asm-ppt

ASM Version 1.0 ASM Version 1.0 11

Assembler/Session 1

Course Title :

ASSEMBLER LANGUAGEDuration : 5 Half - DAYS

Course Title : :

ASSEMBLER ASSEMBLER LANGUAGELANGUAGEDurationDuration : : 5 Half 5 Half -- DAYSDAYS

Assembler/Session 1

Course Title : :

ASSEMBLER ASSEMBLER LANGUAGELANGUAGE DurationDuration : : 5 Half 5 Half - - DAYSDAYS

Course Title : :

ASSEMBLER ASSEMBLER LANGUAGELANGUAGE DurationDuration : : 5 Half 5 Half - - DAYSDAYS

ObjectivesObjectivesObjectivesObjectivesObjectives

• Familiarise with IBM 370 Assembly Language

Assembler/Session 1

SESSION 1SESSION 1Day 1Day 1

Introduction


Addressing

SESSION 3SESSION 3Day 2Day 2 Machine Instructions

Assembler/Session 1 COURSE SCHEDULE COURSE SCHEDULE

Writing a complete program


Program Sectioning


Assembler Directives


Assemble and link program SESSION 7SESSION 7Day 4Day 4

COURSE SCHEDULEAssembler/Session 1

Macro Language SESSION 8SESSION 8Day 4 Day 4

Other Topics SESSION 9SESSION 9Day 5Day 5

Assembler/Session 1

COURSE SCHEDULE

Assembler LanguageAssembler Language

SESSION 1

Assembler/Session 1

ObjectivesINTRODUCTIONINTRODUCTION

• An assembler language is a symbolic form of machine language

• Assembler translates assembler language program to machine language

• An assembler program consists of many statements

• In general, one assembler language statement corresponds to one machine language instruction

Assembler/Session 1

ObjectivesSTATEMENT FORMATSTATEMENT FORMAT

1 10 16 30

label operation operands comments

e.g..

INIT1 LA R5,4 ;INITIALISE REGISTER 5

Rules for choosing labels:Rules for choosing labels:

• maximum 8 characters

• Alphabets, digits, @, #, $

• First character should not be a digit

• label should begin in column 1

Assembler/Session 1


Operation

• One of the 200 M/C instruction mnemonics

Operand

• can be a register or memory location

Continuing a statement

• Place any character in column 72 of the line to be continued

• Continue the statement from column 16 of next line

• Maximum 2 continuation lines for a statement

Assembler/Session 1


Comment Statement

• * in column 1

• Any text in columns 2 - 71

Note : Fields separated by one or more blanks

Assembler/Session 1

ObjectivesTYPES OF INSTRUCTIONS TYPES OF INSTRUCTIONS

1. Machine Instructions

2. Assembler Instructions (Directives)

3. Macro Instructions

Assembler/Session 1

ObjectivesREGISTERSREGISTERS

Registers are storage areas inside the processor

Advantages:

- No need to retrieve data from main storage

(saves time)

- Shared resource that allows inter

communication between programs

Assembler/Session 1

ObjectivesREGISTERSREGISTERS

General purpose registers:

* 16 registers available

* Numbered 0 - 15

* Holds 32 bits (4 bytes) of data

Floating point registers:

* 4 registers available

* Numbered 0,2,4,6

* Holds 64 bits (8 bytes) of data

Note : The registers 0, 1, 13, 14 and 15 are reserved for special purpose

Assembler/Session 1

ObjectivesDATA REPRESENTATIONDATA REPRESENTATION

Binary fields

- Always fixed in length, either 2 or 4 bytes

(Fullword or Halfword)

- Negative numbers stored in 2’s complement form

Examples:

A DC H’295’ 01 27

B DC H’-75’ FF 35

Assembler/Session 1


Characters

- One byte (EBCDIC form)

- Character representation of decimal digits is called

Zoned Decimal (first nibble is zone and next is digit)

Zone digit Zone Code

0 - 9 + C

- D

+, - , blank Blank F

Assembler/Session 1


Floating Point Numbers - Always fixed in length, 4, 8 or 16 bytes

(Full word, double word, double double word)

- Left most bit represents sign

(0 - positive; 1 - negative)

- Next 7 bits represent exponent

- Remaining bytes represent the fraction

Assembler/Session 1


Decimal numbers ( Packed Decimal representation)- Each byte but the rightmost has 2 decimal digits (0-9)

- The right most byte contains a digit in the left half and a sign indicator in the right

Sign indicator: C- Positive

D - Negative

Example: 753 - 7 5 3 C

Assembler/Session 1

ObjectivesAddressing Operands

• Register addressing

• Base, displacement addressing

• Base, index and displacement addressing

Assembler/Session 1

ObjectivesINSTRUCTION FORMATS

RR opcode R1 R2

SI opcode I2 B1 D1

SS opcode L B1 D1 B2 D2

SS opcode L1 L2 B1 D1 B2 D2

RX opcode R1 X2 B2 D2

RS opcode R1 R3 B2 D2

Assembler/Session 6

Assembler LanguageAssembler LanguageAssembler LanguageAssembler Language

SESSION 2SESSION 2

Addressing

Assembler/Session 2

ObjectivesSTORAGE DEFINITIONSSTORAGE DEFINITIONS

Two ways to define fields :

1. Define a field and initialise the data in it using the DC assembler directive

2. Define a field without initialising using the DS assembler directive

Assembler/Session 2


Format:label {DS/DC} dtLn’value’

where : label : Label used to name the field (optional)

d : Duplication factor (optional)

t : Type of data ( required)

Ln : The letter ‘L’ followed by the length of the field in

bytes (optional)

value : Represents the value enclosed in apostrophes

Assembler/Session 2


Examples:ALPHA DC C’ABC EF’

FLDS DS 3CL2

H1 DC H’29’

F2 DC F’-10’

F1 DC X’03’

F3 DC PL4’-72’

Note : for character constants truncation or padding is to the right and for almost all others it is to the left.

Assembler/Session 2


DC TYPES Type Implied Alignment Data Representation

Length

C - None Character

X - None Hex digits

B - None Binary digits

F 4 Full word Binary

H 2 Half word Binary

E 4 Full word Floating point

D 8 Double word Floating point

L 16 Double word Floating point

P - None Packed decimal

Assembler/Session 2


Data Representation in other languages:

Assembler FORTRAN COBOL PASCAL BASIC

Language

DC Type

C Character Display String String

F, H Integer COMP Integer Integer

E Real COMP-1 Real Single

precision

D Double COMP-2 Real Double

Precision Precision

X, B Logical N/A Boolean Hex

P N/A COMP-3 N/A N/A

Assembler/Session 2

ObjectivesSTORAGE DEFINITIONS

Literals• A literal is a constant preceded by an equals sign ‘=‘.

• Can be used as a main-storage operand but not as a destination field of an instruction

• Causes assembler to define a field that is initialised with the data specified

• All constants defined by literals are put by the assembler in a literal pool, usually at the very end of the program

L R4,=F’1’

Assembler/Session 2

ObjectivesExercise 1 Q 1 and Q2.

2.What will happen in the following cases

DC CL5’123’

DC CL5’123456’

DC X’A1245’

DC XL2’A1245’

DC XL5’A1245’

DC F’19’

DC FL1’513’

Assembler/Session 2

ObjectivesEQU (Assembler directive)

• The EQU statement is used to associate a fixed value with a symbol

R4 EQU4

DRBACK EQUOUT+25

Assembler/Session 2

ObjectivesESTABLISHING ADDRESSABILITYESTABLISHING ADDRESSABILITY

• By establishing the addressability of a coding section, you can refer to the symbolic addresses defined in it in the operands of machine instruction

• Assembler will convert the implicit addresses into explicit addresses (base - displacement form)

Assembler/Session 2


To establish the address of a coding section :

• Specify a base address from which the assembler can compute displacements

• Assign a base register to contain this base address

• Write the instruction that loads the base register with the base address

Note: The base address should remain in the base register throughout the execution of the program

Assembler/Session 2


Establishing Base Register

The USING and DROP assembler instructions enable one to use expressions representing implicit addresses as operands of machine instruction statements, leaving the assignment of base registers and the calculation of displacements to the assembler

USING - Use Base Address Register

- allows one to specify a base address and assign one or more base registers

Assembler/Session 2


To use the USING instruction correctly, one should know :

• which locations in a coding section are made addressable by the USING statement

• where in a source module you can use these established addresses as implicit addresses in instruction operands

Format:

symbol USING base address,basereg1| basereg2|,..

e.g. USING BASE,9,10,11

USING *,12

Assembler/Session 2


Range of a USING instruction:• The range of a USING instruction is the 4096

bytes beginning at the base address specified in the USING instruction

Domain of a USING instruction• The domain of a USING instruction begins

where the USING instruction appears in a source module to the end of the source module

Assembler/Session 2


The assembler converts implicit address references into their explicit form:

• if the address reference appears in the domain of a USING instruction

• if the addresses referred to lie within the range of the same USING instruction

Guideline:

• Specify all USING instructions at the beginning of the source module

• Specify a base address in each USING instruction that lies at the beginning of each control section

Assembler/Session 2

ObjectivesRELATIVE ADDRESSINGRELATIVE ADDRESSING

• Relative addressing is the technique of addressing instructions and data areas by designating their location in relation to the location counter or to some symbolic location

ALPHA LR 3,4

CR 4,6 ALPHA+2 or BETA-4

BCR 1,14

BETA AR 2,3

Note : Always avoid using relative addressing

Assembler/Session 2


SESSION 3 & 4

Machine Instructions

Assembler/Session 3 & 4

ObjectivesHANDLING CHARACTER DATAHANDLING CHARACTER DATA

Move Character Instruction (MVC)

• Copy data from one place in memory to another

Format : MVC operand1,operand2

S1(L), S2 - implicit

D1(L,B1),D2(B2) - explicit

e.g...

MVC INPUT(5),OUTPUT



Move Immediate Instruction (MVI)

• Can move only one byte of constant data to a field

Format : MVI operand1,operand2

S1,I2 - implicit

D1(B1),I2 - explicit

e.g..

MVI CTL,C’B’

DBSS TRAINING CENTREDBSS TRAINING CENTRE



Advanced Techniques

1. Explicit lengths and relative addressing

MVC PAD+6(4),=CL4’ ‘

PAD DS CL10

2. Overlapping fields and the MVC instruction

MVC FLDB,FLDA

FLDS DC C’A’

FLDB DS CL3



Comparison Instructions

• Compares 2 values - the values are found in fields, in registers or in immediate data

CLC - Compare logical character

e.g. CLC FLDA,FLDB

CLI - Compare logical immediate

e.g. CLI FLDA,C’K’


ObjectivesExercise 2 Q1 and Q2

2. What will be the effect of the following instructions :

MVI OUTAREA,C’ ‘

MVCOUTAREA+1(132),OUTAREA

OUTAREA DS 133C


ObjectivesBINARY INSTRUCTIONSBINARY INSTRUCTIONS

Three types of binary instructions

•Full word

•Half word

•Register

The Binary Move Instructions

L, LH, LR ,ST, STH

Type : R,X Register and indexed storage

e.g... L 5,FULL LR 5,7

STH 7,HALF



Note : Do not mix up the instruction types and field types

e.g.

LH 5,FULL - right half of Reg 5 gets 1st 2 bytes at FULL

L 6,HALF - Reg 6 gets 4 bytes starting from HALF

ST 3,RES - 4 bytes of reg 3 are stored starting from RES

RES DS H

HALF DC H’15’

FULL DC F’8’



Binary Addition (A, AH and AR)

• Fixed-point overflow occurs when the sum will not fit in the receiving register

• Type R-X

e.g.

A 5,FULL

AH 6,HALF

AR 7,3



Binary Subtraction (S, SH and SR)

• Type R-X

e.g.

S 5,FULL

SH 6,HALF

SR 7,3



Binary comparisons (C, CH and CR)

e.g.

C 5,FULL

CH 6,HALF

CR 7,3

Condition code set as HIGH, LOW or EQUAL


ObjectivesBinary Multiplication (M, MR, MH)

Format : M op1,op2

op1 : An even numbered register; refers to an even-odd pair of registers

(any register in case of halfword format)

op2 : storage area (fullword/halfword/register)


Binary Multiplication (M, MR, MH) ...

Function : The value in OP2 is multiplied by the value in the odd register of the even-odd pair and the result placed in even-odd registers

(For half word format : The half word specified in OP2 is multiplied by the value in OP1 and result stored in OP1.)


Binary Division (D, DR)

Format: D op1,op2

Type : R-X / R-R

Op1 : An even numbered register. It refers to an even-odd pair of registers. The pair holds the double word to be divided. The even register receives the remainder; the odd register receives the quotient.

e.g. D 4,FULL


ObjectivesBRANCHINGBRANCHING

A branch causes execution to continue at some other instruction in the program

• Branch conditions : B, BH, BL, BE, BNH, BNL, BNE, BZ, BNZ, BM, BNM, BO, BNO

e.g : CLI FLDA,C’K’

BNL GOOD


ObjectivesCONDITION CODE PROCESSINGCONDITION CODE PROCESSING

• condition code occupies 2 bits of PSW

• condition code is set by each of a number of instructions

• condition code is an extremely important intermediary between arithmetic instructions and conditional branch instructions

• very important in implementing control structures

0 Zero

1 < Zero

2 >Zero

3 Overflow


ObjectivesBC and BCR Instructions

• instructions that do or do not branch depending on the value of the condition code

Format : BC M1,S2

BCR M1,R2

e.g. BC B’1001’,BRPTA

will cause a branch to the instruction named BRPTA, if at the time the instruction is executed, the condition code is 0 or 3.


ObjectivesBIT MANIPULATIONSBIT MANIPULATIONS

Operation S-I S-S R-R R-X

OR OI OC OR O

AND NI NC NR N

Exclusive OR XI XC XR X

e.g... OI FLDA,X’0F’

NR 5,7

X 9,FULL



Testing individual bits - Test under mask (TM)

TM S1,I2

Function : The bits of S1 ( a single byte) are tested under the control of the mask in I2 and condition code is set as ‘all zeroes’, all ones’ or ‘mixed’

e.g. TM EMP,B’00000101’

BNM NEXT



Bit Shifting Instructions

SLL, SLDL Left logical

SRL, SRDL Right logical

SLA, SLDA Left arithmetic (sign bit not affected)

SRA, SRDA Right arithmetic (& condition code set)

e.g. SLL 5,1

SRDA 4,5


ObjectivesBIT MANIPULATIONS

Translations

• To translate from one bit combination to another

Format : TR S1(L),S2 or S1,S2

S1 : The field whose data is to be translated

S2 : A 256-byte translation table

Function : The value of the original byte is used as a displacement into the translation table. The byte found there replaces the original byte.

e.g. TR WORK,XTABLE


ObjectivesBINARY CONVERSIONSBINARY CONVERSIONS

1. Conversion to binary (CVB)

Format: CVB operand1,operand2

operand1 : Register

operand2 : a double word (containing

valid packed decimal number)

e.g. CVB 5,DOUBLE

Use : character data -(PACK)->packed decimal-(CVB)-> binary


ObjectivesBINARY CONVERSIONSBINARY CONVERSIONS

2. Conversion from binary (CVD)

Format: CVD operand1,operand2

operand1 : Register

operand2 : a double word

e.g. CVD 5,DOUBLE

Use : binary-(CVD)->packed decimal-(UNPK)-> character data


ObjectivesTABLE PROCESSINGTABLE PROCESSING

A table is a named storage structure consisting of subunits or entries

e.g. RATE DS 6F

L 4,RATE+8

Accessing table elements with indexed storage operands:

e.g. LH 9,=H’2’

L 5,RATE(9) (9 - index register)


ObjectivesMulti-purpose branching instructions

Convenient when counted repetition structure (table processing) is needed

• Branch on count (BCT and BCTR)

Format: BCT op1,op2 (R-X)

Function: First the op1 value is decremented by 1. Second the branch is taken to the address specified in op2 only if the value in op1 is not 0.

e.g. LH 9,=H’12’

REPEAT EQU *

..

BCT 9,REPEAT


Objectives• Branch on index high and branch on index low or equal (BXH

and BXLE)

Format: BXLE op1,op2,op3

BXH

op1 : A register known as the index register

op2 : A even-odd pair of registers

Even register - increment register

Odd register - Limit register

op3 : A storage operand. This is the branch addres.


ObjectivesFunction : First, the value in the increment register is added to the indexed register. Second, the branch is taken only when the value in the index register is ‘lower than or equal to’ / ‘higher than’ the value in the limit register

Useful when the same register is to be used as the count and index register


ObjectivesBXLE - ‘DO UNTIL’ repetitions

BXH- ‘DO WHILE’ repetitions

e.g... LH 7,=H’0’ index

LH 2,=H’2’ increment amount

LH 3,=H’18 the limit

---

REPEAT ...

LH 6,TABLE(7)

...

BXLE 7,2,REPEAT


ObjectivesLoad instructions with additional features

• Load and Test (LTR)

e.g... LTR 15,15

BNZ ERROR

• Load Address (LA)

LA R1,D2(X2,B2)


ObjectivesUSING EQUATESUSING EQUATES

• To associate a fixed value with a symbol

• Useful for length and relative address calculation

e.g. TABLE DS 0H

DC C’01

DC C’02’

...

TBLEND EQU *

TBLSIZE EQU TBLEND-TABLE


ObjectivesUSING EQUATESUSING EQUATES

Can be used for the following purposes:

1. To assign single absolute values to symbols.

2. To assign the values of previously defined symbols or expressions to new symbols, thus allowing you to use different mnemonics for different purposes.

3. To compute expressions whose values are unknown at coding time or difficult to calculate. The value of the expressions is then assigned to a symbol.



SESSION 5

Program Sectioning

Assembler/Session 5

ObjectivesBeginning and End of Source ModulesBeginning and End of Source Modules

• Code a CSECT segment before any statement that affects the location counter

• END statement is required as the last statement in the assembly

Assembler/Session 5

ObjectivesCONTROL SECTIONSCONTROL SECTIONS

•A source module can be divided into one or more control sections

•A control section is the smallest subdivision of a program that can be relocated as a unit

Assembler/Session 5

• At coding time, establish the addressability of each control section within the source module, and provide any symbolic linkages between control sections that lie in different source modules.

• Initiated by using the START or CSECT instruction

CONTROL SECTIONSCONTROL SECTIONS

ObjectivesCONTROL SECTIONSCONTROL SECTIONS

• Any instruction that affects the location counter, or uses its current value, establishes the beginning of the first control section.

Assembler/Session 5

Format of CSECT:

Name Operation Operand

Any symbol CSECT Not required

or blank

Note: The end of a control section or portion of a control section is marked by (a) any instruction that defines a new or continued control section, or (b) the END instruction.

CONTROL SECTIONSCONTROL SECTIONS

ObjectivesDUMMY SECTIONSDUMMY SECTIONS

• A dummy control section is a reference control section that allows you to describe the layout of data in a storage area without actually reserving any virtual storage.

Assembler/Session 5

• Use the DSECT instruction to initiate a dummy control section or to indicate its continuation.

Format of DSECT:

Name Operation Operand

Any symbol DSECT Not required

or blank

DUMMY SECTIONSDUMMY SECTIONS

ObjectivesDUMMY SECTIONSDUMMY SECTIONS

To use a dummy section :

• Reserve a storage area for the unformatted data

• Ensure that this data is loaded into the area at execution time

Assembler/Session 5

• Ensure that the locations of the symbols in the dummy section actually correspond to the locations of the data being described

• Establish the addressability of the dummy section in combination with the storage area

You can then refer to the unformatted data symbolically by using the symbols defined in the dummy section.

DUMMY SECTIONSDUMMY SECTIONS

ObjectivesASMBLY2 CSECT

BEGIN BALR 2,0

USING *,2

... Reg 3 points to dataarea

USING INAREA,3

CLI INCODE,C'A'

BE ATYPE

...

ATYPE MVC WORKA,INPUTA

MVC WORKB,INPUTB

. .

Assembler/Session 5

WORKA DS CL20

WORKB DS CL18

...

INAREA DSECT

INCODE DS CL1

INPUTA DS CL20

INPUTB DS CL18

...

END

ObjectivesAssembler DirectivesAssembler Directives

TITLE : To provide headings for each page of the assembly listing of the source modules.

EJECT : To stop the printing of the assembler listing on the current page, and continue the printing on the next page.

ORG : To reset the location counter

Assembler/Session 5

LTORG : A literal pool is created immediately after a LTORG instruction or, if no LTORG instruction is specified, at the end of the first control section.

PRINT : To control the amount of detail to be printed in the listing of programs.

PRINT NOGEN / GEN

Assembler DirectivesAssembler Directives


SESSION 6

Writing a complete program

Assembler/Session 6

ObjectivesProgram Entry and Exit LogicProgram Entry and Exit Logic

Program entry - Preserve register contents

Program Exit - Restore register contents

Register save area

Always calling program provides a savearea of 18 words long used for storage of registers

Savearea address passed through register 13

Assembler/Session 6

ObjectivesA register save area

Word Address Contents

1 SAV

2 SAV+4 Address of calling program’s save area

3 SAV+8 Address of called program’s save area

4 SAV+12 Contents of Register 14



...


Assembler/Session 6

ObjectivesResponsibilities of called program

Program entry conventions

1.Save contents of registers 0-12,14 & 15 in calling program’s save area

2.Establish base register

3.Store calling program’s save area in the 2nd word of its own save area

Assembler/Session 6

ObjectivesProgram entry conventions (contd..)

4. Store the address of its register savearea in the third word of the calling program’s register save area

(The addresses in the 3d word of save area establish a chain of register save areas. This will be useful in reading the dump when program crashes).

Assembler/Session 6

ObjectivesResponsibilities of called program (contd..)

Program Entry

STM R14,R12,12(R13)

BALR R12,0

USING *,R12

ST R13,SAVOWN+4 store calling programs save area

LR R14,R13

LA R13,SAVOWN Reg 13 contains current prog’s SA ...

ST R13,8(R14)

Assembler/Session 6


Program Exit conventions

1. Restore registers 0-12 and 14

2. Place the address of the save area provided by the calling program in Reg 13

3. Place a return code in the low order byte of register 15 if one is required. Otherwise restore register 15.

Assembler/Session 6


Program Exit

L R13,4(R13)

LM R14,R12,12(R13)

BR R4

Assembler/Session 6

ObjectivesResponsibilities of calling program

1. Register 13 must contain the address of a register save area.

2. Register 15 should be set to the beginning address of the subroutine

L R15,=V(SUBENTRY)

where SUBENTRY is the entry address (usually the CSECT name) of the subroutine

Assembler/Session 6

ObjectivesResponsibilities of calling program (contd...)

3. Register 14 should have the return address

4. Register 1 sould have the address of the parameter list

A BALR instruction stores the address of the next instruction in the calling program into register 14 and transfers control to the called subroutine

BALR R14,R15

Assembler/Session 6

ObjectivesPassing parameters to a subroutine

• The standard interface requires that addresses of parameters be placed in a block of storage, and the address of the block be loaded into register 1 as the subroutine is called

• Both input and output parameters are treated the same way

e.g... ADDS DC A(T)

DC A(U)

DC A(V)

LA R1,ADDS

Assembler/Session 6

ObjectivesR1 Main storage

Addr of parmlist Parmlist parm3

Addr of parm1

Addr of parm2 parm1

Addr of parm3 parm2

Assembler/Session 6

ObjectivesCalled subroutine B may get the second parameter by

L R3,4(,R1)

L R8,0(,R3)

Assembler/Session 6

ObjectivesRegisters with special use

R0 : Contains single word output of a subroutine

R1 : contains the address of an area of main storage that contains addresses of parameters

Assembler/Session 6

ObjectivesRegisters with special use (contd...)

R14 : Contains the return address, the address in the calling routine to which a subroutine should return control when finished

R15 : contains the address of the entry point in the subroutine

R13 : contains the address of an area in which register contents can be stored by a subroutine

Assembler/Session 6

ObjectivesThe subroutine RANDOM

RANDOM STM RR14,R12,12(R13)

BALR R12,0

USING *,R12

L R7,RN

M R6,=F’65541’

ST R7,RN

LR R0,R7

LM R1,R12,24(R13)

BR R14

RN DC F’8193’

Assembler/Session 6

ObjectivesSubroutine RDIGIT

RDIGIT STM R14,R12,12(R13)

BALR R12,0

USING *,R12

ST R13,SAV+4

LA R13,SAV

...

L R15,RANDAD

BALR R14,R15

...

L R13,SAV+4

LM R14,R15,12(R13)

LM R1,R12,24(R13)

BR R14

SAV DS 18F

RANDAD DC A(RANDOM)

Assembler/Session 6

ObjectivesLinkage ConventionsLinkage Conventions

•Program divided into 2 or more source modules

•Source module divided into 2 or more control sections

•For link-editing, a complete object module or any individual control section of the object module can be specified

Assembler/Session 6

ObjectivesCommunicating between program parts

• To communicate between 2 or more source modules, symbolically link them together

• To communicate between 2 or more control sections within a source module, establish proper addressability

Assembler/Session 6

ObjectivesEstablishing symbolic linkage

• Identify external symbols in the EXTRN or WXTRN instruction or the V-type address constant

• provide A-type or V-type address constants to reserve storage for addresses represented by external symbols

• In the external source modules, identify these symbols with the ENTRY instruction

(name entry of a START or CSECT instruction is automatically identified as an entry symbol)

External symbol dictionary

Assembler/Session 6

ObjectivesEstablishing symbolic linkage (contd...)

e.g. program A

EXTRN TABLEB

WXTRN TABLEB

TABADR DS V(TABLEB)

program B

ENTRY TABLEB

TABLEB DS ...

Assembler/Session 6

ObjectivesAddress Constants (A and V)

• An address constant is a main storage address contained in a constant

• A V-type constant is the value of an external symbol - a relocatable symbol that is external to the current control section.

Used for branching to locations in other control sections

e.g L 5,ADCON

ADCON DC A(SOMWHERE)GSUBADDC V(READATA)

Assembler/Session 6


SESSION 7

Assemble and Link Program

Assembler/Session 7

ObjectivesProcessing of Instructions

Time/ M/C Assembler ENTRY Macro

Activity instruc. EXTRN Instr.

Code source m/c DC,DS

instruc.

Preassembly Refer to macro

instruc.

Assembly object code

LKED

Prog fetch

Execution data area form data

area in load mod

Processing of Instructions

Time/ M/C Assembler ENTRY Macro

Activity instruc. EXTRN Instr.

Code source m/c DC,DS

instruc.

Preassembly Refer to macro

instruc.

Assembly object code

LKED

Prog fetch

Execution data area form data

area in load mod

Assembler/Session 7

ObjectivesJCL ‘ parm’ processing

EXEC PGM=pgmname,PARM=

When program gets control :

•Register 1 contains the address of a fullword on a fullword boundary in program’s address space

•the high order bit of this fullword is set to 1

(this convention is to indicate the last word in a variable length parameter list)

Assembler/Session 7

JCL ‘ parm’ processing ...• Bits 1-31 of the fullword contain the address of a 2-byte length field on a halfword boundary

• The length field contains a binary count of the no. of bytes in the PARM field which immediately follows the length field

ObjectivesCOBOL to Assembler

CALL asmpgm USING COMM-AREA

PL/I to Assembler

DCL ASMSUB ENTRY OPTIONS(ASSEMBLER)

CHARSTRING CHAR(25);

CALL ASMSUB(CHARSTRING);

Ref : PL/I Programming Guide, COBOL programming Guide

Assembler/Session 7


SESSION 8

Macro Language

Assembler/Session 8

ObjectivesMacros

• Short source routines written and stored in libraries

•Assembler inserts the source statements in the program where the macro appears

Assembler/Session 8

Macro Definition

Format :

•A header statement

•A prototype

•Model statements

•A trailer statement

ObjectivesHeader statement:

MACRO

Prototype:

&name MOVE &TO,&FROM,&LENGTH

Model statements:

A set of machine and assembler instructions

Trailer statement:

&name MEND

Assembler/Session 8

ObjectivesMacro Instruction:

• A statement containing the name of a macro

• when expanded, the symbolic parameters in the model statements are replaced by corresponding parameters from the macro instructions

• symbolic prarameters may be positional or keyword

Assembler/Session 8

Macro Instruction ...MACRO

&LABEL HALFSWAP &REG,&SV

&LABEL ST &REG,&SV

SLL &REG,8

IC &REG,&SV

SLL &REG,8

IC &REG,&SV+1

MEND

ObjectivesSET Symbols (global or local)

3 types :

• arithmetic (SETS)

• binary (SETB)

• character (SETC)

• SET symbols are declared using,

LCLA LCLB LCLC

GCLA GCLB GCLC

Assembler/Session 8

ObjectivesFormat:

Label operation operands

symbol-name SETA An expression

SETB

SETC

e.g.

LCLA &A1

GCLA &A2

&A1 SETA 1

&A2 SETA &A1+3

Assembler/Session 8

ObjectivesAttributes

There are 6 attributes of a symbol or symbolic parameter :

type, length, scaling, integer, count and number

System variable symbols

&SYSINDX, &SYSDATE, &SYSTIME, &SYSECT, &SYSPARM, &SYSLOC

Assembler/Session 8

ObjectivesConditional Assembly

The assembler can be made to branch and loop among assembler language statements using sequence symbols and the assembler instructions AIF and AGO

Sequence symbol : Period followed by 1 to 7 alphabets or digits of which the first is a letter

e.g. .Z23Ab

Assembler/Session 8

ObjectivesFormat:

Label Operation Operand

seq symbo AGO seq. symbol or blank

-do- AIF A logical expression

enclosed in parenthesis,

followed by seq symbol

Assembler/Session 8

A logical expression is composed of one or more relations or values of SETB symbols connected by logical connects AND, OR, AND NOT, OR NOT

A relation consists of 2 arithmetic expressions or 2 character expressions connected by a relational operator EQ, NE, LT, LE, GT, GE

Objectives e.g.

MACRO

PSRCH &PARAMS,&STRING

GBLB &FOUND

LCLA &I

&FOUND SETB 0

.LP AIF ((&I GE N’&PARAMS) OR &FOUND) .E

&I SETA &I+1

&FOUND SETB (‘&PARAMS(&I)’ EQ ‘&STRING’)

AGO .LP

.E MEND

Assembler/Session 8


SESSION 9

Other Topics

Assembler/Session 9

ObjectivesCharacteristics of good assembler program

• has simple, easy to understand logic

• uses mostly simple instructions

• has no relative addressing

• uses subroutines

Assembler/Session 8

Characteristics of good assembler program ...

• uses DSECTs

• has efficient code (LA R10, 4(0,R10 - A R10,=F’4)

• does not abnormally terminate due to user error

• requests and check feedback from macro instructions

• provides meaningful error messages

ObjectivesCharacteristics of good assembler program (contd..)

• lets the assembler determine lengths

• has opcodes, operand and comments aligned

• contains meaningful comments

• uses meaningful labels

Assembler/Session 8

ObjectivesStructured Programming

• To improve design and understandability of a program

• made up of building blocks of subroutines

Conventions for general purpose registers

• Base registers

• Link registers

Assembler/Session 8

ObjectivesThe EXecute Instruction

• the EX instruction is a R-X type instruction that directs the execution of an instruction called the subject instruction, which is addressed by the second operand

• the subject instruction is in effect a one-instruction subroutine

Assembler/Session 9

•The subject instruction is modified before execution (though not altered at its main storage location) : bits 8-15 of the instruction ORed with bits 24-31 of register R1 to form the second byte of the instruction actually executed

e.g. Let reg 9 have the length of string to be moved

EX R9,VARMVC

VARMVC MVC A(0),B

The EXecute Instruction (contd...)

ObjectivesDEBUGGINGDEBUGGING

Exceptions and Interrupts

Interrupts that result directly from attempts at invalid program execution are called program-check interrupts; identified by a code

Interruption code 1 : Operation

Interruption code 2 : Privileged operation

Interruption code 4 : Protection

Interruption code 5 :Addressing

Interruption code 6 :Specification

Assembler/Session 9

ObjectivesDEBUGGING

Exceptions and Interrupts (contd..)

Interruption code 7 : Data

Interruption code 8 : Fixed-Point Overflow

Interruption code 9 : Fixed-Point Divide

Other Interruption codes ( 3, 10, 11, 12, 13, 14, 15)

Assembler/Session 9


Reading dumps

• whenever a program abends an indicative dump is generated

• The completion code is a code furnished by the O/S to designate the reason for the termination of the job step

• In case of program check interruption, the first 2 digits of the completion code is 0C

Assembler/Session 9

• Locate the entry point of your program

Reading dumps ...

DEBUGGINGDEBUGGING


Reading dumps (contd...)

• The register contents are the contents at the point of interruption (the instruction that caused the interrupt is usually the one just before the interrupt address given)

• use address at interrupt and entry address to locate the instruction that caused the program-check interruption

Assembler/Session 9


Full and Partial dumps

• //SYSUDUMP DD SYSOUT=A

• SNAP macro

Assembler/Session 9

Reading the dump

• SAVE AREA trace

• P/P Storage

• Examine register contents, PSW and listed entry point to find the portion of program being executed

• Look at main storage dump to determine the data being used

DEBUGGINGDEBUGGING

ObjectivesSYSTEM MACROSSYSTEM MACROS

Data Management Macros

DCB - Construct a data control block

OPEN - Logically connect a dataset

CLOSE - Logically disconnect a dataset

GET - Obtain next logical record (queued access)

PUT - Write next logical record (queued access)

READ - Read a block (basic access)

WRITE - Write a block (basic access)

Assembler/Session 9


Supervisor Services Macros

ABEND - Abnormally terminate a task

CALL - Pass control to a control section

GETMAIN - Allocate virtual storage

FREEMAIN - Free virtual storage

LOAD - Bring a load module into virtual storage

RETURN - return control to the calling program

SAVE - Save register contents

Assembler/Session 9


Supervisor Services Macros (contd)

SNAP - Dump virtual storage and continue

LINK - Pass control to a Program in Another load module

WTO - Write to operator

Assembler/Session 9


e.g. File I/O

OPEN (INFILE,INPUT)

GET INFILE,RECAREA

PUT OUTFILE,RECAREA

CLOSE (INFILE)

INFILE DCBDSORG=PS,MACRF=GM,DDNAME=IFILE

OUTFILE DCBDSORG=PS,MACRF=PM,DDNAME=OFILE

(RECFM=,LRECL=,BLKSIZE=,)

Assembler/Session 9


Three forms :

Standard form : Results in instructions that store into an inline parameter list and pass control to the required program

List form : Provides asn out-of-line parameter list

Execute form : Provides the executable instructions required to modify the out-of-line parameter list and pass control to the required program

Assembler/Session 9

Thank youThank you

asm-ppt

Documents

statement assemblersession

decimal assemblersession

c assemblersession

objectives objectives

fraction assemblersession

topics assemblersession

apostrophes assemblersession

blanks assemblersession