

routine

name

service

routine.

Then

occurred

routine

the

interrupt
calls

executing,

and

the

conditions

are

completion

routine.

state

the

job

being
executed
determines
what happens when a completion routine is
called.
Under the RT-1l SJ monitor, the completion
routine
executes
immediately.
Under
the
RT-1l1
FB
monitor,
foreground
completion
routine requests have priority over background jobs.
1If a
foreground
job requests a completion routine while a background job is executing,
RT-11

FB suspends the background
3job,
completion routine.
At the end of the
returns

control

foreground

the

and
executes
the
completion routine,

interrupted

background

3job

job
has
requested
service
in
the
job requests a completion routine while

background

executing,

RT-1l1

foreground

queues

the

job

gives up control
completion routine.

background

completion

the

(unless

meantime).
a foreground

routine

system,

foreground
the monitor

and

request

then

the

If
job

until

executes

a
is

the
the

If another completion routine is already executing, the
action
taken
depends
on the RT-1l1l monitor.
Under the KT-11 SJ monitor, the second
completion routine interrupts the first, executes, and returns to
the
first completion routine.
This procedure works only for two different
completion routines because completion
routines
are
not
reentrant.
That

1is,

FORTRAN

completion

itself

the

results

are

Under

RT-11l

FB,

routine

undefined

completion

must

(and

routines

there

not

attempt

interrupt

error

interrupt

each

message).

other.

The

completion routine executing continues
to
completion
regardless
of
whether
it was requested from the foreground or the background.
Then
the second completion routine executes.

Considerations

in Using Completion Routines

Completion routines do not "lock out" interrupts or interrupt
service
routines.
You
do
not
lose
interrupts
or
data
while completion
routines are running.
Completion

routines increase the system
overhead
time
in
interrupt
handling,
and
therefore decrease the rate at which interrupts can be
serviced.
The maximum data-acquisition speed
1is
lower
in
programs
that use completion routines.

Completion routines should
copy
terminal.
Terminal
completion routines.
If

you

anticipate

simultaneously,

that

use

not

several

the

request

T1/0

input

results

completion

SYSLIB

IQSET

1in

output

delays

routines

call

from

could

extend

hard

returning

the

RT-11l

from

active

queue

length.
You

must

you

declare

fail

completion-routine

so,

your

program

name

does

not

EXTERNAL

link

properly.

statement.

A completion routine cannot
in
the
FORTRAN
program.

call a subprogram that is called elsewhere
Because
completion-routine
execution is
asynchronous, the monitor could
interrupt
the
subprogram
to
begin
executing
the
completion routine.
The completion routine would then
attempt

because

the

call

the

FORTRAN

interrupted

subprogram

subprogram,

code

not

with

undefined

reentrant.

results,

RT-11 FORTRAN EXTENSIONS

1.2.2

Software Clock

"the

The FORTRAN extensions library provides a time service known

software clock." The software clock is a memory location that SETR
The
interrupt.
increments by 1 on each hardware clock counter

clock

software

hardware clock

and

when,

operates

when, SETR is running the

only

in repeated-interval mode.

The rate of the software clock is controlled by the rate of <clock
The rate
counter overflow 1interrupts specified in the call to SETR.
frequency
of overflow interrupts is a function of the clock counter
and the 1initial count (rcount), where the interrupt rate is
(irate)
(The time interval between interrupts is equal
equal to irate/rcount.
to

irate*rcount.)

you
Before
integers.
The software-clock values are unsigned binary
can do FORTRAN arithmetic with software-clock values, you must convert
The software clock counts
the values to real number form using FLTle.

177777), which is the largest possible unsigned
(octal
up to 65535
After the next clock-overflow interrupt, the value of
binary integer.
the software clock

The IDIR,

is zero.

IDOR, and DRS routines read the software

routine can change the software-clock value.

1.2.3

clock.

1IDOR

The

SYSLIB Library

SYSLIB
by
The services provided
SYSLIB is the RT-11 system library.
For example,
are described in the RT-11 Advanced Programmer's Guide.
double word
SYSLIB provides character string manipulation routines,

integers,

facility

requests,

programmed

(INTSET).

and

FORTRAN

interrupt

service

the appropriate device
The real-time extensions software protects
interrupt vectors during execution using the RT-11l .PROTECT programmed
Multiple .PROTECT requests on the same vector cause a fatal
request.
you should not use the INTSET call or the
Therefore,
system error.
FORTRAN
the
uses
that
any program
.PROTECT programmed request in
you can use INTSET or .PROTECT
is one exception:
There
extensions.
T1IDIR,
call
for the DR11-K or LPSDR-A vectors if the program does not
IDOR,

1.2.4

DRS.

The RT-1l1l User Service Routine

The User Service Routine

(USKR)

portion of the

RT-1l1

monitor

1is

not

The USR is normally swapped into
1in memory.
resident
permanently
the
supports
USR
(The
memory when any of its services is required.
RT-11

file structure,

and,

for

example,

handles opening files.)

Swapping the USR can slow program execution.

The

you

following
SET

Programs can run

faster

1lock the USR into memory before the program starts executing.

USR

instruction

locks

USR

into memory:

NOSWAP

Locking the USR into memory

possibility

the

recommended because that eliminates the

of the USR swapping over the extensions or the completion

routines.
1If space limitations
prevent
you
locking
the
USR
into
memory,
consult
the
FORTRAN
documentation
to find out the address
where the USR swaps and consult your program link map to see if
there
are

any

conflicts.
1-6

RT-11

1.2.5
The

Action

action

RT-11

Under

EXTENSIONS

CTRL/C

monitor

FORTRAN

running

FORTRAN

RT-11
initializes

SJ,

the
the

CTRL/C

causes
devices.

all

Under

FB,

the

CTRL/C

the

RT-11

program

depends

which

use.

causes

UNIBUS

1INIT,

real-time

which

<clears

extensions

and

software

clear all of the device registers in the device list maintained by
extensions.
Tf your program is using support libraries other than
real-time
extensions
library
(for example, VTLIB or LVLIB), use
monitor INIT command.

1.2.6
The

Programming

FORTRAN

place

extensions

data

collected
array
as

Requirements

High-Volume

data-acquisition

FORTRAN

exceeds the
a circular

for

arrays.

length
array.

Data

routines

the

Acquisition

DRS,

number

the
the
the

HIST,

data

and

points

of the array, these
routines
That is, when the routine has

RTS

treat
filled

be
the
the

array, it continues to put data in the array, starting again from
the
beginning (see Figure 1l-1).
Unless your program removes data from the
array, the routine eventually attempts to
overwrite
data.
When
it
attempts
to
overwrite
data,
the routine sets a flag for this error
condition, which is called "data overrun."

JIBENENNN
Partition 1

HEE

Partition 2
Linear FORTRAN Array

tiz[{
Partition 1
Partition 2

Circular Array

Figure
The

linear

circular
elements

1l-1

Linear

FORTRAN

array

and

Circular

wraps

around

array.
DRS and RTS treat
as array partitions.

Arrays
to

groups

form

array

RT-11

FORTRAN

EXTENSIONS

There
are
four
related
variables
1involved
1in
a
<circular
array
structure:
the
pointer
for
adding
the next data point, the space
remaining before data overrun, the pointer to the next data
point
to
remove,
and the space containing valid data (see Figure 1l-2).
1In the
FORTRAN extensions routines, space can be expressed in units of
array

elements
subset

the

units

total

array

partitions,

1length.

where a partition is a

Space remaining
before data overrun

~— Space containing ___,/
valid data

Pointer for the next

Pointer for adding

data point to remove

the next data point

Figure 1-2
circular
array

pointers

and

Circular Array Pointers
structure
involves
two

available

space

array

variables.

The extensions routines handle
adding
data
to
the
array
but
not
removing
data;
your FORTRAN program must handle removing data.
The
extension routine maintains the
pointer
for
adding
the
next
data
point;
this
pointer
1is inaccessible to your program.
Your program

must define and maintain

the pointer

to the

data

point

removed.

Both your program and the
extension
routine
maintain
the
wvariable
containing
the
space
remaining
before
data overrun (the available
space variable).
HIST defines the available
space
variable
as
the
number
of
array
elements
before data overrun.
HIST decrements the
available space variable each time it adds a data point to the
array.
DRS

and

RTS

define

the

available

space

variable

the

number

complete array partitions available before data overrun.
DRS and
RTS
decrement the available space variable each time an array partition is
filled.
Each time your program removes an array element
(with
HIST)
or the contents of an array partition (with DRS or RTS) thereby making
space available, it must increment the available space variable.
Your

program
pointer does

is also responsible for ensuring
that
the
data-removal
not pass the data-addition pointer.
Your program ensures

proper data
removal
by
starting
to
remove
data
only
after
the
extension
routine
informs your program that data have been received.
(In practice, you can either
and
remove data when

LWAIT

completion

routine

execute

monitor the available space pointer
with
it changes in value, or arrange to have a

after

defined

number

interrupts.)

1.3

RT-11

FORTRAN

define

the

EXTENSIONS

CONVENTIONS

The

following

conventions

1.3.1
The

sections

used

Syntax

this

syntax

conventions

chapter.

and

notation

Conventions

syntax

conventions described below apply to all of the
subroutine
function calls appearing in the chapter.
The discussion of syntax
conventions refers to the following sample sequence:

and

CALL

Any

SUBR([nvaluesi#],rvaluel
[, [rvalue2],cal
,[iflag*]]
l])

capitalized

them

1in

would

the

Punctuation
Any

words

syntax.

words

exactly

required
is

letters

You

can

satisfy

other

tokens

are

For

example,

portion

the

except

lower

that

variable

(Unless

for

when

must

example,

specify

CALL

SUBR

above.

enclosed

square

variable

items

names

with

brackets.
parts

your

the

choosing

of the syntax.
As
far
as
practical,
the function of the argument required.

with

constant

otherwise

You

For

indicate

"number

begins

calls.

sequence

token

constraints
means

obligatory.

all

case

these

mnemonic

word

integer

sequence.

"ncounts"

lowercase

are

way

replace

that

letters

same

obligatory

the

Any

the

of
an

counts."
i1

required

specified,

means

that

1integer

that

the

position

means

name

the

one-word

integer,
the default INTEGER*2 data format.)
The initial letter i is
mnemonic for any integer;
the
1initial
1letter
n
is
nmnemonic
for
“number
of ...."
For example, iflag in the sample sequence refers to
an integer
variable
or
constant;
nvalues
refers
to
an
integer
variable or constant containing the number of values.
Any

lowercase

word

that

begins

with

means

that

the

name

than

real

variable
or
constant
1is
required in that position in the sequence.
(Unless otherwise specified, real refers to the
default
REAL*4
data
format.)
For
example, rvalue in the sample sequence refers to a real
variable.
Any

lowercase

means

that

word

the

sequence.

For

subprogram

name.

Any

item

that

name

example,

square

begins

call

The

syntax

sequence

defaulted.

can

CALL

items

the

sequence,

Any

letter

the

other

required

sample

For

be
defaulted
specifies exactly

example,

that

sequence

can

always

point

refers

(left
out
of
what portions of

",[iflag]"

indicates

regardless of whether iflag appears.
the argument iflag defaulted:

for

square
the

sample

use

this

form:

this

form:

nested

square
The

brackets

trailing

not

syntax.

The

that

the
a

the
the

the

following

SUBR(NCOUNT,SUM, ,SUM2,MEANS)

When

However,

1in

brackets

sequence) .

comma
1is required
call is valid with

with

subprogram

sequence

CALL
CALL

defaulted
are

from

sometimes

the

rightmost

required

the

end

syntax.

above,

SUBR (NCOUNT,SUM)
SUBR(NCOUNT,SUM,,,)

brackets

nesting

are

commas

indicate

rvalue2

and

1-9

more
call

complex
in

the

cases
sample

the

default

sequence

above

RT-11 FORTRAN EXTENSIONS

means that rvalue2 and call cannot be defaulted independently. Either
both rvalue2 and call can be defaulted (indicated by the outerd
specifie
brackets), or rvalue2 can be defaulted while call is
d,
specifie
is
rvalue2
if
However,
).
(indicated by the inner brackets
call must also be specified.

For example, the following calls are valid:
CALL SUBR (NCOUNT,SUM,,SUMZ,MEANS)
, ,MEANS)
CALL SUBR (NCOUNT,SUM,
IFLAG)
CALL SUBR (NCOUNT,SUM,
These calls are not valid:

CALL SUBR (NCOUNT,SUM,,SUM2,)
CALL SUBR (NCOUNT,SUM,,SUMZ2)

argument 1is not part of the
on marks those arguments that
conventi
asterisk
The
.
sequence
calling
can be modified when the routine is called with the first argument
omitted. This capability is referred to as parameter-adjustment mode.
In the following example, the call to SUBR results in a change to the
parameter iflag. According to the syntax, the second argument cannot
Therefore, the wvariable
be modified in parameter-adjustment mode.
call, but according
adjustment
parametera
in
variable
RATE is a dummy

An asterisk following the name of an

to the syntax,

it cannot be defaulted.

CALL SUBR(,RATE,IFLAG)

The character # following the name of an argument is not part of the
The # convention indicates those arguments whose
calling segquence.
Arguments marked
values can change during execution of the routine.
. Under some
constants
than
rather
names
variable
be
with # must
circumstances, if you use a constant instead of a variable name,
FORTRAN changes the value of the constant.

For example, suppose you use the following call:
CALL SUBR(10,SUM,ITOT)

Suppose that SUBR returns 25

(nvalues).

some

the

value

the

first

argument

cases, SUBR would return 25 as the value of the

constant 10. The constant 25 is then substituted for each subsequent
appearance of the constant 10 in the program, with peculiar results
for

the program.

1.3.2

Notation Conventions

Any constant can be specified either as a decimal number or as an
The system assumes that all FORTRAN constants are
octal wvalue.

decimal unless they are marked as octal.
by

prefixing

You mark a constant as octal

the number with a double quote.

octal address 167772, use the constant "167772.

For example, to select

The manual refers to a bit with a value of 1 as being "on" or "set." A
bit with value 0 is referred to as being "off" or "cleared."
The words "defaulted" and "omitted" differ in meaning when applied to
subprogram arguments in this chapter. When the manual states that an
it means that the extensions routines
argument can be defaulted,
for the argument when it is left out
value
numeric
assume a particular
of the argument list. When the manual states that an argument can be
1-10

RT-11

omitted,
for

the

Tnstead,

means

argument
the

that

when

extensions

the

FORTRAN

extensions

does

not

routines

argument is a flag requesting
performed by the routine.

EXTENSIONS

some

routines

appear

1in

not

the

assume
argument

assume

that

the

absence

function

other

than

that

value
1list.
of

the

wusually

RT-11 FORTRAN EXTENSIONS

1.4

DESCRIPTION OF THE

FORTRAN EXTENSIONS

ROUTINES

The
FORTRAN
extensions
routines
are
described
1in
the
following
section.
The
routines
are
arranged
alphabetically for convenient
reference
use.
Section
1.1
contains
functional
groupings
and
descriptions
of
the
routines
so
you
can
use
those
sections to
determine which routines you need to solve your problem.
Appendix
A
contains a reference summary of all the routines and their arguments.
The description of
Name of
Meaning

each

routine

contains

the routine
of the name

Functional operation of
Calling syntax

the

routine

Explanation of the arguments
Error conditions
Interactions with other routines

Special
Examples

cases

restrictions

the

following

information:

RT-11

FORTRAN

EXTENSIONS

CLRD

l.4.1

CLRD

performs

element
one

(Clear

of
the

factor,

two

the

which

fill

CLRD

called

The

Routine

functions

array

arguments.

data

CALL

Display)

on
a

array.

scale

factor

calculates

horizontal

display
width of the

data

multiplying

the

single

routines

DIS

a
and

FSH

each

supplied

(X-axis)

use

screen.

scales

spacing

position

the

subroutine.

CLRD(iarrayname#,npoints,ispace#,rscale)

arguments

are:

larrayname

Name

the

integer

array

containing

the

data

scaled.
Each
array element is multiplied by
rscale.
Scaled values greater than 4095
(1023 for

npoints

the

AR11l)

The

number

are

scaled.

nondisplayable.

elements

CLRD

scales

ispace

The

X-axis

and

FSH

spacing
wuse

the

elements.

factor

1ispace

'
array

returned

that are
npoints

first

CLRD.

‘"stretch"

the

array

DIS

display

horizontally so that the display fills the
screen
with evenly spaced points.
CLRD calculates ispace
as 4096/npoints (1lU24/npoints for the
ARll).
If
npoints
is
greater
than
4096
(or 1024 for the

AR11l),

then

ispace

set

NOTE

The

internal

ispace

The

not

standard
cannot

format.

Therefore

you

own

value

ispace

use

ispace

scale

for

used

stretching
screen
as

zero,

each

then

than
Thus,

multiply

scale

spacing by
fill
the
(rather
value.

displayed.

factor

the

wusing

four

arguments

1-13

are

not

present.

element

the
1If
is

the

Y-axis

values
rscale
set

to
is
-1

a nondisplayable
make
the
whole

CLRD,

the

EKROR

required

each

array

to indicate
1initially
with

cannot

affects

conditions:

The

you

or compressing
you
require.

element

=zero)
you can

array
nondisplayable
selectively
fill
it

SYNTAX

calculate

and

other calculations
normal integer value.

The

1integer

factor

ilarrayname.

Error

representation

your

it were

rscale

machine

and

points

then

later

RT-11 FORTRAN EXTENSIONS

ERROR

ARGUMENT

The argument npoints is less than one.
The argument rscale is negative.

Interactions:

Both DIS and FSH require the spacing factor ispace calculated

CLRD.

for

You can call CLRD to scale either or both of the data arrays
DXY -and

FXY.

Restrictions:
None

Examples:

Set up an array ISCREEN that you will later fill with the data

displayed.

CLRD

returned in TISPACE

makes

is 1.

the

array

nondisplayable.

The

value

DIMENSION ISCREEN (4096)

CALL CLRD (ISCKEEN,4096,ISPACE,0.0)

Suppose you have a data array IDATA of 25V points that you want to

prepare for display. The maximum value in the array is 150U, but you
would like the display to £ill the screen vertically as well as
Calculate your scale factor by dividing the maximum
horizontally.
value you want by the maximum value you have. 1In this case, the scale
factor might be 4095/1500 or approximately 2.67. For an ARll, the
scale factor might be 1VU23/1500, or roughly U.67.
CALL CLRD(IDATA,250,ISP,2.67)

I1f you want the data values scaled more

exactly,

use

the

following

call:

CALL CLRD(IDATA,250,15SP,4095./1500.)

The value returned in ISP is equivalent to 16 (or 4 for the ARll).

RT-11

FORTRAN

EXTENSIONS

CVSWG

1.4.2

CVSWG

converts

format

(Convert

Switched

switched

gain

number

format.

real

Gain

Value)

analog

input

CVSWG

Routine
values

requires

from

packed

input

1integer

value

the

format returned by the RTS and TADC routines.
The value contains
the
digitized
wvalue in the rightmost 12 bits for the ADl1-K and LPS11 (10
bits for the ARll) and the
gain
code
in
bits
12
and
13.
CVSWG
separates this packed information into two values, the digitized value
as a real number, and the gain code as an integer.
CVSWG

called

function.

rvalue=CVSWG(ivalue[,igain#])

The arguments are:
ivalue

The

igain

integer

value

acquired

form
of
the
CVSWG (the value

The

the

name

returned.

variable

The

value

to 4.
The meanings
the descriptions of
Sections 1.4.10 and
Error

using

real-number
returned as

RTS

TIADC.

The

digitized
wvalue
of the function).
which

returned

the
is

gain

the

1is

setting
range

of the gain
codes
appear
1in
the routines IADC and RTS (see
1.4.18, respectively).

conditions:
SYNTAX

ERROR

arguments,

than

two

arguments,

are

present.

Interaction:
The

CVSWG

routine

normally

converts

data

acquired

RTS

TADC.

Restrictions:
None
Examples:

You want to convert a value acquired

from

channel

with

gain

code

(using TADC

function)

channel

VAL=CVSWG
(IADC (0,2))
2.

You

want

convert

value

autogain.
IGAIN
contains
the
the conversion.
Tt is necessary
TGAIN

represents

restore

VAL

acquired

TADC

gain selected
to divide VAL
to

its

original

from

and used by
by the gain

with

IADC during
value
that

scale.

VAL=CVSWG
(IADC (2,0), IGAIN)
VAL=VAL/
(2%* (2% (IGAIN=-1)))
The

expression

and

for

(2** (2% (IGAIN-1l)))
results in scale factors
IGAIN values 1, 2, 3, and 4 (respectively).

16,

RT-11 FORTRAN EXTENSIONS

DIS

1.4.3

DIS

(Display Data)

Routine

initiates the continuous display of Y-axis data on a
example,

(for

scope

the Tektronix 6U4).

refresh-type

DIS displays Y-axis data on the scope.
The routine
assumes
that
if
X-axis values were provided, they would be equally spaced, increasing
Logically, there are two X values for each ¥ value plotted.
values.
One
specifies
the
X
coordinate
on
the
screen
where the Y value
appears, and the other specifies the array subscript where the Y value
The X coordinate and array subscript for a single point
can be found.
X
You do not have to provide explicit
usually have different values.
value is controlled by an X-axis spacing
The X-coordinate
values.
factor, whereas the array subscript value is controlled by an indexing
factor (see the ispace and increment argument descriptions).
DIS displays one data point on every interrupt from the scope

control

DIS constantly plots points,

that

it displays points as gquickly as the scope can receive them.

array as soon as

DIS

called

reaches

starting again at the

the end.

beginning

the

subroutine.

CALL DIS(iydata,ispace,npoints,istart,increment)
The

arguments

iydata

are:

The name
of
the
1integer
array
containing
the
scaled
Y-axis data to be displayed.
The Y values
should be in the range 0 to 4095 (or 0 to 1023 for
(See the restriction on Y values later
the AR1l).
in

ispace

this

section.)

The X-—-axis spacing factor.
controls
the
horizontal

The
1space
argument
spacing (X coordinates)

for the points plotted.
You
must
call
CLRD
to
calculate
1ispace
before you call DIS.
If ispace
is
1
and
increment
1is
1,
then
the
X-axis
for the display assumes values 0, 1, 2
coordinate
ye+., (npoints - 1).
1If ispace is greater than
1,
then the X—axis coordinate for the display assumes
2*ispace*increment,
ispace*increment,
0,
values
..., (npoints—1)*ispace*increment.

npoints

The number of

istart

The array subscript

increment

The indexing factor for the array subscripts.
DIS
adds
increment
to
the
preceding
subscript
to
obtain the subscript of the next point to display.
The starting subscript is istart.
TIf increment 1is

argument must

points

be greater

display.

than

zero.

the

first

displayed
from
1iydata.
The
be a legal subscript
within
array liydata.

then

istart

the

points

through

1-16

displayed

(istart

The

npoints

point

istart argument must
the
bounds
of
the

have

subscript

npoints

values

1).

RT-11

FORTRAN

increment
have

1is

greater

values

1istart

(npoints
array
Error

EXTENSIONS

1).

subscript

than

error

goes

then

through

message

out

the

istart

subscripts
+

increment*

appears

when

bounds.

the

conditions:

SYNTAX

ERROR

The
ARGUMENT

The

required

five

arguments

are

not

present.

ERROR

number

greater

points

than

requested

(npoints)

less

than

4096.

The

subscript

was

less

The

increment

than

the

first

specified

point

was

less

than

displayed

4095,

(istart)

greater

than

Interactions:

DIS

requires

the

horizontal

the

spacing

factor

(ispace)

CLRD.

DIS

displays

scope

control.

displayed by DIS,
receive
the next
of a point to the
22
microseconds.
not

sufficient

only

other

data

point

hardware

every

1interrupt

calculated

interrupt

occurs

from

after

each

the
point

telling DIS that the scope control is ready
to
data point.
The time interval from the display
occurrence of an interrupt may be as
brief
as
Therefore, the remaining machine resources are
allow

routines

much

that

data

are

processing

guaranteed

computation.

running

The

are

other

scopes
1is
narrower
the
D/A converters.

than
For

interrupt handling routines
(for
example,
the
<clock~interrupt
service
routine)
and
completion routines.
Therefore, the most
practical way
to
stop
DIS
displays
1is
to
call
SDIS
in
a
completion routine.
Restrictions:
The

range

the

range

example,

voltages accepted by some
of
voltages
produced
by

full

scale (U to 40Y5) for the converter
may
represent
5.12
volts.
However,
the
scope control may accept
display voltages in the range -5 to +5 volts.
Therefore, digital
values
0
through
47 and 4048 through 4095 are not displayed on
those scopes, and displays including those values
hang
off
the

-5.12

edges
of
the
screen.
arrange the arguments and

To avoid losing
data
so
that

data
the

with these
first
47

scopes,
screen

positions
correspond
to
nondisplayable
Fill
the
data
array
with
displayable
displayable
values
have
X
coordinates

data in the data array.
values
so
that
the
in the range 48 through

4047.

the

The

scaled

values

should

4047.

range

Negative data values, or values greater than 4095 (1023
ARll) are nondisplayable.
A point with a value greater
(1023)
is treated as if its value were -1.

through

for
than

the
4095

Each point is intensified every npoints interrupts.
The
amount
of flicker in the display depends on the number of points and the
persistence of the phosphor in the scope.
Nondisplayable
points
are
not
ignored
but are processed as if they were appearing on

the

screen.

number

1Tf

flicker

nondisplayable

problem,

points

being

first

try

"displayed."

reduce

the

RT-11 FORTRAN EXTENSIONS

Examples:

The horizontal
There is a data array l0u0 points long.
1.
factor ISPACE was calculated in a prior call to CLRD.
(a)

spacing

You want to display the last 8UU points in the array.
CALL DIS(IDATA,ISPACE,800,201,1)

(b)

You want to display a data window of the middle 75U

points,

displaying every fifth point.

CALL DIS (IDATA,T1SPACE,150,120,5)

The following example is a complete short program designed

for

The example calls the routine SETR to control the
refresh-type scope.
duration of the display and to set up the call to the completion
routine containing the command to stop the display.

DIMENSTION ISCREEN (4096)
EXTERNAL STOPS
CALL CLRD(ISCREEN,4096,ISPACE,0.0)
:

DO 10 TI=48,4047,20

;Array to be dispiayed.
;Completion routine name
;declared EXTERNAL.
;Make array

;jnondisplayable and

;ispace=l.

;DO limits chosen to

;£il1l1

screen.

;for

diagonal

sPut values

ISCREEN (T) =1

NPTS=200
IFLG=0
CALL SETR(5,0,1u00.,IFLG,,STOPS)
CALL DIS(ISCREEN,ISPACE,NPTS,48,20)

in ISCREEN
line.

;Specify 200 points.
;Set up SETR flag.
:Time for 10 sec, then
;call the completion
;routine STOPS.

;Display every 20th
;point in ISCREEN
;starting with
;48th

the

point.

PAUSE

:Wait for

keyboard

GO TO 20

;Display

line again.

; input.

END

SUBROUTINE

CALL SDIS
RETURN
END

STOPS

;Stop the display.

RT-~11

FORTRAN

EXTENSIONS

DRS

1.4.4
DRS

DRS

(Digital

mediates

Read-in

repeated

Sampling)

sampling

from

Routine

digital

or
LPSDR~A)
or from a memory location.
digital sampling capabilities beyond the
you can acquire using TIDIK.
DRS

collects

multiple

samples

from

single

location.
Use multiple
memory locations.

calls

DRS

capabilities.

provides

three

basic

DRS

Single-sweep sampling.
Using DRS,
sequence of digital input register
parameters that direct DRS to stop
with data.

Continuous
input data
sampling:
the
total
and

specify
You

input

sample

DRS,

input

(DR11-K

The DRS routine extends
single register reading

wunit

from multiple

you can
readings
after it

your
that

memory

units

acquire
a
short
by specifying the
£fills
an
array

sampling.
Using
DRS,
you
<can
acquire
digital
continuously.
There are two classes of continuous
finite continuous sampling, in
which
vyou
define
number
of
samples to be acquired in the call to

infinite

the

specify

total
the

continuously,

continuous

number

parameters

and

either

sampling,

samples

directing

designate

which

the

call

DRS

you

acquire

completion

not

DRS.
data

routine

arrange your program code to empty the
arrays
periodically.
You
designate the completion routine in the SETR call if you

are acquiring data on clock interrupts;
you
designate
the
completion
routine
directly
from
DRS if you are acquiring
data on interrupts from the digital input interface.
TIf
you
request infinite continuous sampling, you must call DRS again
to stop sampling.
You

provide

the

completion

dispose of the data as
you can write the data
means

Because

routine

empty

the

arrays

and

required by the problem.
For example,
to secondary storage or reduce
it
to

counts.

the

completion

routine

requires

time

execute,

the

maximum
continuous
sampling
rate
(involving
routine) is lower than the maximum single-sweep
(not involving a completion routine).

a completion
sampling rate

End

mode

sampling.

Call

infinite continuous

DRS

DRS is called as a subroutine.
depends on the capabilities you

Single—-sweep
CALL

parameter—adjustment

stop

sampling.
The
number
of
are requesting.

arguments

necessary

sampling:

DRS

(iarrayname,iarraysize,,
[nsamples], [isource],iunit,

imask, [imode],iendflag#,nleft)

RT-11 FORTRAN EXTENSIONS

Continuous

(a)

Clock-interrupt—driven continuous sampling
( iarrayname,iarraysize, [nsubarrays], [nsamples],
[isource],iunit,imask, [imode],iendflag#,nleft#)

DRS

CALL

(b)

sampling:

Event-driven continuous

sampling

( iarrayname,iarraysize, [nsubarrays], [nsamples],
[isource],iunit, imask, [imodel], 1endflag#,nleft#[,

DRS

CALL

[intervall,completel)
End

sampling:

Specify parameter-adjustment mode by omtttlng
imode
can
be
the register or

altered.

iarrayname.

Only

The values of isource and iunit specify

The rest
location for which sampling is to stop.
be the same
of the variable names are dummy arguments that should
in the original DRS call.

as the names

DRS

CALL

(, iarraysize,,, [isource],iunit, imask, imode?*,

iendflag,nleft)
The

arguments

are:

iarrayname

The name of the

integer

array

in which DRS

places

If you omit iarrayname,
1.2.6).
Section
imode
for
call expects a negative value

the DRS
end
to

sampling, DRS
continuous
For
data.
input
the
(see
array
circular
a
treats the data array as
.

sampling.

iarraysize

The
length
of
iarrayname
(in
words) .
The
iarraysize argument specifies the amount of memory
reserved by your program for input data.

nsubarrays

The number
of array partitions in iarrayname.
The
The length of
number of partitions is 1.
default

1in
each partition is equal to the largest integer
with
exanmple,
For
(iarraysize/nsubarrays).
iarraysize = 512, and nsubarrays =
5,
there
are
and 12
1long,
words
100
each
partitions,
five
unused words at the end of iarrayname.

The value

nsubarrays must be

iarraysize.

With

less

continuous

than or

sampling,
there
must
be
at
least
partitions
(see
the
interval
description).
Single—sweep
sampling
require array partitions.
nsamples

equal

double-word
two

array
argument
does
not

The number of samples required.
A
single
sample
is
either
a single data word or a pair of words,
depending on
the
mode
specified.
The
default
value of nsamples is iarraysize.
With single—-sweep sampling, set nsamples equal
iarraysize
for
single—word
sampling modes;

nsamples equal

double-~word

to half

sampling

assign the value
you want to stop

to
the

the value

modes.

With continuous sampling, you
nsamples
depending
on
how

1-20

iarraysize

to
set
for

RT-11

FORTRAN

sampling

process.

set

nsamples

want.

TIn

For

this

then

isource

DKS

with

The

flag

Lnput
value

continuous
number

sampling,

samples

you

iarraysize

negative

reads

calling

the

continuous

any

nsamples
DRS

finite
to

case,

infinite

For

equal

nsamples

equal

EXTENSIONS

sampling,

nsamples

U
data

until

again

1in

negative

set

number.

value

indicating

you

stop

sampling

parameter—adjustment

for

imode.

whether

Value

mode

read

location.

digital

The

default

Meaning

Zero

Read

the

digital

specified

input

1iunit.

1In

the

reading the register, DRS
bits (and only those bits)

process

clears
that it

those
finds

set.

rionzero

Read

the

appears

iunit

The

location

memory
1in

the

location

data

Value
U

address

read.

Meaning

through

or
0

whose

iunit.

through

Logical

unit

digital

1input

U—-through-7

numbers

values

eight

interfaces.
LPSDR~A,

LPSDR-A,

values
DR11-K
(The

select

DR11-K

the

then

and

The
one

digital

1/0

1f
the
includes

configuration
with

for

registers.

system
LPS11

selects

the

1l-through-8

select one of
digital
1TI/0

up to
eight
interfaces.

value

1isource

must

Zero.)

U,2,...,"177776

Memory

address

must

(The

value

selected.
be

even

1isource

The
number.

must

nonzero.)

imask

The

value

performs

value,

used
a

and

For

the

1input

AND

using

imask

the

Request
(decimal)

single-word

format

the

mask

stores

iarrayname.
either as -1

imode

logical

the

interrupts

result

a
or

sampling,
value

driving

1-21

the

mask
with
as "177777
imode
be

the

read

value.
and

data

sampling.

input

array,

all
bits
(octal).

specifies
and

DRS

the

data

source

RT-11

FORTRAN EXTENSIONS

For double-word sampling, imode specifies the data
format
of the value to be read, the source of the
interrupts driving the sampling, and the
contents
of

the

For

second

word.

parameter

adjustment,

imode

specifies

the

stop

code.

The

default

value

imode

1is

zero.

Single-word sampling.
In
single-word
sampling
contents
of
the
digital

location

into

sampling

codes

the

are:

modes,
DKS
reads
the
register or the memory
array.
The
single-word

data

Value

Meaning

Read unsigned BCD data from
or memory location on clock

Read binary data from
the
register
memory location on clock interrupt.

Read

unsigned

memory

BCD

data

1location

from

the register
interrupt.

the

on a digital event

interrupt.

Read binary data from
memory
location
on

the
register
or
a
digital
event

interrupt.

Double-word

sampling.

There
are
two
kinds
of
double-word
sampling:
double-word sampling with bit location, and double
word sampling with time stamping.
In
both,
the
first
word
of
the pair contains the register or
memory

value.

In double—-word sampling
second
word of the pair
of

the

rightmost

with
bit
location,
the
contains the bit position

nonzero

bit

1in

the

nonmasked

portion of the input value.
That is, DRS performs
a logical AND between
the
input
value
and
the
complement
of
the
mask, and then scans right to
left for
a
nonzero
Dbit.
Tt
assigns
the
bit
position
(U
to
15)
as
the value of the second
word.
TIf no bits are on, the second word
of
the
pair
1is
sampling

set
to
-1l.
You
request
double-word
with bit location by adding 2 to
any
of
single-word sampling codes to obtain codes 2,

the
3,

and

double-word

second
word
clock reading

sampling

of
at

with

time

stamping,

the

the
pair
contains the software
the time
of
the
digital
event

interrupt.
The clock value does not change.
The
time—-stamping
option
1is
available
only
with
digital
event
interrupts.
You
use double-word
sampling

with

concerning
register)

time

what
and

when

1-22

stamping

happened
it

happened

obtain

(the
(the

information

digital

input

software-clock

RT-11

FORTRAN

EXTENSIONS

reading).
You
request double-word
time stamping by adding 8 to
modes
obtain codes 12 and 13.
The

double-word

sampling

Value

codes

sampling with
4
and
5
to

are:

Meaning

Double-—word

sampling

Read

and

convert

clock

interrupt.

Double-word

sampling

Read

data

binary

Double-word

Read

and

digital
7

event

bit

with

location.

BCD

bit

clock

sampling

convert

Double-word
Read

with

unsigned

data

location.

interrupt.

with

bit

unsigned

location.

BCD

data

interrupt.

sampling

binary

with

data

bit

location.

digital

event

interrupt.
12

Double-word

Read

and

digital
13

sampling

convert

event

Double-word

Read

with

time

unsigned

stamping.

BCD

data

interrupt.

sampling

binary

with

data

time

stamping.

digital

event

interrupt.

End sampling.
Each
parameter—-adjustment

sampling
If

you

from

are

location,

each

one

sampling

use

input
from

one

unit

The

stop

mode

than

code

values

Finish

call

for

(Request

omitting

the

are:

sampling

any

Finish

for

Stop

the

reading

not

Stop

for

queue

reading

queued

the

array

current

Do
not
honor
routine requests.

completion
do

current

queued
completion
run,
but
do
not

requests.

sampling

partition.
completion

-1

unit

Meaning

gueue

-2

stops

location.

one

stopped.

partition.
Allow
routine requests
to

-3

DRS

memory

argument.)

Value
-4

parameter—-adjustment

parameter—-adjustment
iarrayname

call

unit

data.

routine
any

data.

completion

Allow

requests

any

array

queued

run,

but

requests.

routine

not

honor

requests.

any

RT-11

The

iendflag

FORTRAN EXTENSIONS

completion

iendflag

and/or
=zero

error

flag.

before

you

You
call

must
DRS.

set
DRS

increments
1iendflag
either
when
all
samples
requested
have been acquired or when sampling has
‘been stopped by a stop code
(negative
value
for
imode).
DRS sets

If a data overrun or other error
1endflag to a negative value.

occurs,

The array partition flag.
DRS
1initializes
nleft
with
the
number of array partitions specified by
nsubarrays.
DRS decrements nleft by 1
each
time
an
array
partition
becomes
full.
Thus
nleft
contains
the
number
of
partitions
currently
available
to
DRS.
The
nleft
argument
is
meaningful
only
with
continuous
sampling.
Remember
to
1increment nleft when you remove data
from a full array partition so that the
partition

nleft

can

reused.

nleft

becomes

zero

before

the

number

samples

requested
1in
nsamples
has been acquired, a data
overrun has occurred.
DRS then sets
1iendflag
to
indicate

interval

error.

The number of
interrupts
between
<c¢alls
to
the
completion
routine.
DKS initializes an internal
variable to interval and decrements it after
each

interrupt.

When

the

value

zero,

DRS

reinitializes the internal variable and causes the
completion
routine
to be called.
Therefore, the
completion routine is executed
whenever
interval
interrupts have occurred.
The default value for interval is
the
length
of
the
array partition, iarraysize/nsubarrays.
With
single-word
sampling,
the
default
value
for
interval
causes
the
completion
routine
to
be
executed as each array partition is filled.
With
double-word
sampling,
the
default
wvalue
for
interval
causes
the
completion
routine
to
be
executed
after
two
array
partitions
have been
filled.
Completion~routine name.
The
completion
routine
that
you
provide
1is
executed
after
interval
interrupts have occurred.
There
1is
no
default
completion
routine
name.
If
you
omit
the
completion routine name in
digital
event
driven
modes, DRS does not call a completion routine.
1In
clock—-driven modes, you must omit
the
completion
routine argument and instead specify any necessary
completion
routine
1in
the
<call
to. the
clock
routine, SETR.

complete

Error

conditions:
SYNTAX

ERROR

There
ARGUMENT

are

fewer

than

1U,

than

The

ERROKR
digital

input

The

memory

address

(iunit)

specified

requested

(imode)

The
in

mode

parameter-adjustment

(iunit)

mode.

1-24

12,

arguments

specified
is

negative

does

present.

not

exist.

odd

number.

but

the

call

1is

not

RT-11

DEVICE

FORTRAN

EXTENSIONS

CONFLICT

The

digital

input

use.

Interactions:
In

clock~driven

sampling

completion-routine

name

modes,

the

you

SETK

must

designate

any

call.

Restrictions:
DR5

allows

registers

once
name

you

(and

the

LPSDR-A,

many

eight

present)

DR11-K

each

mode.
the

You

units

can

designate

running

digital

simultaneously.

for each unit you want to read, specifying a
for each unit.
Each unit can run in either

event—driven
for

sample

unique

event-driven

1input

Call

DRS

different array
clock-=driven or

completion

routine

mode.

kxamples:

You

want

minute.
clock

DKS

read

reads

overflow
UDIMENSION

digital
one

word

input
of

binary

data

once

into

each

second

for

one

each

the

array

MIN

data

array.

interrupt.
iSet

TEND=U

MIN (oU)

;End

TENDC=0

;End

flag
flag

CALL

DRS (MIN,00,,,,5,-1,1,TEND,TFULL)

CALL

SETR(5,1,100.,TENDC)

;Request sampling.
;Start clock at 1 Hz.

CALL

LWATT(TEND,
Q)

;Wait

for

60U

CALL

s&8TR(-1,,.,)

;Stop

the

clock.

for
for

DRS.
clock.

samples.

2.
You want to read
60
samples
of
BCD
data
from
digital
input
register
3
on
digital event interrupts.
You are interested only in
the high-order byte, so you mask off the low-order byte.
DIMENSTON

MIN(6U)

iSet
;End

TEND=0

3.
a

CALL

DRS (MIN,60,,,,3,"177400,4,1END,TIFULL)

CALL

LWATIT(IEND,OQ)

You

want

completion

double—-word

read

sampling

status
on

with

appears
twice,
once
completion routines.

(a)

sRequest sampling.
iWait for 60 samples.

the

routine

up data array.
flag for DKS.

every

time

with

digital

stamping

completion

for

DR11-K

event

using

unit

and

interrupt.

You

execute
reguest

mode

13.

The

example

routines,

and

once

without

This example uses a completion routine.
The example assumes
the
interrupts
occur
at
a
rate
slow
enough
to
allow
completion routine to execute fully between interrupts.
DIMENSTON

EXTERNAL

TREG(4)

COMPL

iSet

;Label
;End

the

array.

completion

s EXTERNAL.,

TEND=0

that

flag

for

DRS.

routine

RT=11

CALL

FORTRAN

EXTENSIONS

DRS(IREG,4,2,-1,1,"167770,-1,13,TEND,TFUL,1,COMPL)

;Request sampling.

SUBROUTINE COMPL
for

;jCompletion

routine.

example,

issue control instructions
acquire or process data
compute

examine conditions to decide
IF (TDONE.EQ.1l) GO TO 1lu
TFUL=2

done

:Go to 1lU 1if finished.
;Reset available space

;jpointer.
RETURN
;Return 1f more to do.
CALL DRS(,4,,,1,"167770,-1,-~1,1END,TFUL)
;Call DRS with stop
;code=~1.
RETURN

You

can

achieve

DIMENSTON

the

same

function without

IREG(4)

completion

;Define

TEND=0

;End

DRS(IREG,4,2,-1,1,"167770,-1,13,TERD,TFUL)

CALL

LWATT(IFUL,2)

;Start
10

;Wait

whatever 1is required, for
issue control instructions
acquire or process data

data destination.

flag

CALL

routines.

for

DRS.

sampling.

for

interrupt.

example

compute

examine
go

conditions
if

to decide

finished

TFUL=2

GO TO 1vU
CALL DRS(,4,,,1,"16777v,-1,-1,1TEND,TFUL)

;Reset available space
ipointer.
;Go wait for next event.

;Stop

DRS

with

imode=-1l.

RT-11

FORTRAN

EXTENSIONS

DXY

1.4.5

DXY
The

DXY

(Display

continuously
continuous

Tektronix

6U4).

X-Y

Data

Pairs)

displays X-Y data pairs contained in two data arrays.
display is used for refresh-type scopes (for example,
DXY

plots

are

not

associated

displays

one

data

that

displays

the

values

DXY

so
DXY

constantly

the

arrays

The

data

plots

after

Routine

correlational

point

points
data

has

with
each

displayed

must

any

spaced,

interrupt

from

quickly

points,

data,

equally

the

data

1in

ordered

the

scope

can

starting

again

the

point

the

last

two

arrays.

The

which
values.

control,

receive

them.

beginning

arrays.

X—axis

data

array

contains the horizontal displacement of each point to be plotted.
The
Y-axis data array contains the vertical displacement of each point
to
be plotted.
The displacements are relative to (0,0)
in the lower left
corner of
the
two

value
fifth
The

is the fifth element of
element of the Y array.

data

that
CLRD

elsewhere
is

0
the

40953

range

scale

must

the

called

CALL
The

values

1is,

compress

DXY

the screen.
The X-Y
arrays.
That
1is,

the

array,

the

range

that

1U23

for

and

You

can

program with
a

are in corresponding positions in
the fifth pair to be plotted, the X

for

the

them.

pairs

data

also

and

can

arrays

value

plotted

the AR1ll).

transform

FORTRAN

the

You

the

can

independently
or

scale

the

1is

the

scope,

expand

using

data

arrays

manipulations.

subroutine.

DXY(ixdata,iydata,npoints,istart,increment)

arguments

are:

ixdata

The

name

X-axis

iydata

The

name

Y-axis

values

npoints

The

number

istart

The

array

the
for

1integer
the

the
for

X~Y

array
pairs

1integer
the

points

subscript

X-Y

the

first

increment

and

1iydata.

The

indexing

factor

for

the

array

pair

the

displayed.

containing

be
plotted.
The istart argument
subscript within the bounds of the

array
pairs

(X-Y pairs)

containing

the

displayed.

plotted.

points

must be
arrays

legal
ixdata

subscripts.

DXY

adds
1increment
to
the
preceding
subscript
to
obtain the subscript of the next pair of values to
be
displayed.
Tf increment is 1, then the pairs
displayed have
subscript
wvalues
1istart
through
(istart
+
npoints - 1).
Tf increment is greater
than 1, then the
subscripts
have
values
1istart
through
(istart
+ increment* (npoints - 1)).
The
increment argument must be greater than
or
equal
to
1.
No
error
message appears when the array
subscript goes out of bounds.

RT-11 FORTRAN EXTENSIONS

conditions:

Error

ERROR

SYNTAX

The
ARGUMENT

required

five

arguments

are

not present.

ERROR

greater

than 1 or

than

The array subscript

40UY6.

the

first

(istart) is less than 1.
The increment requested is greater

pair
than

1is

(npoints)

The number of pairs of points to display

less

displayed

4096.

Interaction:

DXY displays the next data point on each interrupt from the scope

interrupt occurs after each point
A scope-control
control.
displayed, telling DXY that the scope control is ready to receive
The time interval from the display command
the next data point.
22
as
interrupt may be as brief
to the occurrence of an

Therefore, the remaining machine resources are not
microseconds.
The
computation.
or
sufficient to allow much data processing
are other
running
be
to
guaranteed
are
that
routines
only
keyboard
or
interrupt handling routines (for example, the clock
Therefore,
service routines) and completion routines.
interrupt
the most practical way to stop a DXY display is to call SDIS from
a completion

routine.

Restrictions:

than
narrower
is
The range of voltages accepted by some scopes
For
D/A converters.
the
by
produced
voltages
of
range
the
represent
may
example, full scale (U to 4095) for the converter
to

-5.12

+5.12

wvolts.

the scope control may accept

However,

values
Therefore, digital
voltages in the range -5 to +5 volts.
U
through
47
and
4U48 through 4095 are not displayed on those
scopes, and displays including those values hang off the edges of
To avoid losing data with these scopes, arrange the
screen.
the
screen positions
47
first
the
that
so
data
and
arguments
Scale both
to nondisplayable data in the data array.
correspond
48
range
the
1in
the X and Y arrays to have displayable values
through

Data

4047,

values

greater

nondisplayable.

than

4095

(1023

for

the

AR11)

A point with a value greater than 4095

treated as

its value were -1l.

Each point

intensified every npoints

interrupts.

(1lU23)

The

are

amount

of flicker in the display depends on the number of points and the
Nondisplayable points
persistence of the phosphor in the scope.
but are processed as if they were appearing on
ignored
not
are
the
reduce
to
try
1If flicker is a problem, first
the screen.
number

nondisplayable points being

"displayed."

Example:

1.
The example draws a horizontal line and a vertical line to
divide
the
screen. into
quadrants.
The
display
1is
stopped by SDIS in a
completion routine that is designated in the call to SETR and executed
after

approximately

one minute.

RT-11

EXTERNAL

FORTRAN

EXTENSIONS

STOPS

TNTEGER
CALL

XDATA (4U96) ,YDATA (40Y6)
CLRD (XDATA,
4096 ,TDUM, 0.0)

;Clear

array.

CALL

CLRD (YDATA,4096,TDUM,0.0)

;Clear

array.

T=44,4047,5U

;Make

horizontal

;Make

vertical

line.

XDATA
(T) =1

1lu

YDATA (T)=2044

T=73,4047,50

line.

XDATA
(1) =2048
20

YDATA
(T) =1

NPTS=160

;Number

;jdisplay.

points

TFLAG=0

;End

CALL SETR(5,0,6000.,IFLAG,,STOPS)

;Time 1 minute; then
;execute completion
;routine STOPS.
;Display the arrays.
;Wait for keyboard input.

CALL

DXY (XDATA,YDATA,NPTS,48,25)

PAUSE

;Go

flag

start

for

SETR.

display

again.

END

SUBROUTTNE
CALL

SDIS

RETURN
END

STOPS

;Completion
;Stop

the

routine.

display.

RT-11

FORTRAN EXTENSIONS

FLT16

l.4.0

FLT16

(l6-bit Floating-Point Conversion)

FLTlo converts

lo-bit

unsigned binary

integer

Routine

real-number

format.

A
le-pbit
wunsigned binary integer (with octal values U to 177777) has
decimal values of 0 to 65535,
(These values are in contrast
to
most
integers
which
are
stored in two's complement binary and can assume
decimal values -32768 through 32767.)
You must use FLT1lé for unsigned
binary
values
1instead
of
the
FORTRAN function FLOAT because FLOAT
converts two's complement binary, not unsigned binary.
FLT16

called

function.

rvalue = FLT1lo (ivalue)
The

arguments

are:

rvalue

The

real-number

ivalue

The

unsigned

real

value.

Error

form of

binary

the

integer

integer
to

argument.

converted

conditions:
SYNTAX

ERROR

There

not

exactly

one

argument

present.

Interaction:

Unsigned

binary

values

are

acquired

binary

value

IDIR,

Restrictions:
None
Example:

Consider
1111

the

111

unsigned

111

IBINRY:

111

VALUE=FLT16 (IBINRY)

then VALUE

contains

the

real-number

form of

65535,

form of

-1.

VALUE=FLCAT (IBINARY)

then VALUE
2.

contains

Consider
1

0uU

the

VU0

the

real-number

unsigned

OVUO

OULU

00U

binary

value

IBIN:

DRS,

and

IDOR.

RT-11

FORTRAN

EXTENSIONS

VAL=FLT16 (IBIN)

then VAL contains

the

real-number

form of

32768.

1f
VAL=FLOAT (IBIN)

then VAL contains the real-number
3.

You

want

DIMENSTON
TFLG=0

determine

the

time

form of -32768.
interval

IEVENT (4)
'

ISFLG=U

CALL

between

two

events.

;Set
;End

up data array.
flag for DRS.

;End

flag

DRS(IEVENT,4,,2,,1,-1,13,1IFLG,NLEFT)
;Read
;word

for

SETR.

;jstamping

event-driven

;mode.

CALL

SETR(4,1,1.,ISFLG)

CALL

LWAIT(IFLG,O0)

TIME=FLTlo (TEVENT (4) )~FLT16 (IEVENT (2))

;SETR runs the software
;iclock in msec.
iWait for end of

;sampling.
;Interval

msec.

1Go on to examine
;digital

the

registers.

RT-11

FORTRAN EXTENSIONS

FSH

1.4.7

FSH

(Flash)

Routine

FSH displays all points

in a data array once,

that

is,

"flashes"

them

on
the
screen.
However, you can use FSH for continuous displays if
you have a storage scope (for example, Tektronix 603), or if you use a
completion routine to refresh the display for refresh—-type scopes (for
example,

Tektronix

604).

FSH displays Y—-axis data on the scope.
The routine
assumes
that
if
X-axis
values
were provided, they would be equally spaced increasing
values.
Logically, there are two X values for each ¥
value
plotted.
One
specifies
the
X
<coordinate
on
the
screen
where the Y value
appears, and the other specifies the array subscript where the Y value
can be found.
The X coordinate and array subscript for a single point
usually have different values.
You do not have to provide explicit
X
values.
The
X-coordinate
value
1s controlled by an X-axis spacing
factor, whereas the array subscript value is controlled by an indexing
factor (see the ispace and increment argument descriptions).
FSH

called

subroutine.

CALL FSH(iydata,ispace,npoints,istart,increment)

The arguments are:
iydata
ispace

‘

The name

scaled Y~axisldata

1integer

array

The

factor.

You must

X-axis

calculate
is
1
and

the

spacing

the

X-axis

(npoints

containing
call CLRD

the
to

1ispace
before you call FSH.
TIf ispace
1increment
1is
1,
then
the
X~axis

coordinate
for
0,1,2,..., (npoints

values

be displayed.

the
-1).

display
If ispace

coordinate

for

the

assumes
values
is greater than

display

assumes

ispace*increment,?2*ispace*increment,...,
1l)*ispace*increment.

npoints

The number of
argument must

points
to
display.
be greater than zero.

The

npoints

istart

The array subscript
of
the
first
point
to
be
displayed from iydata.
The istart value must be a
legal subscript within the
bounds
of
the
array
iydata.

increment

The indexing factor for the array subscripts.
FSH
adds
1increment
to
the
preceding
subscript
to
obtain the subscript
of
the
next
point
to
be
displayed.
The starting subscript is istart.
1If
increment is 1, then
the
points
displayed
have
subscript
values istart through (istart + npoints
-

1).

TIf

increment

greater

subscripts
have
values
increment* (npoints - 1).
if

the

array

subscript

than

then

the

1istart
through istart +
No error message appears

goes

out

bounds.

RT-11

Error

FORTRAN

EXTENSIONS

conditions:
SYNTAX

ERROR

There
ARGUMENT

are

not

exactly

points

five

arguments

present.

displayed

(npoints)

ERROR

The

number

zero

The

greater

increment

greater

The

than

for

than

the
the

less

than

40Y6.

array subscript

1is

less

than

4uYe.

subscript

less

than

first

point

displayed

(ifirst)

Interactions:
A

CLRD

have
You

call

calculating

preceded
can

the

FSH

obtain

FSH

from

calling

the

display

continuous

display

completion

whenever
your
completion
routine
completion-routine
execution
(the
parameters

you

the

routine

SYSLIB

spacing

factor

(ispace)

call.

specify

when

you

refresh-type

routine.

FSH

refresh

initialize

scope

then

executes.

must

executes

The

rate
depends on

rate)

the

clock

(in

ITIMER).

of
the

SETR

Unlike

continuous displays
with
DIS
and
DXY,
the
completion
routine
method
of obtaining continuous displays may not use all
the machine resources.
The amount of time remaining
depends
on
the
number
of
points
1in
the
display
and
on
how often the
completion routine executes.
Your FORTRAN program might be
able
to

process

the

screen.

data

perform

computations

while

the

display

Restrictions:
The

range

the

range

example,

voltages

accepted by some
voltages
produced
by

full

scale

-5.12
to
5.12
voltages in the

range

and

through

scopes,

the

arguments

and

correspond

array
X

with

data

Data

values

the

that

the

range

its

narrower

than

converters.

For

converter

may

represent

value

were

with

these
47

data

that

the

off

screen

Fill

values

scaled

not

the

ignored

screen.

number

1If

but

are

flicker

(1023

value

values

for

greater

the

than

AR11)

are

4095

(1023)

-1.

nondisplayable

processed
a

the

have

4047,

kach point is intensified every npoints interrupts.
The
of flicker in the display depends on the number of points
persistence of the phosphor in the scope.
Nondisplayable
are

problem,

points

the

positions

array.

The

those

edges

arrange

displayable

4047.

accept
values

the

scopes,

through

displayed

hang

first

4095

with

not

the

are

values

data

through

than

point

4095

those

data

values

greater

nondisplayable.
treated

through

losing

displayable

the

1is

D/A

the
scope control may
volts.
Therefore, digital

nondisplayable

coordinates

should

including

avoid

for

the

However,

-5
4048

displays

screen.

4u95)

volts.

and

scopes

being

they

first

were

try

amount

and the
points
appearing on

reduce

the

"displayed."

Examples:

The

storage

following
scope.

example
The

example

complete
erases

1-33

short
the

program

screen

and

designed
plots

for

diagonal

RT-11 FORTRAN EXTENSIONS
at
line of points that remain on the screen until you press aonlykey,
way to
which time it erases the screen and replots the line. The
the
a CTRL/C,

After you enter
1f you want to clear the screen, type
for
system
the
reboot
or
INTT after the monitor dot for RT-11 SJ,

stop the program is to enter a CT

scope

RT-11

in storage mode.

remains

FB.

INTEGER SCREEN (4096)
CALL

ERASE

CALL CLRD (SCREEN,4096,TIDELT,0.U)
DO 10

T=45,4047,3

;Set

space.

array

:Be sure screen is blank.
sMake SCREEN nondisplayable.
;DO limits chosen to fill
;jscreen.

=1
(T) N
SCREE

;Create points for diagonal
;Print

NPT=1333
TYPE 20,NPT
NO.
FORMAT (/"

;line.
;No. of points

to display.

message.

OF POINTS DISPLAYED =',15)
;Display
CALL FSH (SCREEN,IDELT,NPT,4t,3)

array.

for

keyboard

entry.

PAUSE

;Wait

CALL ERASE
GO TO 30

;Go display line again.

;Erase

screen.

END

SUBROUTINE

ERASE

;You write this

routine.

IFLAG=0

;Initialize flag for SETR.

CALL IDOR(1,"170446,2,0)

:Clear erase bit thus starting

CALL IDOR(1,"1704406,"0,"0)

CALL SETR(4,0,250.,IFLAG)
CALL LWAIT(IFLAG,O0)

;Turn on storage mode and
;non—-erase flag.
;jerase

cycle.

;Start 250 msec interval to
;swait for settling time.
;Wait for end of interval.

RETURN
END

2.
on

The following example shows the use of FSH for continuous displays
refresh-type scope (for example, Tektronix 604).

FSH

from within a completion routine which is executed on interrupts

called
from

the programmable clock.
EXTERNAL

COMMON

SHOW

TYDATA (4V96),N,ISP

CALL CLRD(IYDATA,4096,18pP,0.)
DO 1u I=48,4047,20

;Define the completion
;routine.
;Make array available to

;completion routine.

;Make TYDATA nondisplayable.
;DO limits chosen to fill

;jscreen.

TIME=100.

;Fill array with points.
;No. of points to display.
;Time interval between calls

IFLAG=U

;msec).
;Flag for

=1
LYDATA(I)

N=200

CALL SETR(4,1,TIME,IFLAG,,SHOW)
PAUSE

CALL SETR(-1,,,)

;to the completion routine
SETR.

;Get clock going and prepare

;for

;Wait
:Stop

completion routine.
for
the

keyboard
clock.

entry.

END

SUBROUTINE SHOW
COMMON TYDATA (4096),N, ISP

CALL FSH(IYDATA,ISP,N,48,20)
RETURN
END

(in

;Completion routine.
;Display the

array.

RT-11

FORTRAN

EXTENSIONS

FXY

1.4.8
FXY

FXY

(Flash X-Y Data Pairs)

displays

screen).
a storage

X-~Y

data

completion routine
example, Tektronix
You

can

pairs

However, you
scope
(for

use

FXY

associated

plot

any

equally

scatter—-plot

data).

FXY

X-Y

displays

once

data

(that

1is,

"flashes"

can use FXY for continuous
example,
Tektronix
©603),

to refresh
6U04).

with

Routine

the

display

data

1in

spaced,

pairs

for

which

the

two

the

scopes

wvalues

X values

data

if you have
vyou
use
a

refresh-type

increasing

contained

them

displays
or
1if

are

(for

arrays.

(for

not

example,

The

X-axis

data
array
contains
the horizontal displacement of each point to be
plotted.
The Y-axis data array contains the vertical displacement
of
each
point to be plotted.
The displacements are relative to (0,U) 1in
the
1lower
1left
corner
of
the
screen.
The
X~-Y
pairs
are
in
corresponding
positions
1in
the
two arrays.
That is, for the fifth
pair to be plotted, the X value is the fifth element of the
X
array,

and

the Y value

The

data

must

that

values

1is,

compress

the

CLRD

elsewhere
FXY

fifth element of

4095

range

scale

the

range

that

1023

for

and

the

them.

the Y

You

can

data

also

can

array.
be

plotted

the AKll).

arrays

transform

You

the

independently
or

scale

scope,

can expand

the

using

data

arrays

the program with FORTRAN manipulations.

called

subroutine.

CALL FXY(ixdata,iydata,npoints,istart,increment)
The

arguments

ixdata

are:

The

name

X-axis

iydata

The

name

Y-axis

the

values

for

1integer
the

the

values

for

array
pairs

1integer
the

npoints

The number of points

istart

The

array

X-Y

subscript

X~Y

the

array
pairs

first

increment

the

displayed.

to be plotted.
pair

be
plotted.
The istart argument
subscript within the bounds of the

and

the

displayed.

containing

(X-Y pairs)
of

containing

points

must be
arrays

legal
ixdata

iydata.

The indexing factor for the array subscripts.
FXY
adds
1increment
to
the
preceding
subscript
to
obtain the subscript of the next pair of values to
display.
Tf
1increment
1is
1,
then
the
pairs
displayed have
subscript
wvalues
1istart
through
(istart + npoints - 1).
If
increment 1is greater
than

then

the

through
Increment

must

error

message

out:

bounds.

subscripts
(istart

greater

appears

have

values

istart

increment* (npoints
than

equal

the

array

1).

subscript

goes

RT-11 FORTRAN EXTENSIONS

Error

conditions:
SYNTAX

ERROR

There are not exactly five arguments present.

ARGUMENT

ERROR

1is

The number of pairs of points to display (npoints)
than 1 or greater than 40Y6.

The array subscript
(istart)

is less than 1.

the

£first

pair

less

displayed

The increment requested is greater than 4U96.
Interactions:

You can obtain a continuous display on a rerresh-type scope by
FXY then executes
calling FXY from a completion routine.

of
rate
The
whenever your completion routine executes.
completion-routine execution (the refresh rate) depends on the
parameters you specify when you initialize the clock (in SETR or
the SYSLIB routine ITIMEK).

the
DXY,
and
DIS
with
displays
continuous
Unlike
completion-routine method of obtaining continuous displays may
not use all the machine resources. The amount of time remaining
depends on the number of points in the display and on how often
the completion routine executes. Your FORTRAN program may be

able to process data or perform computations while the display is

screen.

the

Restrictions:

than
1is narrower
The range of voltages accepted by some scopes
For
the range of voltages produced by the D/A converters.
represent
may
converter
the
for
4095)
to
(0
scale
example, full
to

-5.12

+5.12

volts.

However,

voltages in the range -5 to +5 volts.
U

through

the scope control may accept

Therefore, digital

wvalues

47 and 4048 through 4095 cannot be displayed on those

scopes, and displays including those values hang off the edges of
the screen. To avoid losing data with these scopes, arrange the
arguments and data so that the first 47 screen positions
Scale both
correspond to nondisplayable data in the data array.
range 48
the
in
values
e
displayabl
have
to
the X and Y arrays
through

Data

4047.

values

greater

nondisplayable.

than

4095

(1023

for

the

AR11)

A point with a value greater than 4095 (1023)

treated as if its value were -1.

Each point is intensified every npoints interrupts.

The

are

amount

of flicker in the display depends on the number of points and the
persistence of the phosphor in the scope. Nondisplayable points
are not 1ignored but are processed as if they were appearing on
1If flicker is a problem, first try to reduce the
the screen.
number of nondisplayable points being "displayed."

Examples:

The following example shows the use of FXY for continuous displays
1.
FXY is called from within a completion
with a refresh-type scope.
The
routine which executes on interrupts from the clock counter.
example program draws a horizontal line and a vertical line to divide
the

screen

into quadrants.

RT-11

COMMON

FORTRAN

EXTENSIONS

TXDATA (40906),IYDATA (4096),N

sMake

arrays

jcompletion
EXTERNAL

SHOW

;Define

available

routine.

completion

routine

sexternal.

CALL CLRD (TXDATA,4096,TISP,U.)
CALL CLRD(IYDATA,40%6,1I5P,U.)
DO 10 T1=48,4047,5u

;Empty X array.
;Empty Y array.
;Create horizontal

line.

IXDATA (T)=1

TYDATA(TI)=2u4s
DO

T1=73,4047,50

;Create

vertical

TYDATA(I)=I
N=loU

iNo.

points

IFLAG=0

;End

flag

line.

IXDATA (T)=204Y

TIME=100.

;Refresh

for

display.

SETK.

rate

luu

clock

iticks.
CALL

SETR(4,1,TIME,TIFLAG,,SHOW)

iStart

clock

msec,

;designate completion

;routine
PAUSE

CALL

SETR(-1,,,)

SHOW.

;Wait

for

keyboard

;Stop

the

clock.

entry.

END

SUBROUTINE

COMMON

SHOW

;Completion

IXDATA (40906),IYDATA (4096),N

;Data

routine.

arrays

from

main

;orogram.

CALL

FXY(IXDATA,TY¥DATA,N,48,25)

;Display

arrays.

RETURN
END

The

following
storage

example

scope.

with sampling started by
pulse).
You can compare
the

scatter

the

data

use

are

FXY

collected

an external event
the two sweeps of

(for
data

display

from

data

analog

example,
a
stimulus
visually by looking at

IDATAL(.100G),IDATA2(100)

;Define

the

data

arrays.

ICLCK=0

;End

flag

for

SETR.

flag

for

RTS.

TEND=0

;End

NPT=100

;Length

CALL

iRequest

CALL

arrays

channel

plot.

DIMENSTON

shows
The

SETR(5,5,1.,LCLCK)

;with

data

clock

RTS (IDATALl,NPT,,,2,,1,2,TEND,IDUMMY)

sweep.

external

start

10UHZz.

;Request single sweep
;100 points on clock

;interrupts.
CALL

LWAIT(IEND,O)

;Wait

for

samples.

CALL

SETR(-1,,,)

;Stop

the

clock.

TCLCK=0

iReset

clock

flag.

CALL SETR(5,5,1.,TCLCK)
iRequest clock
TEND=0
iReset RTS end
CALL RTS (IDATA2,NPT,,,2,,1,2.TEND,IDUMMY)
;Request
CALL

LWAIT(IEND,O)

;Wait

for

CALL

SETK(-1,,,)

;Stop

the

CALL

ERASE

;Erase

CALL

FXY(IDATAl,IDATA2,NPT,1,1)

;Display

again.
flag.

2nd sweep.
2nd sample.

clock.

the

screen.

scatter

character

plot.

PAUSE

:Type

CALL ERASE
GO TO 10

;Clear screen.
sKepeat whole program.

finish.

END

SUBROUTINE
TFLAG=0

ERASE

;Routine
sEnd

flag

clear

for

screen.

SETR.

RT-11

CALL

FORTRAN

IDOR(1,"170446,6,6)

EXTENSIONS

;Turn

snonerase
CALL
CALL
CALL

storage

IDOR(1,"170446,2,0)

;jClear

erase

SETR(4,0,250.,IFLAG)

jerase
;Start

cycle.
250 msec

sto wait

LWAIT(IFLAG,O0)

;Wait

RETURN
END

1-38

for

mode

and

flag.

for

end

bit

start

interval

settling time.
of

interval.

RT-11

FORTRAN

EXTENSIONS

HIST

1.4.9

HIST

(Time

Interval

Sampling Technique)

Routine

HIST reads the current value of the programmable clock counter on each
and
stores
the
time
readings
1in
a
data array.
clock
interrupt
Depending on the clock
mode,
you
<can
time
the
intervals
between
(modes 3 and 7) or you can determine the latencies
successive
events
of a series of events relative to a single event (modes 2 and 6).
HIST

called

CALL
The

subroutine.

HIST (iarrayname,iarraysize,nsamples,iendflag#,nleft#)

arguments

are:

iarrayname

iarraysize

The

name

time

data.

The

length

must

samples

integer

array

iarrayname.

greater

The number
Normally,

nsamples

the

than or

of time
HIST

requested

The

equal

interval
continues
have

which

been

HIST

places

1arraysize

value

values
to
collect.
sampling
until
all
read.

However,

you

can
stop
sampling
before
all samples have been
collected by
calling
HIST
again
with
nsamples
equal to zero.
(In this case, the arguments other
than
nsamples
should
be
the
same
as
1in
the
previous call to HIST.)
The nsamples value may be greater than iarraysize,
in
which
case
HIST
treats
the data array as a
circular array
(see
Section
1.2.6).
You
must
write your program to empty the array periodically
or the data will be overwritten.
HIST itself does
not provide any completion-routine option, but you
can designate a completion routine in the call
to
S5ETR.

iendflag

The

completion

nleft

The

flag

and

error

flag.

You

must

set

HIST
HIST.
<call
you
before
2zero
to
iendflag
increments iendflag
when
all
samples
requested
have
been
collected.
1If a data overrun or other
error occurs, then HIST sets iendflag to -1.
containing

available

for

initializes nleft
HIST
decrements
collected.

the

data

Data

number

before

array

elements

HIST
to
the
wvalue
1in
1iarraysize.
nleft
after
each data point is

overrun

data

occurs

when

overrun.

nleft

becomes

zero
before
all the samples have been collected.
Lf you are removing
data
from
1iarrayname
while
HIST is running, you must increment nleft once for

Error

each

point

not

exactly

you

remove.

conditions:
SYNTAX

ERROR

There

are

five

1-39

arguments

present.

RT-11

ARGUMENT

FORTRAN

EXTENSIONS

ERROR

The
The

array size requested is less than 1.
number of samples requested is less than

zero.

Interactions:

HIST

runs

conjunction

with

SETR.

You

must

call

SETR

set
the clock in an external-event timing mode (mode 2 or 3) and
to set the clock rate.
You connect the sensor for your
external
events
to
Schmitt
trigger
#2
of
the
<clock
counter.
The
occurrence of an external event fires Schmitt trigger #2, causing
the
time of occurrence to be recorded.
TIf you want to start the
clock with an external—event pulse, connect that event to Schmitt
trigger
#1 and specify one of the external-start clock modes for
external—-event

timing

HIST

driven

(mode

interrupts

or
from

7).
the

_
clock

counter.

Therefore,

a
completion routine designated in SETR to service clock Schmitt
trigger #2 interrupts can process the data acquired by HIST.
For
example,
vyou
can
tally
the data for time interval histograms.

You should specify
iarraysize so that
is half full.

an interrupt interval in SETR that is half
of
the completion routine executes when the array

Restrictions:

HIST

does

not

work

for

the

ARI1l.

The

maximum possible time interval that can be
recorded
depends
on
the
clock
counter
rate.
The
maximum
count at any clock
counter rate is 65535.
The faster
the
rate,
the
shorter
the
actual
1interval;
the
slower
the
rate, the longer the actual
interval.

You

cannot

detect

clock

overflow.

When

the

increments
from
65535
to
0
and continues
counter does not give you any signal.

clock

count,

counter

the

clock

The

clock

mode

(mode

Examples:

The

following

counter
3).

1is

When

example

running

event

collects

external

occurs

Schmitt

current
clock counter value in
automatically resets to zero.
DIMENSTION

event
the

time-interval

timing

from

trigger

array

ITIMES (40)

values.

zero

#2,

ITIMES

HIST

and

;Define

base

the

data

places

clock

array.

TFLAG1=0

;HIST

completion

TFLAG2=U

;SETR

completion

CALL

HIST(ITIMES,40,4U,TIFLAGLl,IFULL)

;Request

CALL

SETR(5,3,0.,IFLAG2)

;Start

40U

the

counter

flag.

flag.
values 1in

; ITIMES.

clock

;external
;0 mode.

CALL
CALL

event

base

the

samples

;Wait

until

SETR(-1,,,)

thave
;Stop

been collected.
clock when done.

iprocess

the
or

from

LWAIT(IFLAGLl,0)

;Continue

all

1U0

program

store

the

to
data.

RT-11

FORTRAN

EXTENSIONS

IADC

1.4.10

IADC

(Single

Analog-to-Digital

Conversion)

Routine

IADC acquires a single digitized value from an
A/D
converter.
(You
can
acquire
multiple
digitized
values using the RTS routine.)
The
program requests conversion, pauses within the IADC routine until
the
conversion is complete, and returns the digitized value as an unsigned
integer (0 to 1023 from the ARll, or 0 to
4095
from
the
ADll-K
or
LPS1ll).

If the A/D converter is operating in unipolar mode, a
value
of
=zero
represents
0
volts,
and
a
value
of
4095
(or 1023 for the AR1ll)
represents the full scale voltage.
1If the A/D converter is
operating
in
bipolar
mode,
a
value
of
=zero represents the maximum negative
voltage, and a value of 4095 (or 1023) represents the maximum positive
voltage.
Zero voltage corresponds to a value of 2048 (or 512).
IADC

subroutine:

called

CALL

subroutine

function.

IADC (ichannel, [igain],ivalue#)

function:
ivalue

The

either

arguments

ichannel

IADC(ichannel[,igain])
are:

The

logical

channel

number

The

channel

numbers

are:

the

Value
0

through

A/D

converter.

Meaning
63

Channel
LPS11-E

numbers
or

for

ADl11-K

the

LPS11l

with

AM11-K

used

in
single—ended
pseudodifferential mode.
0

through

Channel

numbers

AD11-K

with

gain

for

AM11-K

the

with

LPS1l1l

switched

operation.

Channel

numbers

for

the

ARll

and

AD11-K,.

0,2,4,,,62

Even-numbered
AD11-K

differential

0,2,4,,,14

channels

with

AM11-K

for
1in

the
true

mode.

Even—-numbered
AD11-K

wused

with

channels
AM11-K

differential
mode
gain operation.

with

for

the

used

switched

RT-11 FORTRAN EXTENSIONS

|is
The default value for igain
The gain setting.
converting the value is
for
used
gain
The
1.
The
1ivalue.
of
12
and
13
bits
in
returned
values of bits 13 and 12 are 00, 01, 10,
possible
These
decimal).
3
and
and 11 binary (0, 1, 2,

igain

and 4
3,
2,
1,
of
to gain codes
correspond
16,
respectively, or to the actual gains of 1, 4,
When you request a gain of 1, bits 13 and
and 64.

12 always
use
must

return as O.
For all other
gains,
you
separate the converted value
to
CVSWG

from the gain bits.

The

gain

codes

are:

Meaning

Value

software
The
autogain.
Request
optimal gain to use for
the
determines
is
Autogain
conversion.
A/D
the
available
only
with
switchable
gain
LPSI11
or
AM11-K,
systems (AD11-K with

with

Request gain of 1.
This value
bipolar mode for the ARlL.

Request gain of 4 to multiply the
input
specifies
value
This
4.
by
signal
bipolar mode for the ARI1l.

Request gain of 16 to multiply the input
specifies
value
This
1l6.
by
signal
unipolar mode for the ARIl.

Request gain of

to multiply

value
This
64.
by
signal
unipolar mode for the ARILI.

specifies
,

the

input

specifies

The name of a variable
containing
the
digitized
value in bits 0 through 11 (or 0 through 9 for the
AR1l) and the gain code in bits 13 and 12.
TIf you
subroutine, you must include the
a
as
IADC
use
use
you
T1f
argument ivalue to recover the data.
Iis
1ivalue
argument
the
function,
a
as
IADC
redundant (because ivalue is returned as the value
of the function) and can be omitted.

ivalue

Error

LPSSG).

conditions:
SYNTAX

ERROR

There

are

arguments,

there

are

than

three

arguments.

ARGUMENT

ERROR

The channel number

is greater

than 15 for

The

1is greater

than

without AMI1-K.

channel number

or the LPS11.
gain setting is negative

for

the AR1l or AD1l1-K
the

AM11-K,

The

greater

than

ADl11-K

with

RT-11

DEVICE

FORTRAN

EXTENSIONS

CONFLICT

Autogain

was

requested

for

the

AR1ll

specifying

1igain

Zero.

ADC

CONFLICT

A/D

conversion

already

being

done

under

control

RTS.

Interactions:
You

RTS

cannot

write

the

hardware,

program

same

and

only

time.

one

that

would

IADC

the

and

attempt

call

RTS

the

same

routines

use

can

execute

TIADC

and

conversion

one

time.

When the value of igain is not 1, TADC returns a value that is
a
combination of the digitized value and the gain setting.
You can
unpack the digitized value in real-number form and the gain
used
by calling the function CVSWG (see Section 1.4.2).

With gains other than 1,
input
signal
by 4, 16,
from CVSWG, you
64)
if you need

the A/D conversion system multiplies the
or 64.
When you get the digitized value

must divide by
to restore the

the appropriate gain
value to the original

(4,
16,
scale.

Restrictions:

Autogain mode does not work with
unipolar

mode.

the

Autogain

AR1l.

When

specify

you

not

requested must

available
any

for

gain

the

other

range

AD1l-K

than

the

LPS11

channel

number

15.

The

AD1l-K with AM11-K hardware can be set up with gains of 1, 5,
and
50
instead
of 1, 4, 16, and 64.
Autogain does not work in
this case.
The meaning of gain codes 2, 3, and 4 depends on
the
actual gain selected for each group of channels in the AM11-K.

Examples:

Collect

digitized

reading

from

channel

2 with

gain

VALUE=CVSWG
(IADC (2))

2.
the

Collect

value

digitized value from
the original scale.

VAL=CVSWG (IADC(1,0) ,IGAIN)

VAL=VAL/
(2** (2* (IGAIN-1)))

channel

1 with

autogain

and

convert

RT-11

FORTRAN

EXTENSIONS

IDIR

1.4.11

1IDIR

(Digital

TDIR provides

The

four

description

Input

distinct

IDTR

Reading)

capabilities,

presented

the classes of capabilities.

IDIK

read

digital

using

IDIKR to

1input

second describes using IOTIR for

describes

Reading

a Digital

Routine

read

the

Input Register

The

two

three

first

which

section

are

sections,

for

describes

a memory

location.

bit and byte manipulation.
software

related.

one

The

clock.

each

using
The

third

a Memory Location

IDIRK reads one data word from a
digital
1input
register
(DR11-K
or
LPSDK~A)
or
from a memory location.
1t performs a logical AND using
the data value and a mask specified in the argument list.
IDTR

ivalue
The

called

function.

IDIR([isource],iunit,imask, [itype][,iwhere#])

arguments

ivalue

isource

are:

The masked register or memory
value
returned
IDIR.
1IDIR performs a logical AND using imask
the register or
memory
value,
and
returns
result as the value of the function.
The code specifying whether to
a
memory
value.
The default

and

the

read a register
or
value of isource is

zero.

The

location

codes

are:

Value

zZero

Meaning

Read

the

logical
process
clears
that it
nonzero

iunit

The
be

specified

Read the memory location
appears in iunit.

number

values

for

the

address

possible

iunit

whose

the

address

value

are:

Value
through

Meaning
7

or
0

the

read.

The

wunit
number
1in iunit.
1In the
of reading
the
register,
1IDIR
those bits (and only those bits)
finds set.
'

through

Logical

unit

digital

input

0-through-7

numbers

values

up
to
eight
interfaces.

for

the

registers.

The

select

one

DR11-K digital 1/0
1t
the
system

RT-11 FORTRAN EXTENSIONS

Value

Meaning

configuration
with

LPSDR-A

one

and

then

up to eight
interfaces.

argument

Memory

must

1isource

nonzero.)

other

with

can

read

the

select

DR11-K

digital

(The

isource

argument
can

the

for

(CSR)

example,
vyou

devices

supported

be
the

associated

AAll-K

bits,

registers

must
read

device.

status
not

LPS1l1

zero.)

You

control/status
some

selects

address
selected.
The
must
be an even number.

address

(The

through

1/0

U,2,...,"177776

includes

LPSDR-A,

vyou

LPSVC

can

read

that

the

are

FORTRAN

extensions.

imask

The

itype

wvalue
register or

wused
to
mask
the
memory location.
A

one

all

either

with
as

The
data
location.

The

data

bits

-1

on.

format
of
The default

types

Value

You

(decimal)

the

Read

unsigned
memory

Read

position
in
the

location

BCD

data
location.

(U to 15)
portion

that

was

IDIR

performs

value

and

the

Error

iwhere

itype

data

of
of
masked

bits

memory

=zero.

contains

the

convert

word is masked
binary.)

the
the
with

1logical

complement

from
and

from

the
result
from right to
If no bits were set in the

then

location

binary

memory

Bit
bit

all

"177777.

are:

binary.
(The data
it is converted to

iwhere

request

the
mask

Meaning

nonzero

can
as

value

zZero

contents
of
commonly used

AND

the

rightmost
nonzero
register or memory
=zeros.
That
1is,
using

the

mask,

left

for

part

masked

value

before

and

scans

nonzero

-1.

with

bit.

zeros,

conditions:
SYNTAX

ERROR

There

are

specified.

fewer

than

four

than

five

arguments

RT-11 FORTRAN EXTENSIONS

ARGUMENT

ERROR

The memory address specified in iunit is an odd number.
The register requested in iunit does not exist.

Interaction:
None

Restrictions:
None

Examples:
1.

input

Reading a digital

IDIR(0,0,"1777,1,ISPOT)

IREG

If the register contains the octal value 51324
0

and

101

the
U

001

mask
000

011

001

vlo

1777
111

10U

(octal)

111

1is

then the logical AND using the register and the mask

Because
1324

its

0V0

conversion

binary

IDIR also

was

specified by imode = 1,

1324.

octal

IDIR returns

value.

Olu

ull

001

examines

complement of

10U

the

the mask:

1 111
0 1ul
0 101

.not. mask
register
result

1logical

AND

using

the

in the word

rightmost bit

Reading a memory

the

110 000 00Qu WOUL
001 011 Olu 1lov
00U V00 000 000

Scanning the result from right to left, IDIR finds the
12 and returns 12 as the value of ISPOT.
bit
in
bit
the

and

is bit 0.)

nonzerc
first
(Remember that

location.

You want to obtain the contents of the status register for DR11-K unit
Zero.

MEMLOC=IDIR(-1,"167770,"177777,1)

Bit and Byte Manipulation of a Single Word
IDIR provides limited bit manipulation capability

the

swap

in FORTRAN.

You can

bytes of a single word or shift bits left or right within a

word.

IDIR

called

ivalue
The

arguments

ivalue

function.

IDIR(,iflag,ibefore,iop[,iresult#])
are:

The byte-swapped value or the logical value of the
last bit shifted, depending on the value of
iop.
1-46

RT-11

iflag

FORTRAN

EXTENSIONS

The

flag specifying the bit-manipulation and clock
functions
of
IDIR.
The
value of ifl ag must be

negative.

ibefore

The

value

which

bit

byte

manipulation

performed.

iop

The

manipulation

operation

The

manipulation

operations

Value

selected.

are:

Meaning
Swaps

the

upper

and

lower

ibefore,

bytes

and
returns
the new value as
the value
of
1IDIR.
The
byte-swapped
value
1is
also returned as the value of
iresult if that argument was
specified.

The

value

ibefore

does

not

change.

Shifts

ibefore m
places to
the right
and
returns
the
shifted
value
in
iresult.
IDIR
fills
1iresult from the
left with m zeros.
The bits shifted out

are

lost.

The

value

ibefore

does

not

change.

The
1logical
shifted
out
value

the

returned

1is

"false"

on,

and

Shifts
to

~16

The
The

ibefore

The

value

1logical

shifted

out

value of
returned

the
1is

on,

Error

function

and

The
value
bit-shifting

resulting
ibefore.

fewer

four

value

1is

off.

the

left

and

1in

iresult.

left

out

does
of

not

-m

1lost.

change.

the

last

bit

ivalue

(the

function IDIR).
or "true" if

from

with

are

The
value
the bit was

-1

was

returned

"false"

bit

shifted

ibefore

the

m places

bits

1last
bit
ivalue (the
The
value

IDIR).

"true"

returns the shifted
value
IDIR fills iresult from the
Zzeros.

iresult

value
of
the
1is returned in

the

bit

was

off.

byte-swapping

conditions:
SYNTAX

ERROR

There

are

present.

Interaction:
None

Restrictions:
None

than

five

arguments

RT-11 FORTRAN EXTENSIONS

Example:

1input
1. You want to find out whether or not bit 13 of digitalregister
whole
the
read
ly
previous
have
You
set.
register 2 has been

value in binary with the following call:
IREG=IDIR(,2,-1,1)

You test bit 13 by shifting the register value three places to the
left so that bit 13 is the last bit shifted out. You are not
interested in the shifted register value itself, so you omit the last
argument,

iresult:

IBIT=IDIR(,~1,IKEG,-3)

Reading the Software Clock

IDIR reads the current value of the software <clock and performs a
logical AND using the clock value and a mask provided in the argument
(See Section 1.2.2 on the software clock.)

list.

IDIR is called as a function.

iclock = IDIR(,iflag,imask,)
or

ivalue = IDIR(,iflag,imask,,iclock#)
The

arguments

are:

The masked clock value returned as

ivalue

the

value

1is returned in iclock if
The same value
IDIR.
IDIR
included in the «call.
that argument 1is
performs a logical AND using the software-clock
The value of the software clock
value and imask.
does

not

change.

iflag

The flag specifying the clock and bit functions of

imask

The value used to mask the

IDIR.

The value of iflag must be negative.

value.

software-clock

the
using
AND
logical
a
performs
IDIR
Request a mask
software-clock value and imask.
with all bits on either as -1 (decimal) or as
"177777.

iclock
Error

The masked

clock

value.

than

four

also

(This

wvalue

1is

than

five

arguments

returned as the value of the function.)

conditions:
SYNTAX

ERROR

There are fewer

present.

Interaction:

SETR is responsible for the software clock. The software clock
is running only if you have called SETR to turn on the

programmable clock counter in repeated interval mode.

RT-11

FORTRAN

EXTENSIONS

Restrictions:
IDIR and IDOR can both read the software clock.
capability
is
necessary
because
there
is
a

subroutines

called

and

call

the

same

you

read

from

within

from

IDIR

maybe

reset

redundant
restriction
on
FORTRAN
program

completion routines.
A
routine designated in the program
subroutine.
The redundancy of IDIR and

completion

This

the

software

completion

read

the

clock

both

routine.
within

within

from

For

the

within

example,

program,

completion

and

cannot

IDOR

both
allows

the

program

and

you

could

use

IDOR

read

routine.

Examples:

Reading

the

software

clock

IFL=0

CALL

SETR(4,1,1.,IFL)

ICLOCK=IDIR(,-1,-1,)

End

flag

for

SETR.

iRun

the

clock

smode

sRead

bits

msec

repeated

interval

rate.

software

clock.

and

RT-11 FORTRAN EXTENSIONS

IDOR

1.4.12

IDOR (Digital Output Register)

Routine

two of which are
IDOR provides three separate output capabilities,
Each of these output capabilities is analogous to an input
related.
The description of IDOR is presented
capability of the function IDIR.
The first section describes using IDOR to load a
in two sections.
The second section
digital output register or memory location.
describes using IDOR to read and reset the software clock.

Loading a Digital Output Register or Memory Location
the

IDOR changes

value

digital

output

memory

The new value of the register or location is determined by
location.
as by the
as well
iset,
iselect and
the value of the arguments
algorithm
following
The
location.
or
register
the
of
value
original

describes the function of the IDOR routine.

location
The bits set in iselect specify the bits in the register or
If the bit in iselect is 0, the corresponding bit
to be altered.
if the bit in iselect is 1,
position in the register is not changed;
the corresponding bit position in the register is altered according to
the value at that bit position in iset.

the bits
The values of the bits in iset specify the action taken for
0, the
1is
1iset
in
selected bit
the
If
1iselect.
in
specified
1if the selected bit in iset is
corresponding register bit is cleared;
1,

the corresponding

bit

set.

are
register
the
in
bits
then all
If all bits in iselect are 1,
If
altered according to the values of the corresponding bits in iset.
all bits in iset are 0, then all bits in the register corresponding to
For example, if iselect is -1 (all
set in iselect are cleared.
those
then the entire
bits are on) and iset is zero (all bits are off),
register

cleared.

The logical relationship among the arguments is shown in the following
equation:

ivalue =

(iset.AND.iselect).OR. (.NOT.iselect.AND. (old register))

IDOK is called as a function or
a

as a subroutine.

function:

ivalue = IDOR([idest],iunit,iselect,iset[,ioutput#])
As

subroutine:

CALL IDOR([idest],iunit,iselect,iset[,ioutput#])
The

arguments

ivalue

are:

The
IDOR.
function
the
by
The value returned
value of the function is the new value loaded into
location.
memory
or
the digital output register
The
and

new
value is a combination of iselect, iset,
or
register
the
of
contents
original
the

memory

location.

1-50

RT-11

idest

The

FORTRAN

destination

location)
of

EXTENSIONS

idest

of
is

(digital

the

output

Load

the

logical
nonzero

Write

logical

location

unit

aadress

specified

number

the

value

number

the

location

whose

iunit.

the

address

the

loaded.

Meaning

through

or
U

memory

iunit.

memory

appears

Value
0

default

Meaning

zero

The

zero.

Value

iunit

value.

through

Logical

unit

digital

output

O-through-7

numbers

interfaces.

0,2,...,"177776

I/0

the

LPSDR-A,

and

eight

I/0

interfaces.

system

through

DR11-K

(The

select

digital

idest

wvalue

zero.)

Memory

address

must

(The

one

digital

includes
an LPS11
then 0 selects
the

one

select

DR11-K

configuration
with LPSDR-A,

must

the

The

values

eight

for

registers.

selected.
be

idest

even

value

The
number.

must

nonzero.)

iselect

The

bits

memory

altered

location.

The

the

bits

output

that

are

select

the

corresponding
bits
1in
the
register
or
memory
location
to
be altered.
To select all bits, you
can use either -1
(decimal) or "1777717.

iset

The

action

for

selected

ioutput

taken

for

alteration.

bit

the

selected

The

value

is
bit

returned

following

bit

the

cleared;

ioutput
argument
the digital output
The

each

selected

bit

the

iselect

iset

bit

iset

the

set.
by

the

contains
register

function

the
or

relationship

IDOR.

The

value loaded into
memory
location.

defines

the

value

ioutput:

ioutput

(iset.AND.iselect).OR. (.NOT.iselect
.AND, (0ld

Error

conditions:
SYNTAX

EKROR

There

are

present.

fewer

than

four

than

five

arguments

RT=-11 FORTRAN EXTENSIONS

ARGUMENT

ERROR

The DR unit addressed does not exist.

The memory address specified was not an even number.
Interaction:
None

Restrictions:
None

Examples:

Clearing a register or memory location.

The call

Clearing a register.

(a)

CALL

IDOR(,1,-1,0)

is
argument
1ioutput
the
that
Notice
1.
clears logical register
argument when you have no need for the
ioutput
the
Omit
omitted.
(b)

the

into

that was loaded

value

Clearing a memory

location.

The

call

IMEM=IDOR (15,"177xxx,-1,U)

clears memory location 177xxx.
zero
2.

result

the

digital

The variable IMEM contains

the

value

call.

output

the
of
byte
You want to set the even-numbered bits in the low-order
clear the even-numbered bits in the high-order byte of
and
register,
The odd-numbered bits in the register remain unchanged.
the register.
Suppose the current value of the register is 177400.
CALL

IDOR(,U,"52525,"377,INEWR)

and
function)
register
Notice that idest is defaulted (indicating a
The value of iselect
zero (indicating logical register 0).
1is
iunit
is

52525

and

iset

377

(both octal).

"The following shows the binary values of the arguments and the
of

the

call

the

Name

Binary Value

0
0
1
1

iselect
iset
0ld register
new register

The new value loaded into
returned

Reading

and

the

value

Resetting

the

101 010
000 000
111 111
010 101

the

1ul
111
vuo
1vul

1uvl 010
011 111
10vu 000
001 010

125125

(octal)

clock

value

and

INEWR.

Software

Clock

IDOR reads the current value of the
software
<clock
and
logical
AND using the clock value and a mask provided in
list (see Section 1.2.2 on the software clock).
IDOR can

software

action

value:

to a value provided

the argument

performs
a
the argument
change
the
list.

RT-11

IDOR

function:

called

ivalue
As

function

FORTRAN

EXTENSIONS

subroutine,

IDOR(,iunit,imask,
[iset] [,iclock#])

subroutine:

CALL IDOR(,iunit,imask, [iset],iclock#)
The

arguments

are:

ivalue

The

value

returned

performs

current
reading

value
of
the
<clock

clock with the
is specified.
iunit

imask

The

flag

The

value

The

value

mask

iset

iunit

used

imask

iset,

the

clock

must

and

the

clock
on,

loaded

into

the

clock

with

iset

change

the

clock

between

wvalue
a

value.
you

IDOR.

the

logical

AND

¢an

software
the

argument

use

reguest

either

-1

clock

value

list,

iset.

IDOR

IDOR.
If

does

you

not

value.

The masked clock value.
AND

argument

"177777.

from

the

After
software

iset

funection

current

bits

The

sets

the

performs

value

IDOR
and

negative,

the

IDOR

all

IDOR,
imask

the
software
clock.
value, IDOR loads the

mask

clock.

Ffuncotion
between

(decimal)

omit

Error

with

IDOR

iclock

the
AND

value

selecting

software

between

logical

imask

and

four

IDOR performs

the

current

a logical

clock

value.

conditions:
SYNTAX

ERROR

There

are

fewer

than

five

arguments

This

redundant

present.

Interaction:
None

Restrictions:
IDOR

and

IDIR

can

both

read

the

software

clock.

function
is
necessary
because
there
1is
a
resgtriction
on
subroutines called from completion routines.
A
FORTRAN
program
and
a
completion
routine designated in the program cannot both
call the same subroutine.
The redundancy of IDOR and IDIR allows
you

from

to
read
within a

the

software clock both from within
completion routine.
For
example,

IDOR

read

and

reset

the

IDIR

read

the

clock

within

clock
the

1-53

within

program.

the
you

completlon

program

and

¢ould

use

routine,

and

RT-11

FORTRAN

EXTENSIONS

Examples:

You want

read

the

lo-bit

software clock

without

changing

its

value.

ICLOCK=IDOR(,-1,-1,)

2.
and

You want to read the rightmost
reset the clock to zero.
ICLOCK=IDOR(,-1,"377,0)

eight

bits

the

software

<clock,

RT-11

FORTRAN

EXTENSIONS

KB2BCD

1.4.13

KB2BCD

KBZ2BCO

converts

(Convert

(decimal)

KB2BCD

called

Binary

lo-bit

4-digit
as

BCD)

binary

unsigned
a

Routine

integer

BCD

the

range

9999

form.

function.

1digits=KBZBCD (ibinary)
The

arguments

are:

idigits

The

4-digit

unsigned

BCD

representation

iboinary.

ibinary

The

standard

The

wvalue

binary

ibinary

integer

exactly

one

argument

value

must

to
in

be
the

converted.
range

Y999,

Error

conditions:
SYNTAX

ERROR

There

not

ARGUMENT

ERROR

The

value

than

converted

output

present.

(ibinary)

negative

BCD

values

greater

9999,

Interaction:
You

can

use

IDOR

unsigned

returned

KB2BCD.
Restrictions:
None

Example:

The

U
You

decimal

UUl

want

Ju0

value

0ul

convert

IBIN

4175

(or

10117

octal):

111

IBIN

BCD

form

for

value

for

4175

(shown

output

BCD

instrument.

IBCD=KB28CD (IBIN)

IBCDL

contains

the

BCD

bits):
0luu

vuul

U111l

vlul

with

BCD

grouping

the

RT-11

FORTRAN

EXTENSIONS

KBCD2B

1.4.14

KBCD2B

(Convert BCD to Binary)

KBCD2B converts a
form.

The

KBCD2B

4-digit unsigned B8CD

result

called

Routine

always

input

positive.

lb-bit binary

integer

function.

ibinary=KBCD2B (idigits)
The

arguments

are:

ibinary

The

l6-bit binary

idigits

The

4-digit

Error

integer

(le-bit)

form of

unsigned

BCD

idigits.
input

value.

conditions:
S5YNTAX

ERROR

There

not

exactly

one

argument

present.

Interaction:
None

Restrictions:
None

Example:

Suppose

IBCD

contains

the

following

bit pattern,

which

BCD:

0000

1001

result

0110

the

1001

tollowing

call,

IBIN=KBCD2B (IBCD)

IBIN

contains
0

0UU

001

the
111

following
001

VUl

bit

pattern,

which

969

decimal.

0969%

RT-11

FORTRAN

EXTENSIONS

LED

1.4.15
LED
LED

LED

displays
without

display
LED

(LED

Dis play)

numbers on the LED panel of the LPS11.
an LPS 11 in the system configuration.

positions
called

CALL

Routine

an d

movable

decimal

You
There

cannot
are six

use
LED

point.

s ubroutine,.

LED(ivalue , 'format')
or

CALL
The

LED(rvalue , 'format'")

arguments

are:

ivalue

The integer
specified.

rvalue

The

real

value

displayed

1in

the

format

displayed

the

format

specified.
format

The

FORTRAN format specification for the value
to
be displayed.
The maximum field width for integer
(I) formats 1is six;
the maximum field
width
for

real
(F)
formats is seven.
The maximum F format
field width is wider to allow for a decimal point.
The
decimal
point
does
not
use an LED display
position as
the
decimal
point
appears
between
display positions.
The minus sign does use an LED
display position
so
that
the
largest
negative
numbers
that
can
be
displayed
have
five
significant digits.
The largest positive
numbers
that can be displayed have six significant digits.
If the
value
exceeds
the
width
of
the
field
specified, the display contains all minus signs.
Error

conditions:
S5YNTAX

ERROR

There
ARGUMENT

The

DEVICE

are

not

exactly

two

arguments

present.

ERROR

format

specified

not

legal.

CONFLICT

The

system

configuration

does

not

include

LPS11.

Interaction:
LED

(for

can

used

example,

to display
IADC,

calibrate your instruments
a data collecti on run.
Restrictions:
LED

requires

LPS11.

data

1IDIR).

values
You

acquired

might

want

Schmitt-trigger

by other
to

routines

this

thresholds

prior

RT-11

FORTRAN

EXTENSIONS

Examples:

Display

negative

integer.

NEG=-123
CALL LED(NEG,'I4')

HEEEE
CALL

LED(NEG,'I3"'")

Display

real

number.

REALV=25,94
CALL LED(REALV,'F5.2")

Display

single

analog

value

that
you want the display value
ADC accepts bipolar input with a

from

to be
range

channel

the

expressed in volts
of -5 volts to +5

following
example
shows
the
transformation
acquired to the corresponding input voltage.

from

CALL LED((FLOAT(IADC(U)-2048)/2048.)*5.,'F7.4')

the

ADC.

Assume

and that
wvolts.

1l2-bit

the
The

value

RT-11

FORTRAN

EXTENSIONS

LWAIT

1.4.16

LWAIT

initiates

the

(Wait)
a

program

Routine

pause

(fcr

example,

the

program.

Any

repeated

activity

execution of the program does not continue past
until
the
arguments
differ.
For
example,
numeric

value

interrupt

LWAIT

subroutine:

called

CALL
Me

service

completion

flag

completion

either

already

sampling)

(that

the
LWAIT
wuse LWAIT to

will

started

continues,

but

statement
compare a

changed

routine).

subroutine

function.

LWAIT (ivaluel,ivalue?2)

function:

ivar=LWAIT (ivaluel,ivalue?2)
The

arguments

are:

ivaluel

The

name

you

wuse

ivar
ivalue2

variable

LWAIT

compared

function

call,

ivalue?2.
LWAIT

sets

ivaluel.

A variable

constant

compared

ivaluel.

NOTE

Neither

the

arguments

changed
within
LWAIT, but
of the arguments
specified
variable
that
changes
as
activity
1in
an
interrupt
completion routine.

Error

LWAIT

1is

at least one
must
be
a
a result of
service
or

conditions:
None

Interaction:
At

least

one

the

arguments

for

LWAIT

must

interrupt
service
or
completion
routine.
argument for LWAIT can be
a
completion
flag
routines SETR, RTS, HIST, or DRS.

changed

For
for

example,
one
of

the

Restrictions:
One

the

CALL
would

arguments

continue.

The

must

eventually

change

call

LWAIT(1l,1)

suspend

program

execution

indefinitely.

program

execution

RT-11

FORTRAN

EXTENSIONS

Example:

You

want

timed

program

delay

50U

msec.

IFLAG=0

;Initialize

CALL

;Start

CALL

SETR(4,0,500.,IFLAG)
LWAIT(IFLAG,U)

flag.

clock

;mode

for

;jWait

until

;honzero.

5UU

single

interval

msec.

clock

endflag

RT-11

FORTRAN EXTENSIONS

REL

1.4.17
REL

REL

operates

LPS11l

the

relays.)

KEL

called

CALL

Control)

two

hardware

the

The

(Relay

for

open

relays

1in

considerations

the

LPS1l.

connecting

(See

the

equipment

subroutine.

REL(irelay,isetting)

arguments

are:

irelay

isetting

Error

normally

manual

Routine

The

relay

number.

irelay

The

relay

The

is
is

an
an

relays

are

numbered

and

odd

number,

REL

selects

relay

even

number,

REL

selects

relay

setting.

Value

Meaning

Zero

Open

nonzero

relay.
relay.

conditions:
SYNTAX

ERROR

There

are

ARGUMENT

ERROR

The

relay

DEVICE

not

exactly

number

two

arguments

present.

negative.

CONFLICT

The

system

configuration

does

not

include

LPS1l1.

Interaction:
None

Restrictions:
REL

meaningful

only

the

system

includes

LPS11l.

Examples:

The

following

call

CALL

REL(2,-5)

Read

digital

value

input
1is

closes

zero;

REL(1,IDIR(,1,-1,1))

number

unit

nonzero.

CALL

relay

Leave
1

relay
the

open

the

value

RT-11 FORTRAN EXTENSIONS

RTS

1.4.18

(Repeated Sampling)

Routine

routine
The RIS
RTS mediates repeated sampling from A/D converters.
sampling capabilities beyond the single reading
analog
your
extends
that

KTS

can

1is

acquired

using

flexible routine providing three basic capabilities.

for
capability
the
provides
RTS
sampling.
Single-sweep
You so specify the
1limited bursts of analog data.
acquiring
array
an
£fills
parameters that RTS stops sampling after it
For the AD11-K and the ARll, RTS provides a fast,
with data.
called
single-sweep sampling mode
capability,
restricted
(described later).

fast-sweep mode

IADC.

Continuous
acquiring

sampling.

continuous

RTS

provides

the

for

capability

streams of analog data.

There are two

sampling,
finite continuous
classes of continuous sampling:
of samples to be
number
total
the
specify
you
which
in
do
you
acquired, and infinite continuous sampling, in which
If
specify the total number of samples in the RTS call.
not
you request infinite continuous sampling, you must call RTS a
second time to stop the sampling process.

fills
RTS
In both cases, you specify the parameters so that
You either designate a completion
data arrays continuously.
routine that executes periodically or

arrange

your

program

structure the completion
You
arrays.
the
to empty
code
routine or program to dispose of the data as you require (for
example, by writing it to secondary storage, or by reducing
require
Because completion routines
it to means or counts).
to execute, the maximum continuous sampling rate (which
time
the maximum
than
lower
is
involves a completion routine)
rate

(which

does

not

involve

single-sweep

sampling

End sampling.

You can call RTS in parameter-adjustment

completion routine).
3.

stop

infinite continuous

mode

sampling.

RTS is called
with
a
subroutine
call.
The
number
necessary depends on the capability you are requesting.

arguments

Single-sweep sampling:

[nchannels],
aysize,,
[nsamples], [ifirst],
CALL RTS (iarrayname,iarr
[igain], [imode] ,iendflag#,nleft)

Continuous

sampling:

CALL RTS (iarrayname,iarraysize,
[nsubarrays], [nsamples], [ifirst],
,nleft#
[imode],iendflag#],
nnels
[igain],
[ncha
[, [interval],complete])

End

sampling:

CALL RTS(,iarraysize,,,,,,imode*,iendflag,nleft)

RT-11

The

arguments

FORTRAN

EXTENSIONS

are:

iarrayname

iarraysize

The

name of the integer array in which
the
analog
data
sampled.
With
sampling, RTS treats the data array as

array.

requests

parameter-adjustment

The

you

1length

parameter

omit

iarrayname,

the

places
continuous

circular

the

RTS

call

mode.

iarrayname

gpecifies

RTS

(in

amount

words) .

This

memory

reserved

for RTS to store data.
For single-sweep sampling,
the
wvalue
of
iarraysize
should
be
calculated
according to the following relationships:
iarraysize

nsamples*nchannels

iarraysize = nsamples*hchannels*z
dual
For

continuous

greater

nsubarrays

The

than

(for

LPS11

sample-and-hold)

sampling,

iarraysize

should

interval*nchannels*nsubarrays.

number

of logical
partitions
in
iarrayname.
length
of
each
partition
is
equal to the
largest integer in
(iarraysize/nsubarrays).
For

The

example,

with

10,

there

are

and

unused

default

iarraysize

1024

and

nsubarrays

partitions,

each

100

words

number

the

The number of partitions
equal to iarraysize.
The
must

channels
nsamples

integer

end

partitions

iarrayname.

1less
each
+the

than
or
partition
number

sampled.

The

number of samples per channel,
The
specifies
the
total
number
of
times
analog channel is sampled.
The default

nsamples

only

1is

with

iarraysize.

single-channel
with the LPS11l
In

sampling,

depends

data

from

each

channel

channels

with

normal

sampling,

nsamples

With

LPS1l

dual

nsamples

continuous

can

the

you

used

value

you

assign

vyou

are

collecting

independently,

LPS11

for

dual

to
from

sample-and-hold.

specify

iarraysize/nchannels

sample-and-hold,

you

specify

iarraysize/(nchannels*2)

The maximum value
with the LPS11l is
In

that each
wvalue
of

sampling

whether

pairs
For

default

parameter

sample.
The default cannot
dual sample-and-hold.

single-sweep

nsamples

The

single-sweep

The

one.

must
be
length of

multiple

1long,

of
nsamples
409e6.

sampling,

nsamples

depends

sampling

process.

the
how

for

value
you

DMA

you
want

sampling

assign
to

stop

to
the

RT-11

EXTENSIONS

FORTRAN

set
you
sampling,
continuous
finite
For
(a)
equal to the number of samples you want.
nsamples
greater
1is
nsamples
of
In this case, the value
are
you
(because
iarraysize/nchannels
than
requesting continuous sampling).

For

(b)

RTS

call

then

negative

set

nsamples

number:

any negative

you

sampling,

continuous

infinite

nsamples equal

samples

data

until

the

program executes a

1in parameter—-adjustment mode with a

RTS

value

for

imode.

must
you
For either type of continuous sampling,
write
the program code and specify the parameters
array
full
from
data
remove
to
necessary
Continuous sampling is not valid for
partitions.
LPS11l

ifirst

DMA

sampling.

The

first

channel

sampled.

The

valid

channel

numbers

you

request

for
the
AM11-K,

are:

Value

Meaning

through

Range of channel
numbers
LPS11
or
AD11-K
with
without switchable gain.

through

Range of channel
numbers
for
the
LPS11
or
ADll-K with AM1l1l-K, with
switchable gain.
Range of channel
AR1ll

numbers

for

the

AD1l1-K.

0,2,...,62

Range of channel numbers for AD1l1l-K
with
AM11-K
with
differential
input, without switchable gain.

0,2,...,14

Range of channel numbers for AD11-K
with
AM11-K
wused
in differential
mode with switchable gain.

0,2,...,7

Range of
with

nchannels

more
than
one
channel, the channels are sampled
consecutively in ascending order so that ifirst is
The
channel to be sampled.
numbered
lowest
the
default value for ifirst is zero.

dual

channel

numbers

for

LPS11

sample-and-hold.

The number of consecutive channels to
be
sampled
in
either single-ended or differential mode.
The
value of nchannels must be
the
total
number
of
channels between the lowest numbered channel to be
sampled and the highest
numbered
channel
to
be
sampled (inclusive).
That 1is,
nchannels

1-64

(highest numbered
numbered) + 1

lowest

RT-11

FORTRAN

EXTENSIONS

This is the case, regardless of
from
any
of
the
intervening
interest.

whether
the
data
channels
are
of

For the LPS1l dual sample-and-hold,

igain

the
number
sampled.

consecutive

The

default

value

The

gain

The

always

for

pairs

nchannels

channels

for igaih
each value

gains,

you

setting.

The default value
used
for converting

is
teturned
in
bits
13
and
12
of
ivalue.
The
possible
wvalues of bits 13 and 12 are 00, 01, 10,
and 11 binary (0, 1,
2,
or
3
decimal).
These
correspond
to
gain
codes
of
1,
2,
3,
and 4
respectively, or to the actual gains of 1, 4,
16,
and 64.
When you request a gain of 1, bits 13 and
mast

gain

1is

return

use

from

the

CVSWG

gain

For

all

separate

other
the

converted

value

bits.

Value

Meaning

Request

autogain.

determines

the

optimal

The
gain

software
to

use

for

each
A/D
conversion.
Autogain
is
available
only
with
switchable
gain
systems (ADll1-K with
AM11-K,
or
LPS11
with
1

Request
bipolar

gain
mode

of 1.
This value
for the ARI1l.

Request
signal

gain
by

of
4.

bipolar

mode

for

imode

LPSS5G).

The

mode

to multiply
This
wvalue
the ARI1l.

the
input
specifies

Request

gain of 16 to multiply
signal
by
1le.
This
wvalue
unipolar mode for the ARI1l.

the

Request gain of 64 to multiply
signal
by
64.
This
wvalue
unipolar mode for the ARI11.

the input
specifies

default

input
specifies

sampling.

The
stop
codes
that
determine what happens
sampling stops.
The

specifies

for

imode

terminate
sampling
to completion routines

zero.

and
when

RT-11 FORTRAN EXTENSIONS

The mode values and their meanings are:
Meaning

Value

For the ADl11-K and the LPS1l,
on

A/D

sample

For

the ARll,

interrupt.

each external

take

one

sample

each Schmitt trigger

take

event.

one

A/D

Take
one
A/D
sample
clock-overflow pulse.

each

For the
ADl1l1l-K
and
the
ARll,
enable
first
the
(see
mode
fast-sweep
one
Take
restriction in this section).
each Schmitt trigger #1
on
sample
A/D
each
on
or
interrupt (for the AD11-K)
external event (for the AR1ll).
For
the
LPS11,
enable
dual
sample-and-hold.
Take one simultaneous

pair of
trigger

A/D

samples

interrupt.

each

Schmitt

For the
ADl1ll-K
and
the
ARll,
enable
fast-sweep
mode
(see
the
first
one
Take
restriction in this section).
A/D sample on each clock overflow pulse.
For

the

LPS11,

sample-~and-hold.

Take

enable

dual

one pair of A/D

samples on each clock overflow pulse.

Enable DMA.
LPS11l
with
sampling.

DMA is valid only
for
the
single-channel single-sweep

Enable DMA
in
burst
mode.
In
burst
mode,
data are acquired as fast as they
one
is,
That
converted.
be
can
conversion
starts
as
soon
as
the
previous one is completed.
Burst
mode
is
valid
only
for
the
LPS1l
with
single-channel single-sweep sampling.

For the
LPS11l,
dual
sample-and-hold,
DMA,
and
clock-overflow
sampling
can
be requested singly
(by using the code in the table) or in combination
by
adding
the
codes
for the desired options to
form the value of imode.
For example, to initiate
clock-overflow
(mode
2),
dual
sample-and-hold
(mode 4) sampling for the LPS1ll, specify imode = 2
+

RT-11

The

stop

FORTRAN

codes

EXTENSIONS

and

their

meanings

Value

are:

Meaning

-4

Continue

sampling

current

any

completion-routine

but

allow

those

-3

sampling

current

any

array

queued

Stop

requests,

already

initiated

any

requests,

already

only

partition.

queued

but

Stop

sampling

honor

any

the

not

honor

requests.

RTS.

not

completion-routine

allow

queued

fill
Do

completion-routine

sampling

queue

-1

requests

the
queue

run.

Continue

-2

only
to
fill
partition.,
Do not

array

those

requests

run.

initiated by RTS.
Do
not
completion-routine

queued

regquests.

iendflag

The

completion and/or error flag.
You
must
set
iendflag
to
zero
before
you
call
RTS.
RTS
increments
iendflag
either
when
all
samples

requested

have

been

acquired

when

sampling

has

been stopped by a stop code
(negative
value
for
imode).
If a data overrun or other error occurs,
RTS sets iendflag to a negative value.

nlett

The

array

with

partition

the

decrements

becomes

nleft

full.

partitions

using

full

array

each

time

RTS
array

of
1

each

Thus

initializes

time

nleft

array

contains
to

RTS.

completion

routine

remove

partition

overrun

has

indicate

you

should

becomes

before
the
nsamples
has

occurred

and

RTS

partition

the

available

partitions,

nleft

partitions.

currently

reaches
zero
requested
in

Note:

flag.

number

number
If

you

data

increment

available.

of
are

from
nleft

nleft

number
of
samples
been acquired, a data

RTS

then

sets

iendflag

error.

nleft

meaningful

only

with

continuous

sampling.

interval

The

number

interrupts

completion

routine.

interrupts,

external

pulses

that

should

You

between

specify

events,
occur

calls

the

number

clock

before

the

of
overflow

completion

routine is executed.
RTS initializes an
internal
variable
to interval and decrements it after each
interrupt.
When
the
wvalue
reaches
zero,
RTS
reinitializes
the
internal
variable,
and
the
completion
routine
executes.
Therefore,
the
completion
routine
executes
every
interval
interrupts or events.

RT-11

FORTRAN EXTENSIONS

The
default
value
for
interval
is
iarraysize/nsubarrays.
If only one data point is
collected on each interrupt, the
default
ensures
that
the
completion
routine
executes when each
array
partition
becomes
full.
With
dual
sample-and-hold or more than one channel, you must
calculate interval
if
you
want
the
completion
routine
called
as
each
array partition becomes
full.

interval

= iarraysize/(nsubarrays*nchannels)

interval = iarraysize/(nsubarrays*nchannels*2)
for

complete

LPS11 dual

sample-and-hold.

Completion-routine name.

You write

completion

routine that executes whenever interval interrupts
have
occurred.
There
is
no
default
completion-routine
name.
When you omit the last
two arguments, RTS does not designate a completion
routine.

Error

conditions:
SYNTAX

ERROR

There are fewer
ARGUMENT

than 10 or more than 12 arguments present.

ERROR

The

starting

channel

the

hardware).

number

(ifirst)

too

large.

The number of channels requested (nchannels) is negative, or
too
large (greater than 16 or greater than 64, depending on
Autogain was requested for the ARIl.
The gain code (igain) is greater than 4.
The mode (imode) requested is out of range
greater than 15).
The sum of ifirst and nchannels
requests

(less

than -4

sampling
from
a
channel with channel number greater than 63 (or

nonexistent
15).

The size of the array partition is not an even
multiple
the number of channels to be sampled (nchannels).
Autogain was

requested with DMA.

Continuous sampling with stop code termination was specified
for

DMA.

Data

acquired

with

DMA

exceeds

the

available

space

(iarraysize).
DEVICE

CONFLICT

An RTS call attempted to initiate
sampling
while
from a previous call to RTS was still in progress.

sampling

Interactions:

RTS

can

request

an A/D

conversion

each

clock-overflow

pulse.

It
does
not
require
clock
interrupts.
You can run the
without interrupts by requesting noninterrupt mode in SETR.

runs

faster

normally

used

without

clock

to process

the

interrupts because you

clock

interrupts.

save

RTS

the

runs

faster
when
SETR
noninterrupt
mode
1is
combined
with
fast-sweep mode
described
in
the
first
restriction
in
section.

clock
RTS

time

even

the
this

RT-11

FORTRAN

EXTENSIONS

You cannot write a program that would
RTS
at the same time.
Both IADC and
hardware and only one of the routines

attempt to
RTS use the
can execute

call
same
at a

IADC
and
conversion
time.

When

the value of igain is not 1,
RTS
returns
a
packed
value
containing
the
digitized
wvalue
and the gain setting.
You can
unpack the digitized value in real-number form, and the
gain
in
integer form by using CVSWG (see Section 1.4.2).

With gains other than 1, the A/D conversion system multiplies the
input
signal
by 4, 16, or 64.
When you get the digitized value
in real-number form
from
CVSWG,
vyou
must
divide
it
by
the
appropriate
gain (4, 16, or 64) if you need to restore the value
to the original scale.
Restrictions:
Fast-sweep mode is a mode available only with the AD11-K or ARIll.
With
fast-sweep
mode,
you
<can
acquire
analog data at a rate
several times faster than normally possible with RTS.
There
are
several
restrictions
on using fast-sweep mode.
You can request
fast-sweep
mode
only
for
single-sweep
sampling
for
a
single-channel
sample
with
the
<c¢lock
running in noninterrupt
mode.
There can be no array partitions, no completion
routines,
and no autogain.
Data

overrun occurs when RTS fills the
array
partitions
faster
than
the
completion
routine
can
empty
them.
This situation
occurs when the clock rate is too fast for the size of the
array
partitions.
When

the

mode,

numbered
DMA

AD1l1-K,

the

channels,

sampling

odd
and

single—-channel
DMA

with

data

without

AM11-K,

numbered

channels

can

sampling.

Call

RTS

only

differential
those

single-sweep

Continuous

once

used

duplicate

even

ignored.

available

sampling.

for

each

sampling
DMA

sweep

mode

not

for

wvalid

required.

maximum
number
operation is 4K
is 4096.

of
data
points that can be
words.
The maximum value for

accepted in one
iarraysize with

The

AM11-K

with

AD11-K

with

hardware

can

set

gains

for
The

DMA
DMA

and
50
instead
of 1, 4, 16, and 64.
Autogain does not work in
this case.
The meaning of gain codes 2, 3, and 4 depends on
the
actual gain selected for each group of channels in the AM11-K.
Autogain

does

Autogain

not

work

with

available

AD11-K

all

for

LPS1ll

the

AR11.

unipolar

mode.

Examples:

l.
You want to collect a
single
sweep
of
200
channels,
starting
with
channel
0.
Because
sampling, you omit nsubarrays,
interval,
and
a
name.

(0).
but

samples
from
four
it
1is
single-sweep
completion
routine

You
wuse
the
default values for ifirst (0), igain
The nleft argument (IDUM)
is not used by single-sweep

must

included

satisfy

the

syntax.

(1),

imode
sampling,

RT-11 FORTRAN EXTENSIONS

:Set up the data array.
;Initialize RTS flag.
:Call RTS.
sWait until all samples

DIMENSION IDATA (4,50)
IFLG=0
CALL RTS (IDATA,200,,50,,4,,,IFLG,IDUM)
CALL LWAIT(IFLG,0)

;read.

(Notice that the data array could have been dimensioned as IDATA (200).
FORTRAN stores the data the same way regardless of how you specify the
data
later
Using IDATA (4,50) can be more convenient for
dimensions.
access, because the value of the first subscript specifies the channel
sequence
the
specifies
number and the value of the second subscript
number

the

sample.)

2.
You want to request continuous sampling of 1300
samples
using
a
You sample channel
400-word data array with four lU0-word partitions.
This
3.
gain of
a
1 100 times per second on clock overflow, with
array partitions without using a completion
example processes the
As soon as a partition is full, the program sums the data in
routine.
that partition, and frees the partition by incrementing the partition
count

(NLEFT).

DIMENSION IDATA(100,4),SUM(13)
DO

;No. of partitions (mod 4).
;Increment index for SUM array.

I=1,100

SUM (N) =SUM (N) +CVSWG (IDATA (I,IPART))/16.
IF (IPART.LT.4)
IPART=0

up arrays.

SUM(I) = 0
:Initialize SUM array.
N=0
;Index for SUM array.
IPART=0
:Index for IDATA array.
IRFLG=0
;RTS endflag.
ISFLG=0
;SETR endflag.
CALL RTS (IDATA,400,4,1300,1,1,3,2,IRFLG,NLEFT)
;Request continuous sampling.
CALL SETR(5,1,1.,ISFLG)
;Start clock at 100Hz.
IF (NLEFT.EQ.4) GO TO 10
;Wait for first partition to
;Eill.

IPART=IPART+1
N=N+1

;Set

I=1,13

GO TO 30

NLEFT=NLEFT+1
IF (IRFLG.GE.(0) GO TO
STOP 'RTS ERROR'
IF (IRFLG.EQ.0) GO TO
CALL SETR(-1,,.,)
TYPE 41,SUM
FORMAT (F12.2)
STOP

:Sum the data in the
;just-filled partition.
:Check to see if end of IDATA.
;Start indexing from start
;of array next time.

;Increment the available

;partition

count.
;Skip if not an error.
:Error if IRFLGKO.
;If still sampling, jump.
;Finished, stop clock.
;Output the data.

'DONE'

END

(Notice that you could specify the data array as IDATA (400).
However,
your
subscript
calculation
would then be more complicated.
FORTRAN
stores its arrays "by columns," so that it puts the first 100
samples
into
IDATA(1,1)
through
IDATA(100,1),
the
second 100 samples into
IDATA(1,2) through IDATA(100,2), and so on.
Therefore, the
value
of
the
first
subscript
specifies
the
sequence
number
within
the
partition, and
the
value
of
the
second
subscript
specifies
the
partition

number.)

RT-11

FORTRAN

EXTENSIONS

SDIS

1.4.19
SDIS
or

SDIS

stops

(Stop

Display)

continuous

Routine

display

that

being

controlled

either

DIS

DXY,

SDIS

called

CALL

subroutine.

SDIS

There

are

arguments.

Error

conditions:
None

Interaction:

While the continuous-display routines (DIS, DXY) are running, the
only
other
activities
for
which
resources are guaranteed are
interrupt service routines and completion
routines.
Therefore,
the most practical way to stop a continuous display is by calling
SDIS

use

from

SETR

after

within

completion

designate

specific

time

routine.

completion

interval,

For

routine

example,

that

would

receipt

you

might

executed

external

event.

Restrictions:
None
Examples:

l.
The following example is part of the example given in full in
description
of
DIS.
The completion routine HOLDIT contains the
to

SDIS.

The

EXTERNAL

completion

routine

specified

HOLDIT

the

call

;Specify

the

SETR.

completion

;routine.

IFLG=0
CALL SETR(5,0,1000.,IFLG, ,HOLDIT)

;End flag for SETR.
;Start the clock.

CALL

;Start

DIS (ISCREEN,ISPACE,NPTS,48,20)

the

display.

PAUSE

SUBROUTINE

CALL

SDIS

RETURN

END

HOLDIT

;Completion

;Stop

the

the
call

routine.

display.

RT-11 FORTRAN EXTENSIONS

SETR

1.4.20

SETR

(Set Rate of

the Programmable Clock)

Routine

SETR provides several
options
for
controlling
time
intervals
and
counting
events.
The hardware clock counter can be used to generate
interrupts to the processor at specified intervals or in
response
to
external

events,

measure

time

intervals,

count

events.

The KWll-K and LPS clock counter run at any of five crystal-controlled
frequencies
(100 Hz to 1 MHz),
at
1line frequency, or at an external
frequency supplied through a Schmitt trigger.
The AR1ll clock
counter
runs
at any of five crystal-controlled frequencies, or at a frequency
supplied through an external input line.
The KW1ll-K and LPS clock counter operate in any of
four
programmable
modes:
single interval, repeated interval, external event timing, or
external event timing from zero base.
The AR1ll clock counter has
two

programmable

clock

modes

modes:

are

SETR

allows you to
synchronization),
completion routine
called.
Call SETR

SETR

called

CALL

The

interval,

later

this

and

repeated

interval.

The

section.

specify

the clock mode, the clock rate (or type
the length of the interval or number of counts,
to be called, and the conditions under which it
to start or to stop the clock.
:

of
the
is

subroutine.

SETR(irate,imode,rcount,iendflag#([, [intervall,complete])

arguments

irate

single

described

are:

The

rate

clock

for

the

counter

oscillator

The
stop
codes
that
determine what happens
the clock is disabled.

Table

rate

disable
the
to completion

controls

the

clock
routines

and
when

codes:

Value
0

that

frequency.

Meaning
Stop
No

the

rate

clock

(without

disabling

selected.

Set

clock

frequency

Set

clock

frequency

100

Set

clock

frequency

Set

clock

frequency

Set

clock

frequency

100

MHz.
KHz.

KHz.
KHz.
Hz.

it).

RT-11

FORTRAN

EXTENSIONS

Value
6

Meaning
Use

Schmitt

input
for an
7

The

stop

Set

clock

(50

codes

trigger

freguency
or

60U

(the

external

Hz).

line

(Not

input

frequency

valid

for

the

are:

Value
-2

line
for
the ARll) as the
external frequency source.

Meaning
Disable

more
allow

the

clock.,

not

gqueue

any

completion~routine
requests,
those requests already
queued

but
to

run.

-1

imode

Disable the clock.
Do
not
honor
queued completion-routine requests.

any

The imode
argument
selects
the
mode
of
clock
operation.
The external event and external start modes (modes
2
through
7,
12,
and 13) are not valid for the
ARL1l.

The

codes

for

clock

modes

Value

are:

Meaning

Single-interval mode.
The clock counter
counts
rcount
times,
generates
an
interrupt, and then stops.
You need not
issue
a
stop
code
hecause
the clock
turns off automatically.
For the ARll, mode 0 can
be
used
only
with rates 1 through 5.

Repeated—-interval
counter

counts

mode.
rcount

The

times,

clock

overflows

(generating
an
interrupt),
begins again to count rcount
and
DRS
require
overflow
synchronize sampling.

and
times.
events

External—-event

The

timing

mode.

the

<c¢lock

counter

(which

you

must

set

zero).

The

counter

runs

the

irate,
A pulse from
causes
the
current
to be transferred to
(You

<can

data

array

counter

read

rate

the

HIST.)

rcount

clock
specified by

Schmitt trigger
#2
clock counter value
a
clock
register.

using

continues

initial

value

then
RTS

count,

value

The

into

clock

RT-11 FORTRAN EXTENSIONS

Meaning

Value

2zero
from
mode
timing
External-event
The initial
to mode 2).
(similar
base
rcount
is
counter
<c¢lock
value of the
The clock
(which you must set to zero).
by
specified
rate
counter runs at the
A pulse from Schmitt trigger #2
irate.
value
counter
causes the current clock
transferred to a clock register.
be
to
(You can read the clock
register
value
The
HIST.)
using
array
data
a
into
clock counter is cleared to zero and the

to count.

clock continues

External

start modes.

you
above,
3
through
When you add 4 to modes 0
that specifies an external
value
a mode
obtain

7,
1In modes 4 through
start pulse for the clock.
clock counter starts counting on receipt of a
the
than
rather
#1,
trigger
Schmitt
from
pulse
as soon as the routine SETR is executed.
starting
the
with
available
Extirnal start modes are not
AR1ll.

The codes for

external

start clock modes are:

Value

Meaning

mode

Single-interval

with

external

with

external

start.

Repeated-interval

mode

start.

External-event timing mode with external
start.

External-event
base

with

Noninterrupt

modes.

When you

obtain

add

mode

timing

external

to modes

value

mode

from

=zero

5 above,

you

start.

and

that specifies running the

clock counter so that clock overflow
generates
a
clock
overflow
pulse
but
does
not generate an

interrupt.
The RTS routine does not
require
the
the
only
but
interrupt,
clock-overflow
clock-overflow pulse.
RTS runs
faster
with
the
turned off because you save the
interrupts
clock
interrupts.
time normally required to process the
The
the

clock-overflow
clock mode is 8,

HIST and

DRS,

which

interrupts
are disabled when
9, 12, or 13.

require

clock

interrupts,

not
work with noninterrupt mode.
1In noninterrupt
mode, SETR does not run the software clock.

RT-11

The

FORTRAN

codes

for

EXTENSIONS

noninterrupt

Value

mode

Meaning

Single-interval

Repeated—-interval

noninterrupt

Single-interval

external

noninterrupt

mode.
mode

with

start.

external
Limited

mode.

noninterrupt

Repeated—-interval

When

are:

noninterrupt

mode

with

start.

mode.

you

add

modes

and

above,

a
limited
service
clock mode.
the only function SETR performs
the

software

only

timing

clock.

TIf

function

the

you

In
is

you

obtain

limited mode,
to
increment

software

clock

require,

wuse

the

limited

mode.
None

the

interrupts

other

services

(HISsT,

DRS,

SETR

routines)

available

limited

The

for

limited

mode

are:

codes

Value

SETR

completion

mode.

Meaning

Repeated-interval

limited

mode.

Repeated-interval

1limited

mode

with

external

(External

start

modes

are
rcount

requiring

and

not

start.

available

Initial

count

value.

rcount

times

the

The

for

the

c¢lock

frequency

AR1l1l.)

counter

specified

counts

irate,

and then overflows
and
generates
Therefore,
the clock generates an

an
interrupt.
interrupt every

rcount

counts,

interrupt

equal

resulting

irate/rcount.

1in

The

value

rate

rcount

must

vte a real number.
For the KWll-K
or
the
LPS11,
the
maximum
value for rcount is 65535 (decimal);
for the ARll, the maximum value for rcount is
255

(decimal).
rcount

iendflag

The

completion

flag

1increments

number

before

external

for

iendflag

overflows
1n
to zero before

completion
specify the
the

the

zero.

counter
iendflag
interval

With
be

SETR

must

the

modes,

interval

mode.

when

the

mode
O.
you call

You
SETR.

nust

between

calls

clock

set

the

With
this
argument,
vyou
clock overflow interrupts or

Schmitt

completion

single

timing

only

interrupts

routine.
number of

event

trigger

routine

pulses

executes.

elapsing

SETR

has

an
1internal variable initialized to interval that
it
decrements
after
each
c¢lock
interrupt
or
Schmitt
trigger #2 pulse.
When the value reaches

1-75

RT-11 FORTRAN EXTENSIONS

zero, SETR
calls
the
completion
routine.
The
completion
routine
thus
executes
after
every
interval
interrupts.
The
default
value
of
interval is 1.
complete

Completion-routine name.
The
completion
routine
that
you
provide
1is
executed whenever interval
interrupts have occurred.
There
1s
no
default
completion-routine
name.
When you omit the last
two arguments, SETR does
not
<call
a
completion
routine.
Since the completion routine is
called
£from
the
clock
interrupt
service
routine,
you
cannot
request a completion routine
when
the
clock
is
running in noninterrupt mode.

Error

conditions:
SYNTAX

ERROR

There

are

fewer

than

requested
requested

is
is

four

than

six

arguments

present.

ARGUMENT

ERROR

The
The

rate
mode

above.
The count
DEVICE

value

less than -4 or greater than
not one of the valid
modes

requested

for

the

ARll

greater

than

255.

CONFLICT

Consecutive
This
error

calls
does

to SETR with
not
occur

rate
for

parameter 1
consecutive

through
7.
calls
with
and
8).
The
a call to SETR

variants of single-interval mode (modes 0, 4,
remedy is to insert (between the other calls)
using a negative rate.
CLOCK

7.
described

ERROR--RATE

TOO

FAST

Clock overrun occurred, meaning that a clock
interrupt
has
been
1lost
and
program
execution stops.
The remedy is to
request a lower clock rate, or,
1if
practical,
¢to
run
in
noninterrupt mode.
Interactions:
Use

SETR

get

the

desired

rate.

You

analog,

digital,

SETR

controls

Section

the

clock

can
time

going

the

then

wuse

RTS,

interval

operation

data

the

desired

DRS,

mode

HIST

the

acquire

(respectively).

16-bit

software

clock

(see

1.2.2).

DRS

and HIST require clock-overflow interrupts.
Therefore
those
routines
do
not
operate
properly
if
the clock is running in
noninterrupt mode.

Restrictions:
You can
0
and
not

use the following call to stop the clock (except in modes
4 in which it stops automatically).
In this case, you do

need

fourth

specify

names

arguments.

CALL

SETR(-1,,,)

values

for

the

second

through

the

RT-11

FORTRAN

EXTENSIONS

Examples:

CALL

SETR(6,0,2.0,IFLAG,
,NAME)

$ Requests
tNAME

completion

after

Schmitt

;#1 has fired twice.

routine
trigger

RT-11 FORTRAN EXTENSIONS
1.5

BUILDING A REAL-TIME SUPPORT LIBRARY

When your system is delivered, you receive the FORTRAN extensions
real-time support routines in object module form, and a configuration
and
the names
lists
(Appendix B
form.
routine in MACRO source
Before you can use the real-time support
contents of these files.)
into a
routines, you must combine the object modules you received
Before you create the library, you must be sure either that
library.
your system configuration matches that specified in the configuration
routine, or that you have changed the configuration routine so that it
describes your

1.5.1

system.

8System Configuration

system
The real-time support library assumes that you have a standard
the necessary source
changing
by
otherwise
it
inform
you
unless
MACRO
a
is
The source program CONFIG.MAC
statements in CONFIG.MAC.
"standard"
a
of
description
hardware
complete
a
containing
program
system.

1.5.1.1

A Standard System -

find

out

what

standard

system

The first few pages of the
contains, list the source file CONFIG.MAC.
Section A
with
block,
set-up
confiquration
the
contain
listing
Section B containing the
and
definition,
hardware
the
containing
statement
Each
addresses.
register
status
and
vector
interrupt

defines some aspect of the

system.

For

example,

SAD11K=0

is the statement defining the absence of an ADl11-K A/D

KWll-K programmable clock

counter

the

converter

and

system.

1.5.1.2
Changing Configuration Specifications Each
statement
in
CONFIG.MAC
is accompanied by comments that explain the statement, and

the changes to make 1if your system differs from the
standard
system.
Read
through
the
explanation,
and,
using
the text editor on your
system, make whatever changes are required to CONFIG.MAC
so
that
it
For example, if you have the AD1l-K and KWll-K
describes your system.
on your system, change the definition statement to
SAD11K=1

Any changes you make will
probably
be
to
the
hardware
definition
Sstatements.
The
interrupt
vector and status register addresses are
factory-set standard addresses.
Do not change any
of
these
address
statements
in
CONFIG.MAC
unless
vyou
have
made
the corresponding
hardware

changes.

RT-11

1.5.1.3
Assembling
to
CONFIG.MAC
are

FORTRAN

EXTENSIONS

the Revised Configuration The changes you
make
«changes
to
a
source
file.
To incorporate the

changes into the system, you must assemble
the
revised
source
file
using
the
MACRO
assembler.
Your dialogue with the computer has the
following sequence (where your input has been underlined):
.MACRO/LIST CONFIG
DETECTED: U

ERRORS

1.5.2

The

Real-Time

Support

Library

Before

you can use the real-time support subroutines, you must
create
the
real-time
subroutine library.
Create the library file using the
librarian system utility LIBR.
(Refer
to
the
RT-11
System
User's
Guide for more information on the librarian.)

1.5.2.1

the

Generating

library

the

Library

real-time

Use

the

LIBR

program

generate

subroutines.

.LIBR/CREATE
Library?

1libnam

Files

LDPOBJ

you

have

library

the

that

standard

you

system

created

configuration,

you

<can

use

the

If you have a nonstandard
system
configuration
and
necessary
changes in CONFIG.MAC, you must perform one

generate
library,

now

(libnam).

the library.
Still
wusing
the
libnam, to include the changes in

have
made
the
further step to

LIBR
program,
CONFIG.

revise

the

.LIBR

Library?
Files
?

1.5.2.2

1libnam
CONFIG/REPLACE

Accessing

the

Library

Write

your

FORTRAN

the real-time subroutine and function calls described
Compile the programs in the normal manner.
(Refer to
User's
Guide for information on the FORTRAN compiler

programs

using

in this chapter.
the RT-11 System
and its optional

switches.)

When you link the programs,
in the list of commands for
.LINK
(1f

you

(1f

FORLIB

name

progrm,libnam

have

.LINK

include the
the linker:

one

system

library

SYSLIB),

progrm,libnam/F
and

SYSLIB

are

separate

libraries).

your

real-time

library

RT-11

1.6

FORTRAN

EXTENSIONS

ERROR MESSAGES

The FORTRAN extensions routines produce five
error
messages.
These
messages
are
SYNTAX
ERROR,
ARGUMENT
ERROR,
DEVICE
CONFLICT, ADC
CONFLICT, and CLOCK ERROR.
Each error message
appears
with
a
line

number,
1indicating the line where the error occurred.
You must refer
to your listing to determine which routine caused the error, and
then
refer
to
the
error
summary
table
(Appendix
C)
to determine the
possible causes of the error.
The error summary
conditions:

table

uses

the

following

Abbreviation
SYN

Error
SYNTAX

abbreviations

DEVICE

subroutine
or
function
number of arguments.

ERROR

least

one

the
routine
message does
in error.
DEV

error

ERROR

ARGUMENT

the

Condition

The
call
to
the
contains an invalid
ARG

for

the

has
not

arguments

the

call

an illegal value.
The error
indicate which
argument
1is

CONFLICT

This

error

arises

when

one

the

routines

tries to use a device that is busy.
This can
happen
under
the
foreground/background
monitor
if
another job is using the device.
It can also occur when you call a routine for
which
completion-routine
requests are still
queued as a result of a previous call to
the
routine.
ADC

ADC

CONFLICT

Device conflict error for the A/D
the A/D converter is busy.
CLK

CLOCK

ERROR--RATE TOO FAST
The clock interrupts are
to be processed.

occurring

Error-causing conditions are described in the
reference
each
routine.
A
summary table of error conditions for
appears

Appendix

converter:

too

fast

section
for
all routines

CHAPTER
FORTRAN

2.1
The

FORTRAN

Debugging

Technique

tool

control of
and change
execution.

the

execution

the

contents

runs

for

any

use

FDT

FORTRAN

PDP-11

operating system.
during a debugging
To

TECHNIQUE

(FDT)

1INTRODUCTION

debugging

FDT

DEBUGGING

(FDT)

sophisticated

IV programs.

FDT

gives

If
an

USING

with

FORTRAN

FDT requlres
session.

successfully,

you

under

approximately

need

know

the

RT-1l1

2K words

FORTRAN

debugging

Compile

have

program that
session:
your

compiler
work with

does

FORTRAN

option
inline

not

work,

program.

that
code.

RSTS/E

memory

space

programming

data

You

Link
to
root

do not

need

your
be

program
1linked.

segment

program

your

generated

units.
If you

your

has

subroutines
Run

the

FORTRAN

formats,

must

you

FDT

V02-01
PAUSE

use

might

program.

because

the

begin

FORTRAN

FDT does not
option
that

FDT

storage

needs them.
map
listing.

listing.

Generate

common

ISN

steps

1Include FDT among the
input
are using overlays, place FDT

automatic
FDT pause before
the program.
The following

FDT

code

program.

named

that

these

produces threaded code;
Do not use the compiler

suppresses internal statement numbers
Obtain a source program listing and a

examine
program

FDT

you
FDT

step-by-step

of your program and the
ability
to
of any variable in your program during

language.
You
do not need to know details of internal
machine operation, or the FORTRAN compilation process.

2.2

interactive

you

blocks

need

FDT

takes

linker

map

assembly

files
in the

vyour

language

examine.
control

and

executes

mainprog

The

variable
XX

The

components
internal

executable

the

message

statement

2-1

the first executable statement of
message appears on the terminal:

are:

number
the

FORTRAN

the
program.

first

FORTRAN DEBUGGING TECHNIQUE

(FDT)

to
refers
FDT
mainprog The main program being debugged.
program with the name assigned to it in
main
the
If you do not
the FORTRAN PROGRAM statement.
name the program,
to
statement
PROGRAM
the
uses the default main program name, .MAIN.

use
FDT

ready

The prompt FDT issues to
accept

indicate

it is

a command.

NOTE

instructions,
the
If you do not follow
FDT
may not be able to get control, and
run
then
will
program
FORTRAN
the
without
FDT.
Sometimes
FDT gets only
the
when
happens
(as
partial control
without
compiled
is
program
main
internal statement numbers)
and
cannot
In this case, FDT issues the
continue.
message FDT START FAIL, and exits to the
monitor.

Begin debugging your program.
Type in the FDT commands
that
you
have
decided
to use to start solving your problem.
At
this point, you must type in at least one command that causes

FDT pause, or else the program runs to completion without

allowing you to enter any
FDT
commands.
If
your program without FDT intervention,
execute

2.3

FDT

There

you
want
to
type CONTINUE

START.

COMMAND TYPES

are

three

classes of

FDT commands:

Program control

Information

transfer

FDT

commands

control

Program Control

commands

Commands

Program control commands allow you to control the
execution
of
your
FORTRAN
program.
You
<can
use the program control commands to halt
program execution, to continue with the next executable
statement
or
restart
from
the
beginning, to step through the program one or more
statements at a time, or to end a debugging session and
exit
to
the
operating
system
monitor.
The
program control commands are START,
STOP,

CONTINUE,

STEP,

Information Transfer

PAUSE,

RESET,

and WATCH.

Commands

Information transfer commands allow you to examine the contents of any
variable or array element in your program and to modify its value.
As
part of the command, you can
define the data type of the variable, so

FORTRAN

DEBUGGING TECHNIQUE

(FDT)

that the contents of the variable appear
1in
familiar
notation
(see
Section
2.4.3
on
mode
codes
and
Section
2.4.2
on
location
specification).
The
information
transfer
commands
are
NAME,
DIMENSION,

TYPE,

ACCEPT,

FDT

Control

Commands

FDT

control

commands

of
FDT
commands.
macros composed of

and

allow

unconditionally,

another

and

WHAT.

2.4

FDT

MACRO,

you

Using
these
FDT commands,

GOTO,

IF,

ERASE.

CONVENTIONS

AND

FDT

control

the

commands,
and
can

command.

operation

sequences

you can define and execute
branch,
conditionally
or

The

FDT

control

commands

are

TERMINOLOGY

Syntax Conventions

The general
parameters.
STEP

form of an FDT command
A simple example is

the

command

name

followed

where

STEP is the command name and 3 is the parameter.
You can
command whenever FDT has issued its exclamation-point prompt.

The

syntax

following

its

conventions

for

entering

FDT

commands

appear

enter

1in

the

paragraphs:

Spaces between the prompt and the command are
optional.
You
can
place
the
command directly after the prompt, or you can
separate the command from the prompt by any number of spaces.

The following forms are correct:
tcommand

command

FDT recognizes only the first three letters
of
any
command.
You
can
abbreviate
any
FDT command name to its first three
letters.
There cannot be any blanks between
the
1letters
in
the

command

name.

The

following

!START
ISTA
ISTAT

The

following

I1ST A
1CO
®

examples

FDT

are

correct:

interprets

examples

are

this

command

STA.

incorrect:

There is an embedded space in the
The abbreviation is too short.

command.

Spaces
are
regquired
between
a
command
name
and
its
parameter (s).
Some
FDT commands can accept parameters.
You
must leave at least one space between the command name and the
parameter.

FORTRAN DEBUGGING TECHNIQUE

The

following

!MACRO

are

correct:

|PAUSE
The

examples

(FDT)

.MAIN.,20 MACRO 2 AFTER 10

following

example

incorrect:

IMACRO1

There are two formats for a series of FDT commands.
You
can
enter the commands as a list with one command on each line, or
you can enter several
commands
on
a
line,
separating
the

commands with

semicolons.

The

following

example

!MACRO

TYPE

The
maximum
continuation

shows

two

commands

line:

I,J

command
length
is
one
line.
There
is
no
character
to
permit
a command longer than one

line.
MACRO definitions
(but
not
the
MACRO
command)
continue onto more than one line (see Section 2.5.7).

can

If you type a command
that
is
incorrectly
spelled
or
incorrectly
spaced,
FDT prints the error message ?UNDEFINED and prompts for a new
command.

2.4.1

Current

Procedure

An important concept in using FDT is the "current procedure."
As
its
name
implies,
the
current procedure is the procedure (main program,
subroutine, or function) beingsexecuted at a given time.
FDT
defines

the
current
procedure
as
the
procedure being executed when an FDT
pause occurs (see Section 2.4.4).
You need to know what
the
current
procedure
1is
when
vyou
are defining locations for FDT (see Sections
2.4.2,

2,4.3,

2.4.2

The

and

2.7.2).

Leccation Specification

The FDT information transfer
commands
require
you
to
specify
location
of
a variable or array element in the FORTRAN program.
location specification consists of two parts, the location itself,
the data type for the location.

the
The
and

There

can

are

several

ways

specify a

location.

For

a variable,

use an offset location, a named location, or a relative
an array element, you use a subscripted name location.
Offset

you

location.

For

Location

The most direct way of specifying a variable's location is to give its
offset.
The
offset
1is
the
difference
between the address of the
variable and the base location of the current procedure's data
block.
The offset
FORTRAN

for

any variable appears

compiler.

the

storage

map produced

the

FORTRAN

You

specify

the

offset

DEBUGGING

TECHNIQUE

location

(FDT)

typing

octal

number:

nnn

where

nnn

storage

map.

bytes.)

You

The

the

need

following

FORTRAN

offset
(The

the

offsets

not
a

enter

Local Variables,

from

Map

for

.PSECT $DATA,

storage

leading

fragment

Storage

variable
the

shown

map

are

the

expressed

FORTRAN
in

zeros.
FORTRAN

Program

Unit

Size = 000322

storage

octal

map:

.MAIN.

(

105.

words)

Name

Type

Offset

Name

Type

DAY

Offset

I*2

000306

Name

Type

I*2

000314

I*2

000316

I*2

000312

MONTH

1I%*2

000304

YEAR

I*2

000310

The

offset

YEAR

You

can

the

specify

variables
in
specification
variables
two

DAY

306;

the

variables
techniques

the

only

for

location

the

variable

unsubscripted

current procedure.
You cannot use a location offset
subscripted array elements, variables in common,
or
FORTRAN routine that is not the current procedure.
The

in
for

this

rule

the
main
specifying

Variable
To

offset

the
for

exceptions

variable

310.

Offset

Offsets

specify

are

for

variables

program.

(Section

the

offsets

Blank

offset

blank

common

and

2.7.1 describes special
variables in named common.)

Common

location

for

type:

variable

blank

common,

.BCOM. +nnn

where

nnn

FORTRAN

Variable
Any

offset

the

wvariable

(as

shown

the

referenced

any

map).

Offsets

variable

time

the

storage

the

main

Main

Program

program

typing:

can

mainprog+nnn
where mainprog is the name
offset of the variable.
Named
You

the

main

program

and

nnn

the

Location

can

specify

the

location

variable

have

associated with the location.
define a name and associate it with
Once
you
have
defined
a
name,

specification.

Relative

The

name

valid

You

using

name

that

vyou

use

the NAME command in FDT to
a location
(see
Section
2.5.8).
you
can
use
it
in any location

any

procedure.

Locations

You

can specify the location of a variable relative
to
a
previously
named
variable.
To
specify
a
location
relative
to a previously
defined name, you enter the name and the displacement of the
variable

from
the
named
location.
This
referencing sequences of variables

practice
in

2-5

1is
consecutive

useful

primarily

locations.

for

FORTRAN DEBUGGING TECHNIQUE

(FDT)

The format of the specification is:
name-+nnn

The components of the specification are:

name

The named location previously defined in a NAME command.

nnn

The displacement in octal between the location to be
(The
specified and the location associated with the name.
displacement is the difference between the address of the
variable and the address of the named location.)

Subscripted Name Locations

You can specify the location of an array element by using a
The FDT DIMENSION command allows you to define a
subscripted name.
subscripted name and associate it with the location of a FORTRAN
Once you have defined a subscripted name and associated it
array.
with an array, you can specify the location of any element within the
array with the subscripted name and the appropriate subscripts in
For

parentheses.

example:

ARRAY (3,7)

There are two ways to specify the subscript of an array element:
an

integer

constant.

Use

Use a name defined in a previous NAME command.

If you

use

as a subscript, FDT assumes that the variable associated

name

with the name has an integer value.

There are certain conventions you must follow for using a
name

subscripted

location:

Each subscript value must be within the range defined for that
If the subscript value is outside this range, FDT
dimension.

prints the following warning message:
$SUBSCR OUT OF

Since the message

BOUNDS

only

warning,

location using the subscripts specified.

FDT

references

the

The number of subscripts you specify 1in a subscripted name
must be less than or equal to the number of dimensions in the
If the number of
DIMENSION command that defined the name.
, FDT uses the
dimensions
of
number
the
to
equal
is
subscripts
the array
FORTRAN subscript reference algorithm to locate
number
the
than
less
is
subscripts
of
number
the
If
element.
of dimensions, FDT assumes that the missing subscripts have a
1is defined as a
if ARRAY
For example,
1.
of
value
to
equivalent
is
ARRAY (3)
array,
three-dimensional
ARRAY (3,1,1).

A subscripted name retains its
regardless

whether

that

association
1location

with
1in

the

location

current

procedure. The name loses its association with the location
only when you redefine the name with another NAME or DIMENSION
command or when you cancel the name with an ERASE command.

FORTRAN

2.4.3
The

second

type

Mode

TECHNIQUE

(FDT)

Codes
major

mode

specify

DEBUGGING

for

component
the

mode

of a
location.

for

the

location

location,

specification

data

the

type:

loc [/mode]
The

components

loc

The

location

mode

The

FDT

mode

similar

location

FORTRAN.

(see

Section

2.4.2).

code

for

the

location.

data

types.

FORTRAN
I,

corresponding

Table

corresponding

specification

2-1

FORTRAN

1lists

data

to
the

type.

are:

the
FDT

The

FDT

modes

default

INTEGER*2
modes

and

mode

data
the

are

for

type

nearest

FORTRAN DEBUGGING TECHNIQUE

(FDT)

Table 2-1
FDT Mode Codes

Mode

FORTRAN Type

INTEGER*2

INTEGER*4

Description

16-bit value displayed in decimal

32 bits,

first 16 bits displayed in

LOGICAL*4

32 bits,

displayed as T or

LOGICAL*1

8 bits,

displayed as T or F

REAL*4

32 bits,

scientific notation

REAL*8

64 bits,

scientific notation

COMPLEX

64 bits,

real and

BYTE

8 bits,

displayed

———

decimal

16 bits,

imaginary parts
in decimal

displayed as three RAD50

characters

———

bits,

displayed

—-——

A string of n ASCII characters
n is in the range 1 to 255)

—-————

ASCIZ

Dummy variable mode

string

in octal

(as used

string handling

(where

in the FORTRAN

package)

When you need to indicate that the associated variable is a
parameter
(dummy
variable),
you
can use the letter P preceding any of the FDT
modes.
In fact, you must specify this form for any variable listed as
a parameter

the attributes

section of

the

FORTRAN storage map.

FORTRAN DEBUGGING

TECHNIQUE

(FDT)

Examples:

Specification

204 /E

l6/PI

POWER/C

:
REAL*4

location

variable

INTEGER*2
COMPLEX

tthe

Any

Meaning

name

specification

offset

parameter

variable

204.

variable

at
a location

offset

lé6.

associated

with

POWERK.

can

include

mode

specification.

For

example:

TYPE
The
202

202/E

offset of the
is REAL*4,

location

If a location specification
possible actions.

the

assumes

the

data

without

does
mode

the

not

For

type

mode,

involve

location

there

name,

offset

are

two

then

FDT

example,

202

same

TYPE

default

and

occurs

specification

TYPE

202

202/I

If the specification includes a name (a named location
or
a
subscripted
name)
then FDT assumes the mode associated with
that name.
(The mode becomes associated with the
name
when
you define the name in an FDT NAME or DIMENSION command.)
For

has

example,

NAME

PI,202/E

TYPE

the

same

TYPE

result

202/E

Similarly,

has

NAME

ANGLE, 202

TYPE

ANGLE

the

same

TYPE

result

202

The original mode definition can be overridden in subsequent commands.
For

example,

NAME

the

PI,202/E

name

definition

1is

FORTRAN DEBUGGING TECHNIQUE

(FDT)

the following commands have the effects shown:
TYPE PI/D

Uses mode D for output.

TYPE 202

Uses the default mode I for output.

Uses originally defined mode E for output.

TYPE PI

The mode code is important for subscripted names 1in a DIMENSION
command because the mode code determines how FDT locates the required
array element. You can use the default mode in a DIMENSION command.
However, it is better practice to specify the intended mode explicitly
in the DIMENSION command.

specification.

You can specify a mode only in a location
other

and

Subscripts

command parameters are not location specifications, so you

cannot associate modes with them.

2.4.4

FDT Pause Definition

FDT contains a pause feature similar in operation to a

FORTRAN

PAUSE

the FDT pause halts execution of the program and allows
statement:
When the pause
the program to be continued by further commands.
you

occurs,

enter

can

with the FORTRAN pause.
FDT

FDT commands.

(Do not confuse the FDT pause

The FORTRAN PAUSE statement cannot

cause

pause.)

There are five ways to cause an FDT pause.
the situation causing the pause:
e

automatic pause

entry

statement

step

watch pause

The pause

name

indicates

pause

The automatic pause occurs before the first executable statement in
(There is no FDT command for invoking the automatic
the main program.

The automatic pause occurs only during the first run of the
pause.)
That
program in a debugging session and not during subsequent runs.
commands
REENTER
or
START
monitor
or
commands
START
FDT
is, subsequent
do not

invoke

the automatic pause.

function.
An entry pause occurs at the entry point of a subroutine or
and FDT
name,
procedure
the
specify
to
command
PAUSE
the
use
You
pauses before the procedure begins executing.

FORTRAN
a particular
A statement pause occurs before execution of
you
where
statement
the
specify
you
command,
PAUSE
a
Using
statement.

want the pause to occur, and FDT pauses when
next one

that

statement

the

to be executed.

A step pause occurs after a defined number of FORTRAN statements have
the STEP
You specify the number of statements using
executed.
command.

2-10

FORTRAN

A watch

pause

changes.

You

When

FDT

when

specify

the

pause

terminal:

FDT

occurs

DEBUGGING

the

wvalue

variable

occurs,

FDT

ISN

TECHNIQUE

watch

prints

the

(FDT)

variable

using

the

following

being

WATCH

watched

command.

message

the

PAUSE

nnn

proc

The

variable

nnn

components

The

internal

FORTRAN

proc

the

message

statement

are:

number

the

executable

statement.

The

name of the
procedure
1in
which
the
FDT
pause
occurred.
A procedure 1is a main program, subprogram,
or
function.
FDT
defines
procedure
proc
as
the
current procedure.
This procedure remains the current

procedure

until

the

FDT

pause

occurs.

FORTRAN

2.5

DESCRIPTION OF

THE

DEBUGGING

TECHNIQUE

(FDT)

FDT COMMANDS

The FDT commands are described in the following
are
arranged
alphabetically
for
convenient
contains a reference summary of the commands.
Useful

commands

for

PAUSE,

STEP,

NAME

TYPE

and

refer

new FDT

user

and

CONTINUE

variables

display

section.
The commands
reference.
Appendix D

are:

control

and

change

execution

values

FORTRAN

DEBUGGING

TECHNIQUE

(FDT)

2.5.1
The

three
three

command

forms
forms

assigns

new

values

arguments

for

freely

within

macro

the

FORTRAN

definition.

valid outside a macro.)
You can specify
single command line.
The arguments must
The

first

form

The

components
The

location

value

The
be

new value to
a
constant

form

'text'

argument

is:

'text'

The

form

text

you

can

whose

value

be
in

string

ACCEPT
of

the

assigned
the same

command
TYPE

apply.

literal
use

form

The

the

form

literals

The

third

literal

analogous

The

form

as
be

many arguments as fit
separated by commas.

are

all
not

changed.

to loc.
The new value may
data format as the mode of

previously

defined

name

is:

text

mix

are:

loc

This

third

There

can

loc=value

argument

second

(The

You

is:

loc,
or
it
may
subscripted name.
The

variables.

command.

the

form
text

printed.
is

identical

command.

The

function

FORTRAN

the

conventions

for

:
of
as

ACCEPT
a

prompt

is
to

useful

for

FDT

enter

new

values.

macros,

where

is:

loc

argument
loc

is:
The

location

specification

for

value

entered

from

the

terminal.

This form of the ACCEPT command is valid only in FDT macros.
It
requires
an
input value from your terminal, and prompts for the
value with a question mark.
The value you enter must follow
the
same conventions required for other forms of the ACCEPT command.
The

following

mode

conventions

apply

the

wvalues

the

command:

The
FDT
the

value may be a previously defined name.
copies its contents into loc.
When you
value, its mode should match the
mode

need

not.
If
the modes of loc and name
the mode of loc determines the
number
of
from name into loc
(see Table 2-1).

2-13

If
it
1is,
use a name as
of
1loc,
but
are

different,
bytes
copied

(FDT)

FORTRAN DEBUGGING TECHNIQUE

For

example:

Mode of loc

Data Type

Number of

I
J
E
D
Z

INTEGER*2
INTEGER*4
REAL*4
REAL*8
ASCIZ

2
4
4
8
1

Bytes

Copied

of
mode
When you use a name as a value, FDT ignores the
and uses name only to find the address of the value
name
mode
one
FDT never does conversions from
to be copied.
to

another.

String-mode constants (/Z or /An) must appear enclosed in
The FORTRAN conventions for text literals
single quotes.
apply.

/M)
and
/L,
RAD50 constants and logical constants (/R,
must be preceded by the character # to indicate that they

not

represent

For

names.

logical

the

example,

constants T and F are represented by #T and #F.

separated

Complex constants appear as two real constants

characters.
40
most
Nonstring constants may include at
The number of characters represented by a string constant
is limited by the length of the input line.

by a comma,

in the order

real,

imaginary.

Examples:

Command

Meaning

ACCEPT 202=1

Sets the INTEGER*2 value at offset 202

ACC DELTA=EPSILON

Replaces the contents

ACC

'DELTA=',DELTA

I=0,J=1,244/E=3,14159

the

value

contents

Prints DELTA=?

the

input value.

waits

for

within

FDT

DELTA

with

terminal

and

EPSILON.

(Valid only

macro.)

Sets the values of locations
named
1
and
J
to 0 and 1 respectively;
sets
the REAL*4 variable at offset
244
to
3.14159.

ACCEPT 1I=12,'J=',d

Sets

the

1location

named

12;

prints
J=?
on the terminal and waits
for input to set the location named J.
(Valid

only

within

FDT

macro.)

FORTRAN

DEBUGGING

TECHNIQUE

(FDT)

NOTE

Both

and

TYPE

use

that

are

loaded

FORTRAN

FORMAT

statements.

you

subroutines

are

process

debugging

program without D, E, F, or G FORTRAN
format specifications, then some of
the
format
conversion
routines
are
not
loaded and FDT
cannot
accept
or
type
variables with modes C, D, or E.
If you
attempt

format

use modes
conversion

present,

FDT

?NO

prints

CONVERSION

Section 2.7.3
this problem.

and

D, or E
routines

the

are

error

continues

describes

when

a way

the
not

message

execution.

avoid

FORTRAN DEBUGGING TECHNIQUE

(FDT)

CONTINUE

2.5.2

CONTINUE

The CONTINUE command resumes program execution after
any
FDT
pause.
The
next statement executed is the statement whose internal statement
number and procedure name appeared in the last pause message.
When it
executes a CONTINUE command, FDT ignores any commands remaining on the
line.

the current FDT pause occurred

the

result of

PAUSE

command,

the
CONTINUE
command
can
have
one
optional
parameter called the
execution count.
FDT ignores the
execution-count
parameter
if
the
current pause is an automatic pause or if it occurred as a result of a
STEP

The

WATCH

form of

the command and parameter

CONTINUE
The

parameter

ntimes

command.

is:

[ntimes]
is:

The execution count.
The execution count is an integer
specifying the number of times the FORTRAN program must
reach this point before another
FDT
pause
can
occur
here.
If you omit the ntimes parameter or if ntimes =
1,

FDT pause

this
point
cancelled by

occurs

the

time

control

(unless
the
PAUSE command
subsequent commands).

reaches

replaced

FORTRAN

DEBUGGING

TECHNIQUE

(FDT)

DIMENSION

2.5.3

DIMENSION

The

DIMENSION

the

FORTRAN

command
array

subscripted

DIMENSION

associates

you

name.

want

FDT

command

1is

name

and

list

dimensions

with

allowing

the

access

FORTRAN

virtual

arrays.

The

equivalent

NAME

cannot
the

define,

subscripted-name

name

used
the

command.

The

form

the

DIMENSION

The

parameters

DIMENSION

command

is:

name(i,j,...)[,loc]

are:

name

(L,F500¢s)

The

FDT

The

list of dimensions associated with

name

There
may
be
dimension value

loc

the

range

The

location

(base)

in
through

associated

most
seven
the list must

the

array.

the

array.

dimensions.

Each

integer

32767.

specification
the

with

array.

for

The

the

first

offset

for

element

the

base of
storage
map.

the array
appears
in
the
FORTRAN
Offsets
for
FORTRAN
virtual arrays are not valid
loc parameters.
If you omit the loc argument,
FDT
erases
Section

The

mode

code

subscripted

DIMENSION

the

subscript
ASCII

you

You

better

(/An)

when

used

specify ASCII

attempt

message

wuse

the

location

the

mode

the

specify

the

FDT

wuses

the

array.

referring
valid

subscript

mode

(see

default

practice

not

specified

part

can

command.

calculations

mode

meaning

name
2.5.4).

important

names.

However,

the

to
a

DIMENSION

a DIMENSION

command

mode

FDT

explicitly

message

However,

responds

for

command.

mode

because

error

command.

name,

DIMENSION

specified

command,

specification
a

intended

calculation.

subscripted

ERASE

all

has

appears

when

you

with

the

error

DATA

with

offset

46.

?UNDEFINED.

Examples:

Command
DIMENSION

DIM

DATA (10,10),46/E

COLUMN (10),DATA(1,3)/E

Meaning
A

REAL*4

array

named

one-dimensional

COLUMN
column

dimensions

equivalenced

the

array

REAL*4

the

DATA.

array

third

FORTRAN DEBUGGING TECHNIQUE

(FDT)

The following example shows a useful trick for naming one
an

element

array.

NAME W,DATA (5,5)/E

A REAL*4 variable W
array
to
NAME
(see
2.5.8).

element
command

equivalenced

(5,5)
1in

of DATA
Section

FORTRAN

DEBUGGING

TECHNIQUE

(FDT)

ERASE

2.5.4

ERASE

The

ERASE

command

and

frees

the

The

ERASE

command

cancels

space

has

the

FDT's
the

association

internal

following

name

and

location,

tables.

form:

ERASE namel[,name2,...]
You

can specify as many names or subscripted names in an ERASE command
as
can
fit on a single line.
The names must be separated by commas.
Subscripted names must
appear
without
subscripts;
subscripts
are

invalid

for

part

the

ERASE

command.

array.

not possible

the

variables

erase

the

name

Examples:
ERASE

TIME

ERASE

SPEED,DIST,DATA,ICOUNT

ERASE

does

names

are

not

change

the

values

erased.

2-19

whose

associated

FORTRAN DEBUGGING TECHNIQUE

(FDT)

GOTO

2.5.5

The GOTO command changes the order of execution of commands within an
FDT macro. It causes an unconditional transfer of control, analogous
(Do not confuse the
to that caused by the FORTRAN GO TO statement.
change the order
cannot
GOTO
FDT
the
TO;
GO
FORTRAN
the
and
FDT GOTO
of execution of FORTRAN statments.) Like the FORTRAN GO TO, the FDT
GOTO requires numeric statement labels.
The form of the GOTO command is:
GOTO

label

(Remember that FDT command names must not contain spaces;

GO TO is an

invalid FDT command.)
The

parameter

label

is:

The numeric label of

the

command

which

control

The label must be an
transferred by the GOTO command.
not
Embedded spaces are
integer in the range 1 to 32767.

valid.

FDT uses the first
labels have the same value,
If the label you reference does not exist
the label.

If two or more
occurrence of

within the current macro, FDT prints the error message ?LABEL.

You can use the IF command and the GOTO command to create loops of FDT

commands analogous to simple FORTRAN loops or to FORTRAN DO loops (see
sections 2.5.6 and 2.5.7). For example, if the first command in the
the 1last command in a loop might be IF
loop has the 1label 100,
The only loops you can do are on conditions or
NEXT<>0;GOTO 100.
values:;

there is no way to increment an index in FDT.

If FDT is executing an infinite loop, you can terminate execution only
The monitor's START
by typing CTRL/C twice to return to the monitor.
or REENTER commands can then operate in the same way as the START
However, in some cases, you will have to use the RUN
command in FDT.
command to reload the FORTRAN program and start the debugging session
from the

beginning.

Defining

Label

You label a command by preceding it with an integer.

The

label

must

the 1left parenthesis that marks the beginning of the
follow either
macro definition or the semicolon that delimits the preceding command.
The label may be set off by spaces in its defining occurrence.
The

following macro definition contains a label:

MACRO 1 (TYPE MAX;
GOTO

10;

10 ACCEPT

'INIT=',INIT

;IF

INIT>MAX;

FORTRAN

DEBUGGING

TECHNIQUE

(FDT)

2.5.6

The

You

command

wuse

The

1IF

given
form
IF

The

requests
specify

condition

the

conditional
that

command

The

<rel>

value

mode

The

logical

one

(not

equal),

equal).

The

value

mode

six

compared

loc,

assumes

ignores

the

FDT

modes
The

the

single

command

must

fit

than),

1loc.

valid.

not)

only

loc

one

the

same

mode

name.

the

also

the

command,
that

discussion

Section
is

the
loc.

loc,

defined
one

mode

mixed

2.5.1.)

executed

specified

of
of

compares

mode
from

same

namne.

mode

has

name

wvalue

FDT

mode

conversions

than

the

mode

and

command

(greater

The

<rel>

(less

the

does

to
J.

(equal),

match

name

actual

defined

are

parameter

previously
need

relations:

(See

The

wvalue

constant

FDT

(less

the
1logical
relation
value is true.

command.

executed

never

another.

either

that

the

contents
loc.

The

FDT

names

should

the

(greater

may

for
must

tested.

logical

(but

loc

equal),

name

and

relation

the

Subscripted
FDT

command

specification

The

parameter

command

location

than

entire

another
is

is:

compared.

The

command

are:

loc

FDT

met.

loc<rel>value;FDT

parameters

execution

another

between

only

loc

and

line.

command compares the value in loc to the value
specified.
relation is true, FDT executes the FDT command specified.
If
relation is not true, FDT does not execute the command
but
skips
the next command.
the

If
the

Example:

PARM/I>5;WATCH

PARM

COUNT,

greater

than

and

executes

the

executes

the

COUNT;STEP

FDT

FORTRAN

sets

FORTRAN

statement.

watch

statement;

the

1location

otherwise,

FDT

named
simply

FORTRAN DEBUGGING TECHNIQUE

(FDT)

MACRO

2.5.7

MACRO

The MACRO command allows you to define,

execute,

delete

FDT

An FDT macro is a sequence of FDT commands that is executed as
macro.
or
subroutine
Thus, an FDT macro is analogous to a FORTRAN
a unit.
subprogram.

function
1.

Defining

a macro

You can create a new macro,
command

typing:

MACRO m(FDT
The

parameters

m
FDT commands

change an existing macro,

with a

MACRO

commands)

are:

The number

range

the macro where m

Any valid FDT command
There

space between

1is

series of
the

value

the

and

the

commands.

macro

number

(m)

left
parenthesis.
The FDT commands appear within the
parentheses, separated by
semicolons.
There
1is
no
limit to the number of commands defining a macro.
The
definition may continue on as many lines as necessary.
FDT
prompts
is defined by

for
the

each new line.
The end of
closing right parenthesis.

the macro

Examples:

The following example prints the values of
the
variables
associated
with
the
names
TIME, SPEED, and DIST, and the first two elements of
the

named

array

MACRO

DATA:

3 (TYPE TIME,SPEED,DIST,DATA(l)
,DATA (2))

The following example shows a
multiple-line
macro
definition.
The
macro
accepts a floating-point value VAR from the terminal, and tests
whether VAR is within the limits required by the program.
If
VAR
is
outside
the limits, FDT goes to the command labeled "100" and prompts
for

another

definitions:

NAME
NAME
Macro

value.

VAR, 234/E
TOP, 240 /E

definition:

MACRO
IF

5(100 ACCEPT

VAR>TOP;

GOTO

'VAR=',6VAR;
1luU;

CON)

1IF VAR<K1l2.;

GOTO

100;

FORTRAN DEBUGGING TECHNIQUE

Executing

You

can

FDT

executes

macro

execute
a

macro

You

macro

execute

MACRO

The

automatically

PAUSE

whenever

command

manually

or
it

manually:
executes

containing

typing

the

MACRO

FDT

pause

parameter

following

(see

command:

parameter

is:

The

either

automatically

that
was
set up
Section 2.5.9).
can

(FDT)

The number of
m is a number

specified

macro

the macro to
in the range

executes

until

FDT

be executed.
0 to 7.

reaches

the

end

The

value

the

macro,

at
which
time
FDT
prompts
you for more commands by printing !, or
until a START, STEP, or CONTINUE command within the MACRO is executed,
at which time FDT resumes execution of the FORTRAN program.
3.

Deleting

You

can

The

form

delete

MACRO

The

a macro
a

the

macro

command

redefining

with

is:

null

command

string.

m()

parameter

is:

The

()

Null

number

not

command
enclose

the

macro

string.
any

The

the

left

characters

range

and

right

parentheses

must

spaces.

Implicit Macro
MACRO

special

macro

number

that

you

cannot

associate

with
a PAUSE.
Any current PAUSE command that you have not
associated with another MACRO is implicitly associated with
MACRO

command
MACRO

usually

without

not
a

defined.

MACRO

1If

parameter

you

define

executes

manually.

2-23

MACRO

You

explicitly

explicitly
MACRO 0.
every
can

PAUSE

execute

FORTRAN DEBUGGING TECHNIQUE

(FDT)

NAME

2.5.8

NAME

The NAME command associates a name with a location and
a
mode.
The
name must be unique in the first six characters.
Any characters after
the sixth are ignored.
The first character of
the
name
must
be
a
letter.
The other characters can be either letters or digits.
The

form of
NAME

The

the NAME

command

is:

name[,loc]

parameters

name

loc

are:

The name to be associated with
the
wvariable.
If
the
name
you
specify
has
already been defined, this NAME
command redefines the name.
The o0ld definition is lost.

The location specification

2.4.2
and
2.4.3).
If
specification, FDT cancels

ERASE

command,

Section

(as

described

Sections

you
omit
the
location
the
name
association
(see

2.5.4).

Examples:

Command

NAME

Meaning

I,202

INTEGER*2 variable I

NAM PARM,16/PI

INTEGER*2 parameter

NAM DELTA,I+2/E

REAL*4 variable DELTA at offset 204.
(Relative location I+2 is location 202+2.)

NAM EPSILON,DELTA+4/E

REAL*4 variable EPSILON at offset 210.

NAM

Erases

PARM

the

at offset

202.

PARM at offset

definition of

PARM.

16.

FORTRAN

DEBUGGING TECHNIQUE

(FDT)

PAUSE

2.5.9

PAUSE

The PAUSE command
Section 2.4.4).

defines

The

PAUSE

marks

want

command

FDT

statement

the

pause

number.

one

the

statement

occur.

The

statement

You

pause

mark

occurs

pauses

and

program

your

the

only

entry

statement

when

the

pauses

before

which

you

statement

using

marked

(see

its

internal

executed.

You can define a statement pause for any executable FORTRAN
statement
in your program, even if that statement is located in a procedure that
is nonresident or in a shared library.
The location of the
pause
is
stored
itself.

internally

You

place

can

entry

FDT.

pause

not

the

stored

entry

the

point

FORTRAN

program

any

FORTRAN

subroutine
or function.
Whenever the procedure is called, FDT issues
the message FDT PAUSE AT ISN 0 IN proc.
You can use this
feature
to
detect
calls
to
a
procedure
without having to determine the first
executable statement of the procedure.
The

PAUSE

There

are

parameters

PAUSE

command

two
MACRO

command

requires

you

specify

optional

parameters.

and

AFTER

either

the

location

You

can

order.

The

specify
complete

the

form

pause.

optional
of

the

is:

PAUSE {[proc],isn} [MACRO m][AFTER ntimes]
proc

The

parameters

are:

proc,isn

The

location

proc

The

name

occurs.

name

printed
you

the

The

current
program,

statement

procedure
default

the name of
procedure

the

specify

procedure

last
a

and

1in

marked.

which

value

for

the

pause

procedure

the current procedure.
is
defined
as
the

subroutine,

name

or function
as
1isn
parameter,

the
FDT

function

FDT pause

subroutine
omit

the

The

main
was

whose

message.

establishes an entry pause at the entry point of
the
FORTRAN
procedure
you
named.
You cannot
place an entry pause on the entry
point
of
an
assembly
language routine.
(Note that "proc,0"
is not equivalent
to
"proc"
because
internal
statement

isn

The

number

internal

source

listing.

undefined.)

statement

occur.
The
the
left

where

internal
of
each

number
you

want

the

FDT

FORTRAN
pause

statement number appears to
statement
in
the
source

FORTRAN DEBUGGING TECHNIQUE

(FDT)

NOTE

FDT does not give an

error

message

or
nonexistent
a
specifies
proc,isn
FORTRAN
the
nonexecutable statement in
an
containing
command
The
program.
is
number
statement
internal
invalid
the nonexistent or
effectively ignored;
by
executed
nonexecutable point is not
the program and thus cannot cause an FDT
pause

MACRO m

occur.

FDT
the
when
executed
1is
The FDT macro that
An FDT macro is a sequence of FDT
pause occurs.
1is
It
unit.
a
as
commands that is executed
in
function
or
subroutine
a
to
analogous
PFor more
FORTRAN.
see Section 2.5.7.

FDT

information

macros,

FDT
the
The m argument specifies the number of
want to associate with the FDT pause
you
macro
an
be
The macro number must
you are creating.
integer in the range 1 to 7.
The MACRO m parameter

is valid

whether

not

macro can be defined or
The
exists.
m
macro
PAUSE
changed at any time without affecting the
the macro does not exist when the
If
command.

operates
FDT
pause occurs,
parameter were not specified.
If

the macro does exist when the

the

pause

macro

occurs,

macro without issuing an FDT
the
executes
FDT
If the macro contains a CONTINUE
pause message.
program
FORTRAN
the
of
execution
command,
a
contain
not
does
macro
the
If
resumes.
command, FDT prompts for more commands
CONTINUE
when

it has

finished executing

the macro.

You can abbreviate the word MACRO to MAC.
Only
for
checked
are
characters
three
first
the
(Remember that there must be
spelling accuracy.
a
space
between
the
word MACRO and the macro
number,
AFTER

ntimes

m.)

The
execution
the
specifies

count.
number

parameter
the program
must
reach
(but
not
execute)
the
marked
statement
before
an
FDT
pause occurs at that
point.
The ntimes value must be an
integer
in
the range 1 to 32767.
The default value for the

The
ntimes
times that

AFTER ntimes parameter is 1, which causes an
pause the first time program control reaches
statement.

FDT
the

FORTRAN

FDT

DEBUGGING

decrements

control

TECHNIQUE

the

reaches

specified.
is
zZero.

automatically
default value

(see

value of
value.)
You

the

abbreviate
first
three

spelling
a

space

count,

can be at most eight active
one
time.
If
you
attempt
to
program, FDT prints out the error
You

cannot

pause

place

you

definition

Statement

pauses

command,

Section

two

pause

try

are

not

definitions
the current

the

initial

the

(Remember
word

that

AFTER

and

there
the

Only

for

must

execution

PAUSE

commands in a program
at
define
more than eight pauses
message ?NO ROOM.

definitions

place

two

the

first.

cancels

the
required
ntimes command

pause
contain

count,

the
the

ntimes.)

There

program.

count
FDT

the word AFTER to AFT.
characters
are
checked

accuracy.
between

specify
CONTINUE

(The
command

execution

can

the

time

number

occurs when
the
pause
occurs,

count,
1in
a

WHAT

each

statement

the execution count to
If you want to override

2.5.2).

the

count

internal

FDT pause
When
the
resets
of 1.

Section

output

execution

the

default execution
execution
count

(FDT)

disabled

when

pauses

FDT

the

same

point

the

same

point,

the

step

mode

(see

any
in a

vyour

second

STEP

2.5.12).

Example:

The

following
PAUSE

places

statement

command

SUBR;PAUSE

entry

pause

the

line
,10

procedure

current

SUBR,

procedure.

and

statement

pause

FORTRAN DEBUGGING TECHNIQUE

(FDT)

RESET

2.5.10

RESET

The RESET command removes pauses created by fhe PAUSE command.
The form of the RESET command is;
|
RESET proc,isn
The

parameters

are:

The 1ocation in the FORTRAN

proc,isn

proc

RESET

the

internal statement

source

command

,10;RES

in which the pause was

statement where

number

the

the pause was

the

FORTRAN

to occur.

line

SUBR

removes the statement pause on statement 10 of the

and

which

occur.

The

isn
example,

for

the pause definition.

The name of the procedure
to

For

program

The location specified
was defined.
pause
the
appeared
must be exactly the same as that which

entry pause on procedure SUBR.

current

procedure

FORTRAN DEBUGGING TECHNIQUE

(FDT)

START

2.5.,11

START

You can issue the
START
command
whenever
FDT
prints
its
command
prompt.
The
FDT START command begins executing your FORTRAN program
at the first executable statement in the main program.
However, START
might
not
work
when
the
system
is
not
in
an
initial or ready
condition.
For example, if there are
open
files,
a
START
command
causes an error message and FDT returns control to the monitor.
The START command has
no
parameters.
When
the
executed, FDT ignores any commands remaining in the

START
line.

command

1is

FORTRAN DEBUGGING TECHNIQUE

(FDT)

STEP

2.5.12

The STEP command continues execution of ybur program in step mode, and

indicates the number of statements that you want to execute before the

The

FDT

pause.

forms of the command and

its optional parameter

are:

STEP [n]
S [n]

(Notice that the STEP command has the special abbreviation S.)
The

parameter

is:

integer

in the range

32767.

FDT executes n FORTRAN

The default value for n is 1,
pausing.
before
statements
which results in single-step operation with a pause
before
each

statement.

Step mode disables all statement pause definitions.
You
cancel
step
mode
and
reenable
the pause definitions by entering any FDT command
(other than another STEP command).
The occurrence
of
an
FDT
pause
also

cancels

step mode.

STEP counts only executable

one

logical

executable
1IF

(depending

statement

the

statements.

statement
may

one

A logical

statement counts

contains a GOTO;
two

outcome of the IF test).

executable

otherwise,

statements

The END statement is an

executable statement.
Nonexecutable statements are FORMAT statements,
declaration statements, and SUBROUTINE and FUNCTION statements.

After
in

the

it executes a STEP command,
line.

FDT ignores any

commands

remaining

FORTRAN DEBUGGING TECHNIQUE

(FDT)

STOP

2.5.13

STOP

You issue the STOP command to
end
a
command
in
FDT
is
equivalent
to
program.
The STOP command closes any

necessary exit
The

STOP

actions,

command

has

and

returns

parameters.

debugging
session.
The
STOP
STOP statement in the FORTRAN
open
logical
units,
performs

control

the

operating

system.

FORTRAN DEBUGGING TECHNIQUE

(FDT)

TYPE

2.5.14

TYPE

The TYPE command prints the values of variables in the FORTRAN program
'

during a debugging session.

The forms of the TYPE command are:
TYPE

loc|[,more]
or

TYPE
The

'text'([,more]

parameters

are:

The location specification of the variable whose value

loc

you want to examine.

A literal text string to be printed on

'text'

the

terminal.

The FORTRAN conventions for text literals apply to FDT
literals.

Another

loc or

text parameter.

Multiple loc or text parameters may appear in a single TYPE command.
The maximum length of the
The parameters must be separated by commas.
TYPE command

is one line.

FDT finds each location specified and prints the contents of the
FDT
location in the mode indicated by the location specification.
the
for
(except
them
type
you
as
exactly
literals
prints text
If multiple parameters appear in a TYPE
enclosing quotation marks).
command, FDT separates the values with commas.
Examples:

Meaning

Command

TYPE 202

Prints INTEGER*2 variable at offset 202.

TYPE

'HI'

Prints HI.

TYPE

'DON''T'

Prints DON'T.

TYPE DELTA,EPSILON

Prints two values separated by a comma.

TYPE DELTA/O

Prints first 16 bits of DELTA in octal.

TYPE

'DELTA=',DELTA Prints DELTA=

TYPE DATA(I,J)

followed

the

value

DELTA.

Prints the value of the
in

DATA.

(i,]j)

array element

FORTRAN

DEBUGGING

TECHNIQUE

(FDT)

WATCH

2.5.15

WATCH

The

WATCH

perform

command

The

form

WATCH

The

directs

watch
the

pause

WATCH

FDT

watch

the

value

whenever

command

contents

that

location

and

changes.

is:

[loc]

parameter

is:

loc

The specification of
the
1location
to
be
watched.
(Secticns
2.4.2
and 2.4.3 describe how to specify a
location.)
Only one location can be watched
at
any
given
time.
A
new
WATCH
1loc command cancels the
previous WATCH command.
A WATCH
command
without
a
locaticn specification cancels the previous WATCH loc
command.
The location to be watched
can
have
any
FDT
mode
except
Alpha (/An).
Only the first character of the
string is watched for locations in ASCIZ string
mode

(/2)
.
When
WATCH

the

value

PAUSE,

statement.

equal
If

The

another

and

value

then

FDT

watched

performs
must

the

command

value

changes,

pause

actually

changes

pause
does occur.
pause by using FDT's

watched

FDT

statement

watch
watch

location
an

change.

IVALUE=5
the

value

FDT

the

does
of

prints

For

example,

not
a

For example, you can
information transfer

the

message

executable

cause
watched

FORTRAN

IVALUE

watch

location,

pauses

are

location.

independent

watch
pause and any
or STEP) on the same
pause

and

another

other

pause

first.

START

before

the

Examples

the

deliberately trigger a
commands to change the

Each time a watch pause occurs, FDT cancels
the
WATCH
command
initiated
it.
You
must
issue
another WATCH command with the
location specification if you want FDT to continue watching
the
location.
Watch

pause.

other

FDT

pauses.

You

can

place

other command causing a pause
(for example,
FORTRAN statement.
If you
do
specify
a
pause

You

watch

WATCH

MAX

WATCH

ARRAY (3,2)

must
pause

for

the

resume
can

command:

same

statement,

execution

occur.

using

FDT

that
same
same

processes

CONTINUE,

PAUSE
watch

STEP,

the

FORTRAN DEBUGGING TECHNIQUE

(FDT)

WHAT

2.5.16

WHAT

The WHAT command displays the current status of the FDT systenm.
you type WHAT, FDT returns the following information:

When

A list of the definitions for all the active pauses.
For each
pause,
FDT
prints
the
1location of the pause (the procedure
name and optional
internal
statement
number),
the
current
value
of the execution count (when it is greater than 1), and
any associated macro number.
The pause listing for
an
entry
pause has no internal statement number.
A list of
of

the

all

macros.

currently defined macro

numbers

and

the

contents

FORTRAN DEBUGGING TECHNIQUE

2.6

CAUTIONS

AND

(FDT)

PITFALLS

The following are some general cautions you should observe while using
FDT:

Correct use of

important

Invalid location specifications
cause

unpredictable

in FDT.

constants

out

range

results.

Under the RT-11 SJ operating system, the ends of long lines of
text
printed
at
the
terminal may be lost when the terminal
reaches its right margin.
FDT
does
not
provide
automatic
carriage

spaces

return or

line

feed.

subscripting
of
purposes
For
ASCIZ mode (/Z)
with
locations
For

example,

the

considers
FDT
copying,
and
to have a length of one byte.

command

NAME/Z=INPUT

copies only the first character from INPUT to NAME.
To
copy
an
ASCIZ string, use mode An, where n is equal to the maximum
possible length of the ASCIZ string, including the terminating
zero

byte.

For

NAME/A25=INPUT

copies

bytes

example,

from

the

INPUT

to NAME.

An FDT error message does not indicate that an entire
command
was
abaorted.
Some action may have occurred before discovery
of the error.
1In this case, side
effects
<can
result.
For
example,

the

command

SUBR+12

PAUSE

produces the error message
PAUSE

command

FDT ignores any commands
For example, if you type
STEP

?UNDEFINED,

but

acts

1like

the

SUBR.

;

TYPE

following CONTINUE,
the command line

START,

STEP.

INDEX

FDT executes three FORTRAN statements and then pauses, without
However, commands following CONTINUE, START,
typing anything.
or STEP are executed if FDT control branches around
CONTINUE,
START,

STEP.

For

example,

you

type

SPY<50;CONTINUE;
TYPE INDEX

then FDT resumes execution if SPY is less than 50,
but
the contents of INDEX and waits for another FDT command
is

greater

than

equal

types
if SPY

50.

You cannot place an FDT pause on the GO TO
of
a
logical
IF
Statement.
For
example,
suppose the FORTRAN source listing
contains
0008

the

following

statements:

NEXT=NOW+1

0009

IF (ITEMS (NEXT) .GT.ITEMS
(NOW)) GO TO

0011

TEMP=ITEMS (NOW)

FORTRAN

You
is,

DEBUGGING TECHNIQUE

cannot place an FDT pause
the FDT command
PAUSE

never

internal

causes

DO
IF

0011

statement 10 would be
wvalid
the following statements:

50 NEXT=1,N
(ITEMS (NEXT).GT.MAX)

ADVANCED

The
use

following
to extend

2.7.1

Named

pause

1if

the

MAX=ITEMS (NEXT)

before
If the

the
value
of
MAX
condition is false,

occurs.

TECHNIQUES

sections describe
power.

some

advanced

techniques

that

you

can

FDT's

Common

FDT can access variables in a FORTRAN named
common
block
define the block using the NAME command.
Use the following
ll

That

CONTINUE

In this case, the FDT pause occurs
changes
1if the condition is true.

2.7

10.

pause.

goos
0009

FDT

statement

,10

However, an FDT pause on
source listing contained

(FDT)

Locate

the

address

the

named

common

after
you
procedure:

block.

The absolute address of the named common block appears in the
link
map.
The name given to the block in the FORTRAN source
code appears
in
the
1link
map
under
the
column
labeled
SECTION.
The
6-digit
number
following
the
name
1is the
address

the

block.

For RT-11 FB foreground
programs,
the
named
common
block
address appearing in the link map is a relative address.
The
desired absolute address is the sum of the link map
relative
address
and
the
foreground
Jjob
load address.
Obtain the
foreground load address by issuing the FRUN command with
the
/P

option.

monitor;
Describe

(The

type
the

named

A NAME command
location:
NAME
The

addr

option

common

the

causes

execute
block

the
to

following

control

foreground

return

the

job.)

FDT.

form

defines

the

block's

block, .ABS.+addr

arguments

block

RESUME

are:

The FDT name of
the
common
block.
If
block
name
is
the
same
as
the
name
variable
in
a
previous
NAME
command,
erases the previous name.
The

6-digit

common

absolute

block.

2-36

address

the

the
of a
FDT

named

FORTRAN

Refer
The

DEBUGGING

locations

location

TECHNIQUE

in named

any

(FDT)

common.

variable

named

common

expressed

follows:

block+nnn
The

components

block

The

nnn

the

FDT

specification

name

the

are:

common

block.

The offset of the variable
FORTRAN storage map.

You can assign FDT names to
variables
using block+nnn as a location in a NAME

given

in
named
command.

the

common

Examples:

Assume

the

that

link

the

name

map gives

its

the

FORTRAN

address

named

common

block

COMBLK

and

063524,

Describe the named common block to FDT:
NAME

COMBLK, .ABS.+63524

the

contents of

INTEGER*2

variable

offset

000010

the

address

and

1in

COMBLK:
TYPE

Name

the

COMBLK+10

variable

NAME

contents

TYPE

FDT

COMBLK+10:

the

variable:

INDEX

How

uses

location

INDEX,COMBLK+1l0

the

2.7.2

FDT Generates

the

location

Addresses

specification

generate

mode

information
of a FORTRAN variable.
The manner in which FDT generates
the 16-bit PDP-11 address depends on the information in
the
location
specification.
The following section describes the address-generation
process

Octal
The

for

the

five

kinds

location

specifications.

Offsets

form

oct

the

location

octal

specification

number

the

is:

range

177777.

When you give an octal number as a location specification, FDT
always
interprets
the
number
as
an
offset
from
the base address of the
current procedure's data block.
In other words, FDT
adds
the
octal
number
to the base address to generate the 16-bit address.
Thus, you
can use the offsets produced
by
the
FORTRAN
compiler
to
generate
addresses

for

any

locations

shown

2-37

the

FORTRAN

storage

map.

FORTRAN DEBUGGING TECHNIQUE (FDT)
Names

The

form of

name

the

location

specification

An alphanumeric
The

first

the

variable.

name

character

is:

unique

must be a

the

first six

letter.

FDT

characters.
1ignores

any

first
number in

six
the

characters after the sixth.
The NAME command associates
a name with a location.
Once the association
has
been
established, the name refers directly to the location of

Relative Addressing

The

form of

the

name+oct

location specification

is:

An
alphanumeric
name
unique
1in
the
characters,
a
plus sign, and an octal
range

177777.

This location specification is a form of relative addressing in
which
the
base
address is the location associated with the name.
FDT adds
the octal displacement to the base address to determine
the
1location
of
the
wvariable.
Displacements
in
the range 100000 to 177777 are
negative displacements.
FDT assumes that all octal numbers
represent
two's
complement
binary
integers.
For
example,
the displacement
177776 represents a displacement of -2 bytes (the word
preceding
the
name) .
You
must
determine
the displacement yourself.
It does not
appear in the storage map.

You could use relative addressing for array subscripts.
However, this
is
unnecessary
because
FDT
allows you to define subscripted names.
Relative addressing allows you to address variables
in
named
common
blocks once you have supplied the base address of the common block.
There are
addresses

several predefined names
for relative addressing.

. MAIN,

in FDT
These

that you
<c¢an
names are:

wuse

Dbase

The default name assigned by FDT to the
base
address
of
the
FORTRAN
main program.
You can reference any
location in the main program using this base.
If
you
named
your
main
program
with
the
FORTRAN PROGRAM
statement, then FDT uses that name to
find
the
Dbase
address.

. BCOM.

The name assigned by FDT to the base
address
of
the
FORTRAN
blank
common area.
Use .BCOM.
to reference
variables in blank common.

.ABS.

The zero address of
base
address
for
location.
For
of

example,
absolute
TYPE

the

memory.
FDT
referencing

following

location

.ABS.+56/0

uses
any

command

(octal)

this name
absolute

prints
octal

the

as the
memory

contents

format:

FORTRAN

DEBUGGING

TECHNIQUE

(FDT)

Subscript Addressing
The

form of

the

location

name (i,j,k,..)

specification

subscripted

most

seven

name

is:

for

specifying

dimensions.

The

form
of
addressing
name (1,1,1,...).
FDT

wuses

the

displacement
element.

FORTRAN

from

subscripting

the

base

address

base

with
for

the

algorithm

and

arrays
address

locate

at
this

address

compute

the

specified

array

Indirect Addressing
When

you

define

FDT

wuses

the

name

indirect address.
name

the

1location

That

address

parameter

is,

FDT

the

specifying

with

regards

the

desired

location of the
parameter
PARM
contains 1234, then the command
TYPE

(by

associated

value

mode

For

2000,

name

associated

value.

contains

its

parameter

with

example,

and

P),
as

location

the
the

2000

PARM/I,PARM/PI

prints
2000,1234

2.7.3

Format Conversion

TYPE,

ACCEPT,

and

conversion

routines.

format

that

FDT

routines
D,

with

modes

you

commands

the

use

requires

can

force

for
the

linker

following

When you link your program,
first linker command line.

The

Type

procedure:

responds

return

end

On-line

If you use
sometimes

the

contains

Debugging

assembly
need
to

linker

and

load

include

carriage

conversion

program

the

library

debugging
does

modes.

1If

the

not

have

not

load

the

require

you

required

format

does

routines

switch

the

request

the

SEARCH:

(followed

only)

FORTRAN

are

with:

floating-point

resulting

2.7.4

RCIS

then

the

linker

the

program you

specifications,

LIBRARY

The

Routines

library

all

Technique

return)

routines

the

and

necessary

null

line

(carriage

list.

conversion

routines.

(ODT)

language routines with
use
FDT
and
ODT at

FORTRAN routines,
you
may
the same time.
There is no

interaction between the two
debugging
techniques
except
when
ODT,
which runs at a higher priority, occasionally interrupts the output of
FDT

traps.

the

FORTRAN

program.

FDT

does

not

use

breakpoint

T-bit

FORTRAN DEBUGGING TECHNIQUE (FDT)
2.7.5

Execution Speed

traceback feature to
FDT uses the FORTRAN internal statement number N progra
m and to execute
gain control at the beginning of a FORTRA some overhea
d to the
adds
FDT
re,
Therefo
PAUSE and STEP commands.
execution of each FORTRAN statement and slightly reduces the execution
speed of any FORTRAN program.

program is difficult
The amount of FDT overhead time in a main FORTRAN
ritical
However, you can get fully debugged, time-croutine
to reduce.
s
these
compile
should
You
speed.
subroutines to run at full
these
within
FDT
use
cannot
You
d.
disable
feature
ck
traceba
the
with
of FDT for
routines. However, you can use the entry-pause feature
traceback
the
require
not
does
pause
entry
the
because
s
routine
these
ine
subrout
a
before
tely
immedia
feature. If you issue a STEP command
the
to
s
proceed
control
,
numbers
nt
stateme
l
interna
without
compiled
ck
traceba
the
with
routine
a
in
next executable FORTRAN statement
feature enabled.

pause
Execution speed with FDT is affected by FDT pauses. Each active
slows
larly
particu
It
nts.
stateme
slows the execution of all FORTRAN
the
of
that
than
greater
numbers
nt
stateme
l
interna
with
those
if
quickly
more
runs
program
The
statement where the pause occurs.
(see
pauses
Entry
y.
necessar
longer
no
are
that
pauses
you reset
Section 2.5.9) do not require overhead for traceback and hence are
more time efficient than other PAUSE commands.

APPENDIX A
ARGUMENT AND DEFAULT SUMMARY TABLE
The table contains arguments and default values for the extensions routines. Arguments in
square brackets can be either defaulted or omitted. If there is a numeric default value, it
appears in the column directly below the argument name. Consult the relevant reference section to
find out what happens when you omit a bracketed argument that does not have an explicit default value.

SECTION

ROUTINE

ARGUMENTS

1.4.1

CLRD

iarrayname

npoints

1.4.2

CVSWG

ivalue

[igain#]

1.4.3

DIS

iydata

144

DRS

[iarrayname]

ispace#

rscale

ispace

npoints

istart

increment

iarraysize

[nsubarrays]

{nsamples]

[isource]

iarraysize

iydata

npoints

istart

increment

i.4.5

DXY

ixdata

1.4.6

FLT16

ivalue

1.4.7

FSH

iydata |

ispace

npoints

istart

increment

14.8

FXY

ixdata

iydata

npoints

istart

increment

1.4.9

HIST

iarr_ayname

iarraysize

nsamples

iendflag#

nleft#

1.4.10

IADC

ichannel

[igain]

ivalue#
litype]

[iwhere#]

[iresult#]

iunit

imask

l[imodeTM]

iendflag#

nleft#

Hintervall

[complete]

iarraysize/nsubarrays

[interval]

[complete]

iarraysize/nsubarrays

1.4.11

IDIR

(a)

[isource]

iunit

imask

1.4.12

(b)

iflag

ibefore

iop

(c)

iflag

imask

[idest]

iunit

iselect

iset

IDOR (a)

(b)

—

iunit

1.4.13

KB2BCD

ibinary

1.4.14

KBCD2B

idigits

1.4.15

LED

ivalue

‘format’

1.4.16

LWAIT

ivalue1

ivalue2

1.4.17

REL

irelay

isetting

1.4.18

RTS

[iarrayname]

iarraysize

1.4.19

SDIS

1.4.20

SETR

lirate]
-

[ioutput#]

imode

imask

[iset]

[iclock#]

[nsubarrays]

[nsamples]

[ifirst]

[nchannels]

[igain]

[imodeTM]

iarraysize

rcount

iendflag#

[interval]

[complete]

iendflag#

nleft#

APPENDIX

OBJECT AND

LDPOBJ.OBJ,

which

you

receive

software,

SOURCE

FILES

part

the

is
a
file concatenating 28 individual
appendix lists the names of all the object modules
the
corresponding
source
files.
The
FORTRAN
software kit includes one source file, CONFIG.MAC.

files

can be

Object

File

CONFIG.OBJ

CLRD.OBJ

ordered

Source

separate

kit.

File

CONFIG.MAC

CLRD.MAC

CVSWG.0OBJ

CVSWG.MAC

DIS.OBJ

DIS.MAC

DRS.OBJ

DRS.MAC

FSH.OBJ

Configuration

CLRD

definition

DIS,

SDIS,

DRS

processor

FSH

and

DXY

processors

for

HIST.OBJ

HIST.MAC

HIST

digital

1I/0

processor
processor

IADC.OBJ

IADC.MAC

IDIR.OBJ

IADC

processor

IDIR.MAC

IDIR

processor

IDOR

processor

IDOR.MAC

all

processor

FLT16

KB2BCD.MAC
KBCD2B.MAC
LAO.MAC
LAl .MAC
LA2.MAC
LDISP.,MAC

module

CVSWG processor

FLLT16.MAC

IDOR.OBJ

files.
This
and
the
names
of
extensions
binary
The
other
source

processor

FLT16.0BJ

KB2BCD.OBJ
KBCD2B.OBJ
LAO.OBJ
LA1.0BJ
LA2,0BJ
LDISP.OBJ

extensions

object

Contents

units
FSH.MAC

FORTRAN

KB2BCD processor
KBCD2B processor
Parameter check routine
Parameter check routine
Parameter check routine
Completion routine dispatcher

LED.OBJ

LED.MAC

LGAO.OBJ

LGAO.MAC

LGA1.0BJ

Argument

LGA1l.MAC

fetch

routine

LICMP.OBJ

Argument

routine

LIN16.0BJ

LICMP.MAC
LIN16.MAC

fetch

Completion

Device

LED

processor

Convert

interrupts
unsigned binary integer
initialization routine

LINIT.OBJ

LINIT.MAC

LPARS.OBJ

LPARS.MAC

LPUT.OBJ

Command

parser

LPUT.MAC

Buffer

REL.OBJ

insertion

REL.MAC

RTS.0OBJ

REL

RTS.MAC

processor

SETR.OBJ

RTS

SETR.MAC

processor

SETR

LPSMAC.MAC

System

DIS,

DXY,

and

FSH

routine

processor
macro

included

command

file

for

line

except

definitions.

in
to

LPSMAC must
MACRO assembler
assemble
any
source

the

CONFIG.MAC.

APPENDIX

ERROR

The table summarizes all of
extensions
routines.
The

ADC,

CLK)

Routine
CLRD

are

described

SUMMARY

TABLE

the conditions
causing
errors
error
name abbreviations (SYN,

Section

1.6.

Error
SYN

Cause
Four

arguments

The

number
(npoints) is

CVSWG
DIS

SYN

from
the
ARG, DEV,

The

scale

One

SYN

Five

ARG

The

required.

el ements
to
be
scaled
less than one.
factor
(rscale)
is negative

two

arguments

arguments
number

required.

required.
points

display

(npoints)

is less than one or greater than 4096.
The first point (istart) is less than one.
The indexing
factor
(increment)
1is
1less
than one or greater than 4096.
DRS

SYN

Ten

ARG

The

digital

arguments

input

does not exist.
The memory
address
odd.
The

mode

The

mode

requested
requested

the
call
mode.
DEV
DXY

Digital

SYN

Five

ARG

The

1is

not

input

unit

arguments
number

required.

unit

requested

specified

(iunit)

1is

(imode) is not valid.
(imode) is negative, but
in parameter adjustment
is

use.

required.
points

display

(npoints)

display

(npoints)

is less than 1 or g reater than 4096.
The first point (istart)
is less than 1.
The indexing factor
(increment) is
greater

FLT16
FSH

SYN

than

4096.

One

argument

SYN

Five

ARG

The

required.

arguments

number

required.

points

is less than 1 or greater than 4096.
The indexing
factor
(increment)
1is
than 1 or greater than 4096.
The first point (istart)
is less than

less
1.

ERROR SUMMARY TABLE

Routine
FXY

Cause

Error

SYN
ARG

Five arguments required.
The number of points to
display
(npoints)
is less than 1 or greater than 4096.
The first point (istart) is less than 1.
greater
The indexing factor (increment) is
than

HIST

SYN
ARG

4096.

Five arguments required.
The length of
the
array
less

than

zero.

The number

IADC

SYN
ARG

One
The

(iarraysize)

(nsamples)

of samples

to three arguments required.
channel
number
1is
greater

(AR11l,

The

AD1l1-K).

channel

greater

1is

number

(AD11-K with AM11-K, LPS1ll).
The
gain
code
(igain)
1is
greater

DEV

IDIR

IDOR

LED

Four or five arguments required.
Memory address (iunit) is odd.
Digital input
register
(iunit)
exist.

SYN

Four or five arguments required.
Memory address (iunit) is odd.
Digital output register
(iunit)
exist.
One argument required.
The value (ibinary) is

negative

9999.

SYN

One

argument

SYN

Two arguments required.
Illegal format requested.

63
or

does

not

does

not

greater

required.

System does not have

an LPS1l.

none

SYN
ARG

RTS

use.

than

DEV

REL

than

1less

the AR1l.

SYN
ARG

ARG

LWAIT

Autogain was requested for

than

negative

A/D converter
is

ARG

KBCD2B

ADC

ARG

KB2BCD

than

1is

|is

Two
The

arguments required.
relay number is less

than 1.
an LPS1l.

DEV

System does not have

SYN

10 to 12 arguments required.
Starting channel number is too large.
The
number
of
channels
(nchannels)
is
negative or too large.
Autogain was requested for the ARII.
The gain code (igain) is greater than 4.
The mode (imode) is less than -4 or greater
than 15.
The sum of ifirst and nchannels requests
a

ARG

nonexistent

channel.

Array partition
nchannels.

size

not a

multiple

ERROR

Routine

SUMMARY

TABLE

Error

Cause

Autogain was requested with DMA.
Continuous
sampling
with
stop
code
requested with DMA.
The array length (iarraysize) with
DMA
greater than 4090.
A/D converter 1s in use.

DEV
SDIS

none

SETR

SYN
ARG

Four

Rate

The

mode

six arguments required.
less than -4 or greater

requested

The

count
value
greater than 255.

DEV

Clock

CLK

Clock

interrupt

(imode)
(rcount)

than

not
for

(clock

1is

valid.
AR1ll

running.

lost

was

overrun).

APPENDIX

FDT

Command

Parameters

loc=value

COMMAND

SUMMARY

Description

Assign value as the new
of loc.
Print text on terminal.

'text'
loc

and

value

assign

from

contents

the

terminal

loc.

CONTINUE

[ntimes]

Resume

DIMENSION

name (i,j,...) [,1loc]

Associate a location specification and a subscript list
with
a

FORTRAN

execution.

name.

ERASE

name [ ,name2,...]

Remove

GOTO

label

Unconditional
branch
command
changes execution sequence within
an

loc<rel>value;FDT

FDT

the

NAME

m(FDT

commands)

macro.

the

Define

FDT

Execute

m()

Delete

name
[ ,1loc]

Associate

proc,isn

[AFTER

ntimes]

FDT

condition

FDT
a

an
statement

m.
m.

location

only

true.

macro

[MACRO

Create

command

macro

FDT

cation with
PAUSE

associations.

command
Execute

MACRO

name

specifi-

name.

m]
FDT

pause

number

isn

internal

procedure

proc.

RESET

proc,isn

START

STEP

Remove

Begin

[n]

Resume

FDT

pause.

execution
execution

statements.

main
and

program.
execute

FDT

Command

COMMAND

SUMMARY

Parameters

STOP

Description
Return

to operating

TYPE

loc
'text'

Print value
terminal.

WATCH

loc

Cause
an
contents
location

WHAT

Print FDT

FDT
of

text

change.

system.
on

pause
the

status.

the

when
the
specified

APPENDIX

FDT

LOCATION

SPECIFICATION

Format

FORMATS

Meaning

XXX

Location

name

Named

name-+xxx

Relative

name (i,Jj,k,...)

Subscripted

.MAIN.

Base

address

the

main

offset

octal

bytes

location

addressing
name
of

the

program

FORTRAN

was

not

main

named

program

statement)
.BCOM.

Base

.ABS.

The

address

zero

address

FORTRAN

blank

memory

common

(if
PROGRAM

APPENDIX
FDT

Mode

FORTRAN

MODES

Type

Description

INTEGER*2

l6-bit

INTEGER*4

bits,

first

LOGICAL*4

bits,

displayed

LOGICAL*1

REAL*4

value

bits,

displayed
16

decimal

displayed
as

as T or F
scientific notation

REAL*8

bits,

scientific

COMPLEX

bits,

real

BYTE

bits,

and

notation

imaginary

displayed

————

bits,

displayed

————

bits,

displayed

octal

—-———

string

255)
string

handling

mode

indicate
variable.

the

that

above

the

table

can

associated

n
(as

RAD50

ASCII
used

parts

decimal

ASCIZ

decimal

displayed

————

in
F

bits,

Any

characters

characters
the

FORTRAN

string

package).

preceded

location

represents

the

1letter

FORTRAN

"P"

parameter

APPENDIX

FDT

ERROR

Message

?BAD

DIM

?BAD

LOC

The

MESSAGES

Description

dimension

limit

invalid

location

correctly

?BAD

MACRO

?BAD

SUBS

Error

was

specification
according

references
bounds.

attempt

was

location

made

specified.

1is

not

1ts

mode

outside

aligned
or

the

redefine

program

macro

while

executing.

Subscripts

are

invalid

format

cause

overflow.
FDT

START

FAIL

Invalid

main

program,

internal
statement
main program.
?FORMAT

The

input

constant

bad

start

numbers

not

address,

not

enabled

the

expected

mode.
?LABEL

?MACRO
?NO

CONVERSION

The

label

specified

does

The

macro

number

not

Floating-point
current

?NO

ROOM

Not

?20NLY

?PAUSE

$SUBSCR

MACRO

NOT

OUT

2UNDEFINED

The

FOUND

BOUNDS

formats

FORTRAN

enough

PAUSE,

operation
FDT

attempt

exist.

the

range

are

not

available

program

memory

MACRO,

not

(see

space

Section

define

0-7.

with

2.7.3).

new

NAME.

attempted

valid

only

within

macro.
was

not

active.

The

specified

made

subscripts

dimensions.

(Warning

Syntax

error

remove

exceed

message

undefined

PAUSE

the

only.)

name.

that

declared

INDEX

A/D, 1-2, 1-3, 1-41, 1-62
Abbreviations, 1-80, 2--3, 2-30
.ABS., 2-36, 2-38
:
Absolute address, 2-36
ACCEPT, 2-3, 2-13, 2-21, 2-39

Circular

ADC

software, 1-2, 1-6,
1-52, 1-~75
CLOCK ERROR~--RATE TOCO

CONFLICT,

1-80

Address,
absolute, 2-36
base, 2-4, 2-17,

2-37,

2-38,

hardware,

Clock

CLrRD,

ARGUMENT

Code,

FDT

circular, 1-7, 1-8, 1-20,
1-39, 1-63
virtual, 1-4, 2-17
Array element, 2-6, 2-10, 2-39
Array partition, 1-7, 1-8,
1-20, 1-24, 1-63, 1-67
ASCII

mode,

2-17

mode, 2-33, 2-35
Asterisk convention, 1-10
Asynchronous, 1-5
Autogain, 1-2, 1-42, 1-43,
1-65, 1-69

pause,

2-1,

1-13

1-72

frequency,

1-3,

1-13,

mode,

2-7,

gain,

1-15,

inline,
mode,
.

2-38,

2-4,

2--17,

FDT,

1-17,

data

type,

1-55,

1-56

2-1

2-2
control,

2-10

2-37,

control,

1-9,

2-19,

to,

2-5,

2-38
2-38

1-1, 1-2,
1-8, 1-19,

1-3, 1-4, 1-5,
1-24, 1-25, 1-28,

1-33,
1-59, 1-68, 1-71,
Conditional transfer of

1-55

control,

1-22,

CONFIG.MAC,

1-39,
1-76

2-21
1-78

Configuration,

to,

1-56

two's complement, 1-30,
unsigned, 1-6, 1-30
Bipolar mode, 1-42, 1-65
Bit location, 1-22, 1-45
Bit manipulation, 1-46
Bit values, 1-10
Blank common, 2-5, 2--38
BOUNDS,

OUT

2-2

2-32

Common,

system,

$SUBSCR

2-2,

2-33

program

Commas,

2-3

transfer,

1-1,

Configuration
BCD

1-65

1-53,

1-2,

1-35

types,

2-4,

Binary,
convert

1-33,

2-8

1-42,

information

2-39

convert binary

1-80

2-17

Command

FDT

BCD,

BCD

FAST,

1-72

named, 2-5, 2-36,
Completion routine,

address,

1-48,

2-1

threaded,

blank,

Base

1-20,

Commands,

ASCIZ

Automatic FDT

1-8,

Clock mode, 1-72
Clock overflow, 1-40
Clock rate,
setting, 1-72

1-80

Array,

1-7,

Clock,

2-39
indirect, 2-39
relative, 2-36, 2-38
/An, 2-13, 2-17, 2=-33
Analog-to-digital, see A/D
ERROR,

array,

1-39, 1-63
Clear display,

OF,

2-6

Brackets,

square, 1-9
Byte manipulation,

1-46

1-78

routine,

1-78

CONFLICT,

2-38

ADC,

1-80

DEVICE,

1-80

Constant,

1-9,

octal,

1-10

string,

2-13

CONTINUE,
2-26,

2-35

2-2,

2=16,

2-23,

Continuous display, 1-3,
1-27, 1-35, 1-71

1-16,

Continuous sampling, 1-19,
finite, 1-19, 1-62
infinite, 1-19, 1-62

Index-1

1-62

INDEX

Control,
relay,

1-61

scope, 1l-16, 1=-27
unconditional transfer of,
2-20

Control commands,
FDT, 2-3
program,

2-2

Convention,
asterisk, 1-10
mode, 2-13
notation, 1-9
syntax, 1-9, 2-3

Conventions, 1-9, 2-13
Conversion,
16-bit floating-point, 1-30
format, 2-15, 2-39
single analog-to-digital,
1-41

Convert BCD to binary, 1-56
Convert binary to BCD, 1-55
Convert switched gain value,
1-15

Digital read-in sampling, 1-19
Digital-to-analog, see D/A
DIMENSION, 2-3, 2-6, 2-9, 2-10,
2-17
Dimensions, see Subscripts
number of, 2-6
pis, 1-3, 1-13, 1-14, 1-1l6,
1-71
Displacement, 2-~5, 2-6, 2-38,
2-39
Display,
clear, 1-13
continuous, 1-3, 1l-l6, 1-27,
1-35, 1-71
LED, 1-57
single, 1-3
stop, 1-=71
Display data, 1-16

execution, 2-16, 2-26, 2-34
Counting events, 1-=72
CTRL/C, 1-7, 2=20
Current procedure, 2-4, 2-6,
2-11,

2-25

1-2,

1-65,

1-3,

1-15,

1-27

Display X-Y data pairs,

DMA sampling, 1-63, 1-69
Double-word sampling, 1-22
DRS, 1-2, 1-6, 1-7, 1-8,
1-19, 1-30, 1-74, 1-75
Dual

Count,

cvVswGg,

(CONT.)

sample-and-hold,

LPS1l, 1-63
Dummy variable,

DXy,

1-3,

1-14,

1-10,
1-27,

2-8
1-71

1-43,

1-69

Element,

array, 2-6, 2-10, 2-39
Entry pause, 2-10, 2-25,

2-34,

2-40

Entry point, 2-10, 2-25
ERASE, 2-3, 2-17, 2-19,

Data,

BCD,

1-22

display, 1-3, 1-16
removing, 1-8, 1-67

Data overrun,

1-7,

1-39,

1-67,

1-69

Data pairs,
display X-Y, 1-27
flash X-Y, 1-35
X-y, 1-3, 1-4, 1-27,
Data type, 2-4, 2-7 to

1-35
2-10
2-1, 2-=2

Debugging procedure,
Defaulted, 1-9, 1-10
Defining a label, 2-20
Defining a macro, 2-22
Deleting a macro, 2-23

DEVICE CONFLICT, 1-80
Differential mode, 1-41,

Digital output register,
1-50

Events,

counting, 1-72
external, 1-40
Executable statement,

2-25,

Executing a macro, 2-23
Execution count, 2-16, 2-26,
2-34

1-64,

Digital input reading, 1-44
Digital input register, 1-2,
1-44

ARGUMENT, 1-80
CLOCK, 1=-80
SYNTAX, 1-80
Error flag, 1-24, 1-39, 1-67
Error message, 1-4, 1-5, 1-80,
2-20, 2-27, 2=35

2-30

1-69

1-19,

2-24

ERROR,

Execution speed, 1-6,
1-19, 1-28, 1-33,

1-17,
1-36,

2-40

External event, 1-40
External event timing mode,

1-2,

1-73

EXTERNAL

Index-2

statement,

1-5

1-62,

INDEX

Factor,

scale,
X-axis

Iapc,

1-13
spacing,

START,

1-13

IDIR,

2-2

1-62,

FDT

l-4,

1-5,

1-7,

2-36

2-1
addressing,

2-37

1-6’
1-53,
1-2, 1-3,

2-4

2-6,

2-39

IF,

2-3,

FDT
FDT

macro,

FDT

mode

2-7,

2-8,

2-33

FDT

pause,

2-1,

2-2,

INIT,

2-4,

2-10,

2-23,

2-30,

2~-33,

Inline

START

FAIL,

2~2

FDT

2-2
commands, 2-3

codes,

Finite

2-22,

continuous
1-19, 1-62
Flag,
error,

Flash,
Flash

1-24,

2=26

2-35

data

Flicker, 1-17,
Floating-point

1-36

2-36

1-13

PAUSE,

1-4,

1-14,
1-14,

2-3,

binary,

1-6,

statement

2-11,
2-40

Interrupt

l1-4,
Interval

2-16,

service

1-30

number,

2-25,

routines,

1-5,

1-17, 1-71
sampling technique,

1-39

Intervals,

time,

1-2,

IQSET,

1-40,

1-72

1-6

1-5

ITIMER,

1-36

1-33,

KB2BCD,

1-2,

KBCD2B,

1-56

?LABEL,
Label,

2-20

defining

numeric,

2-20

Latencies,

1-2,

LDPOBJ,
LED,

LED

1-1

1-74, 1-75
Histograms, 1-40

1-57

1-79

1-78,

1-79,

Limited mode, 1-75
Link, 1-79, 2-39
Link map, 2-1, 2-36
Literals,

Hardware,

1-72
1-8,

1-39

B-1

1-57

Library,

peripherals,

2-20

1-79,

display,

LIBR,

Hardware clock,
Hisr, 1-2, 1-7,

1-55

1-32
1-35

Gain, 1-42, 1-65
convert switched, 1-15
switched, 1-15, 1-41
Gain codes, 1-15, 1-42, 1-65
GOTO,

1l-15

routines,

2-10

clock, 1-72
FRUN, 2-36
FSH, 1-3, 1-13,
1-3,

1-44

2-39

Frequency,

FXY,

1-19,

1-9

2-1,

time,

1-2

conversion

2-1

2-34,

INTSET,

nonstandard integer,
packed integer, 1-15

FORTRAN

code,

Internal

1-35

1-28, 1-33,
conversion,

commands,

1-7

unsigned

Format,

2-15,

2-33

2-4,

Integer,

1-67

1-30
Foreground/Background,
RT-11, 1-4, 1-5, 1-7,

Format

transfer

2=-2,

l6-bit, 1-30
FLT1l6, 1-2, 1-6,

BCD,

Information

packed,

pairs,

sampling,

1-62

Input

sampling,

1-39,

1-55
2-39

continuous

1-19,

1-32
X-Y

1-19’ 1-30,
1-57
1-4, 1-6' 1—30,

1-50,
2-21,

Infinite

types,

2-13,

1-57,

logical, 2-30, 2-35
Implicit macro, 2-23
Indirect addressing, 2-39

command
control

FDT

1-41,

1-2,

IDOR,
1-49,

1-69

FB,

RT-11,

1-15,

1-44,

Fast-sweep mode,

FDT,

1-2,

1-69

FAIL,
FDT

(CONT.)

1_39'

text,

2-13,

2-32

l.ocation,

bit,

1-22,

named,

Index--3

2-4,

1-45
2-5

2-25,

2-39

INDEX

Location (Cont.),
offset, 2-4
relative, 2-4, 2-5
subscripted name, 2-4, 2-6
Location specification, 2-4,
2-24, 2-35, 2-37
Logical IF, 2-30, 2=35
Logical relation, 2-21

NAME, 2-3, 2-5,
2-24, 2-36,

1-3,

1-8,

?NO

1-59

OUT

link, 2-1, 2-36
storage, 2-1, 2-4, 2-8, 2-17,
2-37
Mask, 1-44, 1-45, 1-48, 1-52

1-80,

2-13,

2-21,

1-7,

1-39,

1-67,

1-69

event

fast-sweep,

2-24,
/P

1-64,

timing,

1-62,

noninterrupt,

1-69

1-73

1-69

1-68,

1-74

parameter-adjustment,
1-23,

option,

Packed

FDT, 2-33
limited, 1-75

1-19,

data,

2-20,

2=-35

ASCII, 2-17
bipolar, 1-42, 1-65
clock, 1=72
differential, 1-41,
external

2-6

Overrun,

Message,

2-27,

BOUNDS,

Output register,
digital, 1-2, 1=50
Overflow,
clock, 1-40
Overhead, 1-5, 2-40

Map,

Mode, 2-7,
2=-32

$SUBSCR,

1-5,

1-9
2-6

Object module, 1-78, B-1
Octal constants, 1-10
oDT, 2-39
Offset, 2-4, 2-37
Offset location, 2-4
Omitted, 1-11

2-38

l1l-4,

2=15

Notation conventions,
Number of dimensions,
Numeric label, 2-20

Manipulation,
bit, 1-46
byte, 1-46

error,

CONVERSION,

?NO ROOM, 2-27
Nondisplayable, 1-13, 1-17,
1-28, 1-33, 1-36
Noninterrupt mode, 1-68, 1-74
Nonstandard system, 1-=79

MACRO, 2-3, 2-22
Macro, 2-34
defining a, 2=22
deleting a, 2-23
executing a, 2-23
FDT, 2-13, 2~22, 2-26
implicit, 2-23
Main program name, 2-2
.MAIN.,

2-17,

main program, 2=-2
Named common, 2-5, 2-36, 2-~38
2-4, 2-5
Named location,

1-63

1-1,

2-9,

2-38

Name,

Loops, 2-20
LPS1ll dual sample-and-hold,

LwaIiTr,

(CONT.)

1-10,

1-62

repeated-interval, 1-73
single-ended, 1-64
single-~interval, 1-73
step, 2-27
unipolar, 1-42, 1-65
Mode code,
FpT, 2-7, 2-8, 2-17, 2-33
Mode conventions, 2-13
Mode conversion, 2«14, 2-21
Module,
object, 1-78, B-1

2-36

integer

format,

1-15

Parameter, 2-3, 2-8, 2-39
Parameter-adjustment mode,
1-10, 1-19, 1-23, 1-62
Partition,
array, 1-7, 1-8, 1-20, 1-24,
1-63, 1-67
PAUSE, 2-2, 2-16, 2~23, 2-=25,
2-28, 2-33, 2-40
Pause, 1-59, 2-10
'
automatic FDT, 2-1, 2-10
entry, 2-10, 2-25, 2-34, 2-40
FDT, 2-2, 2-4, 2-10, 2-23,
2-30, 2-33, 2-35
FORTRAN, 2-10
statement, 2-10,

2-25
step, 2-10
watch, 2-10, 2-33
Peripherals hardware, 1l-1
Priority, 1-5, 2-39
Procedure, 2-4, 2-11
current, 2-4, 2-6, 2-11,

Index-4

2-25

INDEX

Program

control

Program

name,

main,

2=2

RSTS/E, 2-1
RT-11, 1-4,

2-2

RT-11

Programmable
set

commands,

clock,

rate,

1-72

Programmed

Prompt,

(CONT.)

request,

2-2,

2-3,

1l-6

2-13,

RT-11

SJ,

1-4,

XM,

1-4

RTS,

1-3,

1-7,

1-43,

1-62

RUN,
1l-6

Punctuation,

1-4,

RT-11
2-22,

2-29
.PROTECT,

FB,

2-20

1-9
Sample-~and-hold,

RAD50,

Rate

2-8,

set,

LPS11 dual,
Sampling,

2-13

the

programmable

clock,

continuous, 1-19, 1-62
digital read-in, 1-19

1=72

RCIS$, 2-39
Read-in sampling,
digital, 1-19
Reading,
digital input, 1-44
Real-time functions,

A/D sampling, 1-2,
D/A output, 1-3

DMA,

1-22
continuous, 1-19, 1-62
infinite continuous, 1-19,
1-62

repeated,

1-3

Reentrant,

Refresh

stop, 1-21, 1-39, 1-67
time interval, 1-39
Scale factor, 1-13

1-1,

2-20

1-5

1-33, 1-36
scopes, 1-3, 1-16,

input, 1-2, 1-19,
1-44
digital output, 1-2, 1-50

refresh,

1-3,

1-16,

storage,

1-3,

1-4,

1-32,

1-16,

1-27

Scope

control,

SpIs,

1-3,

Removing

data,

Repeated

sampling,

1-8,

SETR,

2-38

2-5

1-67

1-17,

programmable

1-2,

1-3,

1-33,

1-36,
1-72

conversion,
Single display,

1-73

Single

1-6,

1-19,

1-39,

1-41

1-3

job,

RT-11,

1l-6

1-4,

1-5,

1-7,

2-35

Single-ended mode, 1-64
Single-interval mode, 1-73
Single-sweep sampling, 1-19,

2-28

1l-1,

the

1-72

1-5,

1-48, 1-68,
Single analog-to-digital

2-36

Routine,
completion,
1-5, 1-8,

1-1,

1-25,

Request,

2-2,

rate

clock,

1-62

Repeated-interval mode,

1-35
1-35

1-71

routine,
interrupt, 1-4,
1-71

Set

logical, 2-21
Relative address, 2-36,
Relative location, 2-4,
Relay control, 1-61

1-32,

Service

REL, l1l-61
Relation,

RESET,

1-13
1-35

1-32,

1-35
Register,
digital

RESUME,

1-3,

Scatter-plot,
Scope,

1-3,

programmed,

1-62

single~sweep, 1-19, 1-62
single-word, 1-21, 1-22

Scaling,

2-10,

1-69

finite

1-2

Refresh,

1-63,

double-word,

digital 1/0, 1-2
plotting, 1-3, 1-4
time interval control,
REENTER,

1-63

1-~2,

1-19,

1-24,

1-28, 1-33, 1-39,
1-59, 1-68, 1-71,
configuration, 1-73

format conversion,
interrupt service,
1-17, 1-71

1-3,

1-4,

1-25,

1-53,

1-4,

sampling,

1-21,

1-22
SJ,

1-76
2-15,

1-62
Single-word

RT-11,

2-39

1-5,

1-4,

1~5,

Software

clock,

1-52,
Spacing,
Y-axis,

1-75

Index-5

1-13

1-7,

1-2,

2-35

1-6,

1-48,

INDEX

Time interval, 1-2, 1-40,
Time interval sampling

Spacing factor,
X-axis, 1-13

Specification,

2-35,

2-24,

2-4,

location,
2=-37

Speed,

execution, 1-6, 1-17, 1-19,
1-28, 1-33, 1-36, 1-62,
2=40

1-9

Square brackets,
Stamping,
time, 1-2,

1=22
Standard system, 1-78

START,

2-2,

START

FAIL,

FDT,

2-10,

2-23,

2-20,

technique,
Time stamping,

1-39
1-2,

1-72

1-22

Timing mode,
external event, 1-73
Transfer commands,
information, 2-2, 2-4,
Transfer of control,
conditional, 2-21
unconditional, 2-20
Two's complement binary,

2-33

1-30,

2-38

TYPE,

2-3,

2-13,

2=32,

2-39

2-35

2-33,

2-~29,

(CONT.)

2-2

Statement,

2-25,

executable,

EXTERNAL, 1-5
Statement number,

2-1,

internal,

2-30

2-16,

2-11,

2-25, 2-34, 2-40
Statement pause, 2-10,

Unconditional transfer of
control, 2-20
?UNDEFINED, 2-4, 2-17, 2-35
Unipolar mode, 1-42, 1-65
Unsigned binary integer, 1l-6,
1-30

2-25
sTEP, 2-2, 2-10, 2-23, 2=27,
2-30, 2-33, 2-35, 2-40

Step mode, 2-27
Step pause, 2-10
sTop, 2-2, 2-31

Stop display, 1-71
Stop sampling, 1-21,
1-67

Storage map,

2-1,

2=37

2-17,

1-32,

1-4,

String constant, 2-13
$SUBSCR OUT OF BOUNDS, 2-6
Subscript, 2-6, 2-19, 2-38
Subscripted name location,
2-4,

2-6

convert,

1-41

1-15,

1-15

Syntax conventions,

sysrnis,

1-5,

1-59

2-2, 2-11, 2-33
Watch Pause, 2-10, 2-33
WHAT, 2-3, 2-27, 2-34

X-Y

2-3,

1-9,

2-4

SYNTAX ERROR,

Wait,
WATCH,

1-35

Switched gain,

2-17

2-8,

2-4,

1-3,

Storage scope,

1-39,

Variable,
dumny, 1-10, 2-8
Virtual arrays, 1-4,

1-80

1-6,

1-33,

1-79

System configuration,

1-36,

1l-1,

data pairs,

1-3,

X-Y data pairs,
display, 1-27
flash, 1-35
X-axis spacing factor,
XM,

RT-11,

1-78

1-4

nonstandard, 1-79
standard, 1-78
Y-axis

Text literals, 2-13,
Threaded code, 2-1
Throughput, 1-3

1-4,

1-27,

1-35

2-32

Index~-6

spacing,

1-13

INDEX

* convention,
#,

1-10,

(CONT.)

1-10

2-13

-ABS.,

2-38

IBCOMO Fi

2-36’
2-5’
oMAIN., 2”38

2-38

/P option, 2-36
/Z, 2—13, 2-33’

2-35

l6-bit floating-point
conversion, 1-~30
?LABEL,

2-20

?NO CONVERSION, 2-15
?NO ROOM, 2-27
?UNDEFINED, 2-4, 2-17,

2-35

Index-7

FORTRAN/RT~-11
Extensions Manual
AA=-2124D=-TC

READER'S COMMENTS

NOTE:

This form is for document comments only.

DIGITAL will

use comments submitted on this form at the company's
discretion. Problems with software should be reported
If you
on a Software Performance Report (SPR) form.
receive
to
eligible
are
and
reply
written
require a
one under SPR service, submit your comments on an SPR
form.

Please cut along this | mne,

Did you find errors in this manual?

If so,

specify by page.

Did you find this manual understandable, usable, and well-organized?
Please make suggestions for improvement.

Is there sufficlent documentation on associated system programs
required for use of the software described in this manual?
If not,

what material is missing and where should it be placed?

Please indicate the type of user/reader that you most nearly represent.

O
O

Assembly

[:]

Occasional programmer

User with little programming experience

(]

language programmer

Higher=-level

language programmer

(experienced)

student programmer

O Non-programmer interested in computer concepts and‘capabilities

Name

Date

Organization
Street

City

State

Zip Code
or

Country

Fold Here

Do Not Tear - Fold Here and Staple

FIRST CLASS

PERMIT NO. 152

MARLBORO, MASS

BUSINESS REPLY MAIL
NO POSTAGE STAMP NECESSARY IF MAILED IN THE UNITED STATES

Postage will be paid by:

Software Documentation

200 Forest Avenue MR1-2/E37
Marlboro, Massachusetts 01752

digital equipment corporation

Printed

U.S.A.