Digital PDFs

AA-H485A-TE

November 1979

112 pages

Original

4.8MB

view download

OCR Version

4.3MB

view download

Document:	VAX-11 PASCAL User's Guide (Ver 1.0-1)
Order Number:	AA-H485A-TE
Revision:	000
Pages:	112
Original Filename:

OCR Text

November 1979

This document describes how to compile, link, and execute VAX-11 PASCAL
programs on the VAX/VMS operating system. It also contains information useful to VAX-11 PASCAL programmers, dealing with input and output, procedure
calling, error processing, and storage allocation.

VAX-11 PASCAL
User’s Guide
Order No. AA-H485A-TE

SUPERSESSION/UPDATE INFORMATION:

This is a new document for this
release,

OPERATING SYSTEM AND VERSION:

VAX/VMS V1.6

SOFTWARE VERSION:

VAX-11 PASCAL V1.0-1

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

Center, Digital Equipment Corporation, Maynard, Massachusetts 01754

digital equipmént corporation - maynard, massachusetts

First Printing,

November

1979

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

The software described in this document is furnished under
a
1license
and
may
only
be used or copied in accordance with the terms of such
license.

responsibility is assumed for

the use or reliability of software on

equipment that is not supplied by DIGITAL or its affiliated companies.

The postage prepaid READER'S COMMENTS form on the last
page
of
this
document
requests
the
wuser's
«critical
evaluation
to assist us in
preparing

future documentation.

The following are trademarks of Digital Equipment Corporation:

DIGITAL
DEC
PDP
DECUS
UNIBUS
COMPUTER

DECsystem-10
DECtape
DIBOL
EDUSYSTEM
FLIP CHIP
FOCAL

MASSBUS
OMNIBUS
0S/8
PHA
RSTS
RSX

VAX
DECnet

VMS
IAS

SBI
PDT

DATATRIEVE

TRAX

COMTEX
DDT
DECCOMM
ASSIST-11

LABS

INDAC
LAB-8
DECSYSTEM-20
RTS-8

TYPESET-8
TYPESET-11
TMS-11
ITPS-10

CONTENTS

PREFACE
b=
*

CREATING

AND

COMPILING

EXECUTING

A PROGRAM

VAX/VMS FILE SPECIFICATIONS AND DEFAULTS

THE
-

PROGRAM
COMMAND

PASCAL

PASCAL
L

NNN

CHAPTER

PASCAL

VAX-11

USING

NN
=

CHAPTER

COMPILER

QUALIFIERS

Specifying Qualifiers with

the PASCAL

AND

- O
N

|
I
A

PROGRAM

CORRECTING

QUALIFIERS

/LIBRARY Qualifier
/INCLUDE Qualifier

[

FILE

wWwWwN

QUALIFIERS

Debugging and Traceback Qualifiers
INPUT

COMMAND

Image File Qualifiers
Map File Qualifiers

ERRORS

Error-Related Command Qualifiers
SHOW CALLS Command
TERMINAL SESSION

DA WWWWND
-

4.1
4.2
4.3

Fixed-Length Records
Variable-Length Records
OPEN PROCEDURE PARAMETERS

Buffer Size
File Status
Record Access Mode

iii

1
.3.2
.

Access

FORMATS

RECORD

Organization

Record

File

NAMES

CHARACTERISTICS

FILE

LOGICAL

OUTPUT

TR A

AND

;oo

SAMPLE
INPUT

BB WWWNDNDN
K

FORMAT

w N
L[]
L]
W N

[]

S
[

Source Code

PROGRAM

COMMAND

LINKER

192

RGNS NGRS RS

the

WWwWwwwwww

LINK

LINKER

L]
N

WwwhhNhDNhND
-

EXECUTING

(NG EGEG NS

FILES

1=Y

W N

.
L]

D
B

O
L
*
L
o
e

WWwwwwwww

LISTING

FINDING

CHAPTER

Qualifiers

OUTPUT

Source Code Listing
Machine Code Listing
Cross—-Reference Listing

THE

CHAPTER

COMPILER

LINKING

CHAPTER

Specifying
SPECIFYING

Command

CONTENTS

Record

Type

Carriage

Control

LOCAL INTERPROCESS COMMUNICATION: MAILBOXES
COMMUNICATING WITH REMOTE COMPUTERS: NETWORKS

(SINGLE,

REAL

(DOUBLE

Type)

Types)

{

A
wWNH-

e Jie) Nie ) Jie) e ) Jie) Je) |
)

|
1
O

WWNDN

—

NN NNNNNNNNN

Double-Precision Floating-Point Data

|
1

»
o

w N+

[

5.2

Single-Precision Floating-Point Data

00 00 00 00 00

OF PROGRAM SECTIONS
STORAGE OF SCALAR AND POINTER TYPES
STORAGE OF UNPACKED STRUCTURED TYPES
STORAGE OF PACKED STRUCTURED TYPES
Storage of Packed Sets
Storage of Packed Arrays
Storage of Packed Records
REPRESENTATION OF FLOATING-POINT DATA

USE

}

SYSTEM ENVIRONMENT

PASCAL

EXAMPLES

s o]

HANDLER

CONDITION

o o]
o

o
o
o
&
o

0 00 00 00 00

o
*

HANDLER

Establishing and Removing Handlers
Parameters for Condition Handlers
Handler Function Return Values
Condition Values and Symbols
Floating-Point Operation

VAX-11
Lt
WwN -

0 00

Condition Signals
Handler Responses
WRITING A CONDITION

W
Nnad

~J
s

NN NNNNNNNIY

RUN-TIME LIBRARY DEFAULT ERROR PROCESSING
OVERVIEW OF VAX-11 CONDITION HANDLING

B WWWwwwNN
-

ERROR PROCESSING AND CONDITION HANDLERS

CHAPTER

CALLING RUN-TIME LIBRARY PROCEDURES
COMPLETE SYSTEM SERVICE EXAMPLE

NANH

Optional Parameters
Passing Character Parameters

OO
L]
L

Calling System Services as Procedures
Passing Parameters to System Services
Input and Output By-Reference Parameters

D WNH

Parameter Passing Mechanisms
By-Reference Mechanism
By-Immediate-Value Mechanism
By-Descriptor Mechanism
Functions and Procedures as Parameters
Function Return Values
Arguments to PASCAL Subprograms
CALLING VAX/VMS SYSTEM SERVICES
Calling System Services by Function
Reference

CHAPTER

VTUNT WD

STANDARD

CALLING

=
.

B WD

VAX-11 PROCEDURE
Argument Lists

S =
R

AN
N
N

CONVENTIONS

CALLING

CHAPTER

CONTENTS

Page

APPENDIX

DIAGNOSTIC

MESSAGES

A-1

A-1
A-2

COMPILER

DIAGNOSTICS

A-1

RUN-TIME

ERROR

A-19

CONTENTS

MESSAGES

RUN-TIME

STACK

DURING

PROCEDURE

CALLS

B-1

INDEX

INDEX-1

FIGURES

FIGURE

1-1

Program

2-1

Compiler

4-1

Source

8-1

Storage

Record

8-4

8-2
8-3

Storage of Sample Record
Storage of Sample Packed Record Containing
Packed Array
Single-Precision Floating-Point Data
Representation
Double-Precision Floating-Point Data
Representation
Contents of Run-Time Stack During Procedure

8-7

Calls

B-2

8-4
8-5

B-1

Development

Listing

Program
of

Process

1-2

Format

Listing

Sample

2-8

Traceback

List

4-3

8-8
8-8
8-9

TABLES

TABLE

1-1

File

2-1
2-2

PASCAL
PASCAL

Specification

Compiler Qualifiers
Command Qualifiers

2-2
2-5

2-3
3-1
4-1

Source
Linker
/DEBUG

Code Qualifiers
Qualifiers
and /TRACEBACK Qualifiers

2-6
3-2
4-2

5-1
6-1
8-1
8-2
8-3

Predefined System Logical Names
Suggested Variable Data Types
Program Section Attributes
Program Section Usage and Attributes
Storage of Scalar and Pointer Types

8-4

Storage

Packed

Defaults

Array

Elements

1-3

5-2
6-10
8-1
8-2
8-3
8-5

PREFACE

MANUAL OBJECTIVES

The VAX-11 PASCAL User's Guide is intended for use in
and 1in compiling and executing
PASCAL programs,

developing new
existing PASCAL

on
supported
PASCAL language elements
programs on VAX/VMS systems.
VAX/VMS are described in the VAX-11 PASCAL Language Reference Manual.

INTENDED AUDIENCE

knowledge
This manual is designed for programmers who have a working
1is helpful but not
Detailed knowledge of VAX/VMS
PASCAL.
of
Some
familiarity with the VAX/VMS Primer is recommended.
essential;
however, (condition handling, for instance)
sections of this book,

In such
require more extensive understanding of the operating system.
sections, you are directed to the appropriate manual(s) for the
information.

required additional

STRUCTURE OF THIS DOCUMENT

This manual

is organized as follows:

Chapter 1 provides an overview of the steps you must follow to

Chapter 2 describes how to compile your program

Chapter 3 supplies information on the VAX-11

Chapter 4 describes execution commands.

Chapter 5
including

create,

compile,

link,

and execute a VAX-11 PASCAL program.

the options available at compile time.

options,

they

apply

provides
details

conventions,

and

PASCAL

programs

information about
the use of
on

record

discusses

and

Linker

explains
and

its

PASCAL input/output,
file
logical names,

structure.

the

conventions

followed

1in

calling

Chapter

how to
Chapter 7 describes error processing, in particular,
use the condition handling facility. This chapter is intended

procedures, especially the conventions for passing parameters.

for users with in-depth knowledge of VAX/VMS.

Chapter 8 describes the relationship between VAX-11 PASCAL and
the VAX/VMS operating system, with particular emphasis on
data
and
allocation,
storage
usage,
program section
representation.

vii

Appendix

summarizes

Appendix

illustrates

during

ASSOCIATED

The

For

documents

contents

are

relevant

VAX-11

PASCAL

Primer

VAX-11

PASCAL

Language

VAX/VMS

Command

Language

User's

VAX-11

Run-Time

Library

Reference

VAX-11

Linker

VAX/VMS

VAX-11

System

list

USED

document

Services

Architecture

complete

uses

Reference

CONVENTIONS

words

letters

THIS

the

words

PASCAL

programming:

Manual

Guide
Manual

Reference

Manual

software

documents,

see

the

VAX-11

DOCUMENT

following

conventions.
Meaning

Uppercase

indicate

words

that
as

and

you

letters,

should

words

and

examples,

indicate

that

value

your

the

1in

examples,

word

used

letter

letters,

you

are

substitute

format

word

type

the

which

one

choice.

Double

brackets

indicate

Square

brackets

indicate

bracket

used

type

shown.

Lowercase
or

stack

Handbook

VAX-11

exactly

letters

run-time

Manual

Convention

and

VAX-1l

Lowercase

the

Primer

Directory.

and

VAX/VMS

Uppercase

calls.

Information

This

the

messages.

DOCUMENTS

following

procedure

diagnostic

optional
that

elements.

you

must

characters.

Braces

are

element

used

enclose
chosen.

1lists

from

A horizontal ellipsis indicates that
the
item(s) can be repeated one or more times.

A vertical

ellipsis

statements

indicates

example

viii

preceding

that

not

all

figure

are

shown.

the

Meaning

Convention
S PASCAL

$_File:

system
and
enter
you
commands
of
examples
In
responses, all output lines and prompting characters

displays are shown in
or
system prints
the
that
in
All the lines you type are shown
black letters.
red

letters.

abbreviation
3-character
to
1a
with
A symbol
that you press a key on the terminal, for
indicates
example,

GO.

CHAPTER

USING VAX-11

PASCAL

the

PASCAL

language.

the

VAX/VMS

operating

VAX-11 PASCAL is an extended implementation of

The

PASCAL compiler executes in native mode under the VAX/VMS

VAX-11

operating

system.

This manual describes how you

with

interact

It contains instructions for compiling,
VAX-11 PASCAL.
using
system
on
information
linking, and executing a PASCAL program, and provides
the

following

topics:

Performing

input and

Calling VAX/VMS system services and Run-Time Ldibrary routines

Using the VAX/VMS condition handling facility

Writing

efficient

output operations

VAX-11

PASCAL

programs

and
creating
1in
steps
the
of
overview
an
This chapter provides
It also describes the standard
program.
PASCAL
VAX-11
a
executing
VAX/VMS

1.1

file

specification and defaults.

CREATING AND EXECUTING A PROGRAM

Figure 1-1 illustrates the program development process, from inception
to
execution.
You specify the steps shown in Figure 1-1 by entering
commands
$
$

to the VAX/VMS operating

system.

These commands are:

EDIT file-spec
PASCAL file-spec

LINK

RUN

file-spec

With each command, you include information that further defines
the
is
importance
Of prime
system to do.
you want the
<can
You
processed.
be
specification, indicating the file to

what
file
also

specify qualifiers that modify the processing performed by the system.

USING

VAX-11

PASCAL

COMMANDS

INPUT/OUTPUT FILES

PAS
$ EDIT AVERAGE.
Use the file type of PAS to

AVERAGE.PAS

indicate the file contains a

Create a

VAX-11 PASCAL program.

source program

$ PASCAL AVERAGE
The PASCAL command

‘ AVERAGE.OBJ

assumes the file type of

Compile the

an input file is PAS.

iy N

source program

(AVERAGE.LIS)

(If you use the /LIST
libraries

qualifier, the compiler

creates a listing file.)

$ LINK AVERAGE
The L/NK command assumes

the file type of an input file

Link the

1s OBJ.

AVERAGE.EXE
(AVERAGE.MAP)

object module

(1f you use the /MAP qualifier,
the linker creates a map file.)

$ RUN AVERAGE

Run the

The RUN command assumes

executable

the file type of an image Is

Image

EXE.

Figure

1.2

VAX/VMS

FILE

VAX/VMS

file

the

output

following

1-1

Program

Development

Process

SPECIFICATIONS AND DEFAULTS

specification

file

indicates

the

produced.

input

File

file

processed

specifications

have

the

form:

node::device:[directorylfilename.filetype;version

The

punctuation
required
syntax
specification.

marks

that

(colons,
separate

brackets,
period,
semicolon)
the
wvarious
components of the

are
file

node

Specifies
systems

network

that

support

node

name.

This

1is

applicable

only

DECnet-VAX.

device
Identifies
written.

the

device

on which

the

file

stored

1is

USING

VAX-11

PASCAL

directory

Identifies the name of the directory
under
which
the
file
is
cataloged,
on
the
device
specified.
You
<can
delimit
the
directory name with square brackets,
as
shown,
or
with
angle
brackets

>).

filename

Identifies the
alphanumeric

file by

its

characters

name;

filename

can

filetype can be up to

long.

filetype

Describes

the kind of

alphanumeric

data

characters

the

file;

long.

version

Specifies which version of the file
1is
desired.
Versions
are
identified
by
a
decimal number, which is incremented by 1 each
time a new version of a file is created.
Either a semicolon or a
period can be used to separate filetype and version.

You need not explicitly state all elements
of
a
file
specification
each
time
you compile, link, or execute a program.
The only part of
the file specification that is usually required is the file name.
If
you
omit any other part of the file specification, a default value is
used.

Table

1-1

summarizes

File

the

default

values.

Table 1-1
Specification

Optional
Element

Defaults

Default
Value

node

Local

network

device

User's

current default

device

directory

User's

current default

directory

filetype

Depends

node

usage:

Input to PASCAL compiler
Output from PASCAL compiler

PAS
OBJ

Input

OBJ

Output
Input

linker

from
to

linker

RUN

command

Compiler source listing
Linker map listing
Input to executing program
Output from executing program

version

Input:

Output:

EXE
EXE

LIS
MAP
DAT
DAT

highest existing version

highest existing version
plus

USING VAX-11l

PASCAL

If you request compilation of a PASCAL program and you specify only
file
name,
the compiler can process the source program if it finds
file with the specified file name that:

stored

cataloged

Has a

file

the

default

under

type

a
a

device

the default

directory name

PAS

If more than one file meets these conditions,
with the highest version number.

the

compiler

chooses

the

one

For example, assume
directory is SMITH,
the

that your default device
Iis
and you supply the following

DBAO,
your default
file specification to

compiler:

$ PASCAL
$_File: CIRCLE
The compiler will search device DBAO in directory SMITH,
seeking
the
highest
version
of
CIRCLE.PAS.
1If you do not explicitly specify an
object file, the compiler will generate the file CIRCLE.OBJ, store
it
on
device
DBAO
1in directory SMITH, and assign it a version number 1
higher than any other version of
CIRCLE.OBJ
currently
cataloged
1in
directory

SMITH

DBAO.

CHAPTER
COMPILING

After creating
compile
time,
source

code

your

and

the

THE

compile

PASCAL
a

which

compiler

are
input
to
the
files.
The listing
about
compilation
listings.

2.1

PROGRAM

VAX-11l PASCAL source program,
you
specify
the
name
of the
indicate

option,

qualifiers

produces

you

one

you
compile
it.
file(s) containing

wish

1linker
(see
Chapter 3),
files contain source
code
errors, and optional items

At
the

use.

object

files,

which

and one or more listing
1listings,
information
such as cross-reference

COMMAND

source

program,

use

the

PASCAL

command

the

following

form:

PASCAL[/qualifiers]

file-spec-listl[/qualifiers]

/qualifiers

Indicate

special

processing

performed

the

compiler.

file-spec-list
Specifies the source file(s) containing the program or module
to
be
compiled.
You
<can
specify
more than one source file.
If
source file specifications are separated by commas, the
programs
are

compiled

separated
as

one

system

signs,

the

If
source
file
specifications are
files are concatenated and
compiled

program.

interactive

separate

separately.

plus

mode,

line

responds

you

typing

with

can
a

also

enter

carriage

prompt

for

the

return

the

file

file
after

specification
you

type

PASCAL.

a
The

specification:

$ PASCAL
$_File:

Type

the

file

specification

the

$File:

prompt.

2.2

PASCAL

COMPILER

many

cases,

the

for
compilation.
compiler qualifiers

and

any

file

the

qualifiers

immediately

after

QUALIFIERS

simplest

form

Sometimes,

specify

PASCAL

however,

special

you

command

will

processing.

need

sufficient

use

PASCAL

COMPILING A PROGRAM

Table 2-1 lists the qualifiers you can

use

with

the

VAX-11

PASCAL

You can specify the qualifiers on the command line or in
compiler.
This section describes the effect of each
source code comments.
Section 2.2.1 describes how to specify
gqualifier on a PASCAL program.
specifying
with
2.2.2 deals
Section
qualifiers.
line
command
qualifiers

in source code comments.

Table

2-1

PASCAL Compiler Qualifiers
Can Be Specified

in Source Code?

Qualifier

Purpose

CHECK

Generates code to perform

CROSS_REFERENCE

Produces a cross-reference

Yes

DEBUG

Generates records for
VAX-11 Symbolic Debugger

Yes

ERROR_LIMIT

Terminates compilation

LIST

Produces source listing

listing of

after

identifiers

errors

Includes representation of

machine

code

source listing

STANDARD
WARNINGS

Yes

file

MACHINE_CODE

OBJECT

Yes

checks

run-time

Yes

file

Specifies name of

Prints messages indicating

Yes

Prints diagnostics for

Yes

object

file

use of PASCAL extensions

warning-level

errors

CHECK

The CHECK qualifier directs the compiler to generate code to perform
run-time checks. This code checks for illegal assignments to sets and
The system
subranges, and out-of-range array bounds and case labels.
issues an error message and normally terminates execution if any of
these conditions occur.

When this qualifier is disabled, the compiler generates no check code.
By default,

CHECK is disabled.

COMPILING A PROGRAM

CROSS_REFERENCE

The CROSS_REFERENCE qualifier produces a
cross-reference
listing
of
all identifiers.
The compiler generates separate cross-references for

To get complete cross-reference listings
each procedure and function.
a program, the qualifier must be in effect for all modules of the
for
being
is
file
1listing
no
if
ignored
This qualifier is
program.
generated.

By default,

CROSS_REFERENCE

disabled.

You can specify this qualifier in the source code,
as
described
1in
Section
2.2.2.
Note, however, that the cross-reference listing for a
portion of a

procedure or

function may be

incomplete.

DEBUG

The DEBUG

qualifier

specifies

that

the

compiler

1is

generate

information
for
use by the VAX-1ll1 Symbolic Debugger and the run-time
error traceback mechanism.
When you enable the option,
the
compiler

generates
some
DEBUG
and
TRACEBACK
records
program for which the qualifier is in effect.

When this qualifier
records.

is disabled,

By default,

DEBUG

for

each procedure or

the compiler generates only TRACEBACK

disabled.

ERROR_LIMIT

The

ERROR LIMIT qualifier

terminates

compilation

after

errors,

excluding warning-level
errors.
If
this
qualifier
1is
disabled,
compilation continues through the entire
unit.
You
cannot
specify
this
By

qualifier

default,

the

source

ERROR_LIMIT

code.

enabled.

Note that after it finds 20 errors (including warning messages) on any
one
source line, the compiler generates error 255, Too many errors on
this source line.
Compilation of the line continues, but
no
further
error messages are printed for that line.
LIST

The

LIST qualifier produces a

source listing file.

It has the

form:

LIST[=file-spec]

You can include a
file
specification
for
the
1listing
file.
The
default
file
specification
designates
the name of the first source
file,

your

default

directory,

and

file

type

LIS.

The compiler does not produce
a
1listing
file
in
interactive
mode
unless
you
specify
the LIST qualifier.
1In batch mode, the compiler
produces a listing file by default.
1In either case, the listing
file
is
not
automatically
printed.
You
must
use
the DIGITAL Command
Language (DCL) PRINT command to obtain a
1line
printer
copy
of
the
listing file.
A sample listing is explained in Section 2.4.
MACHINE

CODE

The MACHINE CODE qualifier places
of

the object

The compiler

code generated

in the listing

by the

ignores this qualifier

enabled.

default

MACHINE CODE

file

representation

compiler.

disabled.
2-3

the

LIST

qualifier

1is

not

COMPILING

PROGRAM

OBJECT

The

OBJECT

qualifier

the

object

file.

can

It has

used

the

when

you

want

specify

the

name

form:

/OBJECT[=file~specl
If you omit the file specification, the object file
defaults
to
name
of the first source file, the default directory, and a file
of OBJ.
You cannot specify this qualifier in the source code.
You can disable this qualifier
when you only want to test the
By

default,

OBJECT

to suppress object
source program for

code;
for
compilation

the
type

example,
errors.

enabled.

STANDARD

The STANDARD
messages
at
features.
Nonstandard

qualifier
tells
the
compiler
each
place
where the program

PASCAL

features

that
are incorporated
VALUE declarations and
VAX-11 PASCAL Language
By

default,

STANDARD

are

the

to
uses

extensions

print
warning-level
"nonstandard" PASCAL

the

PASCAL

1language

in VAX-11 PASCAL.
Nonstandard features include
the exponentiation operator.
Appendix C of the
Reference Manual lists all the extensions.
enabled.

WARNINGS

The WARNINGS qualifier directs the
compiler
to
messages in response to warning-level (W) errors.

generate

diagnostic

default,
WARNINGS
1is
enabled.
A
warning
diagnostic
message
indicates
that
the
compiler
has detected acceptable but unorthodox

syntax or has performed

unexpected
results
messages, disable this
the STANDARD qualifier
Appendix

2.2.1
A

lists

specifying

PASCAL

command

the

some

corrective

action;

either

case,

may
occur.
To
suppress
warning
diagnostic
qualifier.
Note that messages
generated
when
is enabled appear even if WARNINGS is disabled.

compiler

diagnostic

Qualifiers with
qualifier

has

the

messages.

PASCAL Command

form:

/qualifier[[=file-spec]
Table 2-2 lists the qualifiers you can use on the PASCAL command line.
The
optional
file specification indicates the name of an output file
for the OBJECT and LIST qualifiers only.
To
enable
the
qualifier,
specify
1its
name.
To
disable
the qualifier, specify the negative
form.

You can abbreviate all command line qualifiers by truncating
them
on
the
right.
All
qualifiers are unique when truncated to their first
four characters, not including the NO of the negative form.
You
can
truncate
further
as
1long
as
the resulting command is unique.
For
example, you can truncate CROSS REFERENCE to
CR,
CHECK
to
CH,
and

DEBUG

to D.

COMPILING A PROGRAM

When you use PASCAL command qualifiers in command procedure files,
it
is
recommended
that
vyou
use
the
full
qualifier
names to ensure
readability.
To guarantee compatibility with future releases
of
the
system,
fewer

you

than

should

four

not

abbreviate qualifiers

in command procedures

characters.

Table

2-2

Qualifiers

PASCAL Command

Qualifier

Negative Form

Default

/CHECK

/NOCHECK

/CROSS_REFERENCE

/NOCROSS_REFERENCE

/DEBUG

/NODEBUG

/ERROR_LIMIT

/NOERROR_LIMIT

/ERROR LIMIT

/LIST[=file-spec]

/NOLIST

/MACHINE CODE

/NOMACHINE CODE

/OBJECT[=file-spec]

/NOOBJECT

/OBJECT

/STANDARD

/NOSTANDARD

/STANDARD

/WARNINGS

/NOWARNINGS

/WARNINGS

/NOLIST

J/LIST

(interactive)

(batch)

Examples

PASCAL

CALC

The source file CALC.PAS is compiled.
By
default,
the
STANDARD, WARNING, and ERROR_LIMIT options are enabled.
2.

PASCAL/CHECK/NOSTANDARD

The
source
generated.
language

CALC

file
CALC.PAS
1is
compiled,
and
check
code
is
The
compiler does not issue warnings for the use of

extensions.

PASCAL/LIST/CR CALC

The source
CALC.LIS

file
CALC.PAS
is
generated.

cross-reference

2.2.2

Specifying

compiled
and
The
listing

the
file

qualifiers

Qualifiers

have

1listing
file
includes
a

listing.

the

Source Code

You can use qualifiers in
the
source
code
to
enable
special
processing
during compilation.
When specified
code,

OBJECT,

the

form:

(*$qualifier {+}[,qualifier

{i}

;...
]| icomment?)

and
disable
in the source

COMPILING

The

first

character

after

the

comment

PROGRAM

delimiter

must

(S);
the dollar sign cannot be preceded by a space.
the qualifiers you can specify in your source program.
can optionally use a l-character
abbreviation is simply the first
except

for

CROSS_REFERENCE,

abbreviation

Table

Code

sign

2-3 lists
that
vyou

2-3

Qualifiers

Qualifier

Abbreviation

CHECK

CROSS_REFERENCE

DEBUG

LIST

MACHINE

dollar

abbreviation for each qualifier.
The
character
of
the
qualifier
name,

whose

Source

Table
Note

CODE

STANDARD

WARNINGS

enable a qualifier, specify a plus
sign
(+)
after
1its
name
or
abbreviation.
To disable a qualifier, specify a minus sign (-) after
its name or abbreviation.
You can specify any number
of
qualifiers.

You

<can

from

the

also

list

include

text

qualifiers

comment

after

the

qualifiers,

separated

semicolon.

When

specified in the source code, the LIST qualifier cannot contain a
file
specification.
The
listing
file
will
have
the
default
specification as described in Section 2.2 above.

For

example,

program,

and

disable

(X$C+

generate

enable

CHECK

the

end

enable

FROCEDURE

check

the

CHECK

code

for

qualifier

the

for

only

before

procedure,

TESTLI

one

the

procedure

declaration

follows:

onmluX)

TEST1S$

ENIDs
(X$C-idisable

CHECKX)

Command line qualifiers override
example,

the

source

PASCAL/NODEBUG,

the

2.3

OUTPUT

The
the

SPECIFYING

DEBUG

source

code

specifies

option

will

not

code

qualifiers.

DEBUG+,

but

I£f,
you

for
type

effect.

FILES

compiler produces object files and listing files.
You can control
production of these files by using the LIST and OBJECT qualifiers

with the PASCAL command.
Unless you specify otherwise,
the
compiler
generates
an
object
file.
In
interactive
mode, the compiler, by

COMPILING

default, does
qualifier
to

not generate
explicitly

PROGRAM

a listing file;
you
must
specify
a
1listing
file.

however, the opposite is true:
listing file.
To suppress the
qualifier.

by default, the
listing file, you

use
the
In batch

LIST
mode,

compiler
produces
a
must disable the LIST

During the early stages
of
program
development,
you
may
£find
it
helpful
to
suppress the production of object files until your source
program compiles without errors.
To
do
so,
specify
the
NOOBJECT
qualifier
on
the
PASCAL
command
1line.
If
you
do
not
specify
/NOOBJECT,

the

compiler

generates

object

e If you specify one source file,
e

If you specify multiple
the
source
files
are
file is generated.

If you
source

You
to

for

can use
produce

object

follows:

one object file is generated.

source files, separated
by
plus
concatenated and compiled, and one

signs,
object

specify multiple source files, separated by
commas,
each
file
1is
<compiled
separately,
and
an
object
file is

generated
e

files

each

source

file.

both plus signs and commas in the same
command
1line
different
combinations of concatenated and separate

files

(see

Example

below).

To produce an
must
specify

object file with an
explicit
file
specification,
vyou
/OBJECT
on
the PASCAL command line (see Section 2.2).
Otherwise, the object file will have the
name
of
1its
corresponding
source
file
and
a
file
type
of OBJ.
By default, the object file
produced from concatenated source files has
the
name
source
file.
All other file specification attributes
directory, and version) assume the default attributes.

the
first
(node, device,

Examples

PASCAL/LIST AAA,BBB,CCC

Source files
AAA.PAS,
BBB.PAS,
and
CCC.PAS
are
compiled
as
separate
files,
producing
object files named AAA.OBJ, BBB.OBJ,
and CCC.0BJ;
and
1listing
files
named
AAA.LIS,
BBB.LIS,
and
CCC.LIS.

PASCAL

Source

XXX+YYY+ZZZ

files

compiled as
batch mode,

XXX.PAS,

YYY.PAS,

and

ZZZ.PAS

are

concatenated

and

one file, producing an object file named XXX.0BJ.
1In
this command also produces the listing file XXX.LIS.

$ PASCAL/OBJECT=SQUARE

$ File:

CIRCLE

The system issues the $_File:
prompt because the PASCAL
command
does not specify a source file.
The file CIRCLE.PAS is compiled,
producing an object file named SQUARE.OBJ, but no
1listing
file.
(This example applies to interactive mode only.)

PASCAL AAA+BBB,CCC/LIST

Two

object

BBB.PAS)

files
and

are

CCC.OBJ

produced:

AAA.OBJ

(comprising

CCC.PAS).

1In

AAA.PAS

interactive

this command produces the listing file CCC.LIS.
In
it produces two listing files:
AAA.LIS and CCC.LIS.

batch

and
mode,

mode,

COMPILING A

PROGRAM

PASCAL ABC+CIRC/NOOBJECT+XYZ

When you include a qualifier in a list of files that
are
to
be
concatenated, the qualifier affects all files in the list.
Thus,
the command shown above completely suppresses
the
object
file.
That
1is,
source
files
ABC.PAS,
CIRC.PAS, and XYZ.PAS will be
concatenated and compiled, but no object file will be produced.
$

PASCAL/LIST

[DIR]MNP

The source file MNP.PAS in directory [DIR] is compiled, producing
an
object
file
named MNP.OBJ and a listing file named MNP.LIS.
The compiler places the object and listing files in
the
default
directory.

COMPILER

you
listing.

LISTING

FORMAT

specify the

When

LIST qualifier, VAX-11] PASCAL produces a compiler
section
explains
the format of the compiler listing

This

illustrated

EXAMPLE

Figure

11-0CT~-1979

11:35:18

11-0CT-1979

LINE
NUMEERS

2-1.

VAX-11 PASCAL VERSION

11334247

LEVEL
FROC STMT STATEMENT.

100
200

1
2

300
400

03
4q

i
1

rrogram

lo
1

label 103
var AsByCIiREALS

500

]

600

)

700
800

7
8

200

WRITELNC(/Enter

READLNC(AYR)
$

1200

triandle

X¥% WARNING

450!

(SARCA)

SQR(E))

Nonstandard

2
2
1

WRITELNC(‘Hurotenuse
until FALSES:

1700

WRITELNC('Dlone’?#

1800
1900

18
19
20

Comrilation
2 Errors

1 103

*%X%x ERROR
4
Xk% ERROR
20!
1
1
end.
0
time =

")
"»'

Last error(uwarning) on line
end

XX 0.5

~450

isl

Exronentiation

(

12 ==

b 3§

17 ==

//@

"20,*,4<—'@

2036

xK¥k

‘»C)§

exrected
eurected

0.96 seconds

Nonstandard feature

Active ortions at

sides’)$

FPascal!

13
14
15

1600

if EOLNC(INFUT) then goto 107

1300
1400
1500

#*FAS-F-DIAGN

rereat

ZFAS-W~DIAGN

bedin

10
11

PAGE

(1)

EXAMPLE CINFUT»OUTFUT) ¢

1000

1100

V1.0-1

DB2:L200»200IEXAMFLE.PAS34

2.4

lines/mirute).

17.@

of comrilation?

NonEBUG,STANDARD,LIST,NOCHECK,uaRNINGs.CRDSS_REFERENCE,‘g’

MACHINE._CODE
» OBJECT ERROR_LIMIT = 30

Figure 2-1 Compiler Listing Format

COMPILING A PROGRAM

EXAMFLE

11-0CT-1979

1133518

VAX-11

FASCAL VERSION

V1.,0-1

PAGE

MACHINE .CODE

GENERATED

LINE

CODE

ADDRESS

(PRIOR TO BRANCH

0002

MOVAER

XVAR(O»

FUSHL

00 DD

0012
0014
0016
001cC
001E
0020
0023
0025

CLRD
PUSHL
FUSHL
CLRD
FUSHL
FUSHAL
CLRD
FUSHL.

-(R14)
Ri11
#67436548
~(R14)
#0
(R13)BTM-20
-(R14)
#0

7E
Sk
04
7E
00
EC
7
00

002E
0031
0038
003E
0042

ADDL2
#8s+R14
ADDL2
#5636y (R11)
FPUSHAL (R11)RTM8
CALLS
#1sFASSINFUT
FUSHAL (R11)ERTM8

MOVL

0027

0054
0059
005C

00SE
0065
0069
006D
0071
0078

EXAMPLE

00 00 00 00 00 5O DO

#7»8YS$GETJIFI @

#2,FASSOUTFUT
R14y(R13)B"-12
(R11)W"236sR10
#16sR14
R10
XVAR(2s 0)sR9
R?r(R14)B74

MOVL
CALLS
MOVAR

00 00 00 00 00 07 FE

SE
6B
08
00
@08

08
00
AR
00
AR

€O
00 16 04 8F CO
DF
00 00 00 01 FE
IF

00
F4
SA
SE
Sa4
§9
04

00
AD
00
10
Dp
00
AE

00 00 00 02 FE
SE IO
EC CR 9E
€2

#21+(R14)BTM8
#5,PASSWRITESTR
(R11)WTM236yR10

#21s(R14)BTM12

11-0CT-1979

11335218

7C
DD
05 00 04 8F DD
7C
DD
AD DF
7C
DD

00 EC CB DF

CALLS
MOVL.
MOVAR
SUEBL2Z
FUSHL
MOVAE
MOVL

MOVL

(HEXADECIMALS? READ FROM RIGHT TOQ LEFT)

SB 00 00 00 00 00 9E

FPUSHAL (R11)W"236

0049
00350

0)sR11

R13+s%XVAR(O»

CALLS

0045 @
8

BYTESTREAM

0010

0009

OFTIMIZATION)

OFCODE OFERANDS

oC AE
08 AE
00 00
54 00

00
59
15
15
00
EC

VAX~11 PASCAL VERSION V1.0-1

00 00 00 9E
DO
DO
DO
00 00 05 FB
CE 9E

FAGE

CROSS..REFERENCE

CROSSREFERENCE LISTING

BQ)
c

1
1

4
¢D4

INPUT
QUTPUT

GLOBALLY DEFINED

0
0

10
14

REAL

0
0

12
9

Figure

2-1

READLN

SAR
WRITELN

compiler

12@3
13

IDENTIFIERS?

EOLN
FALSE

The

11
12

listing

12
13

@
17

(Cont.)

contains

Compiler

the

Listing Format

following

three

Source code listing -- When you specify
the

source

code

listed

Machine code

Cross-reference listing
all
identifiers
wused

you

must

sections:

the

LIST

listing -- To generate the machine code

specify

CROSS REFERENCE

the

qualifier,

default.

MACHINE CODE

qualifier.

1listing,

-- To generate
cross
references
for
in
the
program, you must specify the

qualifier.

The numbers throughout this section are keyed to the numbers in Figure
2-1.

ronits

[

iin

COMPILING

Title
title

Line -1line

Each
ists

compilation
the

line

Source
page

Code
the

source

code

the date and

listing

lines

contains

line

the

under

Errors

any

module

leftmost

are

with

the

column

irrelevant

S0OS

the

edited

editor,

source

PASCAL

to
the
source
1lines
in
a
PASCAL
traceback that
1is
printed
1if
your
exception at run time refers to these

procedure

Statement level
level
for
delimiter.

1level

@ -- The

nesting
level of

statement

that

and Warnings

source code
diagnostic
description

code

listing.

SOS

1line

numbers

module.
The
symbolic
program
encounters
an
line numbers.

that

1listing

declaration.

declaration
Procedure

block.
nesting

specifies

The
level

statement

The

The
the

it.

source

code

detected

listing

the

includes

compiler.

(°)

that

points

the

line where the error was detected
numeric

code,

particular
listing,
numeric

following

error

the

information

line

beneath

the

character position

circumflex,

the

specifies

the

QD .

that

following

lines

asterisk

(*)

that

shows

source

where

the

compiler

number
You

diagnostics

can

use

the

last

these

backwards

0 .

resumed

@ .

The line number in which the error was detected
.

the

to detect more than one error

translation after the error

line

the compiler prints the text that corresponds to each
code
.
Note that one source program error often

causes the PASCAL compiler

numbers

line in which the error is detected specifies whether
the
1is
a
warning
or
an
error.
In
addition,
the
error
can contain the following information:

A circumflex

source

line

level of PASCAL structured statements.
a comment is the same level as that of

follows

warnings

the

each
1line
of
source
code after the first BEGIN
The statement level starts at 0 and increases by 1

for
each
statement

title

compiler.

Procedure level o -- Each line that contains
the

the

SOS

Line numbers 0 -- The compiler assigns unique

errors

and

a and

level 1 indicates declarations in the
outermost
procedure level number increases by one for each
of functions or procedures.

The
the

The lines of the source code are printed in the
1In addition, the listing contains the following
to the source code:

PASCAL

numbers

lists

1line.
time of

time of source file creation

SOS line numbers " -— If you created or
line

title
date and

file specification of the source file

appear

Listing

Source Code Listing -source
code listing.
information pertaining

contains
,
the

the PASCAL compiler name and version number o ,

specifying

VAX/VMS

PROGRAM

of the listing
module
name

listing page number

2.4.1
Each

page
the

line

error

through

containing

line

the

numbers

source

error

trace

listing.

and

the

diagnostic
the

error

COMPILING

Summary

how

much

generated
all

the

time

end

was

warning

the
errors
.
Finally, the

options

2.4.2

the

required
error

source

for

PROGRAM

listing,

the

messages,

the

and the source
compiler lists the

@ .

Machine Code

the

_compiler

compilation

compiler

prints

line number of
status of all

tells

your
a

vyou

program

summary

the
the

last message
compilation

the
machine

MACHINE CODE
code listing

Listing

The
machine
code
qualifier)
follows

1listing
(if
the
source

requested
1listing.

with
The

contains:

Symbolic representation @
similar

instruction
in

one

performs

VAX-11l

MACRO

generated.

~— The

Because

must

pass,

branch

optimization.

the

generate

Source line number

Hexadecimal address

first
first

‘E’

-- A

appears

these

instructions

optimization

should

these

not

use

the

source

-— The

address

line

addresses

number

hexadecimal

the

for

before
can

object

debugging

marks

generated

for

address

1is

the
the

instruction.
purposes

You

because

they
do
not
correctly
correspond
to
the locations in
executable image.
The branch optimization mentioned above
change the addresses of the object instructions.

cause

deleted.
Therefore, these
to those appearing
in
the

object
instruction that the compiler
PASCAL statement on that source line.

approximation

representation,

for each object
compiler
operates

PASCAL

Branch

certain
BRW
instructions
to
be
instructions will not be identical
executable image.

symbolic

instruction,

Hexadecimal

1instruction

@ -- This

its one-pass
instructions

operation, the compiler must
before
it
can
determine

the

the
can

hexadecimal

rTepresentation of the object instruction.
You should read the
hexadecimal
instruction
from
right
to
1left
because
the
rightmost
byte
has
the smallest address.
Again, because of

their
as

operands.

zeros.

The

After

instruction,

compiler

but

places

addresses

generating

before

the

these

operands

hexadecimal

writing

<correct

generate some object
the address bytes of

the

wvalues

object
1into

are

printed

representation

code
the

file,

binary

the

object

code.

2.4.3

Cross—-Reference

The

cross-reference
qualifier)
appears
sections:

Listing

listing

after

(if
the

requested
with
the
machine code listing.

User-specified identifiers

Globally-defined identifiers @ -- This

identifiers

you

declared.

REFERENCE
contains two

—- This section lists all

the

identifier

The cross-reference listing does not specify pointer
that are used before they are declared.

type

predefined

identifiers

that

the

section

the

1lists

PASCAL

program

Each line of the cross-reference listing contains an
and

CROSS

uses.

1list
of
the source line numbers where the identifier is used
.
The
first
1line
number
indicates
where
the
identifier
|is
declared.
Predefined identifiers are listed as if they were declared

on line 0.
identifiers

2-11

CHAPTER 3
LINKING A PROGRAM

object
the
link
you
PASCAL program
VAX-11
After compiling your
Linking resolves all
produce an executable image file.
to
module (s)
for
addresses
references in the object code and establishes absolute
locations.

symbolic

LINK COMMAND

THE

3.1

To link an object module, issue the
general

LINK

command

the

following

form:

(s)] file-spec[/file-qualifier(s)...]
$ LINK[/command-qualifier
/command-qualifier
(s)

Specify output

file options.

file-spec

Specifies the

input object

file to be linked.

/file-qualifier
(s)

Specify

input

interactive

file options.

mode,

you

<can

accompanying file specification.

issue

the

LINK

command

with

The system responds with the prompt:

$_File:

prompt.
the
The file specification must be typed on the same line as
not fit on one line you can type a
does
specification
file
the
If
next
hyphen (-) as the last character of the line and continue on the
line.

You can enter multiple file specifications separated from each other
When used with the LINK command, the comma
or plus signs.
commas
by
the
no matter which you use,
has the same effect as the plus sign:
1If no
creates a single executable image from the input files.
linker
output file is specified, the linker produces an executable image with
the same name as the first object module and a file type of EXE.
Table 3-1 lists the linker qualifiers of particular

users.

linker.

See

the

VAX-11]

Linker

interest to PASCAL

Reference Manual for details on the

LINKING A PROGRAM

Table

Linker

Type

3-1

Qualifiers

Qualifier

Negative Form

Default

Command

/BRIEF

None

Not applicable

/CROSS_REFERENCE

/NOCROSS_REFERENCE

/DEBUG

/NODEBUG

qualifiers

/EXECUTABLE[=file-spec] | /NOEXECUTABLE

/EXECUTABLE

/FULL

None

Not applicable

/MAP[=file-specl

/NOMAP

/NOMAP
/MAP

(interactive)

(batch)

/SHAREABLE[=£file-spec]

/NOSHAREABLE

/TRACEBACK

/NOTRACEBACK

/TRACEBACK

Ipput

/INCLUDE=module-~name (s)

None

Not applicable

gtiiifiers

/LIBRARY

None

Not applicable

3.2

LINKER COMMAND QUALIFIERS

LINK command qualifiers modify the output of the
1linker
and
specify
whether
the
debugging
or
the traceback facility is to be included.

Linker output consists of an image file and, optionally, a
map
file.
The
following
qualifiers,
described
1in
Section 3.2.1, control the
image

file generated by the

linker:

/EXECUTABLE[=file-spec]
/NOEXECUTABLE

/SHAREABLE[=file-specl

The map file qualifiers,

described

in Section

3.2.2,

are:

/MAP[=file-spec]
/BRIEF
/FULL

/CROSS_REFERENCE

The debugger and traceback qualifiers,
ares.

‘

/DEBUG
/TRACEBACK

described

Section

3.2.3,

LINKING

3.2.1
The

PROGRAM

1Image File Qualifiers

image

file

qualifiers

include:

/EXECUTABLE

/SHAREABLE

If you do
not
specify
an
1image
file
qualifier,
the
/EXECUTABLE;
the
1linker
produces an executable image.
production of an image, specify the negative form, as:

default
Iis
To suppress

/NOEXECUTABLE
For

example:

LINK/NOEXECUTABLE

CIRCLE

The file CIRCLE.OBJ
1is
1linked,
but
no
1image
1is
generated.
The
/NOEXECUTABLE qualifier is useful if you want to verify the results of
linking an object file without actually producing the image.

To designate
/EXECUTABLE

a
in

file
the

specification

for

executable

1image,

use

generated

form:

/EXECUTABLE=file-spec
For

example:

The

LINK/EXECUTABLE=TEST CIRCLE

file CIRCLE.OBJ

named

linked,

and

the executable

image

TEST.EXE.

A shareable image is one that can be used in
a
number
of
different
applications.
It
can
be
a
private
image
vyou
use
for your own
applications, or it can be installed
in
the
system
by
the
system
manager
for
use
by all users.
To create a shareable image, specify
/SHAREABLE.

For

example:

LINK/SHAREABLE CIRCLE

To include a shareable image as input to the linker, you must
use
an
options
file
and
specify
the
/OPTIONS
file qualifier in the LINK
command.
Refer to the VAX-11l Linker Reference Manual for details.

If you specify /NOSHAREABLE,

3.2.2

the linker generates an executable

image.

Map File Qualifiers

The map file qualifiers tell the linker whether to generate a map file
and,
if so, the information the map file is to include.
The map file
qualifiers

are:

/MAP
/BRIEF
/FULL

/CROSS_REFERENCE

3-3

LINKING

The

map

qualifiers

are

specified

PROGRAM

follows:

/MAP[=file-spec] u{/FULL }H[I/CROSS_REFERENCE]]
/BRIEF

Note

that

you

must

specify

/MAP

vyou

specify

/BRIEF,

/FULL,

/CROSS _REFERENCE.
The linker uses defaults to generate
interactive
mode, the default is to
the default is to generate the map.

or
suppress
a
suppress the map;

map
in

file.
In
batch mode,

If no file specification is included with /MAP, the map file
name
of the first input file and a file type of MAP.
It is
the default device, in the default directory.

The
optional
qualifiers
/BRIEF
and
/FULL
define
information included in the map file, as follows:

amount

/BRIEF produces a summary of the
list of contributing modules.

/FULL produces (1) a summary of
the
1image's
characteristics
and
a
1list
of contributing modules (as produced by /BRIEF),
(2) listings of global symbols by name and by value, and (3) a
of

characteristics

1image

characteristics

summary

image's

the

has
the
stored on

sections

the

and

linked

image.

By default,
contains
a

if you specify neither /BRIEF
nor
/FULL,
summary
of
the'
image's
characteristics

contributing

modules

(as

symbols

values,

the

and

map

file's

produced
symbol

contents,

name

/BRIEF),
order.

refer

plus

For

the

the
and

1list

complete

VAX-11]

map
file
a list of

global

description

Linker

Reference

Manual.

The /CROSS_REFERENCE qualifier can be used with either the default
or
/FULL
map qualifier to request cross-reference information for global
symbols.
This
cross-reference
information
indicates
the
object
modules
that define and/or refer to global symbols encountered during
linking.
The default is /NOCROSS_REFERENCE.

3.2.3

Debugging and Traceback Qualifiers

The

/DEBUG

included

generated.
executes

/NODEBUG
When

qualifier

1in
If

you

under

with

the

indicates

specify

the

RUN

that

executable
/DEBUG

control

command.

the

VAX-11

1image

and

The

link

the

symbolic

debugger

symbol

table

the

program

time,

debugger,

default

link

unless

time

you

always
specify

/NODEBUG.

vyou
specify
the
/TRACEBACK
qualifier,
error
messages
are
accompanied
by symbolic traceback information showing the sequence of
calls that transferred control to the program unit in which the
error
occurred.
If
you
specify
/NOTRACEBACK,
this
information
1is not
produced.
The default is /TRACEBACK.
If you specify both /DEBUG
and
/NOTRACEBACK,
the traceback capability is automatically included, and
/NOTRACEBACK has no effect.

LINKING

3.3

LINKER

INPUT FILE

Input file
qualifiers
specification.
Input
specified in an options

3.3.1
The

PROGRAM

QUALIFIERS

are
used
as
modifiers
on
files
can
be
object files,
file, or library files.

the
input
file
shareable images

/LIBRARY Qualifier

/LIBRARY

qualifier

has

the

form:

/LIBRARY

This qualifier specifies that
the
input
file
1is
an
object-module
library
that
the
1linker
must
search
to resolve undefined symbols
referenced in other input modules.
The default file type is OLB.

3.3.2
The

/INCLUDE Qualifier

/INCLUDE

qualifier

has

the

form:

/INCLUDE=module-name
(s)

This qualifier specifies that
the
input
file
1is
an
object-module
library,
and
that
the
modules
named
are the only modules in that
library that are to be explicitly included as
input.
At
1least
one
module name is required.
To specify more than one, enclose the module
names in parentheses, and separate them
with
commas.
The
/LIBRARY

qualifier

can

input

specification.

input

file

file,

references.

be
the

used with
If

specified

the
you

/INCLUDE qualifier
use

1library

both

1is

to modify a

qualifiers

also

searched

for

the

single
same

unresolved

CHAPTER 4
EXECUTING A PROGRAM

can
system
the
program,
your
1linked
and
After you have compiled
It has the form:
The RUN command initiates execution.
execute it.
RUN[/[INOCIDEBUG]file-spec

are
default values
You must specify the file name;
of the file specification.
elements
optional
omit
type

vyou
if
applied
The default file

EXE.

vyou
1if
even
debugger,
the
wuse
to
The DEBUG qualifier allows you
omitted this qualifier from the PASCAL and LINK commands (see Sections
without
executes
If you specify /NODEBUG, the program
2.2 and 3.1).
This qualifier allows you to override a /DEBUG
debugger intervention.
qualifier specified at

4.1

link

FINDING AND CORRECTING

time.

ERRORS

Both the compiler and the
Run-Time
Library
include
facilities
for
detecting and reporting errors.
VAX/VMS also provides the debugger to
you
In addition to the debugger,
help you locate and correct errors.
can
use
a
traceback facility to track down errors that occur during
program execution.

4.1.1

Error-Related Command Qualifiers

At each step in compiling,

linking,

and executing

your

program,

Yyou

qualifiers that affect how errors are reported.
command
specify
can
that
ensure
to
/DEBUG qualifier
the
At compile time, you can use
link
At
debugger.
the
by
use
for
preserved
1is
information
symbolic
time, you can also specify the /DEBUG qualifier to make
the
symbolic
The same qualifier can be
debugger.
the
to
available
information
specified with the RUN command to invoke the debugger at run time.
Table

4-1 summarizes the /DEBUG and /TRACEBACK qualifiers.

the
in
point
any
at
qualifiers
these
of
none
use
you
If
compile-link-execute sequence, and an execution error occurs, you will
receive

traceback

list by default.

EXECUTING

PROGRAM

Table

/DEBUG

Qualifier

Command

/DEBUG

PASCAL

4-1
/TRACEBACK

and

Effect

The

PASCAL

creates
needed

/DEBUG

Qualifiers

LINK

compiler

symbolic
by

Symbolic

Default

the

/NODEBUG

data

debugger.

data

created

the

compiler is passed to the
debugger.
Traceback list
also produced.
/TRACEBACK | LINK

Traceback

information

passed

the

Traceback

/DEBUG

RUN

Invokes
DBG>

/NODEBUG

RUN

the

needed

specified.
/DEBUG
LINK

Not

/TRACEBACK

produced.

was

The

None

displayed.
LINK/DEBUG

specified

command,

suppresses

debugger.

was

the

debugger.

list

prompt

/NODEBUG

the

None

RUN/NODEBUG

DBG>

prompt.

perform symbolic debugging, you must use the /DEBUG qualifier
with
both
the PASCAL command and the LINK command.
It then is unnecessary
to specify it with the RUN command.
If you omit
/DEBUG
from
either
the
PASCAL
command
or the LINK command, you can use it with the RUN
command

invoke

the

not contain debug
will be forced to
symbolically.
If you specify
the
event
of

listing
The

and

the

message
error,
of the

list

error

message,

interpreted

displayed

executable

image

will

in debugging, and you
values,
rather
than

list in
program

follows:

detected,

vyou

receive

the

traceback

information.

the

system,

indicating

the address at which the
Processor Status Longword

traceback

the

list.

condition

followed

However,

and symbol tables used
addresses as absolute

LINK/NOTRACEBACK, you will receive no traceback
error.
Figure 4-1 shows an example of a source

traceback

When

the

debugger.

records
express

the
In

appropriate

this

the

example,

nature

error occurred (PC),
(PSL).
This message

and
is

the

the contents
followed
by

information.

The traceback information is presented in inverse order to the routine
or
subprogram
calls.
Of
particular interest to you are the values
listed under routine name and line, the first
of
which
shows
which
routine
or
subprogram generated the error.
The value given for line
corresponds

listing

(not

compiler-generated

confused

with

line number indicates the nearest
begins.
Using this information,
a

short

time.

you

specify

control

debugger;

either

LINK/DEBUG

execution.

the

traceback

1list

line

number

editor-generated

the

line

source

previous line on which
you can usually isolate

RUN/DEBUG,

error

1is

4-2

not

occurs,

the

automatically

The

a
statement
the error in

debugger

control

program

numbers).

reverts

assumes
to

printed.

the

EXECUTING A PROGRAM

information, you can use the debugger command SHOW

traceback

display

CALLS, as described in Section 4.1.2.
the

For more information

with VAX-11 PASCAL programs,

debugger

using

refer to the VAX-11 PASCAL

Installation Guide/Release Notes.

$ PASCAL/LIST TRACE
% TYFE TRACE

TRACETEST
01

3-0CT-1979
3-0CT-1979

11321108

VAX~-11 PASCAL VERSION V1.0-1
DELICTEST.FASITRACE.PASH1 (1)

112041

PAGE

LEVEL

LINE

FROC STMT STATEMENT

NUMEERS
1

FPROGRAM TRACETEST?

FROCEDURE F1

4
S
&
7

BEGIN
X $¢=
ENDS

8
9

)
1
0

(VAR X

¢ REAL)?

1.0/X3

PROCENURE P2

REAL)?

BEGIN

0
1
0
0
0
1
O
0

10
11
12
13
14
15
16
17

Comeilation time =

P1CY)$
ENDS
REGIN
F2¢0.0)%
END.

0.+68 seconds (

1500 lines/minute).

Active ortions at end of comerilation?

NODEBUG » STANDARD s LIST » NOCHECK » WARNINGS y NOCROSS _REFERENCE »
IT = 30
HINE
_CODE OBJECTyERROR_LIM
NOMAC
$ LINK TRACE
$ RUN TRACE

YSYSTEM-F~FLTDIVs arithmetic trass floating/decimal divide by zero at PC=00000429, FPSL=03C0002A
LTRACE-F~TRACERACKy sumbolic stack dume follows

module name

routine name

TRACETEST
TRACETEST
TRACETEST

F1
P2
TRACETEST

Figure 4-1

4.1.2

lire

relative PC

absolute PC

5
10
i4

00000029
00000030
0000004E

00000429
0000045D
000004AC

Source Program Listing Traceback List

SHOW CALLS Command

When an error occurs in a program that is executing under the

control

To generate a
1is produced.
1list
traceback
no
of the debugger,
traceback list, use the SHOW CALLS command, which has the form:
DBG>SHOW CALLS

EXECUTING

4.2

SAMPLE

TERMINAL

simple

dialog

Username!

SMITH

Fassword!

PROGRAM

SESSION

between you

and

the

system might

appear

RET

WELCOME

(Your
TO

password

VAX/UMS

VERSION

not displayed)

1.6

$ EDIT CIRCLE.FAS 6D
Inerut IDBAZIESMITHICIRCLE
. PAS
00100

(enter source program)
(terminate edit session and write

CORAZ2ICSMITHICIRCLE
.FAS#11]
$
$

FASCAL/ZLIST
LINK CIRCLE

RUN

CIRCLE

file

to disk)

follows:

CHAPTER

INPUT AND OUTPUT

input and output

This chapter describes

(I/0)

for VAX-11

PASCAL.

information about PASCAL I/O in relation to
provides
it
particular,
covered
topics
The
VAX-11 Record Management Services (VAX-11 RMS).
include:

5.1

Logical

names

PASCAL

file characteristics

PASCAL

record

OPEN

Local interprocess communication by means of mailboxes

Remote communication by means

formats

procedure

parameters

of DECnet-VAX

LOGICAL NAMES

a
The VAX/VMS operating system provides the logical name mechanism as
If you use
independent.
file
and
device
programs
making
of
way
on
logical names, your program need not specify the particular device

which

resides

file

the

particular

file that contains data.

Specific devices and files can be defined at run time.

A logical name is an alphanumeric string, up to
63
characters
long,
that
you
specify
in
place
of a file specification.
The operating
system
provides
a
number
of
predefined
1logical
names,
already
Table 5-1 lists the
particular file specifications.
with
associated
logical

names

special

interest

PASCAL

users.

In addition, the system manager defines the
1logical
names
PASSINPUT
and PASSOUTPUT, which default to SYSSINPUT and SYSSOUTPUT, at compiler
installation.

Logical names provide great flexibility because they can be associated
not only with a complete file specification, but also with a device, a
device and a directory,

or even another

logical name.

INPUT AND

Table

Predefined

Name

OUTPUT

5-1

System

Logical

Names

Meaning

SYS$DISK

Default

device

Default

and

directory

As specified

the

user

SYSSERROR

Default

error

message

SYSSCOMMAND | Default

command

SYSSINPUT

input

Default

input

file

User's

stream

User's terminal
(interactive);
batch
command file (batch)

terminal
(interactive);
batch
log file (batch)

file

User's

terminal

(interactive);

command

SYSSOUTPUT

Default

output

file

User's

<can

create

specification

logical

means

name

the

and

terminal

command.

execution, you can associate the logical names in
file specifications appropriate to your needs.

$ ASSIGN

DBAO:[SMITH]TEST.DAT;2

file

associate

ASSIGN

the

This
file

batch

(batch)

(interactive);

log

You

file

with

Thus,

specification

OPEN

when

For

encounters

the

before

file

program

your

program

For

example:

with

DATA

command creates the logical name DATA and associates it
specification DBAO:[SMITH]TEST.DAT;2.
The system uses

execution.

batch

(batch)

logical

name

DATA

with
this

the
file
program

during

example:

(INDATA,

'DATA',

OLD);

In
executing
this
PASCAL
statement,
the
system
uses
the
file
specification
DBAOQ:[SMITH]TEST.DAT;2
for
the logical name DATA.
To
specify a different file when you execute
the
program
again,
issue
anothér ASSIGN command.
For example:

ASSIGN

DBA2:[JONES]REAL.DAT;7

This command associates the
specification
and replaces
OPEN

statement

above

logical
the

DATA
name

will

DATA with
logical

now

different

name

assignment.

refer

the

file
The

file

DBA2: [JONES]REAL.DAT;7.
You

can

also

(see

the

5.2

FILE

clear

and

the

assign

VAX/VMS

logical

Command

names

Language

with

the

User's

MOUNT

and

DEFINE

commands

are

organized

Guide).

CHARACTERISTICS

distinction

must

way

are

records

made

accessed.

between

the

way

files

INPUT AND

The

term

file

arranged

organization
a

storage

applies

device.

method

OUTPUT

the

The

term

way

records

record

are

access

physically

refers

the

used
to
read
records
from
or
write
records
to
a
file,
regardless
of
the
file's
organization.
A
file's organization is
specified when the file is created
and
cannot
be
changed.
Record
access is specified each time the file is opened and can vary.

5.2.1

File Organization

VAX-11

PASCAL

supports

consist
of records
the file (the first
the
second
record
on).

5.2.2

You

specify

Direct
select
the

file

one

record

organization.

1is

the

second

can

added

mode

Sequential

in which they
first record
record

only

the

are
in

the

end

files

written to
the
file,

file,

and

the

file.

two

ways

parameter
accessing

the

OPEN

procedure.

records:

sequential

file,

record

Sequential

access

provides

Sequential

you

written
records

record

PASCAL

file

Access

from

can

result,

Record

VAX-11

sequential

arranged in the order
record written is the

access mode, records are written to
or
read
starting at the beginning and continuing through the
after another.

access

only

after

written

only

select

direct

file

means

reading

the

all

end

that

the

you

records

file

can

read

preceding

that

particular

it.

open

New

for

records

sequential

access.

you

records

are

relative

record

RECORD

Records

You

are

indicating

a
and

only
open for

stored

Fixed

Variable

can

you

can

FIND

the

record

file

input

access

one

fixed-length

When you specify
specifying
that

file
to

order

must

include

read.
fixed-length

which
the

You

can
records,

(reading).

two

formats:

length

Fixed-Length

records,

the
call

length

Variable-length

bytes.

to
contains

records

can

mode.

5.3.1

specify

procedure

FORMATS

mode.

mode,

Each

number

directly
access
resides on disk,

5.3

access

accessed.

sequential

accessed

only

direct

sequential

Records

fixed-length

all

opened

allow

either
be

the

records

for

direct

record

the

(see

Section

file

contain

access

location

5-3

must
be

5.4.4),

the

contain

computed

you

same

are

number of
fixed-length

correctly.

INPUT AND

5.3.2

OUTPUT

Variable-Length Records

Variable-length records can contain any number of
bytes,
up
to
the
buffer
size
specified
when
the
file
was opened.
Variable-length
records are prefixed by a count field, indicating the number of
bytes
in
the
record.
The count field comprises two binary bytes on a disk
device and four decimal digits on magnetic tape.
The value stored
in
the count field indicates the number of data bytes in the record.

5.4

OPEN

PROCEDURE

PARAMETERS

This section supplements the description of the
OPEN
procedure
that
appears
in
the
VAX-11
PASCAL
Language
Reference
Manual.
In
particular, it describes how the
VAX-11
Record
Management
Services
(RMS) affect VAX-11 PASCAL.
For more information, refer to the VAX-1l1
Record Management Services Reference Manual.
The

OPEN

procedure

OPEN

has

the

(file-variable,

following

'VAX/VMS

general

file

format:

specification',

file-status, record-access-mode,
carriage-control);

buffer-size,

record-type,

Buffer size,
file
status,
record
access
mode,
carriage
<control
are
RMS-dependent
attributes

record
type,
described
in

and
this

section.

5.4.1

The

Buffer

buffer

Size

size parameter

record size in bytes for
text file is 133 bytes.
has

is an

integer

a text file.
For a file of

that

specifies

The default for
any other type,

the

maximum

a VAX-11 PASCAL
this
parameter

meaning.

5.4.2

File

Status

The file status parameter indicates whether the specified file
exists
or must be created.
The possible values for this parameter are:
NEW

OLD

A file type of NEW indicates that a new file must be created with
specified characteristics.
NEW is the default value.

If you specify OLD, the system tries to open
an
existing
error
occurs
1f
the file cannot be found.
A type of OLD
for internal files, which are newly created each
time
the
program

5.4.3

unit

Record

The record
records in

DIRECT

file.
An
is illegal
declaring

executed.

Access

Mode

access mode parameter
specifies
the
mode
of
access
the file.
The possible values for this parameter are:

SEQUENTIAL

the

INPUT AND OUTPUT

The
can

default record access mode is SEQUENTIAL.
In SEQUENTIAL mode,
access files that have fixed- or variable-length records.

DIRECT mode

fixed-length
records

5.4.4

The

allows you

records.

DIRECT

Record

mode.

use

the

FIND procedure

You cannot access a

The

read

files

with

file with variable-length

Type

record type parameter specifies

file.

you

possible

values

for

this

the

structure

parameter

records

are:

FIXED
VARIABLE

A value of FIXED indicates that all records in the file have the
same
length.
A
value
of
VARIABLE indicates that the records within the
file can vary in length.
VARIABLE is the default for a new file.
For
an
existing
file, the default is the record type associated with the
file

5.4.5

The
in

its

creation.

Carriage Control

carriage control
effect
for
an

parameter

parameter specifies
output
text
file.

the type of carriage
The possible values

control
for this

are:

LIST
CARRIAGE
NOCARRIAGE

A value of LIST indicates that each record will be preceded by a
1line
feed
and
followed
by a carriage return when the file is output to a
terminal or line printer.
LIST is equivalent to the VAX-1ll RMS record
attribute CR.
LIST is the default for all text files.
A value of CARRIAGE indicates that
the
contains
a carriage control character.
VAX-11 RMS record attribute FTN.

first
byte
CARRIAGE is

of
each
equivalent

record
to the

A value of NOCARRIAGE indicates that the record contains
no
carriage
control
information.
NOCARRIAGE
1is
equivalent
to
the VAX-1l1l RMS
record attribute PRN with all bits equal to zero.

5.5

LOCAL

INTERPROCESS

is often useful

synchronize

A mailbox

COMMUNICATION:

exchange data

execution

record-oriented,

send

MAILBOXES

between processes:

for

example,

messages.

pseudo

I/0

device

that

allows data

be
passed
from one process to another.
Mailboxes are created by the
Create Mailbox (SYSSCREMBX) system service (see Section 6.2.1
for
an
example
using
SYSS$SCREMBX).
This
section describes how to send and
receive

data

using

mailboxes.

Data transmission by means of mailboxes is synchronous;
that
1is,
a
PASCAL program that writes a message to a mailbox must wait until that
message is read, and a program that reads a
message
from
a
mailbox

INPUT AND

must wait until
the mailbox, an

a message is
end-of-file

program.

VAX-1l

before

returns

For

RMS

written.
When
«condition
1is

ensures

control

OUTPUT

the writing program closes
returned
to
the
reading

that

the

message

the

user

program.

transmission

complete

example:

FROGRAM

MAIL

VAR

MEX

(MBXs

OQUTFUT)

TEXTS

OFEN

(MEBXy

REGIN

RESET
WHILE

‘MAILBOX‘y

(MEX)3
NOT EOF

(MEX)

OLDy

SEQUENTIAL)

BEGIN

WHILE

NOT EQLN
BEGIN
WRITE

(MEX)

(MRXTM)3$

GET (MEX)
END
WRITELNS

GET

(MEX)

END?
CLOSE

(MEX)

END,

This

program

reads messages

MAILBOX.

The

the

terminal.

user's

5.6

messages

COMMUNICATING WITH

your

computer

VAX-11

PASCAL

network.
at

the

files

computer

task-to-task

different

allow

from

DECnet

communicate

you

(resource

communication
That

is,

local

which

and

these

the
are

logical
then

name

printed

NETWORKS

network,

with

exchange

(task-to-task

remote computer

transparent.

nodes

the

known by

text,

procedures

remote

are

COMPUTERS:

procedures

the

mailbox

lines

REMOTE

These

Both

one

1I/0

from a

are

you

other

data

with

communication)

can

nodes

and

use

the

program
to

access

sharing).

resource

sharing

intersystem

interprocess

and

between

exchanges

file-access

To communicate across the network, specify a node
element of a file specification.
For example:

name

systems

not

appear

exchanges.
as

the

first

BOSTON: :DBAQ: [SMITH] TEST.DAT;
2

Remote

task-to-task

communication

specification.

You

must

use

name

the

task

name,

and

supply

the
as

requires

notation
in

the

TASK=

special

following

form

place

the

file

device

example:

BOSTON:
: "TASK=UNA"

The

example

specifies

the

task

named

fragment

shows

UPDATE

the

BOSTON

node

the

received

from

network.

The
a

following

remote

program

means

VAX-1l

how messages
PASCAL

I/0

can

procedures.

INPUT AND OUTPUT

OFEN (NETJORy ‘RBOSTONZI:*TASK=UNA"'»OLD)#
RESET (NETJOR)
3
RDAT (= NETJOBTM
S
NET..FROC (RDATsWRTDATY
S
CLOSE

(NETJOR)
S

The effect of these statements is to
establish
a
1link
with
a
Jjob
(TASK)
named
UNA
at the node BOSTON and to receive a component from
the file variable associated with the remote
program.
The
variable
RDAT
contains
the
data.
Then
the procedure NET_PROC is called to
process

the

data

and

the

link

broken.

The next example shows how you can write a
PASCAL

I/0

remote

file

wusing

VAX-11

procedures.

FROGRAM

UFDATE

VAR

(NEWDATy

RBRANCH)
s

NEWDAT

FILE

INTEGERS$

BRANCH

FILE

INTEGERS$

REGIN

OFEN

(NEWDATy "NEWDAT.DAT

»OLIDD 5

RESET (NEWDAT)
OFPEN (BRANCHy 'NASHUA"PLUGH
REWRITE (BRANCH)
WHILE

NOT

EOF (NEWDAT)

XYZZY" ! IMASTER.DAT
s NEW) 5

BEGIN
BRANCHTM
: =NEWLATTM §
GET
FUT

(NEWDAT)
(BRANCH)

END

CLLOSE

(BRANCH)
3

(NEWDAT)

END.

The sample program writes records in a remote file at the node NASHUA.
It
reads
data from a local file known by the logical name NEWDAT and
writes the data across the network to the remote
file
MASTER.DAT
1in
the

directory

[PLUGH]

with

password

XYZZY.

If you use logical names in your program, you can equate
the
logical
names
with
either
1local
or
remote
files.
Thus, if your program
normally
accesses
a
remote
file,
and
the
remote
node
becomes
unavailable,
you
can bring the volume set containing the file to the
local site.
You
<can
then
mount
the
volume
set
and
assign
the
appropriate
Remote

logical

name.

For

example:

Access

$ ASSIGN REM::APPLIC SET:file-name
Local

LOGIC

Access

$
$

MOUNT device-name APPLIC_SET
ASSIGN APPLIC SET:file-name

LOGIC

The MOUNT and ASSIGN commands are described
Command

Language

User's

Guide.

DECnet capabilities are described

in detail

the

VAX/VMS

the DECnet-VAX Reference Manual.

CHAPTER
CALLING

CONVENTIONS

In the context of the VAX/VMS
operating
system,
a
procedure
is
a
routine
entered
by
a CALL instruction.
1In a PASCAL program, such a
routine can be a function or procedure written in PASCAL,
a
function
or procedure written in some other language, a VAX/VMS system service,
or a VAX-11l Run-Time Library procedure.
In
many
cases,
procedures
perform
calculations
that
are
used
widely
and repeatedly in many
applications.
1In PASCAL, you can write each procedure once
and
call
it

from many

other

programs.

This chapter describes how to call procedures that are not written
in
PASCAL and provides information on calling VAX/VMS system services and
VAX-11]

Run-Time

Reference
functions

Library

Manual
and

for

procedures.

See

the

information

defining

VAX-11]

PASCAL

and

Language

invoking PASCAL

procedures.

The material
presented
here
assumes
some
knowledge
of
procedure
calling
and argument passing mechanisms.
You should be familiar with
these subjects before you attempt to use
the
features
described
in
this
chapter.
Refer to the VAX-11 Run-Time Library Reference Manual
and the VAX-11 Architecture Handbook for more information.

6.1

VAX-11l

PROCEDURE

CALLING

STANDARD

Programs compiled
by
the
VAX-11
PASCAL
compiler
conform
to
the
standard
defined
for
VAX-11
procedure calls (see Appendix C of the
VAX-11 Architecture Handbook).
This standard prescribes how arguments
are

passed,
how
function
wvalues
are
returned, and how procedures
receive and return control.
VAX-11] PASCAL also provides features that
allow programs to call system services and procedures written in other
native-mode languages supported by VAX/VMS.

VAX-11

PASCAL

uses

the

VAX-1l

CALLS

instruction

<call

procedures.

The illustrations in Appendix B outline the events that occur during a
procedure call and show the structure of the run-time stack after each
event.

6.1.1

Argument

Lists

Each time you call a procedure, VAX-11 PASCAL constructs
an
argument
list.
The VAX-11 procedure calling standard defines an argument list
as a sequence of longword (4-byte) entries.
The
first
byte
of
the
first entry in the list is an argument count, which indicates how many
arguments follow in the list.

CALLING

CONVENTIONS

The arguments
in
the
1list
are
based
on
specified
in
the
formal parameter list and
parameter list.
The argument list
contains

the
the
the

passed

6.1.2

Parameter

the

procedure.

passing
mechanisms
values in the actual
arguments

actually

Passing Mechanisms

Non-PASCAL
procedures
require
arguments
as
addresses,
immediate
values, or descriptors.
The VAX-11l procedure calling standard defines
three mechanisms by which arguments are passed to procedures:

The

By-reference
value

By-immediate-value

By-descriptor
descriptor of

following

VAX-11

the

argument

the

list

program

describe
to

using
each
of
pertains
only

entry

argument

-- the argument
the value

subsections

PASCAL

subprograms
information

list

what

correctly

the

list

entry

you
pass

must

address

the

passing

PASCAL

Langquage

between

PASCAL

these
mechanisms.
Note
that
to subprograms written in languages

Reference

Manual

for

description

vyour

non-PASCAL

than PASCAL.
For
information
about
passing
arguments
to
subprograms from non-PASCAL programs, see Section 6.1.5.
Refer

VAX-11

the

value

address

specify

arguments

this
other
PASCAL
to the

parameter

subprograms.

6.1.2.1

By-Reference Mechanism - The
by-reference
mechanism
passes
the
address
of
the
actual parameter.
By default, PASCAL uses this
mechanism for everything except dynamic
array
parameters.
You
can
invoke the by-reference mechanism in two ways:
1.

By omitting
list.
This

the mechanism specifier from the formal parameter
syntax invokes PASCAL by-value semantics.

By using the VAR specifier
This syntax invokes PASCAL

in
the
formal
parameter
by-reference semantics.

1list.

If you
actual

omit the mechanism specifier, PASCAL passes the address of
the
parameter.
PASCAL
expects
the called procedure to copy the
value from the specified address to local
storage.
You
should
use
this
method only if the called procedure does not change the value of
the corresponding actual
parameter.
When
vyou
omit
the
mechanism
specifier,

the

actual

For example, the
function call use
FUNCTION
TANH

parameter

must

following
function
this method:

MTHSTANH
MTHSTANH

(ANGLE

expression.

declaration

REAL) :REAL;

and

corresponding

EXTERN;

(RADIANS);

This example declares the VAX-11 Run-Time Library function MTHSTANH as
an
external
subprogram.
The
MTHS TANH
function
returns,
in
floating-point notation, the hyperbolic
tangent
of
an
angle.
The
input
parameter
to
this
function
1is
the
size
of
the angle (in
radians), and it must be passed by-reference.
Because
the
function
MTHSTANH
does
not
change
the
value of the angle, you can omit the
mechanism specifier when you declare the function.
The returned value

CALLING

is
assigned
(See Section

to
6.3

CONVENTIONS

the variable TANH by the assignment statement shown.
for more
information
on
calling
Run-Time
Library

procedures.)

Use the VAR specifier to pass an actual parameter that can
change
in
value
during
execution
of
the
procedure.
You
must use VAR when
passing a file variable as a parameter.
The
VAR
specifier
1is
also
useful to prevent the copying of large parameters.
Specify VAR in the
following
VAR

format:
formal-parm-list

type

;

The formal parameter list specifies one or more formal
parameters
of
the indicated type.
Each of these parameters will be passed using the
by-reference mechanism and with by-reference semantics.

When you call the procedure, the argument list contains the address of
the
value to be passed.
The actual parameter must be a variable or a
component
of
an
unpacked,
structured
variable;
constants,
expressions,

The
pass

procedure

names,

following declarations
an

address

TYPE

BIT64

VAR

SYSTIME

FUNCTION
STATUS

ARRAY

allowed.

function call

show how to

routine.
[1..2]

BOOLEAN;

BIT64;

SYSSGETTIM

function names are not

and corresponding

external

PACKED
:

and

(VAR

SYSSGETTIM

BINTIM

BIT64)

INTEGER;

EXTERN;

(SYSTIME)
;

This example declares the Get Time (SYSSGETTIM) system service,
returns
the
system
time.
The actual parameter SYSTIME is a
variable into which the system service writes the time.

which
64-bit

6.1.2.2
By-Immediate-Value Mechanism - VAX/VMS
system
services
and
Run-Time Library procedures sometimes require that the calling program
pass an immediate value, that is, the value itself.
To direct
PASCAL
to
pass
a
value
instead
of
an
address, use the %IMMED mechanism
specifier,

$IMMED

follows:

formal-parm-list

type

The formal parameter list specifies one or more formal
parameters
of
the
1indicated
type.
Variables
that
require
more than 32 bits of
storage, including all file variables, cannot be passed
as
immediate
values.

You

can use

%IMMED

only with

non-PASCAL

subprograms.

When you call the procedure, the actual parameter
1list
contains
the
value
of
each
parameter for which you specified $IMMED.
The actual
parameter can be a constant, a variable, or an expression.
Note
that
$IMMED
can
also modify procedure and function names, as described in
Section

The

6.1.3.

following declarations and corresponding

pass an
VAR

immediate

value

EVENT_FLAG

FUNCTION
STATUS

function call

show how to

system service procedure.

INTEGER;

SYSSWAITFR
SYSSWAITFR

($IMMED

EFN

(EVENT_FLAG);

6-3

INTEGER)

INTEGER;

EXTERN;

CALLING

This example declares the Wait
system
service,
which
waits
requires

one

value parameter,

CONVENTIONS

for
for
the

to wait.
This number is passed
integer variable EVENT FLAG.

Single
Event
Flag
(SYSSWAITFR)
a
single event flag.
SYSSWAITFR

number

the

event

immediate

flag

value,

for

copied

which

from

the

6.1.2.3
By-Descriptor Mechanism - The by-descriptor mechanism
passes
the
address of a string, array, or scalar descriptor, as described in
Appendix
C
of
the
VAX-11
Architecture
Handbook.
VAX-11
PASCAL
includes
the
$STDESCR
mechanism
specifier
for
passing
string
descriptors and the $%DESCR mechanism specifier for passing
array
and
scalar descriptors.
You cannot pass a component of a packed structure
using either of these specifiers.
You can use these
specifiers
only
with
array

non-PASCAL
descriptor

subprograms.
Note that, by default,
to a formal dynamic array parameter.

PASCAL

passes

To pass a string descriptor, specify %$STDESCR as follows:
$STDESCR

formal-parm-list

type;

The formal parameter list specifies one
or
more
parameters
of
the
indicated
type.
Only string constants, packed character arrays with
subscripts from 1 to n,
and
packed
dynamic
character
arrays
with
subscripts
of
an
integer or integer subscript type can be passed by
string descriptor.
When

you

call
string

each
system

the procedure, the
descriptor.
For

service

TYPE

VAR

requires

PACKED

ARRAY[1..80]

PACKED

ARRAY([1l..6]

FUNCTION

STATUS
The

%STDESCR

string

scalar
in the

The

type.

other

MSGTYPE;

DEV

DEVTYPE)
:

INTEGER;

EXTERN;

(MESSAGE,

list

array

unpacked

array

argument

scalar

list

variable.

TERMINAL);

that

actual

both

high-level

parameters

characters,

formal-parm-list
parameter

CHAR;
CHAR;

MSG

The

written

The

parameters:

$STDESCR

descriptors.
To pass
following format:

scalar

OF
OF

(%STDESCR

indicates

and

formal

array

specifier
80

indicated
type

SYSSBRDCST

$DESCR

the address of
Broadcast (SYS$SBRDCST)

DEVTYPE;

descriptor.

arrays

contains

the

MSGTYPE;
:

SYSS$SBRDCST

list

descriptors

DEVTYPE
MESSAGE

Routines

string

MSGTYPE

TERMINAL

packed

two

argument
example,

MESSAGE

must

and

passed

TERMINAL

are

respectively.
languages

array

may

scalar

require

descriptor,

array

use

$%DESCR

type

specifies

one

type.

The

(fixed

contains

type

dynamic)
the

can be
of

address

parameters

the

any predefined

scalar

predefined
the

descriptor

CALLING

The
a

CONVENTIONS

following example shows how an array descriptor might be passed to

FORTRAN

subroutine.

TYPE FORAY

= ARRAY

PROCEDURE

FORMATRIX

[1..10,1..10]

OF CHAR;

(%$DESCR ARRDES

FORAY);

FORTRAN;

The FORTRAN subroutine
FORMATRIX
expects
the
array
to
by-descriptor.
A call to FORMATRIX might be the following:
FORMATRIX

passed

;
(CHARARR)

The actual parameter CHARARR specifies
a
character
array,
and
the
argument
1list
contains
the
address of a descriptor for this array.
(Note that VAX-11 FORTRAN treats character parameters
as
CHARACTER¥*1
variables.)

6.1.3

Functions and

Procedures

Parameters

You can pass procedure and
function
names
as
immediate
wvalues
routines written in other languages, using the following format:

$IMMED PROCEDURE procedure-name-list
$IMMED FUNCTION

function-name-list

The procedure name list specifies the
procedure
parameters.
The
function
one or more formal function parameters
corresponding
actual
parameter lists
procedures
For

and

functions

passed

FORCALLER

(%$IMMED

type

name
of
one
or
more
formal
name list specifies the name of
of
the
indicated
type.
The
specify the names of the actual
as

parameters.

example:

PROCEDURE

PROCEDURE UTILITY);

FORTRAN;

The FORTRAN subroutine FORCALLER calls a PASCAL procedure and requires
that
the
name of the procedure be passed as an immediate value.
The
argument list contains the address of
the
PASCAL
procedure's
entry
mask.

A call

FORCALLER

the

FORTRAN

(PRINTER);

Any subprogram passed with
and

those

procedure might be:

declared

$IMMED should access only its own variables

program

level.

For information on passing procedure and function names between PASCAL
subprograms, see the VAX-11 PASCAL Language Reference Manual.

6.1.4

Function Return Values

A function returns a value to the calling program
value

the

function's name.

or pointer type;

which a value

The value must be a

structured types are not

returned depends on

its

allowed.

type,

assigning

scalar,

The

that

subrange,

method

as listed below.

CALLING

CONVENTIONS

Type

Return Method

Integer,
single,

real,

General

character,

Boolean, pointer,
user-defined scalar
Double

6.1.5
When

Low-order result
High-order result

Arguments to PASCAL Subprograms
calling

must

RO0:
Rl:

ensure

VAX-11

PASCAL

subprogram

from

that the arguments are
expects

most parameters

non-PASCAL

in the correct
to

subprogram,

form.

you

By default,

passed by-reference.

For a PASCAL value parameter (declared without a mechanism specifier),
the
argument
1list must contain the address of the value.
The PASCAL
subprogram will

copy

the

value

from

the

passed

address

upon

entry.

For

a VAR parameter, the argument list must contain the address of the
variable.
The
subprogram
does not copy the value, but instead uses
the address to access the actual parameter variable.
Actual parameter
variables that can change in value as a result of subprogram execution
must be passed in this manner.
1In addition, all files must be
passed
as

PASCAL

VAR

parameters.

For a
formal
procedure
or
function
parameter
(indicated
by
the
PROCEDURE
or
FUNCTION specifier), the argument list must specify the
address of the bound procedure value, which consists of two longwords.
The
first
1longword
contains
the
address of the entry mask for the

subprogram;
the second longword
contains
the
environment
pointer.
(This
process
implements
the
VAX-11
by-reference
mechanism for a

procedure

For

the

address

6.2

function.)

formal dynamic array parameter,
of

array

CALLING VAX/VMS

the

argument

list

must

contain

descriptor.

SYSTEM

SERVICES

You can declare any VAX/VMS system service as an external function
or
procedure and then call it from your PASCAL program.
When declaring a
system service, specify an identifier in the following form:
SYSS$service-name

For

example,

the

name of

the

S$SFAO

system service

SYSSFAO.

You pass parameters to the system service according to its
particular
requirements:
a value, an address, or the address of a descriptor may
be needed, as described
in
Section
6.1.2.
To
1invoke
the
system
service,

use

function

procedure

call

the VAX/VMS

System Services Reference Manual

each

service.

system

your

for

PASCAL

program.

See

full

description

The
system
provides
three
files
containing
condition
symbol
definitions.
When
vyou
declare a system service or Run-Time Library
procedure, you should
specify
the
appropriate
file
in
the
CONST
section
to
define
the condition values in your PASCAL program.
Use

CALLING

the

$INCLUDE directive

VAX-11

The
and

PASCAL

Language

CONVENTIONS

specify the

file name,

as described

the

Reference Manual.

three files
are
SYSSLIBRARY:LIBDEF.PAS,
SYS$SLIBRARY:MTHDEF.PAS,
SYSSLIBRARY:SIGDEF.PAS, as described below.

SYSSLIBRARY:LIBDEF.PAS

This file contains definitions for
all
condition
symbols
from
the
general
utility
Run-Time Library procedures.
These symbols have the
form:

LIBS xyz
For

example:

LIBS_NOTFOU

SYSSLIBRARY:MTHDEF.PAS

This file contains definitions for
all
mathematical procedures library.
These

condition
symbols
from
symbols have the form:

the

MTHS Xyz
For

example:

MTH$_SQUROONEG

SYSSLIBRARY:SIGDEF.PAS

This

file contains miscellaneous

handlers.

These

symbols

have

the

symbol

definitions

form:

used

condition

SS$_xyz
For

example:

SS$_FLTOVF

6.2.1

Calling System Services by Function Reference

In most cases, you will want to declare a system service as a function
so
that you can check its return status.
Each system service returns
a VAX-11 condition value indicating whether completion was successful.
These
condition
values
can
be
interpreted
as
1integer codes that
correspond to symbolic names such as SS$ ACCVIO.
0dd
codes
indicate
successful completion and even codes indicate failure.
For

example,

Mailbox

the

following procedure

(SYSSCREMBX)

system service.

defines

and

calls

the

Create

CALLING CONVENTIONS

FROCEDURE CREATE_MAILEOXS$

CONST ZINCLUDE ‘SYS$LIBRARYISIGIEF.FAS’
TYFE STATUS = (INT_.STATy» BOOL.STAT)#
WORD = 0..6355357%
SURSZ = 1..637

CHAN_STAT = (INT_.CHAN,DUMMY.CHAN)?
CHAN.TYPE = PACKED RECORD
CASE CHAN.STAT OF

INT.CHAN ¢ (CHAN_NO ¢ INTEGER)S$
DUMMY.CHAN ¢ (BOT_.CHAN ¢ WORID)

ENDs

VAR MEX.REC

RECORD

)

CASE STATUS OF

INT.STAT ¢ (MEX.INT ¢ INTEGER)?
ROOL_STAT ¢ (MEX._.ROOL : ROOLEAN)

ENDs

CHAN_REC

¢!

CHAN._TYFE?

FUNCTION SYS$CREMBX (XZIMMED FPRMFLG ¢ INTEGERS
VAR CHAN : CHAN..TYFE?

»IMMED MAXMSGs BUFQUOy FROMSKs ACMODE : INTEGERS?
YSTOESCR LOGNAM ¢ FACKED ARRAY L[SURG31 OF CHAR):
INTEGER$?

EXTERN#

BEGIN

WITH MBX.REC DO BEGIN

)7
s0s0»
MAILBOX
MEX_INT = SYS$CREMEX(OsCHAN_REC»0+0
IF NOT MBX.EOOL THEN BEGIN

WRITELN (‘Error when truing to create mailbox’)s
HALT
END

END

END§

The function reference allows a return status to be stored 1in the
record MBX REC. If the function's return status is false (represented
by any even integer), indicating failure, an error occurs and error
You can also check for a particular
processing can be undertaken.
s, by comparing the return
privilege
of
return status, such as lack
status to one of the status codes defined by the system. For example:
IF MBX_REC.MBX_INT = SS$_NOPRIV THEN

WRITELN

('No privilege to create mailbox');

Refer to the VAX/VMS System Services Reference Manual for
about

return

status

codes.

The

described with each system service.

relevant

information

return status codes are

CALLING

6.2.2

Calling

CONVENTIONS

System Services as Procedures

If you do not need to check the
return
status,
you
can
declare
a
system
service
as
an
external
procedure
rather
than an external
function.
Procedure calls to system services are made in the same way

that
«calls
are made to any other procedure.
For example,
Create
Mailbox
system
service,
define
and
call
the
SYSSCREMBX,

to use the
procedure

follows:

PROCEDURE

SYSSCREMBX

(%IMMED
VAR

PRMFLG

CHAN

INTEGER;

CHAN_TYPE;

$IMMED MAXMSG, BUFQUO, PROMSK, ACMODE : INTEGER;
$STDESCR LOGNAM : ARRAY [SUB63] OF CHAR);

EXTERN;

SYSSCREMBX

(0,CHAN_REC,0,0,0,0,

You should declare CHAN_REC

MAILBOX')
;

and CHAN TYPE as

the previous section.

This procedure call corresponds to the function
reference,
but
does
not allow you to test the status code returned by the system service.

6.2.3
Most

Passing Parameters
system

services

to System Services

require

input

parameters

passed

immediate
values.
When declaring these parameters, you must use the
$IMMED mechanism specifier.
Some system
services,
however,
require
input
parameters
to
be
passed
by-reference.
For input parameters
passed by-reference, you should use the default
(that
1is,
omit
the
mechanism specifier) so that actual parameters can be expressions.
In addition, most system services
require
output
parameters
to
be
passed
by-reference.
For
these
parameters,
you
must use the VAR
mechanism specifier to ensure that
PASCAL
correctly
interprets
the
output
data.
The VAX/VMS System Services Reference Manual lists the
mechanism by which each parameter to a system service must be passed.

6.2.3.1
Input and Output
By-Reference
Parameters - You
will
often
need
to
tell the system service where to find input values and where
to store output values.
Thus, you must ascertain
the
hardware
data
type of

the parameter:

byte,

word,

longword,

or quadword.

For input parameters that refer to byte, word, or longword values, you
can specify either constants or variables.
If you specify a variable,
it must be of a type that is allocated an equal or greater
amount
of
storage than is allocated to the hardware data type required.

For output parameters, you must declare
a
variable
of
exactly
the
length
required, to avoid including extraneous data.
For example, if
the system returns
a
byte
wvalue
in
a
word-length
wvariable,
the
leftmost
eight bits of the variable will not be overwritten on output
and the variable will not contain the
data
you
expect.
Table
6-1
lists

the

suggested

input and

output

variable

types.

CALLING CONVENTIONS

Table

Suggested

6-1

Variable

Data

Types

Output Parameter

VAX/VMS Hardware | Input Parameter

Declaration

Type Required

Declaration

Byte

INTEGER,

Word

INTEGER

Appropriate

Longword

INTEGER

Quadword

Properly
dimensioned array

CHAR

packed
record

To store the output produced by a system service,
you
must
allocate
You can do so by declaring
contain the output.
to
sufficient
space
Mailbox
Create
the
example,
For
size.
proper
variables of
the
Thus, you can set
(SYSSCREMBX) system service returns a 2-byte value.
up

storage
TYPE

space
WORD

follows:

0,.465535%

CHAN.TYFE

= FACKED RECORD
BOT.CHAN : WORD
END§

+
.

VAR

CHAN..REC

CHAN.TYFE?S

VALUE CHAN.REC t= (0)y
FUNCTION SYS$CREMEX (ZIMMED

FRMFLG ¢ INTEGER?®
VAR CHAN : CHAN.TYFE;

ZIMMED MAXMSGs BUFQUOy FROMSKy ACMODE § INTEGERS$
ZSTDESCR LOGNAM ¢ FACKED ARRAY [SUB&31 OF CHAR):
INTEGERS

EXTERNS

+
*

MEX.REC
. MEX..INT

t= SYS$CREMBX

(OsyCHAN.RECs

Oy ‘MAILBOX)¥

CALLING

CONVENTIONS

If the output is a quadword value, you must declare an array or record
For example, the Get Time (SYSSGETTIM) system
size.
proper
the
of
Thus, you would
service returns the time as a quadword binary value.

need

to specify the

following:

TYFE QUAD = ARRAY

VAR

ISTAT

INTEGERS$

QUADs

SYSTIM

[1..21

INTEGER?

FUNCTION SYS$GETTIM

(VAR F_SYSTIM

{

QUADD! INTEGER?

EXTERN:

ISTAT

(= SYS$GETTIM

(S8YSTIM)}$

to
you
allow
not
Optional Parameters - VAX-11 PASCAL does
6.2.3.2
If you choose not
from procedure or function calls.
parameters
omit

to supply an optional parameter, you should pass the value zero

using

For example, the Translate
(%$IMMED) mechanism.
wvalue
immediate
the
optional
three
has
service
system
(SYSSTRNLOG)
Name
Logical
do not specify values for these parameters, you
you
If
parameters.
must

include

ISTAT

zeros

in their places,

:= SYSSTRNLOG

('CYGNUS',

follows:

NAMLEN,NAMDES,

0,0,0);

system services
Passing Character Parameters - Some VAX/VMS
6.2.3.3
For example,
require character parameters for either input or output.
a
accepts
service
system
(SYSSTRNLOG)
Name
the Translate Logical
name as input and returns the associated logical name or file
logical

specification,

any,

output.

VAX/VMS system services usually require strings to be passed by string
Specify $STDESCR in the function or procedure declaration
descriptor.
On
descriptor.
string
by
parameters
string
to pass the required
packed array of characters must be
or
<constant
character
a
input,
passed to the

are

required:

string and

(2)

the output

system service by descriptor.

(1)

On output,

an INTEGER variable, which is set to the actual

string.

For

example:

TYFPE POS_WORD = 0,.65535%
SURG3

1..635

WORD_TYPE = FPACKED RECORD
SHORTWD : FOS.WORD
ENDvy

STRING..RUF = FACKED ARRAY [1.,.128]1 OF CHAR?
VAR

two parameters

packed array of characters to hold the output

ICODE : INTEGERS
NAMLEN : WORD.TYFES

NAMDES

STRING.EBUF

6-11

1length

CALLING

FROCEDURE

CONVENTIONS

ERROR?$

FUNCTION

SYS$TRNLOG

(XZSTDESCR

VAR

CYGNUS

RSLLEN

PACKED

ARRAY

LSUR&63]

INTEGER)

CHARS

WORD.TYFES$

ZSTDESCR RSLEBUF { STRING.RUF$
ZIMMED TABLEy ACMODE, DSEMSK ¢

INTEGERS

BEGIN
L4
+
*

ICODE = SYS$TRNLOG (“CYGNUS’»
IF NOT ODD (ICODE) THEN ERRORS3

NAMLENsNAMDES»0y0s4)3

The

logical name CYGNUS is translated to its associated name
or
file
specification,
and
the
output values (length and associated name or

file

specification)

NAMLEN

causes

and

the

system

table;

only

Section

6.4

6.3

the

VAX-11

can

call

VAX-11

disable

system

Run-Time

and

the

Library

last

tables

complete

the

are

system

you

parameter,

process

specified
with

1logical

service

example.

provides

mathematical

These

Reference

procedures

procedures
are

SEED.VAL

¢!

that

described

vyou
the

Manual.

can invoke a Run-Time Library procedure from a PASCAL
defining
it
as
an
external
function and including the
function reference.
For example:

RANDL.RSLT

name

searched.

You

VAR

value

PROCEDURES

programs.

Library

locations

The

group

LIBRARY

PASCAL

Run-Time

its

another

RUN-TIME

from

stored

respectively.

presents

CALLING

The

are

NAMDES,

program
by
appropriate

INTEGERS

REALS

+
+

FUNCTION

MTH$RANDOM

(SEED

INTEGER)

REAL$?

EXTERNj

RAND_.RSLT

This
example
(MTH$RANDOM)
.
When

defining

note

the

uses

the

function

uniform

for

pseudorandom

Run-Time

Library

number

generator

procedure,

you

should

following:

The

the

MTH$RANDOM(SEED.VAL)?

mechanism

which

(by-immediate-value,
types

appropriate

pseudorandom

reference

and

the

number

result

each

by-reference,
for

the

generator,

real

parameter

parameters
the

number,

seed

passed

by-descriptor)
and

the

parameter

shown.

result
is

passed

EXTERN?

CALLING CONVENTIONS

6.4

COMPLETE

SYSTEM SERVICE

EXAMPLE

This section presents a sample
PASCAL
procedure
that
declares
and
calls
the
Get
Job/Process
Information (SYSSGETJPI) system service.
The procedure declares SYSSGETJPI as an INTEGER function, so that upon
return, it can check for successful completion.

SYSSGETJPI is used here to get the process identification and name
of
the
current process.
When used for this purpose, SYS$SGETJPI requires
information in only
the
fourth
of
seven
parameters.
The
first,
second, third, fifth, sixth, and seventh parameters must be null.
The
fourth parameter contains the address of a list
of
descriptors
that
describe
the
specific
information requested and point to buffers to
receive the information.
1In this example, the list is constructed
as
a
record.
It contains fields corresponding to each required item as
noted in the description of SYS$GETJPI in the VAX/VMS System
Services
Reference

Manual.

FROCEDURE GETJFINFO

(VAR NAME!

FRCNAM?

VAR

ID!

(XThis rrocedure calls the SYS$SGETJIFI

INTEGER)#

sustem service to

get the process ID and rrocess names which are
used as formal rarameters.
It uses these tures
and

variables?

WORD —- 16 bits to contain buffer lensgth and recuest code
I.0.

FTR_FPID -~ Address of rrocess

FTR.PIDLEN -~ Address of srocess I.Il. lensth
FTR.FRCNAM ~- Address of rsrocess name string

FTR.FPRCNAMLEN ~- Address of length of srocess name string
JPIREL -~ Record containing item list sarameter
ICODE -- Status

returned by SYS$GETIFI function

The following ture is declared in the main srodram?
FRONAM = PACKED ARRAY [1..1%5%1 OF CHARX)
CONST

TYPE

NULL

WORD

0..655353

RECJ

RECORD

FIDINFO

PACKED RECORD
FIDLENy JFIS.FID

WORD

ENDS

PTR..FID

FTROFIDLEN
FRCONAMINFO

TINTEGERS
¢ "INTEGERS
¢ PACKED RECORD

FRONAMLENy

JPIS._FRCNAM

WORD

ENDS

FTR.FRCNAM ¢ "FRUNAMS
FTR.FRCNAMLEN ¢ "INTEGERS
ENDLIST

INTEGER

END¢

VAR

TICODE
JRIREC

:
¢

INTEGERS:
RECJS

FUNCTION SYS$GETJFPI

(ZIMMED A»EBsC

INTEGER?

VAR ITMLST

YIMMED XsYe?Z ¢ INTEGER)Y ¢ INTEGERS
(KAYEByC andgd XsY¥s?Z asre rull rarametersi)

RECJS

EXTERNS

CALLING

CONVENTIONS

BEGIN
(XSet

record rarameterx)

WITH

JRPIREC

REGIN
FPIDINFO.FPIDLEN

¢=

NEW

(FTR.PID)S

NEW

(FTR.FIDLEN)S

FTR..FIDTM

§¢=

FIDINFO.JFIS_PID

(XLength

AX3193%

¢ e

FTR.FIDLENTM

(kZero
03

P seme
2=

FRCNAMINFO.
JFPI$ . FRCNAM

154
3

variablex)

address

(XHex
(kGet

(FPTR.FPRCNAMLEN)

of JPI$_.FRUNAMX)
address of nameX)

PTR.FRCNAMLENTM 3= 03

(¥XZero

name

lendgth

ENDLIST

(XList

must

end

and

returrn

SYS$GETJFI

status

ICODE

ODD ICODRE
BEGIN

THEN

(NULLs

WRITELN

(‘Error

STATUS

NULLs

NULLS

GETJPI

JFIRECY

rrocess’)$

HALT
END

ELSE

(XIf

NAME

InD
END

3
END

JFIREC.FPTR.FPRCNAMTMS

JPIREC.PTR.FIDTM

variableXx)

with

NULLs

OX)

NULLY

errory

(Xrrint
(Xand

BEGIN

lendgthX)

variableX)

(XIf
in

stringX)

address

function

NOT

rname

(XGet

* o

(XCall

lendgthX)
variableX)
bufferXx)

‘%

ENDS

(¥Zero ID length
{(XLength of name

(kRlank-fill

NEW

bufferXx)

AX31CH

(FPTR.FRCNAM)
§

PTR-FRCNAMTM

of JPI$.FIDX)
address of 1DX)

(XGet

FRCNAMINFO.FRONAMLEN

NEW

CkHes:
(XGet

NULL)
#

error

messadeX)

haltXx)

successful
k)

(kassidn rname to NAME
(Xand assidn id to ID

raramX)
raramX)

CHAPTER 7
ERROR

PROCESSING AND CONDITION

HANDLERS

During
the
execution
of
your
VAX-11
PASCAL
program,
various
conditions,
including
errors
and
exceptions,
<can
occur.
These
conditions can result from errors during I/0 operations, invalid input
data,
incorrect
calls
to library routines, errors in arithmetic, or
system-detected errors.
VAX-11 PASCAL provides two methods
of
error
control

and

recovery:

Run-Time

VAX-11

Library default

Condition

condition

error-processing procedures

Handling

Facility

(including

handlers)

user-written
.

These error-processing methods are complementary and can
be
wused
in
the
same program.
Thus, you have two options in dealing with errors.
You can allow the Run-Time Library to provide all
condition
handling
for
vyou.
The
Run-Time Library provides default error processing by
generating error messages for all error or exception
conditions
that
occur during the execution of a PASCAL program.
Section 7.1 describes
error processing by the Run-Time
Library
as
it
applies
to
VAX-11
PASCAL.
Appendix A describes the error messages that can be generated
for

PASCAL

program.

At the lowest level, the VAX-11 Condition Handling
Facility
provides
all
condition handling for the Run-Time Library.
You can also use it
directly to provide procedures (user-written condition
handlers)
for
processing conditions that occur during your program's execution.
The
use of condition handlers, however, requires considerable
programming
experience
and
should not be undertaken by novice users.
You should
understand the condition handling descriptions in the
VAX/VMS
System
Services
Reference
Manual,
the
VAX-11
Run-Time
Library Reference
Manual, and the VAX-11 Architecture Handbook
before
vyou
attempt
to
write

7.1

condition

RUN-TIME

handler.

LIBRARY DEFAULT

ERROR

PROCESSING

By default, the Run-Time
Library
prints
a
message
and
terminates
PASCAL
program execution when an error occurs.
These default actions
occur

unless

Run-time

your

program

errors are

$PAS-F-code,

includes

reported

the

condition handler.

following

format:

text

The code is an abbreviation of the error message text, which
explains
the
error.
VAX-11 PASCAL run-time errors and recovery procedures are
described in Appendix A.
Most
Run-Time
Library
procedures
provide
their
own error messages, as described in the VAX-11] Run-Time Library
Reference

Manual.

ERROR PROCESSING AND CONDITION HANDLERS

7.2

OVERVIEW OF VAX-11 CONDITION HANDLING

When the VAX/VMS system creates
a
user
process,
it
establishes
a
system-defined
condition
handler
which,
1in
the
absence
of
any
user-written condition handler, processes
errors
that
occur
during

execution
of
the
wuser
image.
Thus,
by default, a run-time error
causes the system-defined
condition
handler
to
print
one
of
the
standard
error messages and to terminate or continue execution of the
image, depending on the severity code associated with the error.

When a condition is signaled, the
system
searches
for
a
condition
handler
to
respond.
The
system
conducts
the
search
by calling
handlers in preceding stack frames
wuntil
it
activates
a
condition
handler
that
does
not
resignal.
The
default
handler
calls the
system's message output routine so that it can
send
the
appropriate

message
to the user.
Messages are sent to the SYSSOUTPUT file and to
the SYSSERROR file, if both
files
are
present.
If
the
condition
permits,
program execution continues.
Otherwise, the default handler
forces program termination and the condition value becomes the program
exit

status.

You can create and establish your own condition handlers according
to
the
needs
of
your
applications.
For
example, you can create and
display messages that
specifically
describe
conditions
encountered
during
execution of an application program, instead of relying on the
standard

7.2.1

system

error

messages.

Condition Signals

A
condition
signal
<consists
of
a
call
to
one
of
the
system-supplied
signal
procedures,
LIBSSIGNAL
and
LIBSSTOP.
signal procedures must be declared external, for example:
PROCEDURE

LIBSSIGNAL

PROCEDURE

LIBS$SSTOP

(%IMMED

CONDITION

You signal a condition when you do

not

INTEGER);

want

two
The

EXTERN;

handle

the

routine
in
which
it
was
detected,
but
want
instead
to
pass
notification to other active routines.
1If the current
procedure
can
continue
after
the
signal is made, call LIB$SIGNAL.
A higher-level

procedure can then determine whether program execution is to continue.
If
the
condition
will
not allow the current procedure to continue,
call

LIBSSTOP.

Condition values are usually

expressed

condition

symbols.

For

example:

LIBSSIGNAL

You

<can

(MTHS_FLOOVEMAT);

include

information

about

additional
the

parameters

When called, a signal procedure searches
for
examining
the
preceding
stack frames until
handler

that

does

not

provide

supplementary

error.

resignal.

condition
handlers
by
it activates a condition

ERROR

7.2.2

Handler

PROCESSING AND CONDITION

Responses

A condition handler responds
action in three major areas:
@

Condition

correction

Condition

reporting

Condition

control

First,
If
it

the handler
<can,
the

continues.

1If

the

handler

process

exception

cannot

condition.
the

<correct

the

That

is,

count

Resignaling the same condition
to your terminal or log file

®¢

Changing

the

resignaling

can

exceptions

severity

the

taking

the

that

handler

another

may

condition

encountered

field

involve

send

the

one

during

program

appropriate

condition

message

wvalue

and

condition

©Signaling a different
message oriented to a
be

<condition,

requests

handlers

Maintaining
execution

can

exception.

Execution

<condition

Condition reporting performed
the following actions:

affected

condition, for
example,
specific application
a

number

ways.

Among

was

them

©produce

are:

Continuing from the
a call to LIBSSTOP,

Unwinding to the establisher at the point
of
the
call
that
resulted
1in
the
exception.
The
handler can determine the
function value returned by the called procedure.

Unwinding to the
establisher's
<caller
(the
procedure
called
the
procedure
that
established
the
handler).
handler can determine
the
function
wvalue
returned
by

WRITING

signal.
1If the signal
the program will exit.

called

7.3

determines whether it can
correct
the
<condition.
handler takes the appropriate action, and execution

resignal

HANDLERS

issued

through

that
The
the

procedure.

A CONDITION

HANDLER

The following sections describe how to code and establish a
condition
handler,
and
provide
some
simple
examples.
See Appendix C of the
VAX-11 Architecture Handbook, the VAX-11l
Run-Time
Library
Reference
Manual,
details

and
the
VAX/VMS
§System
on condition handlers.

Services

Reference

Manual

for

ERROR PROCESSING AND CONDITION HANDLERS

7.3.1

Establishing and Removing Handlers

When
a
procedure
is
called,
no
condition
handler
is
initially
established.
To
use a condition handler, you must first define both
the Run-Time Library procedure LIBSESTABLISH and the condition handler
as

follows:

FUNCTION

ERR_HANDLER

PROCEDURE LIBSESTABLISH

INTEGER;

EXTERN;

($IMMED FUNCTION HANDLER :

;
INTEGER)

EXTERN;

To establish the handler, you must call LIBSESTABLISH.

remove

established handler, define the Run-Time Library procedure LIBSREVERT,
and

call

PROCEDURE

follows:

LIBSREVERT;

EXTERN;

LIBSREVERT;

As a

result of this call,

current

procedure

the condition

activation

1is

handler

removed.

the condition handler established by the

established

When a procedure

procedure

1is

1in

the

returns,

automatically

removed.

Note that condition handlers written in PASCAL can access
only
own local data and data declared at program or module level.

7.3.2

Parameters

their

for Condition Handlers

A PASCAL condition handler is an INTEGER function that is called
when
an
exception
condition
occurs.
You
must
define
two
formal VAR
parameters

for

An integer array to refer to the parameter list from the call
That is,
procedure (the signal parameters).
signal
the
to
or
LIBSSIGNAL
to
calls
the list of parameters included in
LIBSSTOP

example,
TYPE

(see Section

7.2.1).

An integer array
to
refer
to
information
concerning
the
procedure
activation
that established the condition handler
(the

For

condition handler:

mechanism arrays).

you can define a

condition handler as

MECHARR = ARRAYLO..41
SIGARR = ARRAY [0..91

OF
OF

follows:

INTEGER?
INTEGER?

FUNCTION ERR.HANDLER (VAR SIGARGS ! SIGARR?
VAR MECHARGS : MECHARR) § INTEGER?
REGIN
*

ENDy

ERROR

PROCESSING AND CONDITION

FROCEDURE LIB$ESTABLISH

LIBSESTARLISH

(ZIMMED

HANDLERS

FUNCTION HANDLER

INTEGER)$#

EXTERNS

(ERR_HANDLER)
§

END
.

The

array

SIGARGS

receives

the

values

listed

below

from

the

signal

being

passed

procedure.

Value

Meaning

SIGARGS[0]

Indicates how many parameters are
in this array (parameter count).

SIGARGS[1]

Indicates

the

(condition
discussion

SIGARGS[2

condition

being

value).
See
Section
of condition values.

signaled
7.3.4

for

Indicates
the
call

optional parameters
as
specified
by
to LIBSSIGNAL or LIBSSTOP;
note that
the
dimension
bounds
for
the
SIGARGS
array
should
specify
as many entries as necessary to
refer to the optional parameters.

The array MECHARGS receives information about the procedure activation
status
of
the procedure that established the condition handler.
The
values

from

this

procedure

are

listed

below.

Value

Meaning

MECHARGS[0]

Specifies

the

number

parameters

this

array

(4).

MECHARGS[1]

Contains the address of the procedure activation
stack frame that established the handler.

MECHARGS[ 2]

Contains the number of
(that
1is,
the
stack

procedure activation,
the

MECHARGS [ 3]

Contains
the

MECHARGS[4]

7.3.3

Handler

exception

listed

the point

been made
from
the

which

occurred.

value

the

time

value

the

time

signal.

Contains the
the signal.

Function Return Values

Condition
handlers
specify
subsequent
execution.
The
are

the

calls that have
frame
depth)

below.

function
function

return
values
return values and

to
their

control
effects

ERROR PROCESSING AND CONDITION HANDLERS

Value

Effect

SSS_CONTINUE

Continues execution from
the
signal.
signal
was
issued
by
a
call
to
however, the program exits.

If
the
LIBSSTOP,

Resignals to continue the search for a condition

SSS$_RESIGNAL

handler to process

the condition.

In addition, a condition handler can request a stack unwind by calling

SYSSUNWIND before returning.

FUNCTION SYSSUNWIND
INTEGER;

‘

Declare SYSSUNWIND as follows:

(%$IMMED DEPTH

INTEGER;

$%IMMED NEWPC :

INTEGER):

EXTERN;

When SYS$UNWIND is called, the function return value is ignored.
The
Rl in the mechanism
and
RO
registers
saved
the
modifies
handler
parameters to specify the called procedure's function value.

A stack

unwind

can be made

resulted

in the
:=

the

Specify:

exception.

SYSSUNWIND

two places:

point

call

the

that

(MECHARGS[2],0);

Unwind to the procedure that called the establisher.
STATUS

7.3.4

one of

Unwind to the establisher,
STATUS

SYSSUNWIND

Specify:

(0,0);

Condition Values and Symbols

has
procedure
VAX-11 uses condition values to indicate that a called
report exception
to
and
failed,
successfully or
executed
either
(usually
records
packed
32-bit
are
values
Condition
conditions.
interpreted 'as
integers),
consisting
of fields that indicate which
was
value
the
reason
the
value,
the
generated
system component
A condition value has
the severity of the condition.
and
generated,
the

form:
31

28 27

Field

Bits

31:28

Facility

27:16

Message

15:3

16 15

FACILITY

MESSAGE

SEV

Meaning
Control

bits

Identifies

the

software

component

wvalue.
condition
the
generated
facility.
indicates a customer
indicates a DIGITAL facility.

that

27 =1
Bit
0
=
27
Bit

Describes the condition that occurred.
Bit
15
1is specific to a
message
the
indicates
1
=
a
indicates
0
=
15
Bit
facility.
single
system-wide

message.

ERROR

Sev

2:0

CONDITION

HANDLERS

Meaning
Specifies

severity

warning

success

error

code,

Bits

AND

NWNOHO

Field

PROCESSING

information

severe

reserved

follows:

error

A warning severity code (0) indicates that output
was
produced,
but
that
the
results
might not be what you expected.
An error severity
code (2) indicates that output was produced even though an
error
was
detected.
Execution
can
continue,
but the results will not all be
correct.
A severe error code (4) indicates that the error was of such
severity
that output was not produced.
A condition handler can alter
the
or

The

severity code of
force an exit,

a condition value to allow execution
depending on the circumstances.

condition

SIGARGS.

value

passed

Occasionally,

your

the

second

<condition

element

handler

may

continue

the

require

particular condition be identified by an exact match.
That
bit
of the condition value (31:0) must match the specified

array

that

1is,
each
condition.

For

example, you may want to process
a
floating
overflow
condition
only
1if
its
severity
code
1is
still
4
(that
1is, if no previous
condition handler has changed the severity code).
As noted
above,
a
typical
<condition handler response is to change the severity code and
resignal.

In many cases, however, you
regardless
of the value of
and control
fields
of
a
LIBSMATCH_COND function, as
FUNCTION

may
want
to
respond
to
a
condition,
the severity code.
To ignore the severity
condition
wvalue,
declare
and
call
the
follows:

LIBSMATCH_COND

(CONDVAL,COMPVAL

INTEGER):

BOOLEAN; EXTERN;

LIBSMATCH_COND(SIGARGS[1],PASS_ERRACCFIL)
THEN...

7.3.5

Floating-Point

Operation

Some conditions involving floating-point
operations
require
special
action
1if
you
want to continue execution.
Operations that involve,
for example, floating overflow, dividing by 0, or computing the square
root
of
a
negative number store a unique result known as a floating
reserved

operand.

subsequent

floating-point

this
result,
a
hardware
reserved
operand
signaled.
This can continue indefinitely
if
does
not change the reserved operand, because
each time the computation is retried.

operation

accesses

fault
1is generated and
the
condition
handler
the operand is accessed

ERROR

PROCESSING AND CONDITION

HANDLERS

To allow computation to continue, you must change the reserved operand
by defining and calling the Run-Time Library routine LIBSFIXUP_FLT, as
follows:

FUNCTION

LIBSFIXUP
FLT

(VAR

VAR MECHADR

MECHVECTOR;

INTEGER;

STATUS:=

SIGADR

SIGVECTOR;

VAR OP

DOUBLE)
:

EXTERN;

LIBSFIXUP_FLT(SIGARGS,MECHARGS,NEWOPERAND);

The types SIGVECTOR and MECHVECTOR in this example refer to the
types
of
the signal and mechanism argument vectors, which are assumed to be
defined in the calling procedure.
Note that you
should
specify
the
third parameter as a double-precision variable.
This ensures that the
reserved
operand
will
be
changed
correctly
regardless
of
its
precision.
If
you
do
not have a special value for this parameter,
specify 0.0.
For more information on LIBSFIXUP_FLT,
see
the
VAX-11
Run-Time

7.4

The

Library Reference Manual.

CONDITION

example

HANDLER

this

EXAMPLES

section

illustrates

how

declare

and

wuse

condition
handler
with a typical PASCAL procedure.
It establishes a
condition handler that is called
when
an
error
occurs
in
a
file

opening
procedure.
opening/creating the
unwind is performed.

If
the
error
is
the
general
"Error
file," signified by the PASS_ERROPECRE
code,
an
Any other error is resignaled.

PROCEDURE OPENHAND}
(XThis erocedure shows how to establish and call a
condition
3 PASCAL rrodsram.
It uses these tures!?
DATAREC ~- a2 record ture for the accounting file

handler

from

DATAFILE -—SIGARR

—--

MECHARR -Arnd

file

ture with comrorents

arraw

ture

an array

for

signal

ture for

mechasnism

it declares these dglobal variables?
FLAG ~—- a Boolean set to TRUE when
NEW_ACCOUNTS

STATUS

all-rurrose

CONST

ZINCLUDE

(XThis
TYFE

file

FLAG

DATAREC

rarameters

an error occurs
DATAFILE

tuyre

return

status

variable

the

PASCAL

declarations

the condition

RECORD
PACKED ARRAY
¢ REALS

COST

REAL3
PACKED

¢
{

ARRAY

[1..301

OF CHARj}

[1..113

OF CHAR

= FILE OF DATARECS
ARRAY [0..9]1 OF INTEGERS$

MECHARR = ARRAY
VAR

function

NAME :
AMOUNT
ESTE
Ds
DATAFILE
SIGARR =

twure

‘SYS$LIBRARY!SIGDEF.PAS’

file contains

DATAREC

varisble

rarameters

[0..41

BOOLEANS?

NEW.ACCOUNTS ¢ DATAFILES
STATUS ¢ INTEGERS

INTEGERS

slsnals*)

ERROR

PROCESSING

AND

CONDITION

HANDLERS

PROCEDURE LIBSESTABLISH (ZIMMED FUNCTION FIXER ¢ INTEGER)S EXTERNS
(XThe sustem service rrocedure LIBSESTABLISH will be called to
establish the condition handlerx)
PROCEDURE OPENER (VAR ACCOUNTS ¢ DATAFILE);
(XThis erocedure will be called to oren the file
records

the

it.

slso

OPEN processing
(XLocal

rerforms

shown

dats

here

for

entry

and

and write
cleanuer.

new
Only

simplicituk)

variable declarationsx)

FUNCTION HANDLER

(VAR SIGARGS ¢ SIGARR? VAR MECHARGS { MECHARR)
!
INTEGERS$
(XThis function will be called to handle 2 file orening error during
the OPENER rrocedure.
It uses only dloballuw declared variablesg»
excert for the SYSSUNWIND and LIBSMATCH.COND functions.X)

FUNCTION

SYSSUNWIND

INTEGERS
FUNCTION

(DEPTH

LIBSMATCH.COND

ROOLEAN?

INTEGER$

ZIMMED NEWPC

INTEGER)!}

EXTERNS$
(CONDVAL»

COMFVAL

INTEGER)$

EXTERN$

BEGIN

END$

(LIBSMATCH._.COND (SIGARGSL11y PAS$_ERROPECRE)) THEN
(XIf error orening filesX)
BEGIN
FLAG := TRUE?
(kset file error flad€x)
STATUS $= SYSSUNWIND(MECHARGSL21+1s 0)
(kand unwindX)
END$
HANDLER (= SS$_RESIGNAL
(XIf some other errorsy resignalXx)
(Xend HANDLERX)

BEGIN

LIBSESTABLISH (HANDLER)$} (Xestablish condition handlerx)
OPEN (ACCOUNTS» ‘LDATAJACCOUNTS.DAT’»0LDs SEQUENTIAL)?
(XData entryr

END}
BEGIN
(XThis

the
FLAG

storadger

and cleanurX)

(XEnd OPENERX)

start

the

outermost

(=FALSE}

OPENER

(kinitialize

(NEW._ACCOUNTS)
3

IF FLAG THEN WRITELN

srocedurex)

(‘Error

(¥oren

Flag

to FALSE

for

test

belowk)

fileX)

in orening file’)
(XFrint

END#

the

(XEnd OPENHANDX)

7-9

messade

error handler

was

calledX)

CHAPTER

VAX-11

This

chapter

system

and

describes
the

PASCAL

the

VAX-11

SYSTEM ENVIRONMENT

relationship
PASCAL

between

compiler.

the

VAX/VMS

covers

operating

the

following

topics:
e

Use

Storage

scalar

Storage

unpacked

Storage

packed

Representation

8.1

USE

The

VAX-11

program

sections
and

pointer

structured

types
types

types

floating-point

data

PROGRAM SECTIONS
PASCAL

sections,

compiler
to

store

uses

contiguous

information

Linker
controls
memory
allocation
and
attributes
of
each
program
section.
program

section

areas

about

the

memory,

program.

sharing
Table 8-1

called

The

according
lists the

VAX-11

to
the
possible

attributes.

Table

Program

8-1

Section

Attributes

Attribute

Meaning

PIC/NOPIC

Position

independent

CON/OVR

Concatenated

REL/ABS

Relocatable

GBL/LCL

Global

SHR/NOSHR

Shareable

EXE/NOEXE

Executable

RD/NORD

Readable

WRT/NOWRT

Writeable

overlaid

absolute

local

scope

nonshareable

position

nonexecutable

nonreadable

nonwriteable

dependent

VAX-11

PASCAL SYSTEM ENVIRONMENT

VAX-11 PASCAL implicitly
declares
three
program
sections:
S$SGLBL,
SCODE,
and
S$PDATA.
Table 8-2 summarizes the usage and attributes of
these

program sections.

Table

8-2

Program Section Usage and Attributes

Program

Section

Name

Usage

SGLBL

Read/write static data
declared at module or
program

Attributes
PIC, OVR, REL, GBL,
NOSHR, NOEXE, RD, WRT

level

SCODE

Read-only generated
executable code

PIC,
SHR,

CON,
EXE,

SPDATA

Read-only
that need

pPI1IC,
SHR,

CON, REL, LCL,
NOEXE, RD, NOWRT

constants
storage

REL, LCL,
RD, NOWRT

Each
module
in
your
PASCAL
program
is
named
according
to
the
identifier specified in the program or module header.
You can use the
module name to qualify the program section name in LINK commands.
For
more information, refer to the VAX-1ll Linker Reference Manual.
When the
1linker
constructs
an
executable
image
into sections.
sections

executable
image,
it
divides
the
Each image section contains program

that have the same attributes.

allocation

arranging

The

1linker

controls

memory

image sections according to program section

attributes.

The linker allows you to use special options to change program section
attributes
and
to influence the memory allocation in the image.
You
include these options in an
options
file,
which
is
input
to
the
linker.
The
options and the file are described in the VAX-1ll Linker
Reference

8.2

Manual.

STORAGE OF SCALAR AND POINTER TYPES

When not part of a packed structure,
allocated

storage

space as

summarized

the scalar types
in Table

8-3.

PASCAL

are

Variables of scalar types, with the exception of DOUBLE variables, are
aligned
on a boundary corresponding to their sizes.
DOUBLE variables
are aligned on longword boundaries rather than on quadword boundaries.
Variables of subrange types are allocated and aligned in the same
way
as
variables of the associated scalar types.
For example, an integer
subrange variable is allocated
one
1longword
and
1is
aligned
on
a
an enumerated type DAYS_OF_ WEEK
of
subrange
A
boundary.
longword
(with values SUNDAY, MONDAY, TUESDAY, and so on) is stored in one byte
and

aligned

A pointer

byte

boundary.

simply a

longword

containing

an address.

VAX-11

PASCAL SYSTEM

Table

Storage

Scalar

ENVIRONMENT

8-3

and

Pointer

Types

Type

Storage

Allocation

Alignment

Character

bits

byte)

Byte

Boolean

bits

byte)

Byte

Integer,

single,

real | 32

bits

longword) | Longword

Double

bits

quadword) | Longword

Enumerated

8 bits (1 byte) if
type contains 256
elements or less;
16 bits (1 word) if
type contains more
than

Pointer

8.3

STORAGE

The unpacked
and
aligned
only to data

array

For

bits

Byte if type contains
256 elements or less;
word if type contains
more than 256 elements

elements

longword) | Longword

STRUCTURED

TYPES

structured types (sets, arrays, and records)
are
as
described below.
Note that this description
items that are not part of another structure.

A set consists
boundary.
An

UNPACKED

256

Boundary

stored

example,

bytes

and

each

aligned

element

longword and aligned on
of
a
character
array
boundary.

1longwords)

according

aligned

the

array of

type

its

integers

stored
applies

longword

elements.

stored

in a

longword boundary.
Similarly, each
element
1is
stored
1in
a
byte and aligned on a byte

Records

The
the
on

are stored field by field according to the type of each field.
type of the first field in the record establishes the alignment of
entire record.
Subsequent fields in the record are always aligned
byte
boundaries,
regardless of the type of the first field.
For

example:
VAR

RECORD

INTEGER

;

BOOLEAN

;

INTEGER

END;

Record

aligned

longword

contains
an integer value, which
shows how this record is stored.
Bytes

which

through
is

contains

through

(bits

integer

through

1longword

the Boolean value of
71) contain the other

boundary

31)

stored

contain

value.

and
integer

because

Byte

its

first

longword.

the
4

first

(bits

bytes
5
through
longword value, Z.

field,

Figure

32
8

8-1

field,

through

39)

(bits

VAX-11

PASCAL

SYSTEM

ENVIRONMENT

7 //////////////////;/;///////////////71 z((?NO::ss:) ]
Figure

8.4

STORAGE

PACKED

Although

you can
for
sets,
storage.

only
their

In general,
the "32-bit

8-1

Storage

Sample

STRUCTURED TYPES

pack

any
arrays,

structured type, packing saves storage space
and records.
Packing files has no effect on

storage space for packed
rule," as follows:

Any data item that is 32
as few bits as possible.

Any data
possible

item over
number of

types

bits

descriptions

apply

less

allocated

length

according

packed

into

32 bits
1long
1is
allocated
the
smallest
bytes and is aligned on a byte boundary.

The subsections below describe exactly how
each
packed
type and note any exceptions
The

Record

only

data

items

storage
1is
allocated
to
to the general 32-bit rule.
that are not part of another

structure.

8.4.1

Storage

Packed

Sets

A packed set that is not a component of another
packed
structure
byte-aligned.
Each
packed
set
1is allocated space according to
32-bit rule, based on the ordinal value of its largest element:
e

If the ordinal value is less than or equal to 31, the
set
is
allocated
the
number of bits equal to the ordinal value plus
one.
For example, a packed set of 2..19 is allocated 20 bits.

If the ordinal value is greater than 31, the set is
allocated
the
number
of
bits
equal
to
the
ordinal value plus one,
rounded up to the
nearest
byte
boundary.
For
example,
a
packed
set
of
subrange
type 100..101 is allocated 13 bytes
(101+1 bits rounded up to a byte boundary).

Since the size of a
values
from
0
to
longwords)

8.4.2

array

aligned
on
the nearest
packed

set is
255,

1limited
packed

to
set

256
elements,
with
can occupy at most 32

ordinal
bytes (8

memory.

Storage of

A packed

1is
the

Packed Arrays

that

not

component

another

a byte boundary.
The elements
bit.
Table 8-4 lists the space

arrays.

packed

of the array
requirements

structure
are
for

packed to
elements

VAX-11 PASCAL SYSTEM ENVIRONMENT
Table

8-4

Storage of Packed Array Elements

Storage

Type

Allocation

Boolean

1l bit

Character

8 bits

(1 byte)

32 bits

(1 longword)

Double

64 bits

(1 quadword)

Subrange of integer,
character, or

Minimum number of bits in which the
largest and smallest possible values

Enumerated types

Number of bits required for
ordinal

value

Pointers

32 bits

(1 longword)

All structured types

Same

Integer,

real,

single

can be expressed

enumerated type

structured

types

1largest

not

However, if the total
packed array.
is
type
structured
the
of
size
array
the
bits,
32
than
greater
minimum
a
allocated
1is
element
number of bytes, that is, the 32-bit
types
Structured
applies.
rule
32 bits or less space are
requiring
bit-aligned.

into the minimum amount of
Note that integer subranges are packed
to hold the largest or smallest value, whichever needs
space needed
OF
[1..10]
For example, each element of PACKED ARRAY
more space.
[1l..64]
ARRAY
PACKED
of
element
Each
bits.
8
allocated
is
-128..127
OF 0..7

is allocated

3 bits.

Enumerated types are packed into the number of bits required to hold
For example, an enumerated type with 16
largest ordinal value.
the
through
0
values is allocated 4 bits, because its ordinal values are
15.

A packed array of an unpacked structured type saves no storage

space.

of such a specification is to byte-align the array.
effect
only
The
The
type.
structured
packed
a
Instead, specify a packed array of
following two examples illustrate this difference.
Examples

TYPE INT_SET = SET OF 0..14;
PACKED ARRAY [1..5]
VAR INT_ARR :

INT SET;

1longwords.
8
as
is stored
INT SET
type
An unpacked set of
Consequently, each element of INT_ ARR requires 8 longwords, for a
total of

longwords

(640

bits)

space.

VAX-11

TYPE
VAR

INT_SET
INT_ARR

packed

set

INT_ARR

the

entire

Storage

for

allocated

=
:

PASCAL

PACKED

SET

PACKED

ARRAY

type

SYSTEM

0..14;
[1..5]

INT_SET

therefore

ENVIRONMENT

requires

INT SET;

allocated

bits,

bits.

total

for

Each
of

element
bits

for

records

array.

packed

arrays

records

and

arrays

packed

similarly.

Multidimensional
arrays
can
also
be
packed.
As
for
the
other
structured
types,
you must specify packing at the innermost level to
gain any significant space advantage.
For a 2-dimensional
array,
an
array

array,

packed
as

array

the

generally
following

takes

less

space

than

packed

array

examples.

Examples
1.

TYPE

INTERNAL
ARR =

VAR

SAMP1_ARR

Each

element

longword.

longwords
INTERNAL

Each

ARRAY

PACKED

array

element

enough

ARR.

The

longwords (800 bits).
2.

VAR

SAMP1 ARR

VAR

SAMP1

ARR

[1..5]

ARRAY

type

entire

PACKED

INTERNAL ARR;

INTERNAL

space

array

ARRAY

[1..5]

0..6;

ARR

SAMP1 ARR, in

storage

ARRAY

[1..5]

for

SAMPl1 ARR

[1..5,1..5]
OF

PACKED

turn,

stored

elements

therefore

ARRAY

requires
of

type

occupies

0..6;
[1..5]

0..6;

Specifying PACKED for an array with multiple
subscripts
results
in packing only at the innermost level.
Therefore, the two array
declarations
in
this
example
are
equivalent.
Each
PACKED
ARRAY[1..5]

0..6

are elements of
even
16
bits.
bits.
Example
allocating this
3.

TYPE
VAR

INTERNAL
ARR =
SAMP2_ARR

requires

bits.

Because

the

packed

arrays

an unpacked array, their size is rounded up to
The total size of each SAMP1_ARR is therefore
3
shows
a
slightly
more
efficient
way
array.

PACKED

ARRAY

[l1..5]

[l..5]

an
80
of

0..6;

INTERNAL_ ARR;

In this example, each element of
INTERNAL ARR
requires
only
3
bits
because the array is packed.
Each element of SAMP2 ARR can
be stored in 15 bits, and the entire array occupies 75 bits.
4.

TYPE

INTERNAL ARR
SAMP3_ARR

VAR

This

SAMPLE

example

PACKED

shows

ARRAY

how

ARRAY

you

[1..5]

can

[1..5]

[1l..5]
OF

0..6;

INTERNAL

ARR;

SAMP3 ARR;

maximize

space

savings

for

arrays

of
more than two dimensions by specifying PACKED at every level.
As in Example 3, each element of INTERNAL
ARR
requires
3
bits,
and each element of SAMP3 ARR requires 15 bits.
The entire array
SAMPLE, then, requires 375 bits.

8.4.3

Storage of

A packed
aligned

record
on
a

allocated

space

Packed

that
byte

Records

is not a component of another packed structure is
boundary.
The
fields
within
the
record are

depending

their

sizes

and

types.

VAX-11

Fields

scalar

packed

the

aligned

unpacked

Except
set,

types,

nearest

byte

PASCAL

less

bit.

boundary

SYSTEM ENVIRONMENT

than or

field

and

equal

that

requires

allocated

space

bits

1long,

than

for

are

bits

field

record.

for

its

alignment,

record

unpacked

record.

the

type
VAR

the

occupies

pack

field

SAMPLEl

such

PACKED

field
the

that
same

contains
amount

field,

packed.

you

For

wunpacked

space

must

array.

packed

explicitly

declare

example:

RECORD

=128

BOOLEAN;

..127;

ARRAY

[1..5]

0..30

END;

This

record

Field

byte-aligned
Storage

storage

follows:

bit

160

integer

bits

subrange,

amount
of
space,
7
data
takes up one bit, leaving 7

which

longword)
Figure

allocated

8 bits

array,

Allocation

Field

and

for

8-2

1is

allocated

each

shows

1is

bits
bits

how

stored

1in

the

smallest

plus a sign bit.
unused.
Field

storage

for

Field
C
is

integers,

B, a
an

possible
Boolean,
unpacked

bits

specifies

elements.

this

record

stored.

ci1l]

unused

Cl2]

cl1]

Ci3]

Cl2]

Cl4]

CI[3]

Cisl

I B I

Cl4]
C[5]
175

Figure

This

record

requires

Compare

the

packed

array.

VAR

preceding

SAMPLE2

8-2

total

Storage

example

PACKED

176

with

Sample

bits

(11

the

Record

words)
one,

-128..127;

BOOLEAN;

PACKED

ARRAY

[1..5]

storage.

which

RECORD

END;

167

0..30

VAX-11

This

record

PASCAL

is byte-aligned

Field

Storage

and

SYSTEM

ENVIRONMENT

allocated

storage

follows:

Allocation
8

bits

1 bit

Fields A and B are allocated the same amount of space in
both
sample
records.
Field
C,
however,
requires
much
less
space in SAMPLE2
because it is packed.
Each element of the
packed
array
C
occupies
only 5 bits.
Figure 8-3 shows how this record is stored.

C[5]

ci4]

CI3]

987

cl2]

Cl1]

C[5]
33

Figure

8-3

Storage

This

record

requires

8.5

REPRESENTATION

Sample

total

Packed

Record

bits

FLOATING-POINT

Containing

Packed Array

storage.

DATA

The
following
sections
summarize
the
1internal
representation
of
single-precision
(REAL and SINGLE types) and double-precision (DOUBLE
type) floating-point numbers.
For more detailed information, see
the
VAX-11

Architecture

8.5.1

Single-Precision Floating-Point

Handbook.

single-precision

contigueus
as shown in

bytes.
Figure

floating-point

The
8-4.

bits

are

15 14

Data

value

numbered

(SINGLE,
1is

EXPONENT

represented

from the

REAL Types)

right

four

through

31,

FRACTION

Figure

8-4

Single-Precision Floating-Point Data Representation

single-precision

floating-point

value

specified

its

address

the
address
of
the byte containing bit 0.
The form of the value is
sign magnitude with bit 15 the sign bit, bits 14 through 7
an
excess
128
binary
exponent,
and
bits
6
through
0
and
31 through 16 a
normalized
24-bit
fraction
with
the
redundant
most
significant

VAX-11

fraction

bit

not

significance

PASCAL

represented.
from

SYSTEM

Within

through

ENVIRONMENT

the

and

The 8-bit exponent
field
encodes
the
exponent
wvalue
of
0,
with
a
sign
floating-point value has a value of 0.

255 indicate binary exponents
of 0, with
a
sign
bit
of
Floating-point
instructions
reserved

The

operand

value

floating

point

through

single-precision

value

8.5.2

-127
is

bits

increasing

wvalues
bit
of
Exponent

0
through
255.
An
0, indicates that the
values of
1
through

through +127.
taken
as
a

processing

An exponent value
reserved
operand.
reserved
operand
take a

fault.

.29% (10**-38)
decimal

of
1

fraction,
through

number

the

1.7*%(10*%*38).

approximate

The

approximately

one

range

precision

part

2**23,

digits.

Double~Precision Floating-Point

double-precision

floating-point

contiguous

bytes.

The

Figure

8-5.

shown

bits

are

wvalue

numbered

1514

Data

(DOUBLE

1is
from

represented
the

EXPONENT

Type)

right

eight

through

63,

FRACTION

FRACTION
FRACTION

FRACTION

63
Figure

8-5

Double-~Precision

double-precision

the
address
of
double-precision

Floating-Point

floating-point

value

the
byte
containing
floating-point
value

Data

specified
bit
is

single-precision
floating-point
value
except
low-significance
fraction
bits.
Within
the
increasing
significance are numbered 48 through
through

The

31,

exponent

and

through

conventions

#
Representation

its

address A,

0.
The
form
of
identical
to

a
a

for
an additional
fraction,
bits
63, 32 through 47,

32
of
16

and

approximate

range

values

are

the

same

for
double-precision
floating-point
wvalues
as for single-precision
floating-point
wvalues.
The
precision
of
a
double-precision
floating-point value is approximately one part in 2*¥*55, or 16 decimal
digits.

APPENDIX
DIAGNOSTIC

MESSAGES

This appendix summarizes the error messages that can be generated by a
PASCAL program at compile time and at

A.1

run time.

COMPILER DIAGNOSTICS

listing
source
VAX-11 PASCAL reports compile-time diagnostics in the
(in
terminal
the
them on
summarizes
and
created)
is
one
(if
interactive
mode)
or
in
the
batch
1log
file
(in
batch
mode)
.
Compile-time diagnostics are preceded by the following:
$PAS-1-DIAGN

The 1 represents the severity level
indicates

warning-level

the

error which will

error.

level

not prevent your

program

from linking or executing.
A level of F indicates a fatal error which
you must correct for your program to link and execute properly.
The diagnostic messages that the PASCAL compiler can print are
below.

All messages are printed with both number and text.

with numbers less than 400
must
correct
for
©proper

you
that
errors
syntax
indicate serious
compilation.
Messages with numbers greater

than 400 indicate the use of
compiler

listed

Messages

VAX-11

PASCAL

extensions

and

illegal

options.

Error

in simple

type

The declaration for a base type of a set or the index type of
array

contains

syntax

error.

Identifier expected

The statement syntax requires an

identifier,

but

can

PROGRAM

none

found.

PROGRAM or MODULE expected

The statement

syntax

requires

the

reserved

word

MODULE.

')'

expected

The statement syntax requires the right-parenthesis character.

':'

expected

The

statement

Illegal

The

colon,

The

statement

'('

expected

Error

contains an illegal symbol,
illegal character.

'['

expected

The

statement

']'

expected

The

statement

END

expected

';'

expected

The

statement

Integer

The

such

misspelled

list

list contains a syntax error,
or semicolon character.

such

syntax

requires

the

reserved

syntax

requires

the

left-parenthesis

syntax

requires

syntax

requires

the

left

syntax

requires

the

right

The compiler cannot find
of a compound statement,

character.

missing

word

OF.

character.

type

The statement
is present.

colon

expected

The statement
10

requires

parameter

comma,

MESSAGES

symbol

The statement
reserved word

Error

syntax

DIAGNOSTIC

syntax

the

data

type,

delimiter

subprogram,

but

square

type

bracket

square

END,

identifier

character.

bracket

which

marks

character.

the

end

program.

requires

the

semicolon

requires

integer,

character.

expected

statement

syntax

label.

for

example,

DIAGNOSTIC

'='

MESSAGES

expected

The statement syntax
requires
the
equal
constant
identifier
from a constant value
identifier from a type definition.

BEGIN

beginning

find

the

executable

delimiter

BEGIN,

marks

the

expected

'..'

Error

The

field

','

expected

The

statement

list

syntax

record declaration contains

requires

Empty parameter

(successive

The

1list

parameter

Illegal

which

required

field-list

(nonprintable)

The program contains

comma

',')

attempts

parameter,
or
contains
omit optional parameters.

which

section.

The compiler cannot
£find
the
..
symbol,
between the endpoints of the subrange.

to
separate
a
separate a type

expected

The compiler cannot

sign
or to

ASCII

syntax

error.

character.

not allowed

specify

null

extra

comma.

PASCAL,

missing

you

cannot

printable

ASCII

character

a character

that

not

character.

Error

in constant

A constant

contains

':='

expected

The

statement

THEN

expected

syntax

The compiler cannot
IF-THEN

UNTIL

illegal

requires

character

the

assignment

is‘improperly

formed.

operator.

find the

reserved word THEN to

complete

the

find the

reserved word UNTIL to complete

the

statement.

expected

The compiler cannot
REPEAT

statement,

DIAGNOSTIC

expected

The

compiler

cannot

statement

TO/DOWNTO

expected

the

The

compiler

FOR

statement.

Invalid

the

reserved

word

complete

the

FOR

word

DOWNTO

the

statement.

find

the

reserved

expression

statement

the

compiler

Error

find

WHILE

cannot

The

MESSAGES

syntax
finds

requires

not

expression,

legal

element

record

but

the

first

symbol

expression.

variable

reference

array

field

contains

syntax

error.

ARRAY

The

expected

compiler

cannot

find

the

reserved

word

ARRAY

the

type

definition.

Strings

excess

65535

Relational operators
65535 bytes.

Parameter

count

The number
255,

end

program

100

Array
A

101

cannot

exceeds

255

parameters

End of input encountered
aborted.
The

the

had

size

input

been

too

array

1is

bits

for

You

declared

identifier
<can

sections.

applied

strings

procedure

function

was

end

comparisons
1longer

cannot

program.

encountered

than

exceed

Compilation

before

entire

large

declared

allowed

before

file

not

parsed.

2,147,483,647

Identifier

characters

1larger

than

2,147,483,647

bytes

array.

twice

declared

redeclare

packed

twice

identifiers

within

only

declaration
different

section.

declaration

DIAGNOSTIC

102

MESSAGES

Lowbound exceeds highbound

The lower limit of a subrange is greater than the upper limit
of
the subrange, based on their ordinal values in their base type.

103

Identifier

is not of appropriate class

The identifier names the wrong class of data.
For
example,
it
names
a
constant
where
the syntax of the statement requires a
procedure.

104

Identifier not declared
The

105

107

program uses

Sign not

minus

nonnumeric

type.

sign

has

Incompatible

subrange

types

The subrange

that has not

been declared.

types are

not

occurred

compatible

before

expression

according to the

rules

compatibility.

File not allowed

A file

109

identifier

allowed

A plus or

type

108

in variant part

type cannot appear

in the variant part of a

record.

Type must not be REAL or DOUBLE

You cannot specify a real value here.
Real values cannot be used
as
array subscripts, control values for FOR loops, tag fields of

variant records,

110

elements of set expressions,

subrange

types.

Tagfield

type must be

The

tag

field

for

scalar or

a variant

boundaries

subrange

record must

scalar

subrange

type.

111

Incompatible with

tagfield

The case label and the
tag
These
two
items
must
be
compatibility

112

Index

type

field
are
compatible

of
incompatible
types.
according to the general

rules.

type must not be REAL or DOUBLE

Array subscripts cannot be
integer

integer

subrange

real values;
values.

if numeric,

they must be

DIAGNOSTIC

113

Index

type must

Array
of

114

subscripts
structured

Base

The

type

base

must

scalar

must

MESSAGES

subrange

scalar

subrange

values,

and

cannot

the

type.

not

type of

REAL

this

set

DOUBLE

subrange

cannot

one

real

types.

115

Base

The

type must

base

values,

116

Actual

The

type of
and

this

cannot

parameter

actual

scalar

set

subrange

structured

must

set

parameter

must

scalar

subrange

type.

correct

size

when

passed

VAR

parameter.

117

Undefined
The

118

119

base

VALUE

forward
type

pointer

was

initialization must

type declaration:

<name)>

not

TYPE

defined

initialization statement can
block;
vyou cannot initialize

Forward

declared;

cannot

Function

repetition

repeat

declaration

result

the

section.

in main program

A VALUE
program

You

120

reference

appear
only
variables in

parameter

list

1list

not

after

in
the
main
subprograms.

allowed
the

type must

scalar,

subrange,

pointer

The function specifies a result that is not a
scalar,
or pointer type.
Function results cannot be structured

121

File value
A

122

file

Forward

parameter not

cannot

declared

The
result
declaration
can

123

appear

Missing

of
and
only

result

The

function

the

function.

forward

subprogram.

passed

allowed
as

function;

the
in
the

type

heading

value

parameter.

repetition

function
the later

subrange,
types.

forward

appears
complete

result

type

not

allowed

in
both
the
forward
declaration.
The result

declaration.

function declaration

does

not

declare

the

type

the

result

DIAGNOSTIC MESSAGES

124

F-format

for REAL and DOUBLE only

You can specify two integers in the field width
for

125

real,

single,

correct

predeclared

not

function

the

Number of parameters does not agree with declaration
different

subprogram.

parameters

passed

subprogram

the

from the number of formal parameters declared for that
You cannot add or omit parameters.

Parameter cannot be element of a packed structure

pass
you

cannot
You
subprogram;
use

128

type.

The number of actual

127

R:3:2)

Error in type of predeclared function parameter

A parameter passed to

126

(such

and double values only.

a
to
structure
packed
a
of
element
one
pass the entire structure if you want to
must

it.

Result type of actual function

parameter

does

not

agree

with

the

type

declaration

The result of an actual function parameter is
specified in the

129

not

Operands are of incompatible types

Two or more of the operands in an expression are of

types.
and a

130

formal parameter list.

For

incompatible

example, the program attempted to compare a numeric

character variable.

Expression

is not of set type

The operators you specified are valid only for set expressions.
131

Type of variable

is not set

The statement syntax requires a set variable.

132

Strict

inclusion not allowed

inclusion.
to test set
You must use the <= and >= operators
allow you to use the less than (<) and greater
not
PASCAL does
than

133

(>)

signs.

File comparison not allowed

Relational operators cannot be applied to file variables.

DIAGNOSTIC

134

Illegal
You

type

cannot

specified

135

136

138

perform

This

operation

Set

element must

The

elements

operand

have

must

be
a

types

The

elements

square

Index

operation

set

that

bracket

is
or

Type

not

variable

A period

141

Type

Illegal

subrange

scalar

structured

set

not

are

not

all

appears

comma

array

inside

not

must

parameter

Loop control
scalar

the

same

type

square

followed

variable

file

The

specified

this

left

compatible

with

the

type

that

not

record

type.

pointer

after

the

name

variable

that

substitution

variable must

type.

declaration

compatible

unstructured,

variable.

type

Sets

brackets.

control variable in a FOR loop must be
subrange,
or
user-defined
scalar
type;

Illegal

types.

record

following

variable

the

array

compatible with

subrange

The

144

types.

The type of an actual parameter is not
of the corresponding formal parameter.

143

items

operand.

A circumflex character appears
is not a file or pointer.

142

data

compatible

not
a

Boolean

must
of

not

The specified array subscript is not
specified in the array declaration.

140

Boolean

scalar

this

variable

type

elements

element

specified

requires

Set

Type

the

types.

A variable

139

operand(s)

Type

cannot

137

MESSAGES

with

the

type

non-floating

point

integer,

cannot

integer

real

expression

position.

evaluates

type

that

incompatible

DIAGNOSTIC

145

Type

conflict

between control

MESSAGES

variable

and

loop bounds

The type of the control variable in a FOR
with the type of the bounds you specified.

146

Assignment of
You

assign one file to
give values to files.

type of a case label is
selecting
expression

expressions

Subrange
You can
specify

149

Index

must

bounds

must

type must

not

Output

procedures

incompatible with the type
evaluates.
Case labels and

compatible

must

Assignment

this

types.

Assignment

You

assign

cannot

value

formal
a

types

only.

You

cannot

integer

function
a

to
which
selecting

scalar

type of a nondynamic
an integer subrange.

You cannot assign
identifier.

151

another.

specify subranges
of
scalar
a real or string subrange.

The index
it can be

150

incompatible

Label types incompatible with sélecting expression
The
the

148

1is

allowed

cannot

used

147

files not

1loop

array

not

cannot

integer,

although

predeclared

function

allowed

external

function parameter
value

the

name

not
of

allowed
a

formal

function

parameter,

152

153

such

field

this

You attempted
field name.

access

Error

count

The

number
option.

154

Type of

exceeds
of

errors

record
a

record

error

limit.

exceeds

parameter must be

The actual parameter
an integer.

30,

incorrect

nonexistent

Compilation aborted
the

limit

set

the

ERROR_LIMIT

integer

passed

this

function

procedure

must

DIAGNOSTIC

155

Recursive

$INCLUDE

not

allowed.

The

$INCLUDE directive
directive appears.

156

Multidefined case
The

157

same

case

label

can

Case

label

MESSAGES

cannot

Compilation aborted
1include

a call
specified

label
used

refers

only

once

range exceeds

which

Double,

NEW

than

within

real

162

Each

case

smallest

case

one

statement.

CASE

statement.

1000
the

largest

and

cannot

tagfields
real

declaration was
of

procedure

Type

The

type

record

allowed

string

variables,

but

must

forward

already

procedure or
or

not

procedure

already

never

were

the

function

that

was

not

block

was

illegal.

The

Undeclared

not

has

subprogram has

constants

refers.

Procedure/function

163

the

than

DISPOSE,
more
tagfield
number of nested variants in

string

The
reiteration
forward-declared

161

which

the

the pointer

Tag fields
scalar.

160

Missing corresponding variant declaration
In

159

file

label

The range of ordinal values between
labels must not exceed 1000.

158

the

been

forward

function:

function

was

forward

declared

declared.

<name>

forward-declared

but

its

declared.

parameter

subprogram

must

requires

real

integer

real

integer

expression

parameter.
14

164

This
The

procedure/function cannot
specified

actual

predeclared

parameter.

you

an actual

procedure
must

directly.

A-10

use

or
it

parameter

function
in

the

cannot

subprogram,

call

DIAGNOSTIC MESSAGES

165

Multidefined label

A label appears in front of more than one statement in
executable

166

single

section.

Multideclared label

The program declares the same label more than once.
167

Undeclared label

The program contains a label that has not been declared.
168

Undefined label:

<label>

The program defines a label, but does not use the
executable

169

label

the

between

and

section.

Set element value must not exceed 255

The ordinal value of an element of a set must be
255.

170

Value parameter expected

A subprogram that is passed as an actual parameter can have
value

171

Type of variable must be textfile

file

The program heading specifies an external file that has not

been

variable

subprogram

operation

operand

parameter.

requires

Undeclared external file

declared at program or module

173

(FILE OF CHAR)

text

The specified

172

only

parameters.

level.

Negative set elements not allowed

The value of an integer set element must be between 0 and 255.
174

Type of parameter must be file

The specified subprogram requires a file as a parameter.
175

INPUT not declared as an external file

file variable
The program makes an implicit reference to the
redeclared
been
has
or
declared,
not
either
is
but INPUT
INPUT,
at

inner

level.

A-11

DIAGNOSTIC

176

MESSAGES

OUTPUT not declared as an external

file

The program makes an implicit
reference
to
the
file
wvariable
OUTPUT,
but OUTPUT is either not declared or has been redeclared
at

177

inner

Assignment to

Assignment
function

178

level,

function identifier not allowed here

to a

function

Multidefined

File

You

181

value
variant.

type not

bounds
to

appears

that

has

array

once

the

file

type.

are

too

large

allow

the

elements

the

correctly.

The

expression must specify

are

not

scalar

value;

structured variables

legal.

$IMMED,

$DESCR,

External

procedure

Formal

$STDESCR

most

Formal

has

allowed

only

procedure

name

as main program

names

must

name
value

that

most

is defined

Illegal

dynamic

passed

for

can

have

parameters

formal

parameter

have

dynamic array parameters

parameter.

parameter

to 'a

procedure

assignment

involving

array assignment

The program attempts
dynamic

only

parameters.

procedures may not

itself

external

unique.

You cannot pass a dynamic array as a
that

allowed
EXTERN,

same

procedures may have

A procedure

187

the

Expression must be scalar

Program and

186

than

components

These 'extended
parameter
specifiers
are
procedures and functions which are declared

185

within

allowed

file

accessed

procedure/function

184

only

Array bounds too large

array

183

file

field
record

cannot declare

The

182

is allowed

record variant

A constant
tag
definition of a

179

identifier

block.

to perform an

arrays.

A-12

illegal

DIAGNOSTIC MESSAGES

188

Parameter must be scalar and not real or double

The parameters to the predeclared functions SUCC and PRED must be
scalar types, and cannot be one of the real types.

189

Actual parameter must be a variable

When you use VAR with a formal parameter, the corresponding
actual parameter must be a variable and not a general expression.
190

READLN/WRITELN/PAGE are defined only for textfiles

The predeclared procedures READLN, WRITELN, and PAGE operate only
on

191

text

files.

READ/WRITE require input/output parameter list
The READ and WRITE procedures require
you cannot omit the parameter list.

192

least

one

parameter;

Illegal type of input/output parameter

Arrays, sets, records, and pointers cannot be parameters
READ

193

and WRITE

the

procedures.

Field width parameter must be of type INTEGER

The field width you specify must be an integer.
194

Variable must be of type PACKED ARRAY[l..11l]

OF CHAR

The DATE and TIME procedures require a parameter of this type.
195

Type of variable must be pointer

The statement syntax requires a variable of pointer type.
196

Type of variable does not agree with tagfield type

The type of a variable in a tag value list is
the

197

tag

field

with

(single-

Type of parameter must be REAL or DOUBLE

The

statement

double-precision)

198

incompatible

type.

syntax

value.

requires

real

Type of parameter must be DOUBLE

The statement syntax requires a double-precision value.

A-13

DIAGNOSTIC

199

Parameter must
The

procedure

be
or

numeric

function

MESSAGES

type

requires

integer

and

not

real

number

value.

200

Parameter

The

must

procedure

scalar

function

scalar,
Boolean, integer
or pointer parameter.

201

Error
A

real

202

real

constant:

constant

pointer

requires

subrange,

digit

contains

real

integer,

user-defined

scalar

subrange,

expected

nonnumeric

character

where

numeral

required.

String constant must not exceed
The

end

string.

the

line

Make

occurs

sure

that

source

before

the

second

line
apostrophe

apostrophe

that

has

not

range

closes

been

left

out.

203

204

Integer

constantvexceeds

An integer constant
(that is, -2%*3]1 to

2**31-1),

Actual

not

The

parameter

actual

Zero
You

206

Illegal
An

207

length
cannot

Real

specify

numbers

not
a

the

correct

1is

not

permitted

integers

type

compatible

type

with

the

parameter.

allowed

string

in octal

hexadecimal

double

A
singlepermitted

formal

string

digit

octal

outside

parameter

corresponding

205

range

that

hexadecimal

constant

has

out

characters.

constant

contains

illegal

digit.

range

double-precision
real
number
is
outside
the
-- 0.29*10**(-38)
to
1.7*(10**38) for positive
-0.29*10**(-38)
to
-1.7%(10**38)
for
negative

range

and

numbers.

208

Data

type

This

variable contains

cannot

initialized

type

that

file.

A-14

cannot

initialized,

such

DIAGNOSTIC MESSAGES

209

Variable has been previously initialized

You can specify only one VALUE declaration for a variable.
210

Variable

is not array or record type

The VALUE initialization for a variable that is not a

an array contains

211

record

constructor.

Incorrect number of values for this variable

The VALUE declaration contains too many or too few values for the
variable

212

being

initialized.

Repetition factor must be positive

The optional type identifier must be compatible with the type

The repetition factor
positive

213

integer

array

initialization

must

constant.

Type identifier does not match type of variable

variable

214

integer constant

initialized.

Incorrect type of value element

other
type
A constant appearing in a VALUE initialization has a
of the variable, record field, or array element to be
that
than
initialized.

215

RMS

record size

is out of

range

The record size specified in the OPEN procedure call exceeds

the

maximum,

216

Type OLD is not allowed for this
You cannot

217

specify OLD

for

file

internal file.

$DESCR, %STDESCR not allowed for procedure or function parameters

The only extended mechanism specifier

that

PROCEDURE and FUNCTION parameters is $IMMED.

218

may

applied

Array must be unpacked

An array parameter to PACK or UNPACK is not unpacked correctly.
219

Array must

packed

An array parameter to PACK or UNPACK is not packed correctly.

A-15

DIAGNOSTIC

220

Packed

The

221

bounds

$STDESCR

This

222

not

must

not

the

packed

allowed

mechanism

packed

dynamic

$DESCR

not

This

223

bounds

mechanism
types

$IMMED

not

for

this

specifier

may
CHAR

for

this

specifier

and

allowed

exceed

bounds

the

unpacked

bounds.

type

applied

indexed

only

strings

and

may

this

applied

arrays

only
these

the

predefined

types.

type

This mechanism specifier may be applied only to types that
4 bytes or less, or to PROCEDURE or FUNCTION parameters.

224

$DESCR,

%IMMED,

You

cannot
specifiers

225

combine

with

Too

many

You

can

occurs

251

255

Too

have

with

many

for

VAR

%DESCR,

VAR

scopes

only

each

$%STDESCR,
specifier.

can

nested

This

is generated

error

function

with

Too

many

errors

The

PASCAL

parameters

and

%IMMED

mechanism

same

statement

that

too

expression

1is

not

identifiers
of

nesting.

and/or
no

not

new

nesting

functions
than

allowed

when

source

diagnoses

level

statement.

levels

here.

function

deep.

Probable

name/scope

nested

inside

name.

this

compiler

OPEN

line

only

the

first

errors

line.

Expression

should

the

WITH

nested procedures

The

levels

block

Assignment
conflict

source

259

allowed

an attribute in
the compiler.

nested

Subprograms

252

the

not

occupy

Iilegal file attribute specification
You specified
recognized by

250

3%STDESCR

integer.

type

unpacked

for

unpacked

array

arrays

allowed

scalar

exceed

MESSAGES

each

complicated

too

deeply

separately evaluate

nested.

some

A-16

parts

correct

the

this

error,

expression.

vyou

DIAGNOSTIC MESSAGES

260

Too many nonlocal

labels

The subprogram contains more than 1000 labels that
that is,

at a higher level,

261

not locally declared.

declared

are

Declarations out of order or repeated declaration sections
labels,
Only the

in the following order:
The declarations must be
variables, values, and subprograms.
types,
constants,
main program can contain value declarations.
263

Program segment too large:

branch

displacement

32767

exceeds

bytes

A statement is too large to allow the generation of a branch
Use subprogram calls to break
instruction to span the statement.
the program into smaller units.

300

Division by zero

The program attempts

302

Index expression out of bounds

The

value

subscripts

303

to divide by zero.

the

1is

expression

outside

the

range

the

this array.

is out of bounds

Value to be assigned

The value to the right of the assignment operator is out of range
for the variable to which it is being assigned.
304

Element expression out of

range

The value of the expression is out of range for the array element
to which you are assigning

305

it.

Dimension specification out of range

The

second

dimension

argument

greater

than

LOWER

UPPER

array

parameter

the

number

array

specifies

exceeds

range

of dimensions of the first

argument.

306

Index

type

declaration

dynamic

The index type of the actual dynamic array parameter
beyond the range declared in the formal parameter list.
401

Warning:

extends

1Identifier exceeds nn characters

Identifiers can be any length, but PASCAL scans only the first 15

characters

for

uniqueness.

A-17

DIAGNOSTIC

402

Warning:

Error

A compiler

403

404

Warning:

Source

compiler

the

program.

Warning:

in option specification

option

The

MESSAGES

incorrectly

input

ignores

Duplicate

after

any

specified

"END."

the

source

the

END

that

code.

ignored

characters

external

after

procedure

terminates

name

Two external procedures or functions have been declared with
the
same
name,
They
refer
to
the
same
externally
compiled
subprogram.

405

Warning:
The
a

LABEL Declaration

compiler

ignores

label

in module

ignored

declarations

the

outermost

level

module.

450

Nonstandard

Pascal:

Exponentiation

451

Nonstandard

Pascal:

Value

452

Nonstandard

Pascal:

OTHERWISE

453

Nonstandard Pascal:

$INCLUDE directive

454

Nonstandard

MODULE

declaration

456

Nonstandard Pascal:

'$'

'_'

457

Nonstandard

Pascal:

Dynamic

arrays

458

Nonstandard

Pascal:

$IMMED,

$%DESCR,

459

Nonstandard

Pascal:

Octal

460

Nonstandard

Pascal:

Double

461

Nonstandard

Pascal:

External

procedure

462

Nonstandard

Pascal:

Octal

hexadecimal

463

Nonstandard

Pascal:

Output

464

Nonstandard

Pascal:

Input of

Pascal:

declaration

clause

identifier(s)

$STDESCR

hexadecimal

precision

constant

declaration

data

user-defined

string

A-18

parameter

output

scalar

user-defined

scalar

DIAGNOSTIC MESSAGES

465

Nonstandard Pascal:

Input/output of double precision data

466

Nonstandard Pascal:

Implementation-defined

type,

function,

procedure

A.2

RUN-TIME

ERROR MESSAGES

message
the

run-time,

When an error occurs

aborts execution.

and

VAX-11]

PASCAL

issues

error

The run-time error messages appear in

format:

$PAS~-F~-code,

Text

code

An abbreviation of the message text.
by

this

Messages

are

alphabetized

code.

Text

The

explanation

the

error.

messages
several
Some conditions, particularly I/O errors, may cause
The first message is a general diagnostic specifying
printed.
be
to
the file being accessed (if any) when the error occurred.
Then a more
be issued to clarify the nature of the
may
message
PASCAL
specific
most
In
Finally, a VAX-11 RMS error message may be printed.
error.

be able to understand the error by looking up the
should
you
cases,
VAX/VMS
the
to
If not, refer
first two messages in the list below.
and Recovery Procedures Manual for an explanation of
System Messages
the

VAX-11l

RMS

error

message.

ATTDISINV Attempt to dispose invalid pointer value xxx at PC = XXX

The DISPOSE procedure was called with an illegal parameter value,
You should use the
probably because of an uninitialized pointer.
NEW procedure to correctly allocate the pointer.
CASSELBOU CASE

selector out of bounds at PC = xxXx

not
does
expression
selector
case
the
statement,
In a CASE
case label values, and no OTHERWISE
the
of
one
to
correspond
CHECK
the
only when
This message occurs
clause is specified.
option

ERRACCFIL Error

effect.

in accessing file nnnnnn

This message identifies the file being accessed when an I/O error
occurred.

ERRCLOFIL Error closing

file

an
1is
This
closed.
being
An error occurred while a file was
Please
PASCAL error and should be reported to DIGITAL.
internal
example
submit a Software Performance Report (SPR), including an
program

possible.

A-19

DIAGNOSTIC

ERROPECRE Error opening/creating
An

error

file.

occurred

The

defaults,

the

Error

ERRREWFIL

resetting

program

Error

An error
This
1is

the

was

system

could

This

are

FILNOTCLO Files

not

allocated

not

find

are

the

probably

procedure.
reported to
Report
(SPR),

enough

that

space

too

many

files

closed

allocate

are

open

the

too

file

many

use.

call

OUTPUT cannot
for

opened

type

not

the

predeclared

by user

file

wvariables

INPUT

with

invalid

parameters

to open for direct access (DIRECT)
with variable-length records.

Input

system

conversion

found

Invalid

same

option when

element

open

the

file

type TEXT,

error

erroneous

assignment

dimension.

you

appropriate

The program tried to
another.
For
the
the

used)

the

(or

file

means

OUTPUT

You tried
or a file

each

create

execution of the RESET procedure.
This
error and should be reported to DIGITAL.
Performance Report (SPR),
including
an

not

INPUT and

cannot

FILTYPNOT File

INVASGINC

procedure

file

You can specify only a carriage control
predeclared file variable OUTPUT.

The

open

OUTPUT.

FILOUTINV File

INPCONERR

OPEN

possible.

rewriting

buffer

buffer.

You
and

1in

Please submit
a
Software
Performance
an example program if possible.

FILBUFNOT File

pointers

attempted

occurred during
execution
of
the
REWRITE
an
internal
PASCAL
error
and
should be

DIGITAL.
including

The

system

specified

An error occurred during
is
an
internal
PASCAL
Please submit a Software
example

file

if the
OPEN
procedure
for this file.

incorrect

ERRRESFIL

when

parameters

MESSAGES

input

when

reading

incompatible

text

dynamic

arrays

assign incompatible dynamic
assignment to be legal, the

type

and

This

message

the

same

upper

appears

only

and
if

file.

arrays
to
arrays must

lower
CHECK

bounds
is

xxx

one
have
for

enabled.

DIAGNOSTIC MESSAGES

LINLENEXC Line length exceeded,

The

line

length = xxx

length of an output line was greater than the maximum allowed

Check to be sure that you did
the record size for this file.
by
write
must
If you
procedure.
not omit a call to the WRITELN
size in the OPEN
record
the
increase
1length,
this
of
lines
statement.

LINLIMEXC LINELIMIT exceeded,

LINELIMIT = XXX

the
exceeds
file
specified
the
The number of lines output to
limit.
Make sure that the excessive output was not caused by an
infinite

loop and

increase the

line

limit

if necessary.

OUTCONERR Output conversion error

text
a
to
The program tried to write data of an incorrect type
file.
Make sure that all output values are properly defined.

PROEXCHEA Process exceeds heap maximum size at PC = XXX

The system could not find enough space to allocate storage for
a
pointer
wvariable.
This error is probably caused by an infinite
loop that calls the NEW procedure, thus attempting to allocate an
infinite amount of heap storage.
If the program does not include
an infinite loop, your process may
actually
require
more
heap
1In this case, try
than the maximum process size allows.
storage
to make your program smaller;
1if you
<cannot,
ask
mahager about increasing the maximum process size.

your

system

PROEXCSTA Process exceeds stack maximum size at PC = xxx

The system could not expand the stack to make room for
the
last
procedure
or
function called.
This error is probably caused by
infinite recursion, where a number of
procedures
and
functions
call
each
other
without returning.
Make sure that the program
logic
If the program
does not include this type of logic error.
is
sound,
the
process may actually require more space than the
your
make
to
try
case,
1In this
maximum process size allows.
you cannot, ask your system manager about
if
smaller;
program
increasing

the maximum process

size.

RESREQACC RESET required before accessing

are

open

for

before

trying

You can use the FIND procedure only on files that
input.

RESREQREA RESET required before reading the file

The program did not call the RESET
read

the

file.

A-21

procedure

DIAGNOSTIC

REWREQWRI

REWRITE

required

The program did not
write to the file.

SETASGBOU
The

Set

assignment

program

tried

before

call

out

The

Subrange

writing

REWRITE

procedure

bounds

assign

Make sure that all set
the bounds of the set.
enabled.

SUBASGBOU

the

MESSAGES

trying

xxx
set

variable.
are within
CHECK
is

out

illegal

before

assignments specify values that
This message appears
only
1if

assignment

file

program tried to
assign
variable.
Make sure that all
that are within the bounds of
only if CHECK is enabled.

bounds

value

xxx

an
illegal
value
to
a
subrange
subrange assignments specify values
the subrange.
This message appears

A-22

APPENDIX

CONTENTS OF RUN-TIME STACK DURING PROCEDURE CALLS

Figure B-1 outlines the events that occur during a procedure call
shows the contents of the

run-time stack after each event.

and

CONTENTS OF RUN-TIME STACK DURING PROCEDURE CALLS

0 Before procedure call:
lower addresses

:SP

Stack grows

calling

procedure local
data area

toward lower
addresses

:FP

higher addresses

SP = Stack Pointer
FP = Frame Pointer

9 The calling procedure’s actions:

First:
Decrements SP by 4 times

number of parameters &
stores actual parameters

on stack.
parameter 1
called
procedure

—

parameters

parameter n

calling
procedure

local data
area

Second:
Calculates “static link”

‘

to allow the called procedure
W

to find the stack frame
of its declaring procedure.

Stores static link in R1.

Finally:
Issues CALLS instruction.

Figure

B-1

Contents

Run-Time

Stack

During

Procedure Calls

CONTENTS OF RUN-TIME STACK DURING PROCEDURE CALLS

9 The CALLS instruction’s actions:

Finally:
Creates and pushes
stack frame.

First:

Condition handler

Pushes parameter count &

fills with zero-bytes to a
longword size.

procedure

Second:

called

stack

frame

mask

old AP

Sets new argument pointer

(AP) equal to current SP.

R2-R11 saved

parameter count

as needed

and zeros

PSW

calied

procedure

Third:

parameters

Adjusts stack to a

longword boundary.

calling

procedure

data area

e The called procedure’s actions:
]

First:

Loads base register R11 with the

Third:

address of the program-level
variables.

Allocates space for
local data and

copies value parameters.

Second:

ocal

local AP on stack.

data

I
—

oca

"
Pushes static link (R1) and

local AP

tati

static link

condition handier

& local data

Finally:
Saves SP at -12(FP).

PSW

mask

called
procedure
stack frame

-12(FP)

local AP

—

value parameters

old AP
FP

R2-R11

as needed
.

parameter count
and zeros

calling procedure
parameters

calling procedure
data area
o

Figure

B-1

(Cont.)

——

N ——

Contents of Run-Time Stack During Procedure Calls

INDEX

A
Access,

Communication,

direct

(random),

remote,

5-6

sequential,

5-3,

DECnet, 5-6
interprocess,
network, 5-6

5-4

5-3,
5-4

Actual parameter list, 6-2, 6-3
Address parameters—--see By-reference mechanism
count, 6-1
Argument list, 6-1, 6-6
Arguments to PASCAL subprograms,

Argument

6-6

program section,

task-to-task, 5-6
Compiler error messages,

A-1

Compiler listing, 2-3, 2- 8
Compiler qualifiers, 2-1
Compiling a program, 2-1
Concatenating source files,
Condition handler, 7-1
data accessible to, 7-4
7-2
establishing, 7-4
example of, 7-8

2-7

default,

Array descriptor, 6-4, 6-6
Assigning logical names, 5-2
Attributes,
file, 5-4

5-5

function return
parameters for,
removing, 7-4

8-1

7-5

values,
7-4

request of stack unwind by,

6-6

Bound procedure value,
Bounds

checking,

Buffer

size,

BRIEF qualifier,

2-2

3-2,

7-6
responses,

3-3

5-4

Condition handling, 7-1
Condition signal, 7-2
Condition symbol, 7-2, 7-6

6-2,

By-descriptor mechanism,
6-4

Condition symbol definition

By-immediate-value mechanism,
6-2,

6-9

By-reference mechanism,
6-9

By-reference semantics,
By-value semantics,

6-2

files,

6-2,

6-6,

6-2,

6-3

1-1

6-1
Library proce-

6-12

Calling standard, procedure,
Calling system services, 6-6
Calls, procedure, 6-1
CALLS instruction, 6-1, B-2
CARRIAGE attribute, 5-5
Carriage control, 5-4

Character parameters,
6-11

CHECK qualifier, 2-2,
Checking, bounds, 2-2
Commands,
ASSIGN,

EDIT,
LINK,

PASCAL,

6-1

2-10

CROSS REFERENCE qualifier
(Tinker), 3-2, 3-3

D
Data representation,
floating-point, 8-8

DEBUG qualifier

2-6

DEBUG qualifier
DEBUG qualifier

2-3,

2-6

3-4,

4-2

2-3,

Debugger,

Command line qualifiers,

2-2,

(execution), 4-1
(linker), 3-2,
3-4,

4-1,

4-2,

Declaring Run-Time Library procedures,

2-1

RUN, 1-1, 4-1
SHOW CALLS, 4-3

(compiler),

4-3

4- 4
3- 1

1-1,

2-3,

CROSS REFERENCE qualifier
(compiler), 2-2, 2-3, 2-6,

passing,

5-2

1-1,
1-1,

listing,

2-12

2-10,

CALL instruction, 6-1
Calling conventions, PASCAL,
dures,

6-6

Condition value, 6-7, 7-2
format, 7-6
Count field, 5-4
Creating and executing a program,

Cross-reference

C
Calling Run-Time

7-3

system—-defined, 7-2
user written, 7-2, 7-3

6-12

Declaring system services,
6-7,

2-4

6-9

DECnet communications,

Index-1

5-6

6-6,

INDEX

Descriptor,
Device,

6-4,

6-2,

6-6

Files,
concatenating

1-2

Diagnostic messages,
compiler, A-1 to A-19
format of compiler, A-1l

6-6,
Filetype,

format of run-time, A-19
run time, A-19 to A-22
Direct access, 5-3, 5-4
DIRECT attribute, 5-4, 5-5
Directory, 1-3
Divide by zero,

format,

8-9

command, 1l-1
Entry mask, 6-6
Environment, PASCAL system,
Environment pointer, 6-6

Format,

8-1

carriage

control, 5-4
compiler listing, 2-8
condition value, 7-6
double-precision data, 8-9
file specification, 1-2
floating point, 8-8
listing file, 2-8
OPEN procedure, 5-4
record, 5-3, 5-4
single-precision data, 8-8
FULL qualifier, 3-2, 3-3

Error,

compiler, A-1
correction, 4-1
default processing, 7-1
messages, A-1 through A-22
A-1l

processing, 7-1, A-1
run time, 7-1, A-19

2-2,

2-3

Establish a condition handler,

Function,
arguments to, 6-6
as parameters, 6-5

7-4

Executable image, 3-3
EXECUTABLE qualifier, 3-2, 3-3
Executing a program, 1-1, 4-1
Extensions
2-4

External

standard

subprograms,

declaring

reserved

PASCAL,

external,

status, 6-8
return values, 6-5
Run-Time Librar Y , 6-1,
return

7-7 ’

8-9
File,
attributes,
image,

executable, 1-2,
shareable, 3-3

6-7

library, 3-5
listing, 2-2, 2-8
map, 3-3, 3-4
MTHDEF.PAS, 6- 7
object, 2-2, 2 -4,

status,

3-3

$IMMED FUNCTION specifier,

$IMMED mechanism specifier,

6-5
6-3,

6-9

2-6

specifier,

Immediate

6-2,

value,

and

6-3,

function

6-5
6-9

names,

6-5

$INCLUDE directive, 6-7
INCLUDE qualifier, 3-2, 3-5
Input by-reference parameters,

5-4

defaults,

$IMMED PROCEDURE
procedure

organization, 5-3
parameter, 6-3, 6-6
sequential, 5-3
SIGDEF.PAS, 6-7
specification, 1-2
File specification
Filename, 1-3

6-12

Image,

5-2

3-3

LIBDEF.PAS,

6-2

MTHSRANDOM, 6-12
MTHSTANH, 6-2

6-2

operand,

5-4
characteristics,

system service as,

6-7

F
Fault,

5-5

Formal parameter,
function, 6-5, 6-6
list, 6-2, 6-3
procedure, 6-5, 6-6

EDIT

severity code, 7-6
ERROR LIMIT qualifier,

6-7
1-3

Floating-point data,
errors involving, 7-7, 7-8
representation of, 8-8
Floating-point operation, 7-7

numbers,

2-7

FIND procedure, 5-3, 5-5
FIXED attribute, 5-5
Fixed-length records, 5-3,

7-4

Double-precision

source,

condition symbol definition,

1-3

Input/output, 5-1
Interprocess communication,

Index-2

5-5

INDEX

N
LIBSESTABLISH, 7-4
LIBSREVERT, 7-4
LIB$SSIGNAL, 7-2, 7-4
LIBSSTOP, 7-2, 7-3, 7-4
LIBDEF.PAS file, 6-7
Library files, 3-5
LIBRARY qualifier, 3-2,
Library, object-module,

Line length,
size),

LINK
LINK

maximum

Names,

logicai, 5-1, 5-2
program section, 8-2

NEW file attribute, 5-4
NOCARRIAGE attribute, 5- 5
Nonstandard features, 2- 4

3-5
3-5
(buffer

5-4

(o)

command, 1-1, 3-1, 8-2
command qualifiers, 3-2,

input

Object code listing, 2-10, 2-11
Object file, 2-2, 2-4, 2-6
OBJECT qualifier, 2-2, 2-4, 2-6
OLD file attribute, 5-4
OPEN procedure parameters, 5-4
Optional parameters, 6-11
Options--see qualifiers
Output by-reference parameters,

file qualifiers,

3-5

Linking the object modules, 3-1
List, argument, 6-1
LIST carriage control format,
5-5

LIST qualifier,*2-2,
command line, 2-5
file specification

2-3

6-9

for,

2-3

code,

Object

4-2

Linker

floating,

Overflow,

2-3,

files produced by, 2-3, 2-6
specified in source code, 2-6
cross~reference,

2-3,

Parameter lists,
Parameters,

character,

2-10,

2-12

6-9

OPEN procedure,

optional,

Mailbox,

Map

2-11

passing mechanisms,

6-3

by-descriptor,

6-2,

default,

Passing

2-2

6-6
6-4

by-immediate-value, 6-2,
by-reference, 6-2, 6-9

6-3

6-2

parameters,

by default mechanism, 6-2
by-descriptor, 6-2, 6-4
by-immediate value, 6-2, 6-3
by-reference, 6-2, 6-9
to OPEN procedure, 5-4
to PASCAL subprograms, 6-5
to Run-Time Library procedures,

6-2

Messages,

compiler, A-1
run-time, A-19
MTHDEF.PAS file,

subprograms,

Passing mechanisms,

$IMMED FUNCTION, 6-5
$IMMED PROCEDURE, 6-5
$STDESCR, 6-4
VAR,

6-2

PASCAL command, 1-1, 2-1
PASCAL command qualifiers,

arguments passed to,

3-3

MAP qualifier, 3-2, 3-3
Mechanism arrays, 7-4
Mechanism specifier,
default, 6-2
$DESCR, 6-4
$IMMED,

6-5, 6-6

signal, 7-4, 7-5
VAR, 6-2, 6-6, 7-4
PASSINPUT, 5-1
PASSOUTPUT, 5-1

PASCAL

5-5

file,

5-4

6-11

PASCAL subprogram,

Machine code listing, 2-10, 2-11
MACHINE_ CODE qualifier, 2-2,
2-10,

6-11

formal procedure, 6-5, 6-6
input and output by-reference,

Listing file, 2-2
format of, 2-8, 2-9, 2 -10
when produced, 2-3, 2- 6, 2-7
Logical names, 5-1, 5-2

2-6,

6-3

6-2,

7- 5
4
condition handler, 7-4,
formal function, 6-5, 6-6

machine code, 2-10, 2-11
source code, 2-10, 2-11
traceback, 4-1

2-3,

7-4

2-6

Listing,

5-6

Network communications,

6-7

Index-3

6—-12

system services,

6-9

INDEX

Procedure

calling

Procedure

calls,

contents

B-1,
arguments

to,

6-6
6-5

system

external,

Run-Time
Program

stack,

service,

6-1,

development
sections,

names,

7-3
5-5
6-1

process,

1-1,

2-2,

2-1

2-6

2-2,

2-3,

OBJECT, 2-2,

2-3

2-6

2-2,

2-3,

2-4

2-6

Sequential

access,

SEQUENTIAL

attribute,

Sequential

Shareable

files, 5-3
image, 3-3

SHAREABLE

qualifier,

2-4,

SHOW CALLS

command,

WARNINGS,

2-2,

2-4,

22—

SIGDEF.PAS

file,

BRIEF,

3-3

CROSS

3-2,

REFERENCE,

DEBUG, 3-2,
EXECUTABLE,
FULL,

3-2,

3- 5

LIBRARY,

3-2,

3- 5

3-2,

TRACEBACK,

3-2,

3-3

3-2,

3-4,

4-1

command,

source

code,

2-4,

2-6

(direct)

access,

Record

access

Record

formats,

Record

Management

mode,

listing,

qualifiers,

Source

files,

5- 4

size,

maximum,

5-4,

Record

type,

5-4,

Records,

fixed-length, 5-3, 5-5
variable~length, 5-3, 5-4,

arrays,

packed

sets,

packed

records,

pointer

types,

5-5

2-7

2-4,

8-4

types,

8-4,

8-2

arrays,

unpacked

records, 8-3
sets, 8-3

Index-4

8-3

6-4

6-1

arguments

to,

external,

6-2

Subprograms

8-6

8-2

unpacked

Subprograms,

5-5

2-2,

allocation,

unpacked
String descriptor,

5-4

2-11

2-6

concatenating,

qualifier,

packed

scalar

5-5
Record

2-10,

2-6
Storage

5-3,

5-4
Services,

represen-

code

5-3

5-3,

7-5

8-9

code

R
Random

data

Source

2-5

Qualifiers,

7-4,

Source

STANDARD

3-3

7-4
7-3, 7-4

7-2,

condition, 7-2
Single-precision
tation, 8-8,

PASCAL

3-2,
4-3

6-7

parameters,

LIBSSTOP,
Signals,

3-3

SHAREABLE,

Qualifiers,

3-3

3-2,

5- 4

5-4

Signal procedure,
LIBSSIGNAL, 7-2,

3-3

INCLUDE,

MAP,

Signal

3- 1

3-4, 4-1

6-11

5-3,

2-2,

linker,

6-1

7-4
7-8

STANDARD,

Qualifiers,

A-19

2-2

2-3,

CODE,

4-1

7-4

2-6

LIMIT,

7-4
8-9

A-22

LIBSREVERT,

ERROR

7-7,

messages,

MTHSRANDOM, 6-12
MTHSTANH, 6-2
optional parameters,

REFERENCE,

LIST,

1-1,

error

5-7

handler,

fault,

LIBSESTABLISH,
LIBSFIXUP FLT,

compiler,

DEBUG, 2-2,

6-6,

Run-Time Library procedure,
declaring, 6-12

8-1

CROSS

MACHINE

command,

through

6-9

8-1

2-2,

5-4,

Run-time

6-12

Q
CHECK,

6-2,

5-6,

operand

RMS,

8-2

Qualifiers,

by,

condition

Resignal,

RUN

6-2

Library,

attributes,

access,

Routines,

6-9

1-2
Program

Remote

Remove

Reserved

parameters,

declaring

pass

6-9

run-time

6-1

system

Reference,

6-1

B-2

Procedures,
as

standard,

6-6

parameters,

6-5

INDEX

Symbol, condition,
System services,

7-2,

Broadcast (SYS$BRDCST), 6-4
Create Mailbox (SYSSCREMBX),
5-5, 6-7, 6-9, 6-10
declaring as functions,

(SYSSGETTIM),

User-written condition handler,

6-7

declaring as procedures, 6-9
Get Job/Process Information
(SYSSGETJPI), 6-13

Get Time

7-6

6-3,

vV
Value semantics,
Values,

condition, 7-6
function return,

6-11

naming,

6-6

optional
output

parameters,

from,

parameters

6-1 1

6-10

to,

6-9

Translate Logical Name
(SYSSTRNLOG), 6-11
unwind (SYSSUNWIND), 7-6
Wait for Single Event Fl ag
(SYSSWAITFR), 6-3
T

Task~to-task communication, 5~ 6
Text file,
carriage control, 5-5
maximum line length, 5-4
Traceback information, 3-4 [ 4- 1
TRACEBACK qualifier, 3-2, 3_ 4 4
4-1

6-2

signal
VAR

procedure,

parameters,

6-5
7-5

6-2,

6-3,

6-6,

7-4

VARIABLE

attribute, 5-5
Variable-length records,
5-4,

5-5

Version,

1-3

5-3,

W
Warning messages,
A-17

WARNINGS

2-2,

qualifier,

2-4,

2-2,

A-1,

2-4,

2-6

Writing

condition

y4
Zero

Index-5

divide,

7-4

handler,

7-3

VAX-11 PASCAL

User's Guide

AA-H485A-TC
READER'S

NOTE:

COMMENTS

This form is for document comments only.
DIGITAL will
use comments submitted on this form at the company's
discretion.
If you require a written reply and are
eligible to receive one under Software Performance
Report (SPR) service, submit your comments on an SPR
form.

Did you find this manual understandable, usable,

Did you find errors in this manual?

If so,

and well-organized?

specify the error and the

page number.

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

Oodood

Please cut along this line.

Please make suggestions for improvement.
-

Assembly

language

Higher-level

programmer

language programmer

Occasional programmer

(experienced)

User with little programming experience
Student programmer

Other

(please specify)

Name

Date

Organization
Street

City.

State

Zip Code
or

Country

—

Do Not Tear - Fold Here and Tape —

—

_—

—

—_

]

No Postage

Necessary

if Mailed in the |

United States

BUSINESS REPLY MAIL
FIRST CLASS PERMIT NO.33 MAYNARD MASS.

POSTAGE WILL BE PAID BY ADDRESSEE

RT/C SOFTWARE PUBLICATIONS

TWwW/A14

DIGITAL EQUIPMENT CORPORATION
1925 ANDOVER STREET

Do Not Tear - Fold Here

@~ @

—

Alnsvva Maddnd ¥ 2an

—

01876

TEWKSBURY, MASSACHUSETTS