Digital PDFs

DEC-XV-UPUMA-A-D

May 2000

58 pages

Original

16MB

Document:	PATCH XVM Utility Manual
Order Number:	DEC-XV-UPUMA-A-D
Revision:
Pages:	58
Original Filename:	http://bitsavers.org/pdf/dec/pdp15/XVM/DEC-XV-UPUMA-A-D-PATCH_XVM_UTILITY_MANUAL.pdf

OCR Text

PATCHXVM
UTILITY MANUAL
DEC-XV-UPUMA-A-D

digital equipment corporation · maynard. massachusetts

First Printing, December 1975

The information in this document is subject to change without notice
and should not be construed as a commitment by Digital Equipment
Corporation. Digital Equipment Corporation assumes no responsibility
for any errors that may appear in this document.
The software described in this document is furnished under a license
and may be used or copied only in accordance with the terms of such
license.
Digital Equipment Corporation assumes no responsibility for the use or
reliability of its software on equipment that is not supplied by
DIGITAL.

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

The following are trademarks of Digital Equipment Corporation:

DIGITAL
DEC
PDP
DECUS
UNIBUS
COMPUTER LABS
COMTEX
DDT
DECCOMM

DECsystem-lO
DECtape
DIBOL
EDUSYSTEM
FLIP CHIP
FOCAL
INDAC
LAB-8

MASSBUS
OMNIBUS
OS/8
PHA
RSTS
RSX
TYPESET-8
TYPESET-IO
TYPESET-ll

'"- " .

CONTENTS
Page
PREFACE
CHp.PTER

CHAPTER

vii
1

INTRODUCTION

1.1
1.2
1.3
1.4

PATCH PROGRMl
MANUAL ORGAt~IZATION AND USE
MAJOR PATCH FUNCTIONS
PATCH, I/O DATA FLOW PATHS

SELECTION OF ITEM TO BE PATCHED

2.1
2.1.1
2.1. 2

SELECT FUNCTION
Program Select Command
Block Select (E)

LOADING RELOCATADLF PROGRAMS AS SYSTEM
PROGRAHS

3.1
3.2
3.2.1
3.2.2
3.3
3.3.]
3.4
3.5
3.6

PATCH LOAD FUNCTION
PATCH READR LOAD CDrn"lAND
Procedure
Input File Requirements
RFLOCATION
PATCH READR File Relocation
BANK BIT INITIALIZATION (BBI) ROUTINE
READR ESTABLISHED SYSTEJ'.l COMIVlUNICATIOKS AREA
LOADING PATCH-RELOCATED CORE H'lAGE SYSTEJ'.1
PROGRAJ'.1S

LOADING ABSOLUTE PROGRAJ'.'IS AND DATA

4.1
4.2
4.3
4.3.1
4.4
4.4.1
4.4.2
4.5

READ AND PATCH FILE FUNCTIONS
READ LOAD FUNCTION
READ COMMAND
READ Load Operation
WRITING PATCH PROGRAJ'.1S
Partial Overlay Patch Programs
Replacing Complete Programs or Data Blocks
PRECAUTIONS TO BE OBSERVED FOR PATCH READ
OPERATIONS

LISTING AND MODIFYING PROGRAM AND DATA
BLOCK REGISTERS

5.1
5.2
5.2.1
5.2.2
5.2.3
5.3
5.3.1

REGISTER MODIFICATION FUNCTIONS
PATCH LIST FUNCTION
List Operation
Expressions
List Operation, Line Terminator
L LIST COMMAND
Example:
Select, List and Modification
Procedures

iii

1-1
1-1
1-2
1-2

2-1
2-1
2-3

3-1
3-1
3-2
3-2
3-3
..,
~

~)-.j

3-4
3-5
3-6

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

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

CONTENTS (Cont.)

Page
5.4
5.4.1
5.4.2
5.5
5.6
CHAPTER

LR LIS':' COMMAND
LR, Register Modification Procedure
Relocation and LR List Operation Exa~ples
SYSTEr-1 PROGRAN SYSRLK PARN1ETER COMMMJDS
MODIFYING REGISTERS IN DATA BLO~KS

5-6
5-7
5-8
5-10
5-12

SUMMl>.RY, OPERATING PROCEDURES AND ERROR
r~ESSAGES

6.1
6.2
6.2.1
6.2.2
6.3
6.4
6.4.1
6.4.2
6.5
6.6
APPENDIX A
A.l
A.1.1
A.1. 2
APPENDIX B

CONTENTS
PRE-LOAD OPERATIONS
.DAT Slot Assignments
I/O Device Preparation
CALLING PATCH
PATCH OPERATIONS
Replacing Absolute Programs froD External
PATCH Files
Installation of ~elocatacle Programs as
Absolute SYS Files
ERROR DETECTION
ERROR ivlESSAGES

6-1
6-1
6-1
6-2
6-2
6-2
6-2
6-3
6-3
6-3

SYSTEM BLOCK (SYSBLK) DESCRIPTION
SYSTEM BLOCK. (SYSBLK)
SYSBLK Parameter Tahle
Communications Table

P.-]

SYSTEM PROGRAMS, GENERAL DESCRIPTION

B-1

p.-2

A-2

Index-l

INDEX
FIGURES

Page
Figure

1-1
3-1

A-I
A-2.
B-1

PATCH Operations, Information Flow Diagram
Memory Map, Loading a READR-Installed
SYS-file
System Block, Overall Configuration
System Program EDIT SYSBLK Parameter Group
Memory Map, Loading System Programs

1-3
3-7

A-I
A-2
B-1

TABLES
Page
Table

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

PATCH Accessible System Programs
Operators Recognized by PATCH
List Operations, Command String Terminators
Required .DAT Slot Assignments
Procedure for Modifying Registers in
Absolute System Programs
Procedure for Modifying Registers in PATCHREADR-Installed System Programs
Error Messages

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

LIST OF ALL XVM MANUALS

The following is a list of all XVM manuals and their DEC numbers, including the latest version available.

Within this manual, other XVM

manuals are referenced by title only.

Refer to this list for the

DEC numbers of these referenced manuals.

BOSS XVM USER'S MANUAL

DEC-XV-OBUAA-A-D

CHAIN XVM/EXECUTE XVM UTILITY MANUAL

DEC-XV-UCHNA-A-D

DDT XVM UTILITY MANUAL

DEC-XV-UDDTA-A-D

EDIT/EDITVP/EDITVT XVM UTILITY MANUAL

DEC-XV-UETUA-A-D

8TRAN XVM UTILITY MANUAL

DEC-XV-UTRNA-A-D

FOCAL XVM LANGUAGE MANUAL

DEC-XV-LFLGA-A-D

FORTRAN IV XVM LANGUAGE MANUAL

DEC-XV-LF4MA-A-D

FORTRAN IV XVM OPERATING ENVIRONMENT MANUAL

DEC-XV-LF4EA-A-D

LINKING LOADER XVM UTILITY MANUAL

DEC-XV-ULLUA-A-D

MACll XVM ASSEMBLER LANGUAGE MANUAL

DEC-XV-LMLAA-A-D

MACRO XVM ASSEMBLER LANGUAGE MANpAL

DEC-XV-LMALA-A-D

MTDUMP XVM UTILITY MANUAL

DEC-XV-UMTUA-A-D

PATCH XVM UTILITY MANUAL

DEC-XV-UPUMA-A-D

PIP XVM UTILITY MANUAL

DEC-XV-UPPUA-A-D

SGEN XVM UTILITY MANUAL

DEC-XV-USUTA-A-D

SRCCOM XVM UTILITY MANUAL
UPDATE XVMUTILITY MANUAL
VPl5A XVM GRAPHICS SOFTWARE MANUAL

DEC-XV-USRCA-A-D
DEC-XV-UUPDA-A-D
DEC-XV-GVPAA-A-D

VTl5 XVM GRAPHICS SOFTWARE MANUAL

DEC-XV-GVTAA-A-D

XVM/DOS KEYBOARD COMMAND GUIDE
XVM/DOS READER'S GUIDE AND MASTER INDEX

DEC-XV-ODKBA-A-D
DEC-XV-ODGIA-A-D

XVM/DOS SYSTEH MANUAL

DEC-XV-ODSAA-A-D

XVM/DOS USERS MANUAL
XVM/DOS VlA SYSTEM INSTALLATION GUIDE

DEC-XV-Om1AA-A-D
DEC-XV-OD5IA-A-D
DEC-XV-IRS~ffi-A-D

XVM/RSX SYSTEM MANUAL

DEC-XV-XUSMA-A-D

XVM UNICHANNEL SOFTWARE MANUAL

--.. .

PREFACE
This manual describes the operation and use of the PATCH XVM Utility
Program.
NOTE
PATCH XVM (PATCH) should be used only by
the System Manager.
It was assumed in the preparation of this manual that the user is familiar with DIGITAL XVM hardware and operating system.

If there are

any questions in this area, consult the XVM/DOS User's Manual.
The following is a list of XVM documents which either support directly
or contain information useful in understanding PATCH and its functions.
XVM/DOS User's Manual
XVM/DOS Keyboard Command Guide
SGEN XVM Utility Manual
PIP XVM Utility Manual
MACRO XVM Assembler Language Manual
XVM/DOS VIA Syst.em Installation Guide

vii

CHAPTER 1
INTRODUCTION
1.1

PATCH PROGRAM

The PATCH Utility program enables the DIGITp_L XVf,1

(XV~1)

user to view

and modify System Programs and data on specific device data blocks.
The commands provided by PATCH may be entered via the console keyboard
or, in the monitor Batch mode, via either paper tape or card reader
unit.
This manual assumes that PJ..TCP commands have been entered via the
keyboard.

The same commands can be entered under Batch mode: see the

XVH/DOS User's Hanual for details.
1.2

JVlANUAL ORGANIZATION AND USE

This manual is intended for users who are familiar with:
a)

The general operating procedures (i.e., use of equipment and startup procedures), and

the elements, structures and use of the particular
monitor software system in which Ph'J'CH is to be used.

Introductory information and detailed description of each PhTCH function
and their applications are given in Chapters 2 through 5.
Chapter 6 surrmarizes the PATCH startup and operating procedures.
New users of PATCH should familiarize themselves with the contents of
Chapters 1 through 5; thereafter, they

nee~

only refer to Chapter 6

for concise information.
Two Appendices are included which contain supplementary System Block
(SYSBLK) and System Program descriptions.
SYSBLK is described in Appendix A, System Programs are discussed in
Appendix B.

1-1

Introduction

1.3

MAJOR PATCH FUNCTIONS

The major functions for which PATCH may be used are:
a)

the loading of relocatable programs assembled by ~~CRO XVM
into a system as a System Program (described in Chapter 3);

the loading of absolute programs assembled by MACRO XVM and
data onto a System Program area (described in Chapter 4) i

the modification of registers located in
1)

System Programs,

Data blocks,

System SYSBLK (system block) •

The items on which PATCH functions may be performed together with a
description of the commands and procedures needed for their selection
are given in Chapter 2.
1.4

PATCH, I/O DATA FLOVl Pl,THS
,._-..,.

Figure 1-1 illustrates the primary data flow paths employed during
PATCH operations.
User command/response operations are carried out via a console/printer
unit.

Peripheral devices/UFD's containing the files to be patched

must be assigned to .DAT -14; those devices containing auxiliary input
files must be assigned to .DAT -10.

1-2

Introduction

Input Patches &
Patch Programs

CONSOLE TERMINAL

1--------,

MONITOR
Outputs

PATCH
PROGRAM

G·1D~A~T~-]3j-----~::~~~-t~PRINTOUTS

I
I
I

I
I

Commands

.DAT-2

L _______I

Inputs/Outputs

Figure 1-1
PATCH Operations, Information Flow Diagram

1-3

CHAPTER 2
SELECTION OF ITEM TO BE PATCHED
All operations explained in this section are done under control of the
PATCH program.
2.1

SELECT FUNCTION

The first step required in any PATCH procedure is the selection of the
item (.OAT -14) to be patched. T,vo basic commands "Program Select",
and "Block Select" are provided for this purpose.
2.1.1

Proqram Select Command

The System Programs located on the device assigned to .OAT -14 are
selected for PATCl! operations by the entry of a unique name which is
stored in a monitor table; these unique names enable the monitor to
identify the requested program.
The format for a Program Select command is:
(The right angle bracket is the PATCH program user-prompt character.)
>name)
For example, to select the Editor program the entry
>EOIT)
is entered.

The entry of a System Program name merely identifies the

item to be operated upon~ no further action is taken until a PATCH
function is initiated by another command.

NOTE
The Program Select command may be used only if
the device assigned to .OAT -14 contains a System Block (SYSBLK).
In addition to System Programs, system Device System Block (SYSBLK)
and CTRL Q (tQAREA) areas may also

be selected for PATCH operations.

Table 2-1 lists the unique names of the programs normally supplied as
System Programs with the XVM/DOS software system.

2-1

Selection of Item to be Patched

Table 2-1
PATCH Accessible System Programs
PROGRAM
1)

XVM/DOS Resident Monitor

RESMON

System Loader

.SYSLD

3)
4)

Control Q Area

tQAREA

Standard Editor Program

EDIT

Text Editor for VT15 Graphics Display

EDITVT

Text Editor for VP15 Storage Tube Display
Unit

EDITVP

Peripheral Interchange Program

PIP

Macro-Assembler Program for DIGITAL XVM
(MACRO)

MACRO

Cross Reference Program

CREF

10)

System Overlay Build Program

CHAIN

11)

System Generator

SGEN

12)

FORTRAN IV

13)

; Core Dump Program

QFILE

14)

General Dump Program

DUMP

15)

System Patch Program

PATCH

16)

Library UPDATE Program

UPDATE

17)

Source Compare Program

SRCCOM

18)

XVM/DOS non-resident MONITOR

DOS15

19)

Macro Assembler Program for PDP-ll MACll

MACll

20)

Input/Output SPOOLER Program

SPOOL

21)

Magtape File Manipulation Utility

MTDUMP

22)
23)

Spooler Disk-Area Generator
Install Spooler as System Program

1
SPLGEN
1
SPLOAD

24)

BOSS pre-processor

B.PRE

25)

BOSS non-resident monitor

BOSS

26)

PDP-8 to XVM Translator

8TRAN

27)

Install ~ACll as System Program

MCLOAD

lUNICHANNEL-15 systems only.

2-2

1
1

Selection of Item to be Patched

2.1.2

Block Select (B)

A single 4g~8-word data block located on the device assigned to .DAT -14

is selected for PATCH operations, using a "B": Block Select conunand.
is not necessary to select a system program before performing this
operation.

FORMATS:

>B'-In)
or

>B+ ....... n) (DECtape only)
or

>B-L."In) (DECtape only)

where:
11

B identifies the Block Select command

+ indicates the selected block should be read
in the forward direction.

- indicates that the selected block should be
read in the reverse (counter-clockwise) direction.

n represents an octal number which identifies
a logical block on the PATCH I / O device (.DAT -14).
The value of n must be greater than or equal to
~.

DESCRIPTION:
The entry of a "B" select command relocates the identified word block
into a PATCH Block Buffer in core.

In relocating the block, the load

address is set to ~, the size is set to 4~~8' and the block number is
as specified.
If the block number (n) in any of the select commands
is followed by a space, any data on the remainder of the input line is
treated as a comment and is ignored.

For example:

>B L."I l~~ L."ICOMMENT)
is permitted.
The

ILl

operation (refer to Section 5.3)

"B" command.

2-3

is permitted under the

CHAPTER 3
LOADING RELOCATABLE PROGRAMS AS SYSTEM PROGRM1S
3.1

PATCH LOAD FUNCTION

A PATCH load function enables the user to convert a relocatable file
into an absolute SYS file format and load the converted file onto a
system device as a System Program.

This section describes in detail

the PATCH command used, the procedure required, the relocatable file
requirements, and the operations performed by PATCH.
The PATCH load function can be used only when the system on .DAT -14
has been prepared by SGEN to receive the converted file.

During SGEN

procedures (refer to the SGEN XVf1 Utility Manual) the user must enter
the name of each program to be installed, and specify .DAT slot usage.
The program name, number of system blocks required, and .DAT slot usage
information entered during SGEN is stored in the system device SYSBLK
(see appendix A) •
3.2

PATCH READR LOAD CO:1l1AND

The command "READR" initiates the PATCH relocatable-to-SYS file program load function.
>READR
where:

lNOTE:

nnnnn

The general format of this command is as follows:

[ext]
[comments])
...... [nnnnn] ...... [filename] ............
the highest 13-bit core address to be occupied by
the relocated file when it is loaded into core as
a SYS file.
If an address is not specified, the
first free register below the bootstrap is assumed
by PATCH.
If specified, the value of nnnnn must
be ~176368; a larger value will cause an error.

filename

1
the name of the'relocatable file •

ext

filename extension; if not given, DIN is assumed.

comments

if desired, a comment may be added to the command
string, but only if the extension is given
explicitly.

Filename may be omitted for nonfile oriented auxiliary
devices (.DAT -l~).
3-1

Loading Relocatable Programs as System Programs
3.2.1

Procedure

The READR command must be preceded by a program select command which
identifies the system device area (named in SGEN) into which the modified file is to be written.

For example, the sequence:

>STRAN.J
>READR

STRAN..J

must be entered to install the file 8TRAN onto a prepared system device at .DAT-14, given the name STRAN.
On the execution of a READR command, PATCH obtains the named file from
.DAT-l~

~ee

and checks it for required size and program characteristics

paragraph 3.2.2}.

If the file is not acceptable, the load opera-

tion is halted and an appropriate error message is output to the user.
If the file is acceptable, PATCH relocates addresses within the input
file, builds a table of the transfer vectors contained by the file,
and appends the table and a Bank Bit Initialization routine to the
file.

The modified file and its appended routine and vector table are

then loaded as a unit, in core image form, onto the system device on
.DAT-l4.

PATCH automatically changes the patch parameters in SYSBLK.

Relocation

and the Bank Bit Initialization routines are described in paragraphs
3.3 and 3.4, respectively.

3.2.2

Input File Requirements

In order to be installed onto a system as a SYS file, a relocatable
file on .DAT

-1~

must:

not contain more than 256 transfer vectors

not contain external .GLOBL references

not use indexed instructions

not have a core image size greater than the number of
system blocks allocated multiplied by 4~~8

not cause the total program size to exceed 8K (see the note)

not contain COMMON references

3-2

Loading Relocatable Programs as System Programs

NOTE
The total installed size of a program may be computed by adding:
1) the size of the binary program, 2) the number of transfer vectors, and, 3)
32 8 locations which are occupied by the Bank Bit
In1tialization routine.
3.3

RELOCATION

MACRO outputs relocatable programs addressed as if they were assembled
starting in location 0.

Relocatable programs are normally loaded

into core for execution by the Linking Loader System Program which,
at load time, computes the difference between the MACRO-assigned addresses and the addresses of the actual core area into which the program is to be loaded for execution.
the relocation factor,

This computed

difference, called

is then added to each location address contained

by the program as it is loaded into memory.

The relocation factor

properly orients the relocatable program within memory.
3.3.1

PATCH READR File Relocation

During the execution of a READR command, PATCH calculates the required
13-bit relocation factor by subtracting the program size from the
final

(highest) address the program is to occupy.

The final address

is either specified in the READR command by the user or, by default,
17636

is assumed.
For example, the relocation factor (#) for a re8
locatable program 400 words in size with no address specified in the
8

READR command would be calculated:

# = 17636 8 -400 8 + 1

= 17237 8

In order to relocate the program addresses

(0 through 377) for con-

version to the absolute values required for SYS files, PATCH then
adds the relocation factor to each program address, i.e.,

3-3

Loading Relocatable Programs as System Programs

= address of first word location (17237 8 )

# + ~

= address of last program word location

# + 377

(17636 )
8

PATCH also relocates the addresses contained by the memory reference
instructions and transfer vectors found in the input program.
3.4

BANK BIT INITIALIZATION (BBI) ROUTINE

Transfer vectors contain lS-bit addresses which include bank identity
bits and the l3-bit, 8K word location addresses.

The format of a

Transfer Vector data word and the interpretation of its address portion are:

BANK
0
1
2

I
I

BANK
BITS

'13-bi t BANK WORD LOCATION ADDRESSES
I
I

... 10
... 10
.. 1 1
... 11
1

0
1
0
1

,,--.. 09000 to 17777

- . . 20000 8 to 3777788

I~ 40000 8 to 577778
I~ 60000 8 to 777778
I
8

During READR relocation operations the relocation factor is added to
the least significant 13 word bits of the Transfer Vector to relocate
the bank word address which it contains.

The bank bits of a Transfer

Vector cannot be relocated at the time the file is installed as a System
Program since the number of the highest bank of a system varies according to the overall size of the system.
To ensure that transfer vector bank bits are relocated prior to run
time, PATCH, during READR operations, develops a table of transfer
vector locations, adds the table to

a Bank Bit Initialization routine,

and appends the routine to the input program.

The BBI routine is

installed onto the system device medium as part of the new System
Program.
When a READR-installed program is loaded into core, its associated
BBl· is also loaded immediately below the program (see Figure 3-1).
The loader, .SYSLD, on completion of its functions, turns control
over to the BBI routine.

The BBI routine determines what bank the

System Program has been loaded into and updates the bank bits of all
Transfer Vectors contained by the program.

3-4

The routine identifies

Loading Relocatable Programs as System Programs

the vector words from the Transfer Vector table developed during
PATCH READR operations.
On completion of its initialization function, this routine sets the
contents of the .SCOM+3 pointer to the address of the last location occupied by the routine and turns control over to the System
Program.
The modification of .SCOM+3 at the end of BBI functions frees the
core occupied by this routine for use by the System Program.
The basic portion of the BBI routine requires 328 word locations.
The routine Transfer Vector taQle requires an additional location for
each Vector in the program.

A maximum of 400

vectors is permitted.

The PATCH appended BBI routine relieves the user of the responsibility
of including a routine to perform this function.
3.5

READR ESTABLISHED SYSTEM CO~~UNICATIONS AREA

The ability to specify in the READR command the highest load address
for the program being loaded permits a buffer area to be set up
between the last program address and the first address of the bootstrap.

Such a buffer would provide READR-loaded programs with a

common core-resident buffer for intra-program communications and/or
common data storage. See below and Figure 3-1.

r'"'B-O-O-T-S-T-RA-P--.... "::-577778
~------4~576368

BUFFER AREA
- -- -- - -

~Specified

SYSTEM PROGRAM
BBI

3-5

Address

Loading Relocatable Programs as System Programs

3.6

LOADING PATCH-RELOCATED CORE IMAGE SYSTEM PROGRA~S

READR-installed Systems Programs and their associated Device Handlers
are loaded into core from the system device by the System Loader
(.SYSLD) program.

The placement of READR-installed System Programs and their device
handlers in core is illustrated in Figure 3-1.

3-6

Loading Relocatable Programs as System Programs

LEGEND:

N is:
40000

for 24K

60000~ for 32K

..---- 177778 + N

r-----------------~~
BOOTSTRAp l

+ N

17636

.SCOM + 3 Adjusted value
after BBI initialization
operation

.SCOM + 3 Initial Value

SYS-FILE
(except MACll) 1

BANK BIT
INITIALIZATION
ROUTINE (BBI)

-------

...~.....--- ~~~2~8 + N Lowest Limit for
SYS file BBI initialization routine

- - - - - - - - - ...-ca----

.SCOM + 2

DEVICE
HANDLER(S)

.......---------~ ~

.SCOM + 1

RESIDENT
MONITOR
~

1~1ACll, supplied with UC15 systems
always runs in Bank "I" due to
restrictions in the Common
Memory Space.

Figure 3-1
Memory Map, Loading a READR-Installed SYS-file

3-7

CHAPTER 4
LOADING ABSOLUTE PROGRAMS AND DATA
4.1

READ AND PATCH FILE FUNCTIONS

This section contains detailed descriptions of the READ load function,
the function user commands, and the load operation.

Patch files, the

input sources to PATCH during READ load function, are also described
in detail.
4.2

READ LOAD FUNCTION

The "READ" load function enables the user to transfer absolute "Patch
files" from the auxiliary input device (.DAT-IO) into a user-selected
program or data block located on the device on .DAT -14.
A Patch file is an absolute program written by the user to operate with
the PATCH READ load function; its purpose is to direct the PATCH program
in loading information contained by the file into a specific register
or series of registers located within a program or data block of the
device on .DAT-14.

Entire programs or data blocks, as well as selected

items, may be loaded from Patch files onto the device on .DAT -14, using
the PATCH READ load function.
Patch files are used as a means of making changes (i.e. patches) to
existing SYS files or a device data block or for installing a new
version of a System Program onto a system device.

They are particu-

larly useful when the same patches are to be made to a SYS file
contained by more than one system device.
NOTE
When patching or replacing a System Program, care
must be taken to ensure either that the existing
SYSBLK parameters are not affected or th~t ~hey
are changed to comply with the character~st~cs of
the replacement version of the program (see
paragraph 4.5).

4-1

Loading Absolute Programs and Data
4.3

READ COMMAND

The command READ initiates the performance of the absolute load
function.

This command may have any of the following formats:

>READ)

>READ~[filename])

>READ~[filename] L...I [ext]&...\[comment] )

where:

(for non-directoried devices only

filename is the name of the Patch file to be input
from .DAT-10. The use of a filename extension
(format c) is optional. If not given, the extension
is assumed to be ABS.

Comments may be added to command strings having
format £.

Patch files on paper tapes do not require a
filename (format £) .
NOTE

The READ command must be preceded by either a
Program Select or Block Select command. The
patch file read must be a .ABS macro program.
4.3.1

READ Load Operation

During READ command load operations, PATCH loads into core one 400 g word block at a time from both the Patch File and the selected program
or data block.

Once both blocks are loaded, PATCH processes in

sequence each word of the Patch File block.

Using reference data

supplied in the Patch File, PATCH determines the address within the
user selected program or data block which corresponds to that
addressed by the Patch File word being processed.

If the addressed

register is within the selected program or data block currently in
core, the contents of the Patch File word are loaded into that
register by PATCH.
If the addressed register is not in the block currently in core, the
block in core is written back into the device at .DAT-14 and the block
containing the addressed register is loaded.

After the new block is

loaded, PATCH locates the addressed register and writes the contents
of the Patch File word into it.
A user-selected data block is handled, during READ operations, as if
it were an absolute program having addresses ranging from ~~~~~
through

~~377g.

If more than one data block is to be patched, a

separate patch File is required for each block and individual Select
Block and READ commands must be issued for each.
4-2

Loading Absolute Programs and Data

If a Patch File word

addres~es

a location not within the limits of

the selected program or block, PATCH outputs an ADDRESS OUT OF
RANGE error to the user console teleprinter (refer to Section 6
for description of error messages) .

4.4

WRITING PATCH PROGRAMS

The following is an' example of the proper format for a PATCH assembly
source program to be assembled and output by MACRO XVM:

PATCH FILE
.TITLE
.ABS
. LOC
734777

MEANING AND OPERATION PERFORMED
anything
NLD
234

~4~674

No Loader .
Set location counter to address 234.
Set contents of 234 to 734777, and
go to next locat~on.
Set contents of 235 to 6~~261 .
Advance counter to location 35~.
Set contents of 35~ to -3 .
Advance location counter to
address 371.
Write ~4~674 into LOC 371

.END

End program.

6~~261

. LOC
-3
. LOC

35~

371

The commands available for patching system programs assembled under
MACll are the 'B'
Section 5.6.

(block) command and the 'L'

(location) command.

See

As illustrated in the above example, .LOC statements in the patch
program are recognized by PATCH as pointers.

These pointers direct

PATCH to start a replacement operation at a specified address in the
program or data block being patched.

In the replacement (overwrite)

operation, the contents of the Patch File location immediately following the .LOC statement is written, by PATCH, into the selected program
or data block location pointed to by the .LOC address.

PATCH then

writes, in sequence, the contents of the following Patch File locations
into the registers of the file being patched immediately following the
one pointed to by the patch file .LOC statement.

All patches are

written as II the program is to be loaded into the first bank (13
bit addresses).

All transfer vectors in a .ABS program must be

bank-bit initialized before use.

4-3

Loading Absolute Programs and Data

The current replacement operation continues until another .LOC statement is reached in the input Patch File.

The new .LOC statement ends

the current replacement operation and directs PATCH to go to a new
start location and to start another replacement operation.
The Patch File operations are terminated when PATCH reaches the PATCH
File .END statement.
4.4.1

Partial Overlay Patch Programs

A Patch Program written to change the contents of specific locations
or groups of contiguous locations throughout a selected program or
data block acts as a partial patch to the file being modified: only
those items specified in the Patch Program are affected.

The Patch

Program in the example in paragraph 4.4 represents a partial patch.
Patch programs may be used only to change the contents of registers
already existing within a selected program or data block; they cannot
be used to insert information between existing program lines.
4.4.2

Replacing Complete Programs or Data Blocks

Patch Programs may be written to replace an entire absolute System
Program or 400 -word data block.
B

To accomplish this, the user sets

the first Patch File .LOC statement to point to the first location
of the program or data block to be replaced and follows it with the
replacement program or data block.
.TITLE
.ABS
.LOC
~
Replacement
Program or
Block

For example:

(Patch Program name)
NLD
(First Location of selected program or
data block)

.END

4.5

PRECAUTIONS TO BE OBSERVED FOR PATCH READ OPERATIONS

The PATCH READ load operation does not modify (i.e., update) SYSBLK on
completion of the function.

If any changes are made to a system Program

which affect the parameters stored for it in SYSBLK, the user must
4-4

Loading Absolute Programs and Data

change these parameters using the PATCH List function (described in
Section 5).

Indexed instructions are not permitted.

When overlaying a complete program with a new version, the user must
ensure that the new version will fit into the area allotted to that
program during SGEN.

All .ABS program transfer vectors must be bank-

bit initialized by the program itself before it uses them if the core
image is loaded above 8K.

4-5

CHAPTER 5
LISTING AND MODIFYING PROGRAM AND DATA BLOCK RFGISTERS
5.1

This section describes the functions, commands and procedures which
enable the user to select and modify individual registers within:

5.2

SYS files initially written as absolute programs,

READR-installed SYS files,

a system device SYSBLK Program Parameter Table, and

a system save, tQAREA, area.

PATCH LIST FUNCTION

PATCH register modification is carried out as a list operation in
which the address and contents of each register selected by the user
are listed (printedl on the user's console teleprinter.

On completion

of the print operation, the user may enter, if desired, an expression
to replace the contents of the currently listed register.

When the

user is through with the currently selected register, he terminates
that operation and either terminates the function or directs PATCH
to list the contents and address of another register.
or data block containing the register(s)

The program

to be modified must be

identified by the entry of a Select command (see Section 2); the
specific register to be viewed and modified (if desired)
by the user with a "List" command.

is selected

Three types of List commands are

provided by PATCH:
al

L command, used to select and initiate the list
operation for registers within any selected SYSfile, data block, SYSBLK or QAREA contained by
the device on .DAT -14.

LR command, used only for the selection and listing of registers within READR-installed SYSfiles.

5-1

Listing and Modifying Program and Data Block Registers

SYSBLK Program Parameter Commands, five separate
commands, each of which initiates the list operation for a specific SYSBLK parameter for the currently selected SYSfile. Used only with SYS files.

The format and use of the above List commands are described in paragraphs 5-3, 5-4, and 5-5, respectively.
5.2.1

List Operation

The PATCH operations performed during a List operation are:

5.2.2

On the entry of a list command, PATCH loads the 400
8
word data block or selected Program block containing
the user identified register into core. The selected
register is opened by PATCH and its address and contents are printed on the console teleprinter unit.

PATCH waits for the entry of a modifier and/or a
terminator.

PATCH writes the input modifier (if made) into the
opened register and examines the line terminator to
determine the next operation to be performed. Depending on the terminator entered (see 5.2.3) PATCH
terminates the List function or opens another
register.
Expressions

Expressions, as defined for MACRO XVM, are strings of symbols and
numbers separated by arithmetic or Boolean operators.

Octal numbers

of from one to six digits (¢ to 777777 ) or alphanumeric symbols of
8
from one to three characters may be used as expressions or as com-

ponents of an expression.
rn PATCH List operations, expressions are entered as replacements for
the contents of a program or data block register.

Expressions entered

to change or modify a data word consist of up to six octal digits,
originally preceded by a minus sign; those entered for an instruc·tion
word consist of an instruction operation code and an operand.

PATCH

contains symbol tables for all of the XVM basic instructions op-codes
and operate group instructions with the octal values of each.

The

user can, therefore, patch registers using symbolic rather than octal
representations of ~~CRO XVM instructions (e.g., an expression may be
entered as LAC 17536 instead of its octal form,

217536).

The operators recognized by PATCH are listed in Table 5-1.

5-2

Listing and Modifying Program and Data Rlock Registers

Table 5-1
Operators Recognized by PATCH
OPERATOR

FUNCTION

Two's complement subtraction
Inclusive OR

Two's complement addition

The asterisk (*) in addition to its use as an operator, also sets the
indirect bit of the

.
express~on

value of the expression.
symbol

.
caus~ng

2~~~~
PPPPS

to b e XORe d

.
~nto

the

The XOR operation occurs each time that the

(*) is encountered; two sequential asterisks negate the setting

of the indirect bit.

**
The value of a PATCH expression is null (no modification is made to
the opened register)

unless the expression contains a number, a symbol

or an asterisk.
Expressions are evaluated from left to right assuming an initial value
of zero followed by a + operator

(i.e., 0 + User's expression).

Lead-

ing and trailing operators are legal in an expression but the latter
are ignored.

Whenever a string of consecutive operators is used, only

the last one in the string is used by PATCH in evaluating the expression.
NOTE
The LAW is a special case in PATCH; it should be
used only in the following ways:
a}

LAw~n

which is equivalent to 76~~~~+n

LAW~-n

which is equivalent to -no

The use of LAW (LOAD ACCUMULATOR WITH this number)
in any manner other than that described above will
result in an error.

5-3

Listing and Modifying Program and Data Block Registers

List Operatio~, Line Terminator

5.2.3

When all list operations for an opened register are completed, one of
four possible Line Terminators is entered to close the register and
indicate the next list operation.

The line terminators are described

in Table 5-2.
Table 5-2
List Operations, Command String Terminators

ENTRY

DESCRIPTION

)

Carriage Return

Open and list the next sequential
register

Up Arrow and Carriage
Return

Open and list the register which
precedes the current register.
Treat any entries after t as a
comment .

Back Arrow and Carriage Return

Open and list the contents of the
register pointed to by the address
portion of the word contained in the
currently opened register.
Treat
any entries after + as a comment.

ALT MODE

Terminate current list operation;
write PATCH buffer block onto
.DAT -14 and wait for next entry.

....)

OPERATION

5.3 L LIST COMMAND
The L List command has the following format:

where

L initiates the list operation, and

n represents the octal form of the 13-bit address of
the location to be opened.
When the L command is used
to open registers of READR-installed SYS files the
address n must be the relocated (i.e., address after
relocation) value.

5-4

Listing and Hodifying Program and Data Block Registers

The response to an L command is a printout having the following format:
>address of opened location) /

(contents of opened location»

EXAMPLE:
>L

132

L-I.0~132/77 7435 >

(location 132 8 ; contents 777435 )
8

After outputting the address and contents of the accessed location,
PATCH prints its go-ahead symbol (»
1)

and waits for:

the entry of an expression to replace the current contents, to modify the opened location.
For example:
> (address)/(current contents»(expression or new contents)

5.3.1

the entry of selected string terminator to indicate the
next sequential list operation.
Example:

Select, List and Modification Procedures

The following illustrates the use of PATCH select and list commands in
the modification of a hypothetical system program named JOVE.
PROCEDURE

DESCRIPTION
Select System Program JOVE.
Initiate List operation starting at location 1.0.0

>~~l.0~/777435> )

PATCH prints address and contents of
opened location; user enters ) to go to
and open next sequential location.
Address and contents of location l¢l are
printed, user changes contents from
6.0¢2.0.0 to 213775 and opens the next sequential location.
User modifies contents of location 1.02 and
commands PATCH to go to and open the location addressed by the contents of location
1¢2 (i.e., location 3.0~).
User enters t ) to open and examine contents of preceding location (i.e., 277).

>~~277/7~3112>~

.User terminates current list operation
by entering ALT MODE (symbol $).

PATCH prints a go-ahead symbol to in?icate
that it is ready for the next operat~on.

5-5

Listing and Modifying Program and Data Block Registers

5.4

LR LIST COMMAND

The LR List command has the following format:
>LR ~ n}
where LR initiates the listing operation for a previously selected
READR-installed Systems Program, and

represents the 13 bit "unrelo-

cated" address (in octal) of the program location to be opened.

NOTE
PATCH automatically calculates and adds the needed
relocation factor to the user's entry (i.e., n).
This enables the user to work directly from original unrelocated listings.
The major difference between the "L" and "LR" list operations is that
in L operations only absolute

~ddress

information is output in responses

and is required in modification procedures.

In LR operations, however,

the user must deal with both relocated and unrelocated addresses in
list printouts and in register modification procedures.
When the LR command is used to open word locations containing either
data or non-memory reference instructions, the responses output are
similar to those described for the L command.

The only difference is

that in the LR responses the address of the accessed location is given
in relocated form relative to the original MACRO XVM assigned address.

NOTE
In the following paragraphs the symbol # is used to
denote the relocation factor used by PATCH in the
conversion of an unrelocated program into a system
program.

To illustrate, assume that 6 is the unrelocated address of a userselected location containing a non-memory reference instruction or data
and that the relocation factor (#)

is 17344.

command:

5-6

The response to the

Listing and Modifying Program and Data Block Registers

is printed in the following form:
>17344 + 6

nnnnnn

Relocated /contents of location (6 digits)
Address
Op-Code and Operand or Data
For example:
>17352/017206>
To illustrate, assume that a relocation factor of 17642 is current and
that the selected program location

l~l

contains a JMP

.+1 instruction.

The response to the command:

is printed in the following form:
>17743/6l7744<~~1~2>

Relocated /
Location
Address
(17642+101)

Contents, Op-Code + # + Operand
(6~~~~~ + 17642 + 1~2
JMP=6~~~~~

.+1 =

l~l+l

( Unrelocated Operand ~
Referenced Address /
(Le., ~~1~2)

1~2

Printing an address operand in its unrelocated form enables the user to
easily recognize the instruction containing it, and the referenced location without calculating and subtracting the PATCH relocation factor
(#) •

The unrelocated referenced address within a memory-reference type instruction is not printed by PATCH if it is less than the first address of the
program; the address printed in angle brackets is always positive, though
it may be zero.
5.4.1

LR', Reqister Modification Procedure

The contents of registers opened during LR operation are changed, as
for L list operations, by the entry of an expression which is to replace the contents of the current location.

New contents are entered

immediately after the listed response and replace the current contents
on termination of the user input.

5-7

Listing and Modifying Program and Data Block Registers

Input expressions are formed using:
a.

Octal numbers up to six digits in length;
PATCH always assume

a six-digit input and

right-justifies all numeric inputs
an entry of 300

(e.g.,

is recognized as 000300 ).
S

Operators (symbols)

as lis ted in Table 5-l.

Symbolic representation of all DIGITAL XVM
basic instruction op-codes and operate
group instructions.

The symbol # to represent the relocation
factor of the currently selected, READRinstalled System Program.

Command string terminators indicate the next sequential operation to
be carried out (refer to Table 5-2).
5.4.2

Relocation and LR List Operation Examples

The following examples use a single memory reference instruction to
illustrate the relocation operations performed by READR in the installation of a SYS file and the manner in which the instruction may subsequently be examined using the LR command.
EXAMPLE:
a.

Assume the instruction LAC~300 located in MACRO assigned location 200 of an unrelocated 400 S -word program.

The program is selected by the user and is patched
onto a prepared system device medium as a SYS file
using the READR function/command.
Program
Name

>Filename (name given at SGEN time)
>READR Filename (current name of program)

During READR operations, a relocation factor is calculated and added to all location addresses contained
by the program. For a 400 -word program # = 17237 S '
Therefore, the relocatable Sinput memory reference
instruction
Instruction
Address
LAC 300
is relocated as
(#+200)

(LAC #+300)
5-8

Listing and Modifying Program and Data Block Reqisters

which equals

Address

Instruction

17437 8

LAC 17537 8

or, in full octal form:
17437

217537

The relocated instruction is then output to the system
device.
d.
r

To examine the instruction given in step c. in a READRinstalled SYS file, the user calls PATCH, selects the
program, and uses the LR command
$PATCH
>filename )
>LRL...I20a)
The responses output on the I/O console will be:
>17437/217537<~~3~~>

To change the contents of the location, the user enters
the desired expression and terminates the command string.
If the contents contain an address, the relocation factor
must be added to the new entry.
For example, to change the address references in the contents of the location openedm step
from the 300 8 to
3?5
(unrelocated values) ,the user must enter the expres8

s~on:

217544

if he knows the value of the relocation factor # (i.e.,
17237 + 305).
or

EXAMPLE:
>17436/217537< ~~3~~> LAC#+3~5}
changes the location to
17436/217544
and the terminator )
causes the next sequential location to be opened and a response output.

5-9

Listing and Modifying Program and Data Block Registers

5.5

SYSTEM PROGRAM SYSBLK PARA'1ETER CO~1ANDS

PATCH provides, as a convenience for the user, five separate commands
which obtain and modify, if desired, SYSBLK parameters for the user
selected program currently in core.

By entering the appropriate com-

mand from the following list and using suitable terminators as listed
in Table 5-2, the user may start the listing at any specific point and
proceed subsequently through the desired parameters.
The commands, the parameters opened, and the word positions are as
follows:
Command

SYSBLK Parameter Opened

Word Position

FB)

Number of first block used

NB.)

Number of blocks allotted to
program

FA..)

First core address occupied
when the SYSfile is loaded
into core (13-bits)

PS)
SA)

Program Size
Starting Address (13-bits)

When examining the parameter group of a

READR-installed System Pro-

gram, the user must remember that:
FA

will give the address of the first (lowest)
core locations occupied by the System Program/BBl combination when it is loaded.

will give the TOTAL size of the System
Program/BBl combination.

will give the starting address of the BBl
routine (first address above Transfer Vector address table).
NOTE
The program system parameter commands listed above
may not be used:
a.

when a program has not been selected or
the command CTRL P (tP) has been entered;

when SYSBLK has been selected;

when the commands B, B+, or B- are in effect.

5-10

Listing and Modifying Program and Data Block Registers

The five SYSBLK Parameter commands operate in a manner similar to the
List (L) command.

When entered, the corresponding parameter location

address and contents are printed out followed by a go-ahead symbol (»
as for the L command.

As in other list operations, the user may, ac-

cording to the manner in which terminators are used:

1) modify the

opened register, 2) view the next sequential register, 3) view the
preceding register, 4)

jump to and view the location addressed by the

contents of the last opened register, or 5) terminate the operation.
The operations initiated by a Parameter command are carried out on the
r

SYSBLK information contained by the PATCH command table.

This SYSBLK

data is obtained by PATCH when it is first loaded into core and prior
to the start of user PATCH operations.

On completion of each SYSBLK

Parameter command operation (ALT MODE terminator used), the modified
(or unchanged) command table SYSBLK is copied onto the system on .DAT -14.
The five separate commands are provided by PATCH as a convenience for
the user; they permit the user to obtain a specific parameter or to
start listing operations at a specific point with a simple straightforward command.

The complete series of five program parameters may

be obtained simply by starting with the FB command and advancing the
listing operation through the next four sequential word locations.
This method is the only way to modify SYSBLK parameters without reloading PATCH since it also modifies the PATCH command table.
EXAMPLE:

The following example illustrates the use of a PATCH SYSBLK parameter
command to obtain and modify the first address occupied, program size
and starting address parameters for the system program currently in
core.
> FA..)

00066/013 670>13 660..)
00067/003747>3757 )
00070/0l3671>1366l(ALT MODE)
>

5-11

parameter command
first address opened and modified
program size opened and modified
starting address opened and
modified and operation terminated.

Listing and Modifying Program and Data Block Registers

5.6

MODIFYING REGISTERS IN DATA BLOCKS

Users can modify registers in data blocks (on the device assigned to
.DAT -14) by using the 'B' command like the 'L' operation.
This facility is utilized in a UC15 system to patch PDP-ll programs,
through octal expressions.

5-12

CHAPTER 6
SUMMARY, OPERATING PROCEDURES AND ERROR MESSAGES
6.1

CONTENTS

Brief procedural descriptions of the operations needed before loading
r

the PATCH program and for the performance of each PATCH function are
given in this Section.
described.
6.2
6.2.1

Error detection and error messages are also

PRE-LOAD OPERATIONS
.OAT Slot Assiqnments

The user must make the .OAT slot assignments given in Table 6-1
immediately prior to loading the PATCH program.
Table 6-1
Required .DAT Slot Assignments
Used to

.OAT Slot
-14

Input from and output to the device
on Which patches are to be made.
The device handler is required only to
perform .TRAN.

-10

Input from the auxiliary device.
The device handler must handle Oump
Mode input and, if it is for a nonfile oriented device, must handle
image alpha mode.

-3
Permanently
Assigned (
-2

Output to the teleprinter.
Input from the console keyboard or batch
processing device .

. OAT slot -10 can be assigned to no device handler (NONE) if auxiliary
input is not required.

.DAT slots -3 and -2 cannot be changed.

6-1

Summary, Operating Procedures and Error r1essages

6.2.2

I/O Device Preparation

The user must ensure that the I/O device assigned to .DAT-14
(material to be patched) is WRITE ENABLED before starting PATCH
operations.

If the device is a

6.3

DECtape unit - set the unit's WRITE ENABLED/WRITE LOCK
switch to the WRITE ENABLED position.

Disk - set the assigned logical disk unit's WRITE
LOCKOUT switches to their ENABLED positions.

CALLING PATCH

PATCH is called by issuing the command PATCH from the system I/O
console.

When PATCH is loaded and running, it outputs its name

and version number.

For example:

XVM/DOS Vnxnnn
$PATCH..I
PATCH XVM Vnxnnn
>

6.4

PATCH OPERATIONS

The operations which may be performed using PATCH fall into four (4)
functional groups.

Procedures, in table form, are given for the List

"L" and "LR" functions.
The Examination and modification of registers in absolute programs are
summarized in Table 6-2.

The examination and modification of registers

in Patch-READR-installed system programs are summarized in Table 6-3.

6.4.1

Replacing Absolute Programs from External PATCH Files

The required procedure is to:
a.

Select Program or data block to be patched

Specify READ operation and input file with the command
string
> READ~[filename])

6-2

(e.g.,

(e.g. >EDIT)).

> READL.-AEDPCH))

Summary, Operating Procedures and Error Messages

6.4.2

Installation of Relocatable Proqrams as Absolute SYS Files

The required procedure is:
a.

Select the system device area prepared (by SGEN)
receive the new SYS file (e.g.,
>NUFIL)).

Enter the PATCH READR command using the following
format:
>READR~nnnnn]L...I[filename])

(e. g. ,

>READRL...Il76~~L...INUFIL))

NOTE
"nnnnn" represents the highest register the new SYS
file will occupy when loaded into core.
In the·
example, this is specified as l76~~8 which would
leave a 36 a-word buffer between the top of the SYS
file and the first Bootstrap location. The default
value of nnnnn=17636a.
6.5

ERROR DETECTION

PATCH, on the detection of an error, causes the following:
1)

the current function is terminated;

the data block (program or selected block) current in
the PATCH Block Buffer is output to the device on
.DAT-14 if modifications had been made to the block;

an appropriate error message is output at the us.er I s
I/O terminal.

In the event of an .IOPS error, control is not automatically returned to
the monitor.

PATCH can be restarted by the entry of a CTRL P (tp)

command.
CTRL P is also useful in terminating a read operation in the
event of an equipment malfunction (e.g., paper tape reader jams).
In the event of an I/O device not ready error (IOPS4), CTRL R may be
used to continue the current operation after the device involved has
been made ready.

CTRL C may also be used to return control to the

monitor; however, care must be taken not to use this command when a
register is open since any modification made to the current block in
core will be lost.
6.6

ERROR MESSAGES

The error messages output by PATCH are listed and described in
Table 6-4.

6-3

summary, Operating Procedures and Error Messages

Table 6-2
Procedure for Hodifying Registers in Absolute System Programs

Item to be Patched
BLOCK

PROGRAM

Operation
SELECT ITEM
TO BE
PATCHED

> BL-In.J

> Prog. Name.J

PROGRAM
SYSBLK
PARAMETERS
> Prog. Name)

>B+L-II n.)or
> B-L-I n)

SELECT
REGISTER OR
PARAMETER

> FA.J
> FB )
>NB'
> PS.3
> SA

or
or
or
or

ENTER NEW CONTENTS IMMEDIATELY AFTER RESPONSE
OUTPUT WHEN LOCATION WAS OPENED.

MODIFY
OPENED
ITEM
e.g.,

SELECT
NEXT
OPERATION

n.J
L-I

>Address/ Old Contents> New Contents

TERMINATE RESPONSE LINE ACCORDING TO DESIRED NEXT
OPERATION:

)

go to next sequential location open
and list address and contents

go to preceding register, open and
list address and contents

+- )

go to register identified by contents
of current register, open and list
address and contents
terminate current list operations

6-4

Summary, Operating Procedures and Error Hessages

Table 6-3
Procedure for Modifying Registers in PATCH-READRInstalled System Programs
Item To Be Patched
PROGRAM

PROGRAM
SYSBLK
PARAMETERS

SELECT ITEM

> Prog. Name)

>Prog. Name.,)

SELECT REGISTER OR
PARAMETER

>LR

>FA)
>FB .J
>NB" .)
>PS)
>SA.)

Operation

MODIFY OPENED
ITEM

L...I

or
or
or
or

ENTER NEW CONTENTS IMMEDIATELY AFTER RESPONSE
OUTPUT WHEN LOCATION WAS OPENED.
USE RELOCATION
FACTOR (#) WHEN CHANGING MEMORY REFERENCE OR
TRANSFER VE.CTOR REGISTERS.
e.g., standard register
relocated
address

old
new
contents > contents

memory reference instruction or transfer
vector
>

SELECT NEXT
OPERATION

relocated
address

old
<unrelocated>
contents address of
contents

new
contents
+ #

TERMINATE RESPONSE LINE ACCORDING TO DESIRED
NEXT OPERATION:

)

go to next sequential location
open and list address and contents

t.J

go to preceding register, open
and list address and contents
go to register identified by
contents of current register,
open and list address and contents.
terminate current list operations

6-5

Summary, Operating Procedures and Error Messages

Table 6-4
r -____________________~------~E~R~R~O:~R~M=E~S~S~A~G~E~S~--______________________~

Printout
ILLEGAL COMMAND

Meaning
a.
b.

An attempt was made to issue a command
before a file or block was selected.
Command unrecognizable.

NOT OCTAL DIGIT

A character other than an octal digit was
encountered where an octal digit was
expected.

TOO MANY DIGITS

The command string contains an octal number
with more than 6 digits.

ADDRESS OUT OF RANGE

a.
b.

The address requested is not within the
legal range of the currently selected
program or block.
Address or size argument of a SYSBLK
comma~d is greater than 17645
or is
8
negat1ve.

CHECKSUM ERROR

Bad data read in from auxiliary input device
(checksum, parity, or validity bits) - try
again.

FILE NOT FOUND

The file named in a READ or READR command does
not exist on the auxiliary device.

ILLEGAL BLOCK #

The block number specified in a Block
Selection command was less than ~.

ILLEGAL SIZE

a.
b.

The program size (PS command) exceeds the
number of blocks allotted for the current
program.
The size of the program (including the
bank bit initialization routine supplied
by PATCH) input by a READR command exceeds
the number of blocks allotted (during
system generation) for the SYS file or
is greater than nnnnn-2~8.

MORE THAN 256
TRANSFER VECTORS

The program being input by a READR command
contains more than 256 transfer vectors.

NOT RELOCATABLE BINARY

The program being input by a READR command
is not a relocatable binary file.

.GLOBL NOT ALLOWED

The program being input by a READR command
contains an external .GLOBL (internal .GLOBL
ignored) .

LAST ADDRESS GREATER
THAN 17636

The last address of the program input by a
READR command is greater than 17636 9 .

6-6

APPENDIX A
SYSTEM BLOCK (SYSBLK) DESCRIPTION
A.l

SYSTEM BLOCK (SYSBLK)

In DIGITAL XVM systems the monitor contains a System Block table
(abbreviated as SYSBLK) which contains the name, physical parameters
and I/O information for each system program in the software system.
The information contained by SYSBLK in the initial operating system is
determined during System Generation procedures; refer to the SGEN XVM
Utility Manual.

SYSBLK is used by the system loader programs in

locating and loading selected (called) System Programs into core.
PATCH, when first loaded into core, checks for a SYSBLK on .DAT -14
and, if one is present, it loads the SYSBLK Parameter Table into core
as a PATCH Command Table.

This command table is used unchanged during

all subsequent PATCH System Program operations, including SYSBLK modification operations.
SYSBLK, itself, is divided into two distinct areas (see Figure A-I),
a Parameter Table, and a Communications Table.

PATCH operations nor-

mally concern only the SYSBLK Parameter Table.

PARAMETER
TABLES

}

SYSBLK

FREE AtEA

t
f--

COMMUNICATIONS __
DATA GROUPS

COMBLK

Figure A-I
System Block, Overall Configuration

A-I

System Block (SYSBLK) Description
A.I.I

SYSBLK Parameter Table

The SYSBLK parameter tables contain separate seven-word parameter
groups for each System Program in the operating system.

Each para-

meter group consists of the following:
WORD

CONTENTS

I & 2

Name of the System Program or overlay (i.e., PATCH,
DDT, EDIT, etc.) in .SIXBT form.

The number of the first system device block occupied
by this System Program or overlay.

The amount of system device blocks allotted to contain the program. For PATCH-installed programs, this
number is allocated during user SGEN procedures.

The address (13-bit) of the first core locations that
will be occupied by the System Program when it is
loaded.

The size of the program.

The program starting address (13-bit).
OCTAL
PRINTOUT
INTERPRETATION

WORD

CONTENT

~~~l~
~~~ll

~5~411
249191~~

~~~12

~91~63~

~91~13

91~91~12

Twelve blocks (octal) are allotted to
hold the program EDIT.

~~.£l14

911391~91

The first core location! occupied by
EDIT is 13~Y'Y'8.

~.£l~15

91914636

EDIT consists of 4636

~~.£l16

911391~.£l

Program start address! is at core location 13~.0.08.

Program name "EDIT" in .SIXBT code.
The first block occupied by EDIT is
63~8·

words.

Figure A-2
System Program EDIT SYSELK Parameter Group

A.I.2

Communications Table

The system block COMBLK area contains individual communication data
groups which store .DAT slot, buffer, and overlay information for
each system program.
!At load time, the bank bits of the highest 8K system bank are added to
this address to properly locate the program in the highest bank.
A-2

APPENDIX B
SYSTEM PROGRAMS, GENERAL DESCRIPTION
System Programs have a unique feature in that they are loaded into core
and are started by the entry of a single user command (i.e., program
name) at the system console keyboard.

This feature is implemented by

name, control parameters, and I/O device information contained for
each System Program in the system SYSRLK Parameter and Communication
(COMBLK) tables.

The information contained in SYSBLK enables the

Monitor to recognize System Program names

(as stored in SYSBLK) and

causes it to initiate the loading of the identified program into core.
System Programs and their associated Device Handlers are loaded into
core from the system device by the System Loader (.SYSLD) program.
System programs may not be loaded above 32K and should not use wide
addressing.
The placement of System Programs and their device handlers in core is
illustrated in Figure B-1.

As shown:

System Programs are always loaded immediately below the
Bootstrap in the highest portion of the highest 8K bank
in the system, except for ~l.AC11 which is always loaded
into 'Bank l' of the system.

The lowest register which a System Program may occupy
is 208 in the highest 8K bank of the system; handlers
and free core may be located anywhere in the system.

System Programs are stored on the system device as absolute executable
files and are conunonly referred to as "SYS files".

Software is sup-

plied to the user and contains a standard group of conunonly used
utility and language programs installed on some medium (e.g., DECtape)
as SYS files.

The SGEN and PATCH utility programs can be used to

delete the supplied SYS files or replace them with other DIGITAL-supplied programs or with user programs.

Table 2-1 lists the names and

SYSBLK entry names of the System Programs supplied with DIGITAL XVM
software systems.

B-1

System Programs, General Description

~177778

+ N

BOOTSTRAP
SYSTEM
PROGRAM
(SYSfile)
except MACll

1------- --

.SCOM + 3

.....-f3f3f32f38 + N lower limit
for SYSfile

- - --DEVICE-- ~.SCOM

+ 2

HANDLERS
RESIDENT
MONITOR

LEGEND:

~.SCOM

4~~~~
6¢¢~¢

+ 1

for 24K systems
for 32K systems

Figure B-1
Memory Map, Loading System Programs

B-2

INDEX
Absolute load function, 4-2
Absolute SYS files, 6-3
Absolute system programs, 6-2,
6-4
Addition operator, 5-3
Address of relocated file, 3-1
Alphanumeric symbols, 5-2
Asterisk (*) usage, 5-3

Installation of programs, 6-3
I/O device preparation, 6-2

LAW instruction, 5-3
Line terminator, 5-4
Linking Loader, 3-3
List operation, 5-1, 5-2
L (List) command, 5-4
Load function, 3-1
Loading relocated programs, 3-6
.LOC statements, 4-3
LR (List) command, 5-6

Bank Bit Initialization
routine, 3-4
Blocks, 4-2
Block Select command, 2-3
Bootstrap, 3-5

Memory, 3-2
Memory map, 3-7
Modification registers in data
blocks, 5-12

Calling PATCH, 6-2
Command string terminators, 5-4
Comment, 2-3, 4-2
Common references, 3-2
Communications table, A-2
Complete replacement, 4-4
Core image size, 3-2
CTRL C, 6-3
CTRL P, 5-10
CTRL R, 6-3

Numbers, 5-2
Number (#) symbol usage, 5-6, 5-8

Data block register modification,
5-12
Data block replacement, 4-4
.DAT slots, 1-2, 3-1, 6-1
Device assignments, 1-2

Error messages, 6-3, 6-6
Examples, 5-5, 5-8
Exclusive OR operator, 5-3
Expressions, 5-2
External .GLOBL references, 3-2

Filename, 3-1, 4-2
Functions, 1-2

Inclusive OR operator, 5-3
Indexed instructions, 3-2, 4-5
Input expressions, 5-8
Input file requirements, 3-2

Octal numbers, 5-2, 5-8
Operating READ load, 4-2
Operations, PATCH, 6-2
Operators, 5-3, 5-8
OR operator, 5-3

Partial patch, 4-4
Precautions for read operations,
4-4
Procedures for modifying
registers, 6-4, 6-5
Programs accessible to PATCH, 2-2
Program Select command, 2-1, 3-2
Program size, 3-2

READ command, 4-2
READ load function, 4-1
READR command, 3-1, 3-2
READR file relocation, 3-3
READR-installed system programs,
6-5
Registers, 4-2
modification of, 5-1, 5-7
Relocatable programs, 3-3
Relocation factor (#), 6-5

Index-l

INDEX (Cont.)
Replacing absolute programs,
6-2
Replacing programs, 4-3, 4-4

Select command, 5-1
Selecting item for patching,
2-1
Subtraction operator, 5-3
Switches, 6-2
Symbolic representation,
5-2, 5-8
SYSBLK (system block table) ,
A-I, A-2
commands, 5-2, 5-10, 5-11
System communications area,
3-5, A-2
System loader (.SYSLD) program,
3-6
System program, B-1

Terminators, command string, 5-4
Transfer vectors, 3-2, 3-4, 4-3,
4-5
TWo's complement, 5-3

Vectors, 3-4

XOR operator, 5-3

Index-2

PATCH XVM
Utility Manual
DEC-XV-UPUMA-A-D

READER'S COMMENTS
NOTE:

This form is for document corrments only. Problems
with software should be reported on a Software
Problem Repcrt (SPR) form.

Did you find errors in this manual?

If so, specify by page.

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

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

Please indicate the type of user/reader that you most nearly represent.
[] Assembly language programmer
[]

Higher-level language programmer

[] Occasional programmer (experienced)
[] User with little programming experience
[] Student programmer
[] Non-programmer interested in computer concepts and capabilities
Name

Date __________________________

Organization _______________________________________________________________
Street ________________________________------------------------------------------------------City ___________________________ State _____________ Zip Code ______________
or
Country
If you require a written reply, please check here.

[]

-------------------------------------------------------------Fold lIere-----------------------------------------------____________ _

------------------------------------------------ Do Not Tear - Fold Here and Staple -----------------------------------------------

FIRST CLASS
PER:V1lT NO. 33
MAYNARD , MASS.
BUSINESS REPLY MAIL
NO POST AGE STAMP NECESSARY IF MAILED IN THE UNITED STATES

Postage will be paid by:

Software Communications
P. O. Box F
Maynard, Massachusetts 01754

digital equipment corporation

PRINTED IN U.S.A.