advertisement
i. iceenuKraTBH
JtlCR«CMH»IITEKS
MICIWCOMPUTEKS jm cm co m piitek s
Publication N um ber 6500-50A
MCS6500
MICROCOMPUTER FAMILY
PROGRAMMING MANUAL
JANUARY 1976
The information in this manual has been reviewed and is believed to be entirely reliable. However, no responsibility is assumed for inaccuracies. The material in this manual is for informational purposes only and is subject to change without notice.
Second Edition
© M O S TECHNOLOGY, INC. 1976
“ All Rights Reserved”
MOS TECHNOLOGY, INC.
950 Rittenhouse Road
Norristown, PA 19401
TABLE OF CONTENTS
CHAPTER 1 INTRODUCTORY REMARKS
1.0 Manual Introduction......................................... 1
1.1 Microprocessor Architecture................................. 2
CHAPTER 2 THE DATA BUS, ACCUMULATOR AND ARITHMETIC UNIT
2.0 The Data B u s ............................................... 3
2 .1 The Accumulator............................................. 4
2.1.1 LDA— Load Accumulator with Memory..........................4
2.1.2 STA— Store Accumulator in Memory ........................ 5
2.2 The Arithmetic Unit......................................... 6
2.2.1 ADC— Add Memory with Carry to Accumulator.................. 7
2.2.1.0 Multiple Precision Addition............................. 8
2.2.1.1 Signed Arithmetic.................................... 10
2.2.1.2 Decimal A d d i t i o n ....................................13
2.2 .1.3 Add Summary....................................... 14
2.2.2 SBC— Subtract Memory from Accumulator 14
2.2.2.0 Multiple Precision Subtraction ...................... 16
2.2.2.1 Signed Arithmetic.................................... 18
2.2.2.2 Decimal Subtract....................................19
2.2.3 Carry and Overflow During Arithmetic
2.2.4 Logical Operands .
2.2.4.1 AND— "AND" Memory with Accumulator .................. 20
2.2.4.2 ORA— "OR" Memory with Accumulator.................... 21
2.2.4.3 EOR— "Exclusive OR" Memory with Accumulator.......... 21
CHAPTER 3 CONCEPTS OF FLAGS AND STATUS REGISTER
3.0 Carry Flag ( C ) ........ ........................... .
3.0.1 SEC— Set Carry Flag.................................... 24
3.0.2 CLC— Clear Carry Flag.................................. 25
3.1 Zero Flag (Z)............................................ 25
24
3.2 Interrupt Disable (I).................................... 25
3.2.1 SEI— Set Interrupt Disable ............................ 26
3.2.2 CLI— Clear Interrupt Disable .......................... 26
3.3 Decimal Mode Flag (D)................ ................... 26
3.3.1 SED— Set Decimal Mode.................................. 26
3.3.2 CLD— Clear Decimal Mode................................ 27
3.4 Break Command (B)........................................ 27
3.5 Expansion Bit............................................ 27
3.6 Overflow ( V ) ............................................ 27
3.6.1 CLV— Clear Overflow F l a g .............................. 28
3.6.2 Determination of Overflow.............................. 28
3.7 Negative Flag (N)........................................ 29
3.8 Flag S u m m a r y ............................................ 30
CHAPTER 4 TEST, BRANCH AND JUMP INSTRUCTIONS
4.0 Concepts of Program Sequence ............................ 31
4.0.1 Use of Program Counter to Fetch
4.0.2 JMP— Jump to New Location.............................. 36
4.1 Branching................................................ 37
4.1.1 Basic Concept of Relative
4.1.2 Branch Instructions.................................... 40
4.1.2.1 BMI— Branch on Result Minus..........................40
4.1.2. 2 BPL— Branch on Result P l u s ..........................40
4.1.2.3 BCC— Branch on Carry Clear .......................... 40
4.1.2.4 BCS— Branch on Carry S e t ............................40
4.1.2.5 BEQ— Branch on Result Z e r o ..........................41
4.1.2.6 BNE— Branch on Result Not Z e r o ......................41
4.1.2.7 BVS— Branch on Overflow Set.......................... 41
4.1.2.8 BVC— Branch on Overflow Clear........................ 41
4.1.3 Branch S u m m a r y ........................................ 42
4.1.4 Solution to Branch Out of Range........................ 42
4.2 Test Instructions........................................ 45
4.2.1 CMP— Compare Memory and Accumulator.................... 45
4.2.2 Bit Testing............................................ 47
4.2.2.1 BIT— Test Bits in Memory with Accumulator. ........ 47
CHAPTER 5 NON-INDEXING ADDRESSING TECHNIQUES
5.0 Addressing Techniques.................................... 50
5.1 Concepts of Pipelining and Program Sequence.............. 52
5.2 Memory Utilization ...................................... 56
5.2.1 1/0 Control............................................ 56
5.2.2 Memory Allocation...................................... 57
5.3 Implied Addressing ...................................... 57
5.4 Immediate Addressing .................................... 59
5.5 Absolute Addressing...................................... 59
5.6 Zero Page Addressing.................................... 61
5.7 Relative Addressing...................................... 63 iii
CHAPTER 6 INDEX REGISTERS AND INDEX ADDRESSING CONCEPTS
6.0 General Concept of Indexing.............................. 69
6.1 Absolute Indexed ........................................ 79
6.2 Zero Page Indexed........................................ 81
6.3 Indirect Addressing...................................... 83
6.4 Indexed Indirect Addressing.............................. 85
6.5 Indirect Indexed Addressing.............................. 87
6.6 Indirect Absolute........................................ 92
6.7 Application of I n d e x e s .................................. 92
CHAPTER 7 INDEX REGISTER INSTRUCTIONS
7.0 LDX— Load Index Register X from M e m o r y .................. 96
7.1 LDY— Load Index Register Y from M e m o r y .................. 96
7.2 STX— Store Index Register X in Memory.................... 97
7.3 STY— Store Index Register Y in Memory.................... 97
7.4 INX— Increment Index Register X by O n e .................. 97
7.5 INY— Increment Index Register Y by O n e .................. 97
7.6 DEX— Decrement Index Register X by O n e .................. 98
7.7 DEY— Decrement Index Register Y by O n e .................. 98
7.8 CPX— Compare Index Register X to Memory.................. 99
7.9 CPY— Compare Index Register Y to Memory.................. 99
7.11 TAX— Transfer Accumulator to Index X .................. 100
7.12 TXA— Transfer Index X to Accumulator.................. 100
7.13 TAY— Transfer Accumulator to Index Y .................. 101
7.14 TYA— Transfer Index Y to Accumulator.................. 101
7.15 Summary of Index Register Applications and 102
CHAPTER 8 STACK PROCESSING
8.0 Introduction to Stack and to Push Down .
103
8.1 JSR— Jump to Subroutine................................ 104
8.2 RTS— Return from Subroutine............................ 108
8.3 Implementation of Stack in MCS6501 Through 112
8.3.1 Summary of Stack Implementation........................115
8.4 Use of the Stack by the Programmer.................... 116
8.5 PHA— Push Accumulator on S t a c k ........................ 117
8.6 PLA— Pull Accumulator from S t a c k ...................... 118
8.7 Use of Pushes and Pulls to Communicate
Subroutine Operations ............................... 119
8.8 TXS— Transfer Index X to Stack P o i n t e r .................. 120
8.9 TSX— Transfer Stack Pointer to Index X .................. 122
8.10 Saving of the Processor Status ....... 122
8.11 PHP— Push Processor Status on Stack...................... 122
8.12 PLP— Pull Processor Status from Stack.................... 123
8.13 Summary on the S t a c k .................................. 123 iv
CHAPTER 9 RESET AND INTERRUPT CONSIDERATIONS
9.0 Vectors.................................................. 124
9.1 Reset or R e s t a r t ........................................ 125
9.2 Start F u n c t i o n .......................................... 126
9.3 Programmer Considerations for Initialization Sequences . . 127
9.4 Restart.................................................. 129
9.5 Interrupt Considerations . .............................. 129
9.6 RTI— Return from Interrupt.............................. 132
9.7 Software Polling for Interrupt
9.8 Fully Vectored Interrupts................................ 140
9.8.1 JMP Indirect........................ .................. 141
9.9 Interrupt Summary........................................ 142
9.10 Non-Maskable Interrupt .................................. 142
9.11 BRK— Break C o m m a n d ...................................... 144
9.12 Memory M a p .................. .......................... 146
CHAPTER 10 SHIFT AND MEMORY MODIFY INSTRUCTIONS
10.0 Definition of Shift and R o t a t e .......................... 147
10.1 LSR— Logical Shift Right ................................ 148
10.2 ASL— Arithmetic Shift L e f t .............................. 149
10.3 R0L— Rotate L e f t ........................................ 149
10.4 ROR— Rotate Right........................................ 150
10.5 Accumulator Mode Addressing.............................. 150
10.6 Read/Modify/Write Instructions ......................... 151
10.7 INC— Increment Memory by O n e ...................... ..
10.8 DEC— Decrement Memory by O n e ...................... ..
154
155
10.9 General Note on Read/Modify/Write Instructions . . . . . . 155
CHAPTER 11 PERIPHERAL PROGRAMMING
11.0 Review of MCS6520 for 1/0 Operations.................... 156
11.1 MCS6520 Interrupt Control................................ 158
11.2 Implementation Tricks for Use of the MCS6520 Peripheral
Interface Devices . ................................. 161
11.2.1 Shortcut Polling Sequences .................... ...... 161
11.2.2 Bit Organization on MCS6520s .......................... 162
11.2.3 Use of READ/MODIFY/WRITE Instruction for Keyboard
Encoding................................... ....... 163
11.3 MCS6530 Programming.............................. . 166
11.3.1 Reading of the Counter Register................ .
11.4 How to Organize to Implement Coding.............. .
11.4.1 Label Standards........................................ 168
11.5 Comprehensive 1/0 Program................................ 170
166
166 v
APPENDICES
A. Instruction List, Alphabetic by Mnemonic, Definition of Instruction Groups......................... ....... A-l
MCS6501 - MCS6505 Microprocessor Instruction Set—
Alphabetic Sequence................................... A-2
A.l Introduction................................ ....... A-3
A.2 Group One Instructions...................... ....... A-3
A. 3 Group Two Instructions.............................. A-4
A. 4 Group Three Instructions.................... ....... A-5
B. Instruction List, Alphabetic by Mnemonic, with OP CODEs,
C. Instruction Addressing Modes and Related Execution Times. .
C-l
D. Operation Code Instruction Listing Hexidecimal Sequence .
D-l
E. Summary of Addressing Modes
E.l Implied Addressing.......................... ....... E-2
E.2 Immediate Addressing........................ ....... E-3
E.3 Absolute Addressing........................ ....... E-3
E.4 Zero Page Addressing........................ ....... E-4
E.5 Relative Addressing.......... ............. .. E-4
E.6 Absolute Indexed Addressing ........................ .. . .
E.7 Zero Page Indexed Addressing................ .. E-6
E.8 Indexed Indirect Addressing ................ . . . . E-7
E.9 Indirect Indexed Addressing ................ . . . . E-8
F. MCS650X Programming Model ......................... . . . . F-l
G. Discussion— Indirect Addressing ......................... G-l
H. Review of Binary and Binary Coded Decimal Arithmetic. .
. H-l vi
I n
L b
LIST OF EXAMPLES
CHAPTER 2 THE DATA BUS, ACCUMULATOR AND ARITHMETIC UNIT
2.1 Add 2 Numbers with Carry; No Carry Generation............ 7
2.2 Add 2 Numbers with Carry; Carry Generation .............. 8
2.3 Adding Two 16-Bit Numbers......................... .. 9
2.4 Add Two 16-Bit Numbers, No Carry from Low Order 9
2.5 Add Two 16-Bit Numbers, with Carry from Low Order
2.6 Add 2 Positive Numbers with No Overflow........... ..
2.7 Add 2 Positive Numbers with O v e r f l o w ................... 12
11
2.8 Add Positive and Negative Number with Positive Result. .
12
2.9 Add Positive and Negative Number with Negative Result. .
12
2.10 Add 2 Negative Numbers without Overflow................. 12
2.11 Add 2 Negative Numbers with O v e r f l o w ................... 13
2.12 Decimal Addition ....................................... 13
2.13 Subtract 2 Numbers with Borrow; Positive Result.......... 15
2.14 Subtract 2 Numbers with Borrow; Negative Result.......... 16
2.15 Subtracting Two 16-Bit Numbers ......................... 16
2.16 Subtract in Double Precision Format; Positive Result . . . 17
2.17 Subtract in Double Precision Format; Negative Result: . . . 18
2.18 Decimal Subtraction.............................. ....... 19
2.19 Clearing a Bit with "AND"........... ........... ....... 21
2.20 Setting a Bit with "OR".......................... ....... 21
2.21 Complementing a Byte with "EOR".................. .......
CHAPTER 4 TEST, BRANCH AND JUMP INSTRUCTIONS
4.1 Accessing Instructions with the P-Counter Value......... 33
4.2 Accessing Data Addressing with P-Counter Value ......... 34
4.3 Use of JMP Instruction................................. 36
4.4 Illustration of "Branch on Carry Set"................... 38
4.5 Sequencing Two Branch Instructions ..................... 39
4.6 Use of JMP to Branch Out of Range....................... 43
4.7 Using the CMP Instruction............................... 46
4.8 Sample Program Using the BIT Test....................... 48 vii
CHAPTER 5 NON-INDEXING ADDRESSING TECHNIQUES
5.1 Using Absolute Addressing................................ 51
5.2 Demonstration of "Pipelining" Effect .................... 54
5.3 Illustration of Implied Addressing ...................... 58
5.4 Illustration of Immediate Addressing .................... 59
5.5 Illustration of Absolute Addressing...................... 60
5.6 Illustration of Zero Page Addressing.................... 62
5.7 Illustration of Relative Addressing; Branch Not .
63
5.8 Illustration of Relative Addressing; Branch
Taken, No Crossing of Page Boundaries................64
5.9 Illustration of Relative Addressing; Branch Negative
Taken, Crossing of Page Boundaries....................65
CHAPTER 6 INDEX REGISTERS AND INDEX ADDRESSING CONCEPTS
6.1 Moving Five Bytes of Data with Straight Line Code........ 70
6.2 Moving Five Bytes of Data with Loop...................... 72
6.3 Coded Detail of Moving Fields with Loop.................. 73
6.4 Moving Five Bytes of Data with Index Register............ 76
6.5 Moving Five Bytes of Data by Decrementing the Index
Register. .......................... .......... .. 77
6.6 Absolute Indexed; with No Page Crossing............ .
6.7 Absolute Indexed; with Page C r o s s i n g .............. ..
6.8 Illustration of Zero Page Ind e x i n g ................ ..
6.9 Demonstrating the Wrap-Around............................ 83
79
80
82
6.10 Illustration of Indexed Indirect Addressing........ .. 86
6.11 Indirect Indexed Addressing (No Page Crossing) . . ...... 88
6.12 Indirect Indexed Addressing (with Page Crossing) . ...... 89
6.13 Absolute Indexed Add— Sample Program ........... ....... 90
6.14 Indexed Indirect Add— Sample Program ........... ....... 90
6.15 Move N Bytes (N < 2 5 6 ) .......................... .
6.16 Move N Bytes (N > 2 5 6 ) .......................... .
94
95
CHAPTER 8 STACK PROCESSING
8.1 Basic Stack Map for 3-Deep JMP to Subroutine............ 104
8.2 Basic Stack Operation. .................................. 105
8.3 Illustration of JSR Instruction.......................... 106
8.4 Illustration of RTS Instruction.......................... 109
8.5 Memory Map for RTS Instruction.......................... Ill
8.6 Expansion of RTS Memory Map.............................. Ill
8.7 Call-a-Move Subroutine Using Preassigned Memory Locations. 116
8.8 Operation of PHA, Assuming at 0 1 F F ................ 118
8.9 Operation of PLA Stack from 8.8.................. 119
8.10 Call-a-Move Subroutine Using the Stack to Communicate. .
. 119
8.11 Jump to Subroutine (JSR) Followed by Parameters.......... 121 viii
CHAPTER 9 RESET AND INTERRUPT CONSIDERATIONS
9.1 Illustration of Start Cycle.............................. 127
9.2 Interrupt Sequence ...................................... 131
9.3 Return from Interrupt.................................... 133
9.4 Illustration of Save and Restore for Interrupts...........133
9.5 Interrupt Polling........................................ 137
9.6 Illustration of JMP Indirect............................ 141
9.7 Break-Interrupt Processing .............................. 145
9.8 Patching with a Break Utilizing PROMs.................... 146
CHAPTER 10 SHIFT AND MEMORY MODIFY INSTRUCTIONS
10.1 General Shift and R o t a t e ................................ 147
10.2 Rotate Accumulator Left. ................................ 150
10.3 Rotate Memory Left Absolute,X............................ 151
10.4 Move a New BCD Number into F i e l d ........................ 154
CHAPTER 11 PERIPHERAL PROGRAMMING
11.1 The MCS6520 Register M a p .......................... .. 156
11.2 General PIA Initialization........................ ..
11.3 Interrupt Mode S e t u p .............................. ..
157
159
11.4 CA2; CB2 Output Control. .......................... ..
11.6 Polling the MCS6520................................ ..
11.7 Coding for Strobing an 8 x 8
159
11.5 Routine to Change CB1 or CB2 Using
161
164
11.8 Polling for Active Signal.......................... . 172
rr
LIST OF FIGURES
CHAPTER 2 THE DATA BUS, ACCUMULATOR AND ARITHMETIC UNIT
2.1 Partial Block Diagram of MCS650X............................. 3
2.2 Partial Block Diagram Including Arithmetic of M C S 6 5 0 X .............................................. 6
2.3 Byte Orientation with Sign Po s i t i o n ........................ 11
CHAPTER 3 CONCEPTS OF FLAGS AND STATUS REGISTER
3.1 Partial Block Diagram of MCS650X Including
3.2 Processor Status Register, "P"............................. 24
CHAPTER 4 TEST, BRANCH AND JUMP INSTRUCTIONS
4.1 Partial Block Diagram of MCS650X Including Program
Counter and Internal Address Bus ...................... 31
4.2 Use of Conditional T e s t .................................... 37
CHAPTER 5 NON-INDEXING ADDRESSING TECHNIQUES
5.1 Address Bus and Relation to Memory Field.................... 54
5.2 Example of Timing— MCS650X Family ......................... 55
CHAPTER 6 INDEX REGISTERS AND INDEX ADDRESSING CONCEPTS
6.1 Moving Five Bytes of Data with L o o p ........................ 72
6.2 Moving Five Bytes of Data with Counter...................... 75
6.3 Partial Block Diagram of MCS650X Including 78
6.4 Indirect Addressing— Pictorial Drawing...................... 84
6.5 Indexed Indirect Addressing ............................... 85
6.6 Indirect Indexed Addressing ............................... 87
CHAPTER 8 STACK PROCESSING
8.1 Partial Block Diagram of MCS650X Including Stack
Pointer, S ......................................... 113
CHAPTER 10 SHIFT AND MEMORY MODIFY INSTRUCTIONS
10.1 Flow Chart for Moving in a New BCD N u m b e r .............. 153
CHAPTER 11 PERIPHERAL PROGRAMMING
11.1 Keyboard Encoding Matrix Program........................ 163
11.2 Keyboard Strobe Sequence................................ 165
11.3 Program Flow-Polling for Active Signal.................. 171
I
CHAPTER 1
INTRODUCTORY REMARKS
1.0 MANUAL INTRODUCTION
Welcome to the MCS650X product family. This manual is designed to work in conjunction with the Hardware Manual which describes the basic hardware considerations when using the MOS Technology, Inc. microcomputer family.
Before reading this manual, it is suggested that the reader acquaint himself with the Hardware Manual in order to understand the components available in this system, how these their basic architecture. Developed in this manual is the concept of microprocessor internal architecture and how it is used, with attention given to input/output considerations. Familiarity with the hardware will facilitate easier understanding of these important concepts.
In order to best serve the total customer base, this manual is written in two levels. The first is a very ily, and the second level is for the user who has to refer to the manual on more than an occasional basis and who wants to rapidly scan and find specific sections. For the user who is quite familiar with programming and the MCS650X instruction set, the appendices are the best reference in the sense that all the data which is discussed in detail in the manual is summarized in a series of tables for convenience.
It is recommended that the user who is an experienced programmer and familiar with microprocessors still take the time to read through the manual in detail. Some of the architectural concepts are different from those found in second generation machines and this manual instructs the user how to optimize the utilization of the microprocessor while providing an introduction of its basic concepts.
1
Criticism of this manual is welcomed at all times. Of particular interest are cases where one could not, by use of the index and appendix, rapidly find the answer to a question which developed in the course of designing a microprocessor system. Welcomed are any comments which will enhance the content and format of this manual in future editions or adden- dums.
1.1 MICROPROCESSOR ARCHITECTURE
The MCS6501, MCS6502, MCS6503, MCS6504, and MCS6505 are all 8-bit microprocessors. That means that 8 bits of data are transferred or oper ated upon during each instruction cycle or operation cycle.
All devices in the MCS650X family operate on data 8 bits at a time, although some of the operations will look like serial or 16-bit wide oper ations. In a future section, discussed will be the use of sequential operations on an 8-bit basis and how one can accomplish 16-bit effective operands and addressing.
The computer industry, for some time, has been treating 8-bit combina tions of data by a term known as a "byte." In many large computers which operate simultaneously on multiple bytes of data, the number of bytes which are transferred and operated on by the machine in parallel are called a
"word." Because these microprocessors are 8-bit microprocessors, the words and bytes are of equal length. Therefore, for convenience through the dis cussion of the basic 8-bit processors, "byte" and "word" will be used synonymously although in some of the expanded versions there will exist a
16-bit word composed of two 8-bit bytes.
2
CHAPTER 2
THE DATA BUS, ACCUMULATOR AND ARITHMETIC UNIT
2.0 THE DATA BUS
Although most of the following discussion will consider how one operates with a general purpose register called the accumulator, it must be understood that data has to transfer between the accumulator and out side sources by means of passing through the microprocessor to 8 lines called the data bus. The outside sources include the program which con trols the microprocessor, the memory which will be used as interim stor age for internal registers when they are to be used in a current opera tion, and the actual communications to the world through input/output ports. Later in this document performance of transfers to and from each of these devices will be discussed. However, at present, discussion will center on the microprocessor itself.
Partial Block Diagram ofMCS650X
FIGURE 2.1
The only operation of the data bus is to transfer data between mem ory and the processor's internal registers such as the accumulator. Fig ure 2.1 displays the basic communication between the accumulator, A, and the memory, M, through the use of 8 bi-directional data lines called the data bus.
3
2.1 THE ACCUMULATOR
The accumulator is a register in which data is kept on which opera tions are performed. All operations between memory locations must be communicated through the accumulator or one of the auxiliary index reg isters. The accumulator is used as a temporary storage in moving data from one memory location to another. Therefore, the first use for the accumulator (A) is just in transferring data from memory to the accumu lator or from the accumulator to memory. One can bring data into the accumulator, perform operations such as AND/OR on it, test the results of those operations, set new bits into it, or transfer it back out to the outside world. It serves as an interim storage for a series of oper ations such as adding 2 values together; where one of them is loaded into the accumulator, the second one added to it, and the results stored in the accumulator. The accumulator really acts as two functions: 1) It is one of the primary storage points for the machine; 2) It is the point at which intermediate results are normally stored.
2.1.1 LDA— Load Accumulator with Memory
When instruction LDA is executed by the microprocessor, data is transferred from memory to the accumulator and stored in the accumulator.
Rather than continuing to give a word picture of the opera tion, introduced will be the symbolic representation M -* A, where the arrow means "transfer to." Therefore the LDA instruction sym bolic representation is read, "memory transferred to the accumulator."
LDA affects the contents of the accumulator, does not affect the carry or overflow flags; sets the zero flag if the accumulator is zero as a result of the LDA, otherwise resets the zero flag; sets the negative flag if bit 7 of the accumulator is a 1, other wise resets the negative flag.
Although yet to be developed is the concept of addressing modes, for reference purpose, LDA is a "Group One" instruction and has all of the major addressing modes of the machine available to it as stated in Appendix A. These addressing modes include Immed iate; Absolute; Zero Page; Absolute,X; Absolute,Y; Zero Page,X;
Indexed Indirect; and Indirect Indexed.
4
2.1.2 STA— Store Accumulator In Memory memory.
This instruction transfers the contents of the accumulator to
The symbolic representation for this instruction is A -* M.
This instruction affects none of the flags in the processor status register and does not affect the accumulator.
It is a "Group One" instruction and has the following address ing modes available to it: Absolute; Zero Page; Absolute,X; Abso lute,Y; Zero Page,X; Indexed Indirect; and Indirect Indexed.
5
2.2 THE ARITHMETIC UNIT
One of the functions to be expected from any computer is the ability to compute or perforin arithmetic operations. Even in a simple control problem, one often finds it useful to add 2 numbers in order to determine that a value has been reached, or subtract 2 numbers to calculate a new value which must be obtained. In addition, many problems involve some rudimentary form of decimal or binary arithmetic; certainly many applica tions of the microprocessor will involve both. The MCS650X has an 8-bit arithmetic unit which interfaces to the accumulator as shown in Figure 2.2
Partial Block Diagram including Arithmetic Logic Unit o f MCS650X
FIGURE 2.2
The arithmetic unit is composed of several major parts. The most important of these is the circuitry necessary to perform a two's comple ment add of 8-bit parallel values and generate an 8 parallel bit binary result plus a carry. A review of binary and binary coded decimal (BCD) arithmetic is presented in Appendix H. However, a quick review of the concept of "carry" is in order. The largest range than can be repre sented in an 8-bit number is 256 with values ranging between 0 and 255.
If we add any 2 numbers which result in a sum which is greater than 255, we represent the result with a ninth bit plus the 8 bits of the excess over 255. The ninth bit is called "carry."
6
2.2.1 ADC— Add Memory to Accumulator with Carry
This instruction adds the value of memory and carry from the previous operation to the value of the accumulator and stores the result: in the accumulator.
The symbolic representation for this instruction is
A + M + C A.
This instruction affects the accumulator; sets the carry flag when the sum of a binary add exceeds 255 or when the sum of a decimal add exceeds 99, otherwise carry is reset. The overflow flag is set when the sign or bit 7 is changed due to the result exceeding +127 or -128, otherwise overflow is reset. The negative flag is set if the accumulator result contains bit 7 on, otherwise the negative flag is reset. The zero flag is set if the accumulator result is 0, otherwise the zero flag is reset.
It is a "Group One" instruction and has the following address ing modes: Immediate; Absolute; Zero Page; Absolute,X; Absolute,Y;
Zero Page,X; Indexed Indirect; and Indirect Indexed.
The ninth bit of the result is stored in the carry flag and the remaining 8 bits reside in the accumulator. The carry flag can be thought of as a flag bit which is remote from the accumulator it self but which is directly affected by accumulator operations as though it were a ninth bit in the accumulator. The primary reason for not viewing the carry bit as merely a ninth bit in the accumu lator is that one has program control over its state by being able to set (to "1") or clear (to "0") the bit and, of course, it is not part of the 8-bit accumulator in data transfer operations. Examples employing the Add with Carry operation follow.
__
Carry = /0/
0000 1101 13 = (A)*
1101 0011 211 = (M)*
__________ 1 _____1 CARRY
1110 0001 225 = (A)
* (A) and (M) refer to the "contents" of the accumulator and
"contents" of memory respectively.
Example 2.2: Add 2 numbers with carry; carry generation
Carry = flf
1111
0000
0000
1110
0110
___ 1
0101
254 = (A)
6 = (M)
5 = (A)
While the accumulator contains "5," the carry flag signals the user that the result exceeded 255 and, therefore, the result can be properly interpreted as 256 + 5 = 261.
2.2.1.0 Multiple Precision Addition
To perform the addition of 2 numbers, one issues to the microprocessor an ADC instruction which adds the memory and the accu mulator and stores the results in the accumulator with the carry bit going set if the results exceeded 255.
To add numbers which had significantly higher value than
255, it would be necessary to represent these numbers by a series of serial 8-bit numbers. With the 16 bits in 2 serial 8-bit numbers, it is possible to represent binary numbers of greater than 65,000 in value. In order to add two 16-bit numbers together and thus accomplish double precision addition, one first loads the lowest byte of one number into the accumulator, clears the carry flag and then adds the second number to the first number in the accumulator using the ADC command. One would then store this result into another memory loca tion using the STA command. The carry flag would now represent the , carry from the lowest byte to the highest byte. One could then load the high order byte of the first number, add with carry again to the high value of the second number, and store the result in the high order byte of the result. Thus, it can be seen that the carry allows us to perform as much precision arithmetic as is necessary. The example listing below displays the commands used to execute the addi tion of two 16-bit numbers.
8
Example 2.3: Adding two 16-bit numbers
High Order Byte
First Number
Second Number
HI
H2
Result of Addition H3
LI LDA
CLC
ADC L2
Low Order Byte
LI
L2
L3
Load low order byte, first number
Clear carry flag (carry = 0)
Add Ll to low order byte, second num ber
STA L3
LDA
ADC
STA
HI
H2
H3
Store result in memory, carry flag is still set if set in ADC operation
Load high order byte, first number
Add HI and carry value from first ADC operation to high order byte, second number
Store result in memory
In this example it was necessary to clear the carry flag before starting the add instruction. This, of course, means that commands exist that set and clear the carry flag allowing for addi tion without values generated from the prior operation. One could also, at the end of the program, check to see if the result exceeded
16 bits by testing the carry flag. Exactly how one alters and tests flags will be discussed in the Flag and Branches Section. The examples below display the concept of carry from the addition of the low order bytes.
Example 2.4: Add two 16-bit numbers, no carry from low order add
0000 0001 0000 0010 258
0001 0000 0001 0000 4112
Add low order bytes: (clear carry)
0000 0010 (A)
__ 0001 0000 (M)
Carry = /0/ 0001 0010 (A)
Add high order bytes (carry = 0):
0000 0001 (A)
0001 0000 (M)
__ __________ 0 CARRY
Carry = _/0/ 0001 0001 (A)
Result = 0001 0001 0001 0010 = 4370
9
Example 2.5: Add two 16-bit numbers, with carry from low order add
0000 0001 1000 0000
0000 0000 1000 0000
384
128
Add low order bytes: (clear carry)
1000 0000
__ 1000 0000 (M)
Carry = /1/ 0000 0000 (A)
(A)
Add high order bytes: (carry = 1)
0000 0001
0000 0000
__ __________ 1
Carry = [0/ 0000 0010 (A)
(A)
(M)
CARRY
Result = 0000 0010 0000 0000 = 512
2.2.1.1 Signed Arithmetic
It is possible to look at the add operation and the way data is represented in memory in a different way. If, in the 16-bit problem (Examples 2.4 and 2.5), one were working with 15 bits of pre cision (in other words, 15 bits of valid data) plus 1 bit of sign (0 for positive and 1 for negative), it would be possible to perform tate this concept, the microprocessor has the ability to represent positive or negative numbers by means of a sign flag which will be discussed at length in Section 3.7. In the MCS650X family, bit 7 is the sign position bit. This means that the highest order byte in a series of bytes should have the sign in the eighth position. If, for simplicity, one talks about signed 8-bit numbers, it would mean that one was allowed only 128 combinations of each sign because that is the most that can be represented in 7 bits, with the eighth bit or the highest bit reserved for the sign position.
^BIT POSITION
T
7 6 5 4 3 2 1 0
SIGN POSITION
Byte Orientation with Sign Position
FIGURE 2.3
“T = NEGATIVE
“O’ = POSITIVE
In the following examples of signed arithmetic it should be noted that operations are occurring on a 7-bit field of numbers and that any carry generated out of that field will reside in the eighth bit— not in the carry flag discussed during the add operations. The generation of a carry out of the field is the same as when adding two 8-bit numbers, except for the fact that the normal carry flag does not correctly represent the fact that the field has been ex ceeded,, This is because the true carry from adding the two 7-bit numbers resides in the sign bit position. Therefore, the carry flag has no real meaning. Instead, there is a separate flag, the over flow flag, used to indicate when a carry from 7 bits has occurred and allows the user to write correction programs.
In each example, the negative numbers are in two's comple ment form. Also included in each result will be the status of the carry and overflow flags. The overflow flag is set whenever the sign bit (bit 7) is changed as a result of the operation.
0000 0101
__ 0000 0111
+5
+7
(A)
(M)
Carry = /0/ 0000 1100 +12 (A)
Overflow = /0/ "0" in bit 7 indicates positive result.
Note that both the carry and overflow flag remain cleared.
11
Example 2.7: Add 2 positive numbers with overflow
0111 1111 +127 (A)
0000 0010 + 2 (M)
Carry = /0/ 1000 0001 "-127" (A)
Overflow = /1/ "1" in bit 7 indicates negative result and the two's complement of the result is 127; however, the overflow flag is set indicat ing the allowable range was exceeded in the addition.
Therefore, examination of the overflow indicated that the result was in fact not negative but that the bit 7 position represented an over flow beyond the value of 127. Hence the user is flagged of an incor rect result and a correction routine (program) must follow.
Example 2.8: Add positive and negative number with posi tive result
0000 0101 +5 (A)
__ 1111 1101 ^3 (M)
Carry = /I/ 0000 0010 +2 (A)
Overflow = /0/ "0" in bit 7 indicates positive result.
(Recall that though the carry flag is set, it has no meaning in signed operations.)
Example 2.9: Add positive and negative number with negative result
0000 0101 +5 (A)
__ 1111 1001 ^7 (M)
Carry = /£/ 1111 1110 -2 (A)
Overflow = /0/ "1" in bit 7 indicates negative result.
Example 2.10: Add 2 negative numbers without overflow
1111 1011 -5 (A)
__ 1111 1001 -7 (M)
Carry = /l/ 1111 0100 -12 (A)
Overflow = /0/ "1" in bit 7 indicates negative result.
12
Example 2.11: Add 2 negative numbers with overflow
1011 1110 -66 (A)
_ 1011 1111 -65 (M)
Carry = /l/ 0111 1101 "+125" (A)
Overflow = /!/ "0" indicates positive result, but the overflow flag is set indicating that the allowable range was exceeded in the opera tion. Without the overflow indication, the result would be interpreted as +125. The overflow, however, indicated that the result was negative and exceeded the value -128.
Hence the user is flagged of an incorrect result, indicating the need for a correc tion routine.
2.2.1.2 Decimal Addition
There is a way for the user to organize data for decimal operations. The MOS Technology, Inc. MCS650X microprocessors have a modified adder which allows the user to represent his numbers as two 4-bit binary coded decimals (BCD) numbers packed into a single byte. This is a unique feature of the MCS650X family in that the operation in the following example can be performed.
Example 2.12: Decimal addition
CLC
SED
LDA 0111 1001
Clear Carry Flag
Set Decimal Mode
79
ADC 0001 0100 +14
STA 1001 0011 93
The microprocessor adder has the unique capability of per forming real time correction to the normal expected binary result without any direct interference from the programmer. Other popular microprocessors require a separate instruction (Decimal Adjust) which corrects the direct binary result of the arithmetic unit to obtain the same final results as are available on this microprocessor directly.
In order to make the same arithmetic unit perform either as a binary adder or as a decimal adder, the user chooses the mode in which he is going to operate (either decimal or binary) by setting
13
another flip-flop in the microprocessor called the decimal flag. As shown in this example, one not only initializes the adder by clearing the carry flag, but also puts the processor into decimal mode with the SED instruction. Even though this also requires 1 instruction, it is possible to put the machine in decimal mode once and perform many long strings of decimal numbers without further user interven tion. The "Decimal Adjust" feature on other microprocessors requires programming subsequent to each binary operation.
2.2.1.3 Add Summary
In summary, the basic arithmetic unit is a binary adder which, under control of the ADC command, performs binary arithmetic on the accumulator and data, storing the result in the accumulator.
Depending on the way the user looks at the data which is presented to the adder and the results which are obtained from it, the user can determine whether or not the result exceeds 255 binary or 99 decimal; he can perform precision arithmetic by use of the ninth bit or carry flag; he can control whether or not the microprocessor is a decimal adder by setting the decimal mode; and he can represent his numbers as signed binary numbers by analyzing other flags that are set in the machine.
2.2.2 SBC Subtract Memory from Accumulator with Borrow
This instruction subtracts the value of memory and borrow from the value of the accumulator, using two's complement arithmetic, and stores the result in the accumulator. Borrow is defined as the carry flag complemented; therefore, a resultant carry flag indicates that a borrow has not occurred.
The symbolic representation for this instruction is
A - M - C ■* A.
This instruction affects the accumulator. The carry flag is set if the result is greater than or equal to 0. The carry flag is reset when the result is less than 0, indicating a borrow. The over flow flag is set when the result exceeds +127 or -127, otherwise it
14
is reset. The negative flag is set if the result in the accumulator has bit 7 on, otherwise it is reset. The Z flag is set if the result in the accumulator is 0, otherwise it is reset.
It is a "Group One" instruction. It has addressing modes
Immediate; Absolute; Zero Page; Absolute,X; Absolute,!; Zero Page,X;
Indexed Indirect; and Indirect Indexed.
In a binary machine, the classical way to perform arithmetic
In using two's complement notation, any subtraction operation becomes a sequence of bit comple mentations and additions. This reduces the complexity of the circuits required to perform a subtraction.
When the SBC instruction is used in single precision subtrac therefore, the programmer must set the carry flag, by using the SEC (Set carry to 1) instruc tion, before using the SBC instruction. The microprocessor adds the carry flag to the complemented memory data, resulting in a true two's complement form of the memory value with its sign inverted.
Example 2.13: Subtract 2 numbers with borrow; positive result
Assume a single precision subtraction where A contains 5 and M con tains 3. The carry flag must be set to a 1 using the SEC instruc tion, thereby representing the no-borrow condition.
The adder changes the sign of M by taking the two's complement of M. This involves complementing M and adding the carry bit.
M = 3 0000
Complemented M 1111
Add C = 1
-M = -3
0011
1100
_____ _ 1 _
The adder adds A and the two's complement -M together. This operation occurs simultaneously with the complement operation.
A = 5 0000 0101
Add -M = -3_ 1111 1101
Carry = /I/ 0000 0010 = +2
The presence of the carry flag after this operation indicates that No Borrow was required, therefore the result is +2.
15
Example 2.14: Subtract 2 numbers with borrow; negative result
Assume a single precision subtraction where A contains 5 and M con tains 6. Set the carry flag to a 1 with SEC to indicate No Borrow.
M = 6
Complemented M
0000
1111
0110
1001
Add C = 1 1
-M = -6 1111 1010 2
1 S Cot wr*
A = 5 0000 0101
Add -M = -6 1111 1010
Carry - JQJ 1111 1111 = -1
The absence of the carry flag after this operation indicates that a borrow was required, therefore the result is a -1 in two's complement form. The absolute (unsigned) result in straight binary could be obtained by taking the two's complement of this number.
2.2.2.0 Multiple Precision Subtraction
Double precision subtraction is implemented in a fashion similar to addition. An example for subtracting a 16-bit number and storing the result follows:
Example 2.15: Subtracting two 16-bit numbers
First Number
Second Number
Result of Subtraction
SEC Set Carry
LDA
SBC
LI
L2
STA
LDA
SBC
STA
L3
HI
H2
H3
High Order Byte Low Order Byte
HI
H2
H3
LI
L2
L3
Load Low Order Byte, First Number
Subtract with Borrow, Low Order Byte of Second
Number from LI
Store Result in Memory
Load High Order Byte, First Number
Subtract with Borrow, High Order Byte of Second
Number from HI
Store Result in Memory
16
Example 2.16: Subtract in double precision format; positive result
Assume a double precision subtraction where 255 is to be subtracted from 512 for an example. Since there has been no borrow coming into this subtraction operation, the carry flag must be set.
Following are the 2 numbers in binary form:
High Order Byte Low Order Byte
A field = 512
M field = 255
0000
0000
0010
0000
0000
1111
0000
11.11
Since the adder can only operate on single byte, numbers, the programmer must operate on the low order bytes first.
M = 1111
Complemented M = 0000
Add C = 1 __________ 1
-M 0000
1111
0001
A = 0000
Add -M_= 0000 0001
Carry = /0/ 0000 0001
0000
The carry is brought over to the subtract operation on the high order bytes.
M = 0000
1111
Add C = 0 ___________0
-M 1111
A = 0000
Add -M_= 1111 1111
Carry = /l/ 0000 0001
The result in binary form follows:
0000
1111
0010
Carry = j\J 0000 0001 0001 +257
The presence of the carry flag after the highest order byte subtraction indicates that the entire number required No Borrow, therefore it is a positive number in straight binary form.
17
Example 2.17: Subtract in double precision format; negative result
Now assume a double precision subtraction where 512 is to be
Folio wing are the two numbers in binary form:
= 255
M fi Bid = 512
High Order Byte
0000 0000
0000 0010
Oper ating on the low order byte:
0000 0000
The
The
Add C = 1 larry = /!/
1111 1111
1
0000 0000 dd -M = /!/ larry = /!/
1111 1111
0000 0000
1111 1111 the carry = 1 brought over tract operati an:
Add C = 1
0000 0010
1111 1101
1
1111 1110
Low Order Byte
1111
0000
1111
0000
0000 0000
1111 1110
Carry = /0/ 1111 1110
The result in binary form is:
Carrkr = /07 1111 1110 1111 1111 = -257
Carry = /0/ indicates the presence of a borrow, therefore the number is negative and is in two's complement form.
2.2.2.1 Sign ed Arithmetic just as easily as they can be added. The microprocessor converts the numbers from memory to its two's complemented form and then adds it to the value of the accumulator just as it does in an unsigned
18
subtract described in Section 2.2.2. The addition operation is identical to that described, and to the examples given in Section
2.2.1.1
It should be remembered that before using the SBC instruc tion, either signed or unsigned, the carry flag must be set to a 1 in order to indicate a no borrow condition. The resultant carry flag has no meaning after a signed arithmetic operation.
2.2.2.2 Decimal Subtract
As indicated in the Section 2.2.1.2, it is possible to repre sent numbers as packed 4-bit BCD numbers. In this case, which is again unique to this microprocessor, it is possible to make the adder act as though it is a decimal adder. In this case, the function of the machine is one of correcting for the subtraction of positive num bers by complementing the number, setting the carry and performing binary arithmetic with an automatic correction at the time the result is stored in the accumulator. The unique capabilities of this adder give the results as shown in the next example.
Example 2.18: Decimal Subtraction
SED
SEC
LDA
SBC
STA
0100
0010
0001
0100
1001
0101
Set Decimal Mode
Set Carry Flag
44
29
15
By setting the decimal mode and setting the carry flag, one can sub tract number 29 from number 44 with the results in the accumulator automatically being 15.
As has been indicated, one can perform both addition and subtraction when the machine is set in decimal mode, treating the bytes to be added as unsigned, positive, binary coded digits. The carry flag in addition represents the case when the result in the number exceeded 99 and in subtraction the absence of the carry flag represents a true borrow situation.
2.2.3 Carry and Overflow During Arithmetic Operations
It is necessary to set or reset the carry flag prior to the beginning of any arithmetic instruction. Because the carry flag is set or reset as a result of the arithmetic operation at the end of the loop, one can test the flag to determine whether or not a carry or a borrow occurred in the operation. By proper use of the overflow flag one can treat the high order bit of any set of bytes as a sign bit as long as the results of the negative numbers are carried in two's complement form. The microprocessor also sets the overflow flip-flop to indicate when a result larger than can be stored in a
7-bit field has occurred and when the resultant sign is incorrect.
In binary arithmetic the carry flag set indicates results in excess of 256, and in decimal arithmetic indicates results in excess of 99.
Although the input carry is very important to these operations, a simple rule is: set the carry flag prior to subtract; clear the carry flag prior to add.
2.2.4 Logical Operands
In implementing a parallel binary adder there are several use ful logic functions which are subsets of a binary add operation. In the MCS650X family, these subsets are used to implement the logical operands "AND," "OR," and "EOR" (Exclusive Or). These operations are used to test and control bit manipulations.
2.2.4.1 AND— Memory with Accumulator r
The AND instructions transfer the accumulator and memory to the adder which performs a bit-by-bit AND operation and stores the result back in the accumulator.
This instruction affects the accumulator; sets the zero flag if the result in the accumulator is 0, otherwise resets the zero flag sets the negative flag if the result in the accumulator has bit 7 on, otherwise resets the negative flag.
This is symbolically represented by A A M -»■ A.
20
AND is a "Group One" instruction having addressing modes of
Immediate; Absolute; Zero Page; Absolute,X; Absolute,Y; Zero Page,X;
Indexed Indirect; and Indirect Indexed.
One of the uses for the AND operation is that of resetting a bit in memory. In the example below,
LDA 1100 Xlll, where X is 0 or 1
AND 1111 0111
STA 1100 0111 a byte is loaded into the accumulator and the AND instruction resets the accumulator bit 3 to 0. The accumulator is then stored back into memory, thereby resetting the bit.
2.2.4.2 ORA "OR" Memory with Accumulator
The ORA instruction transfers the memory and the accumulator to the adder which performs a binary "OR" on a bit-by-bit basis and stores the result in the accumulator.
This is indicated symbolically by A
V
M A.
This instruction affects the accumulator; sets the zero flag if the result in the accumulator is 0, otherwise resets the zero flag; sets the negative flag if the result in the accumulator has bit 7 on, otherwise resets the negative flag. ORA is a "Group One" instruction.
It has the addressing modes Immediate; Absolute; Zero Page; Absolute,X
Absolute,Y; Zero Page,X; Indexed Indirect; and Indirect Indexed.
To set a bit, the OR instruction is used as shown below:
Example 2.20: Setting a bit with OR
LDA 1110 Xlll, where X is 0 or 1
ORA 0000 1000
STA 1110 1111
2.2.4.3 EOR— "Exclusive OR" Memory with Accumulator
The EOR instruction transfers the memory and the accumulator to the adder which performs a binary "EXCLUSIVE OR" on a bit-by-bit basis and stores the result in the accumulator.
This is indicated symbolically by A ^ M -* A.
This instruction affects the accumulator; sets the zero flag if the result in the accumulator is 0, otherwise resets the zero flag sets the negative flag if the result in the accumulator has bit 7 on, otherwise resets the negative flag.
EOR is a "Group One" instruction having addressing modes of
Immediate; Absolute; Zero Page; Absolute,X; Absolute,?; Zero Page,X;
Indexed Indirect; and Indirect Indexed.
One of the uses of the EOR instruction is in complementing bytes. This is accomplished below by exclusive ORA-ing the byte with all l ’s.
Example 2.21: Complementing a byte with EOR
LDA 1010 1111
EOR 1111 1111
STA 0101 0000
22
CHAPTER 3
CONCEPTS OF FLAGS AND STATUS REGISTER
One can view each of the individual flags or status bits in the machine as individual flip-flops. The carry flag can be considered the ninth bit of an arithmetic operation. The decimal mode flag is set and cleared by the user and used by the microprocessor to select either binary or decimal mode. For programming convenience the microprocessor treats all of the flags or status bits as component bits of a single 8-bit reg ister. In Figure 3.1 the processor status register (or "P" register) is added to the block diagram.
DATA BUS
C=^
PROCESSOR
REGISTER
Partial Block Diagram o f MCS650X including P Register
FIGURE 3.1
23
Each of the individual flags or bits has its own particular meaning in the microprocessor as defined in Figure 3.2.
N V B D I Z C PROCESSOR STATUS REGISTER
CARRY
ZERO RESULT
INTERRUPT DISABLE
DECIMAL M O D E
BREAK C O M M A N D
EXPANSION
OVERFLO W
NEGATIVE RESULT
Processor Status Register
FIGURE 3.2
3.0 CARRY FLAG (C)
The carry bit which is<modified as a result of specific arithmetic operations or by a set or clear carry command has been discussed previously.
In the case of shift and rotate instruction, the carry bit is used as a ninth bit as it is in the arithmetic operation. The carry flag can be set or reset by the programmer. A SEC'instruction will set and a CLC instruc tion will reset the carry flag. Operations which affect the carry are ADC,
ASL, CLC, CMP, CPX, CPY, LSR, PLP, ROL, :RTT, SBC, sfc. f ~
This instruction initialized'the carry flag ta a 1. This op eration should normally precede a SBC loop. It is also useful when used with a ROL instruction to initialize a bit in memory to a 1.
This instruction affects no registers in the microprocessor and no flags other than the carry fldg which is set. '
SEC is a single-byte instruction and its addressing mode is
Implied.
24
3.0.2 CLC— Clear Carry Flag
This instruction initializes the carry flag to a 0. This op eration should normally precede an ADC loop. It is also useful when used with a ROL instruction to clear a bit in memory.
This instruction affects no registers in the microprocessor and no flags other than the carry flag which is reset.
CLC is a single-byte instruction and its addressing mode is
Implied.
3.1 ZERO FLAG (Z)
This flag is automatically set by the microprocessor during any data movement or calculation operation when the 8 bits of results of the opera tion are 0. Therefore, the bit is on ("1") when the results are 0, and off ("0") when the results are not equal to 0. The feature of the machine is similar to that of the PDP11 in the sense that operations which are decrementing index registers or memory locations have a built-in test for
0 as a result of decrementing to the 0 condition. It is also possible to test for 0 condition immediately following load and other logical opera tions, as opposed to processors which have to do a test and branch instruc tion. The Z flag is not directly settable or resettable by an instruction but is affected by the following instructions: ADC, AND, ASL, BIT, CMP,
CPY, CPX, DEC, DEX, DEY, EOR, INC, INX, INY, LDA, LDX, LDY, LSR, ORA, PLA,
PLP, ROL, RTI, SBC, TAX, TAY, TXA, TYA.
3.2 INTERRUPT DISABLE (I)
The interrupt disable is a flip-flop made use of by the programmer and by the microprocessor to control the operations of the interrupt re quest pin. A more detailed discussion of the effects of the interrupt disable are given in the discussion under interrupt control. However, the purpose of the interrupt disable is to disable the effects of the interrupt request pin. The interrupt disable, I, is set by the microprocessor dur ing reset and interrupt commands. The I bit is reset by the CLI instruc tion or the PLP instruction, or at a return from interrupt in which the interrupt disable was reset prior to the interrupt. The interrupt flag may be set by the programmer using a SEI instruction and is cleared by the
25
programmer by using a CLI instruction. Instructions which affect the interrupt disable are BRK, CLI, PLP, RTI and SEI.
3.2.1 SEI— Set Interrupt Disable
This instruction initializes the interrupt disable to a 1. It is used to mask interrupt requests during system reset operations and during interrupt commands.
It affects no registers in the microprocessor and no flags other than the interrupt disable which is set.
SEI is a single-byte instruction and its addressing mode is
Implied.
3.2.2 CLI— Clear Interrupt Disable
This instruction initializes the interrupt disable to a 0.
This allows the microprocessor to receive interrupts.
It affects no registers in the microprocessor and no flags other than the interrupt disable which is cleared.
CLI is a single-byte instruction and its addressing mode is
Implied.
3.3 DECIMAL MODE FLAG (D)
As discussed, the use of the decimal mode flag is to control whether or not the adder operates as a straight binary adder for add and subtract instructions or as a decimal adder for add and subtract instructions. The
SED instruction sets the flag and the CLD instruction resets it. The only
3.3.1 SED— Set Decimal Mode
This instruction sets the decimal mode flag D to a 1. makes all subsequent ADC and SBC instructions operate as a decimal
This arithmetic operation.
SED affects no registers in the microprocessor and no flags other than the decimal mode which is set to a 1.
3.3.2 CLD— Clear Decimal Mode
This instruction sets the decimal mode flag to a 0. This causes all subsequent ADC and SBC instructions to operate as simple binary operations.
CLD affects no registers in the microprocessor and no flags other than the decimal mode flag which is set to a 0.
3.4 BREAK COMMAND (B)
The break command flag is set only by the microprocessor and is used to determine during an interrupt service sequence whether or not the inter rupt was caused by BRK command or by a real interrupt. A more detailed discussion of BRK is in the interrupt section. This bit should be con sidered to have meaning only during an analysis of a normal interrupt se quence. There are no instructions which can set or which reset this bit.
3.5 EXPANSION BIT
The next bit in the flag register is an unused bit. It is most likely that this bit will appear to be on when one is analyzing the bit pattern in the processor status register; however, no guarantee as to its state is made as this bit will be used in expanded versions of the microprocessor.
3.6 OVERFLOW(V)
As discussed in the section on arithmetic operations, if one is to look at the binary arithmetic operations as signed binary operations, there needs to be some indication of the fact the result of the arithmetic opera tion has a greater value than could be contained in the 7 bits of the re sult. This bit is the overflow bit and during ADC and SBC instructions represents a status of an overflow into the sign position. The user who is not using signed arithmetic can totally ignore this flag during his programming; however, this flag has the same meaning as the carry to correction routine must be used if this bit is on after an add or subtract using signed numbers.
27
In addition to its use to monitor the validity of the sign bit in ADC and SBC instructions, the overflow flag in the MCS650X products is dramat ically changed from PDP11 and the MC6800. In those systems the overflow flag was very carefully controlled so as to allow certain signed branches for analysis of signed numbers. These branches have been deleted from the
MCS6500 series because of confusion and difficulty often associated with using them, and so therefore, the overflow flag is applicable only to the operation of ADC and SBC. and then only when using signed numbers.
However, in order to maximize the effectiveness of this testable flag the BIT instruction which may be used to sample interface devices, allows the overflow flag to reflect the condition of bit 6 in the sampled field.
During a BIT instruction the overflow flag is set equal to the content of the bit 6 on the data tested with BIT instruction. When used in this mode, the overflow has nothing to do with signed arithmetic but is just another sense bit for the microprocessor. Instructions which affect the V flag are
ADC, BIT, CLV, PLP, RTI and SBC. On certain versions of the microproces sor the V bit will also be available for stimulus from the outside world.
3.6.1 CLV— Clear Overflow Flag
This instruction clears the overflow flag to a 0. This com mand is used in conjunction with the set overflow pin which can change the state of the overflow flag with an external signal.
CLV affects no registers in the microprocessor and no flags other than the overflow flag which is set to a 0.
3.6.2 Determination of Overflow
To briefly recap the concept of overflow detection, one must understand that the machine signals an overflow based on the data entered to the operation and the final result. Since, with signed arithmetic, the range of numbers that be represented is +127 to -128, the overflow flag will never set when numbers of opposite sign are added, since their result will never exceed that range. The machine deals with this by recognizing that for any 2 positive numbers, the
"bit 7" of each is a "O" and that for any arithmetic operation
yielding a result less than or equal to +127, the resultant "bit 7" must be a "0." If it is a 1, the overflow flag is set.
Similarly, when two negative numbers are added, the "bit 7" of each is a "1" and for any result yielding a value less than or equal to -128, the resultant "bit" must be a "1." If it is a 0, the over flow flag is set.
Therefore, the machine recognizes by knowledge of the "bit 7" of each of the numbers to be added what the resultant "bit 7" must be in a non-overflow situation. If these conditions are not met, the overflow flag goes set.
3
.7
NEGA TIVE FLA G (N)
As already discussed, one of the uses of the microprocessor is to per form arithmetic operations on signed numbers. To allow the user to readily sample the status of the sign bit (bit 7), the N flag is set equal to bit 7 of the resulting value in all data movement and data arithmetic. This means, for instance, after a signed add one can determine the sign of the result by sampling the N flag directly rather than finding a way to iso late bit 7. Although signs were the primary purpose for which the N flag was intended, its usefulness far exceeds that of strictly a sign bit.
Because of every operation including simple moves and add operations the N bit is equal to the status of bit 7 as a result of the pri mary use becomes that of an easily testable bit. Almost all single-bit in structions, all interrupts and all I/O status flags use bit 7 as a sense bit.
This allows the user to perform some type of memory access operation such as Load A followed by immediate conditional branch based on the status of bit 7 as reflected in the N flag. Like the Z bit, this flag is not settable or controllable by the programmer and represents the status of the last data movement operation. Instructions which affect the negative flag are ADC,
AND, ASL, BIT, CMP, CPY, CPX, DEC, DEX, DEY, EOR, INC, INX, INY, LDA, LDX,
29
3.8 FLAG SUMMARY
To summarize, the microprocessor treats a series of flags or status bits as a single register called the "P" or "Program Status" register.
Some of these flags are controllable only by the programmer (such as the D flag); others are controllable by both the user program and microprocessor
(such as the interrupt disable flag). Some of them are set and reset by almost every processor operation, such as the N and Z flags. Each of these flags has its own meaning to the programmer at a particular point in time.
When combined with the concept of conditional branches, they represent a powerful test and jump capability not normally found in a machine of this magnitude. Other than perhaps the carry flag which is used as part of the arithmetic instructions, the flags by themselves have relatively little meaning unless one has the ability to test them. For this purpose there is a series of conditional branch instructions designed into the machine.
CHAPTER 4
TEST, BRANCH AND JUMP INSTRUCTIONS
4.0 CONCEPTS OF PROGRAM SEQUENCE
In all the discussions up until now, there has been little discus sion about how the microprocessor understands the instructions used to perform various arithmetic and accumulator manipulations. However, it is appropriate that the concept of a program and how the microprocessor determines each instruction be developed. More registers are required in the machine as shown in the figure below.
Partial Block Diagram o f MCS650X Including Program
Counter and Internal Address Bus
FIGURE 4.1
31
Although two 8 bit registers have been added, they are the only registers in the machine that act as though they are one 16 bit reg ister. They implement a concept known as program count or program sequence and subsequently their value will be referred to as PC or program count. In certain operations it may be convenient to talk about how one affects the program count low (PCL) which will be the lower 8 bit register or the program count high (PCH) which will be the higher 8 bit register. The reason for this register being 16 bits in length is that if it had only 8 bits it would only be able to reference
256 locations. Since it is through the address bus that one accesses memory, the program counter which defines the addressable location, should be as wide a word as possible.
The accessing of a memory location is called "addressing". It is the selection of a particular eight-bit data word (byte) out of the
65,536 possibilities for memory data locations. This selection is trans mitted to the memory through the 16 address lines (ADH, ADL) of the micro processor.
For a more detailed discussion of how an individual memory byte is selected by the address lines, the reader is referred to Chapter 1 of the Hardware Manual.
If the program counter was only 1 byte and if the bit pattern which allows the microprocessor to choose which instruction it wants to act on next, such as "LDA" as opposed to an "AND", was contained in one byte of data we could only have 256 program steps. Although the machine of this length might make an interesting toy, it would have no real practical value. Therefore, almost all of the competitive 8 bit microprocessors have chosen to go to a double length program counter. Even though some of the microprocessors of the MCS650X family do not have all of the out put address lines necessary to allow the user to address 65K bytes of program (due to package pinout constraints), in all cases the program counter is capable of addressing a full 65K by virtue of it's 16 bit length.
32
4.0.1 Use of Program Counter to Fetch an Instruction
The microprocessor contains an internal timing and state con trol counter. This counter, along with a decode matrix, governs the operation of the microprocessor on each clock cycle. When the state of the microprocessor indicates that a new instruction is needed, the program counter (program address pointer) is used to choose
(address) the next memory location and the value which the memory sends back is decoded in order to determine what operation the
MCS650X is going to perform next.
To use the program counter to perform this operation cor rectly, it must always be addressing the operation the user wants to perform next. This operation may be an instruction or may be data on which the instruction will operate.
In the MCS650X family, the program counter is set with the value of the address of an instruction. The microprocessor then puts the value of the program counter onto the address bus, trans ferring the 8 bits of data at that memory address into the instruc tion decode. The program counter then automatically increments by one and the microprocessor fetches further data for address operation necessary to complete the instruction. In the simple example below,
P Counter* Location Contents
0100**
0101
0102
LDA
ADC
STA
*Program Counter
**Hexadecimal
Notation one can see how the program counter is used to access the instruc tion sequence load A, add with carry, and store the result. In this example, the program counter would start out containing 0100. The microprocessor would read location 0100 by using the program counter to access memory and would then interpret and implement the LDA in struction as previously described. The program counter will auto matically increment by one on each instruction fetch, stepping to
0101. After performing the LDA, the microprocessor would fetch the
next instruction addressing memory with the program counter. This would pick up the ADC instruction, the add would then be performed, the program counter which has been incremented to 0102 would be used to address the next instruction, STA. The P counter incrementing once with each instruction is an oversimplified view of what actu ally transpires within the microprocessor.
The MCS650X processors usually require more than one byte to correctly interpret an instruction. The first byte of an instruction is called the OP CODE and is coded to contain the basic operation such as LDA (load accumulator with memory) and also the data neces sary to allow the microprocessor to interpret the address of the data on which the operation will occur. In most cases, this address will appear in memory right after the OP CODE byte. This allows the micro processor to use the program counter to access the address as well as the OP CODE.
The following example shows how the program counter picks up the instruction and the address of data located at address 5155.
Example 4.2: Accessing Data Address With P Counter Value
P Counter Location Contents
0100
0101
0102
0103
LDA
55
51
Next Instruction
The OP CODE appears in Location Address 0100. The code for the 55 would appear next in Location Address 0101 and the 51 would appear in Location Address 0102, and the OP CODE for the next instruction appears in Location Address 0103. In this example, we see that the program counter is used not only to pick up the operation code, LDA, but is also used to pick up the address of the memory location from which the LDA is going to obtain its data. In this case, the program counter automatically is incremented three times to pick up the full instruction with the microprocessor interpreting each of the indivi dual fetches as the appropriate data. In other words, the first
34
fetch is used to pick up the OP CODE, LDA, the second fetch is used to pick up the low order address byte of the data and the third fetch is used to pick up the high order address byte of the data. This is the form in which many of the microprocessor instructions will appear as it is the most simple form of addressing in the machine and allows referencing to any memory location.
Assuming that the microprocessor has the ability to start the program counter at a known instruction, it should be fairly obvious that the program counter would then continue to advance from that location' up to the maximum memory location, roll over to the least memory location and continue incrementing through the memory, fetch ing instructions and addresses as it went. This would give us an interesting sequential program but one which lacked one tremendously powerful concept. The program would have no ability to perform tests or implement various options based on the results of those tests.
In the previous section, the concept of flags which are set as a result of the microprocessor operations was developed.
To use these flags, the program should be able to test them and- then change the sequence of operations which are being performed depending on the result of the test. The program counter is going to continually put out an address, the microprocessor is going to fetch the instruction stored at that address and perform operations based on that instruction. In order to change a sequence of perform ed instructions by the microprocessor, the programmer must change the value in the program counter. Therefore, test instructions are in corporated which may result in a change of program count sequence as a result of performing one of the tests. The simplest way to change program sequence is to substitute a new value into the program counter location. In the MCS650X microprocessors the simplest way to change the program count sequence is with a JMP instruction.
4.0.2 JMP— Jump to New Location
In this instruction, the data from the memory location located in the program sequence after the OP CODE is loaded into the low order byte of the program counter (PCL) and the data from the next memory location after that is loaded into the high order byte of the program counter (PCH).
The symbolic notation for jump is (PC + 1)-H?CL, (PC + 2)-H?CH.
As stated earlier, the "( )" means "contents of" a memory location.
PC indicates the contents of the program counter at the time the
OP CODE is fetched. Therefore (PC + 2)->-PCH reads, "the contents of the program counter two locations beyond the OP CODE fetch location are transferred to the new PC high order byte."
The addressing modes are Absolute and Absolute Indirect.
The JMP instruction affects no flags and only PCL and PCH.
The JMP instruction allows use of the program counter to access the new program counter value as illustrated by the following example:
Address
0100
0101
0102
3625
Data Comments
JMP
25
Jump to Location 3625
(New PCL byte)
36 (New PCH byte)
OP CODE Next Instruction
The program counter in the example starts out at location 100. The microprocessor loads a jump instruction. The program counter auto matically increments to 101 where the microprocessor picks up and temporarily stores the 25. The program counter automatically in crements to 102 where the microprocessor picks up the 36.
The 3625 is substituted into the program counter and is used to address the next instruction. Therefore, the JMP instruction contains within its address the new program counter location.
Although the jump allows the change of program sequence, it does so without performing any test. So it is a JMP instruction that is employed when it is desired to change the program counter no matter what conditions have occurred.
36
Another JMP addressing Mode in the Indirect Addressing Mode.
Before this technique can be understood, the basis of indirect addressing found in Chapter 6 must be reviewed. The JMP Indirect instruction is detailed in Chapter 9, page 141.
4.1 BRANCHING
To allow for conditional program sequence change, there are a series of branch instructions which test and perform optional changes of the pro gram counter based on the status of the flags. To perform a conditional change of sequence, the microprocessor must interpret the instruction, test the value of a flag, and then change the P counter if the value agrees with the instruction. If the condition is not met, the program counter continues to increment in its normal fashion. Figure 4.2 illustrates how a conditional test might be used.
Use o f Conditional Test
FIGURE 4.2
37
In this example, it is seen that generation of a carry from the add operation will allow an out-of-sequence branch to a new location.
4.1.1 Basic Concept of Relative Addressing
If one considers that the instruction JMP required three bytes, one for OP CODE, one for new program counter low (PCL) and one for new program counter high (PCH) it is seen that jump on carry set would also require three bytes. Because most programs for con trol require many continual jumps or branches, the MCS650X uses
"relative" addressing for all conditional test instructions. To perforin any branch, the program counter must be changed. In rela tive addressing, however, we add the value in the memory location following the OP CODE to the program counter. This allows us to specify a new program counter location with only two bytes, one for the OP CODE and one for the value to be added.
To illustrate this, in the following example, the branch on carry set (BCS) illustration is followed by a value of 50. If the carry is set, the new program location would be 108 + 50 = 158; in other words, it will take the branch.
Address
0100
0101
0102
0103
0104
0105
0106
0107
0108
0109
010A
0158
Data Comments
LDA
ADL1
ADH1
ADC
ADL2
ADH2
BCS
Load First Value
First Number, low byte
First Number, high byte
Add Second Value
Second Number, low byte
Second Number, high byte
Test for Carry Set. If yes, branch to 0158
+50
STA If not, store results
ADL3
ADH3 of add
Result, low byte
Result, high byte
OP CODE New Instruction
38
r
The 0108 represents the value of the program counter after reading the offset value. The program counter automatically incre ments so it can reference the next memory location on the next cycle.
The add of the offset is a signed binary add as discussed in the arith metic section. A positive branch is indicated by a 0 in bit 7 of the relative value, and a minus branch is in two's complement form and is indicated by a 1 in bit 7. The inherent capabilities of this type of notation system allow branch conditionally forward 127 bytes from the next instruction and back 128 bytes from that instruction. All branches in the MCS650X series are conditional relative branches and all have the form shown above. The advantage of relative ad dressing is best shown in the following example:
Example 4.5: Sequencing Two Branch Instructions
Address
0100
0101
0102
0103
0104
0105
0106
Data
LDA
ADL1
ADH1
ADC
ADL2
ADH2
BCS
Comments
Load First Value
Add Second Value
Test for Carry Set. If yes,branch to 0158
0107
0108
+50
BMI Test for Minus Number.
If yes, branch to 0095
0109
010A
010B
010C
-75
STA
ADL3
ADH3
If not, Store
In this example, the previous single-branch example was modi fied to also test the resulting number to see if it is negative. In sequencing two-branch instructions, this loop is 2 bytes shorter by use of relative branches rather than 3 byte branches.
39
4.1.2 Branch Instructions
4.1.2.1 BMI - Branch on Result Minus
This instruction takes the conditional branch if the N bit is set.
BMI does not affect any of the flags or any other part of the machine other than the program counter and then only if the
N bit is on.
The mode of addressing for BMI is Relative.
4.1.2.2 BPL
This instruction is the complementary branch to branch on result minus. It is a conditional branch which takes the branch when the N bit is reset (0). BPL bit 7 was off (0) and branch the previous result was minus or bit 7 was on (1).
The instruction affects no flags other than the P counter and only affects the P the N bit is reset.
The addressing mode is Relative.
4.1.2.3 BCC
This instruction tests the state of the carry bit and takes a conditional branch if the carry bit is reset.
It affects no flags or registers other than the program counter and then only if the C flag is not on.
The addressing mode is Relative.
4.1.2.4 BCS
This instruction takes the conditional branch if the carry flag is on.
BCS does not affect any of the flags or registers except for the program counter and only then if the carry flag is on.
The addressing mode is Relative.
40
i
F
4.1.2.5 BEQ - Branch on Result Zero
This instruction could also be called "Branch on Equal."
It takes a conditional branch whenever the Z flag is on or the previ ous result is equal to 0.
BEQ does not affect any of the flags or registers other than the program counter and only then when the Z flag is set.
The addressing mode is Relative.
4.1.2.6 BNE - Branch on Result Not Zero
This instruction could also be called "Branch on Not Equal."
It tests the Z flag and takes the conditional branch if the Z flag is not on, indicating that the previous result was not zero.
BNE does not affect any of the flags or registers other than the program counter and only then if the Z flag is reset.
The addressing mode is Relative.
4.1.2.7 BVS - Branch on Overflow Set
This instruction tests the V flag and takes the conditional branch if V is on.
BVS does not affect any flags or registers other than the program, counter and only when the overflow flag is set.
The addressing mode is Relative.
4.1.2.8 BVC - Branch on Overflow Clear
This instruction tests the status of the V flag and takes the conditional branch if the flag is not set.
BVC does not affect any of the flags and registers other than the program counter and only when the overflow flag is reset.
The addressing mode is Relative.
41
4.1.3 Branch Summary
To summarize, the MCS650X branches have two characteristics; each of them tests the state of a flag and then either accesses the next instruction in program sequence if the flag is not in the test state or adds the offset value to the PC value at the OP CODE of the next instruction (PC + 1) to allow the program to change operations.
This allows the programmer the full ability to make decisions. By writing a sequence of branch instructions, any combination of condi tions of the microprocessor may be determined and new action taken as a result of the tests.
There are four branch conditions in the MCS6501-5 micropro cessors. These are branch on carry flag, branch of overflow flag, branch on N flag, and branch on zero flag. Each of the branches has a branch on flag set (1) or branch on flag clear (0).
4.1.4 Solution to Branch Out of Range
The branch relative instruction is unlike the jump instruc tion which can reach anywhere in memory, since branch relative is limited to +127 or -128 from'the current program counter location.
Although for many loops and many tests this is sufficient range, longer programs will occasionally find it necessary to condition ally branch to a location that is significantly further away than the branch command will directly reach. This is one of the uses of complementary branches. If a program should find it necessary to branch to a location which was significantly further away than
127, the following solution would facilitate the branch:
42
Address
100
101
102
103
104
105
106
107
108
Data
LDA
ADL1
ADH1
ADC
ADL2
ADH2
BCC
+3
JMP
Comments
Load First Value
Add Second Value
Branch, if no carry, ahead 3 (to Point 2)
If carry set, jump to location specified by
ADH4, ADL4
Point 2
109
10A
10B
10C
10D
10E
ADL4
ADH4
BMI
Offset
STA
ADL3
ADH3
Check for Minus
If not minus, Store
Result
10F
In this example, carry set is being checked. In order to accomplish this when the branch command would have to reach outside of the 128 range, the use of a complementary branch is required. Instead of doing the "branch on carry set" to the location, the "branch on carry clear" is utilized (a complementary instruction) which branches past the jump. If the complementary branch is not taken, the jump is the "branch on carry set" function.
This technique of branching past a jump with the complementary branch is a universal solution to the branch out of range problem.
Another solution is to find a like branch to the same location that is within range and although this involves two branches to trans fer control, it does save memory locations.
By use of the relative branch less bytes of code are used than if a conditional jump had been used. However, in large programs, the branch out of range occurs more frequently. If the user can determine that a branch will be out of range by inspection, he should use the jump solution at the time he is writing the code. Otherwise, the
43
various assemblers indicate an out of range branch which will require recoding to use the jump solution.
NOTE: The jump solution causes 5 bytes of code to be substituted for 2 bytes of branch which in a symbolic assembly may force other branches to go out of range.
This might cause several consecutive reassemblies but this technique will solve the problem.
4.2 TEST INSTRUCTIONS
Although most of the normal operations of the microprocessor involve setting of flags, there are specific instructions which are designed only to set flags for testing with the branch instruction.
4.2.1 CMP - Compare Memory and Accumulator
This instruction subtracts the contents of memory from the contents of the accumulator.
Its symbolic notation is A - M.
The use of the CMP affects the following flags: Z flag is set on an equal comparison, reset otherwise; the N flag is set or reset by the result bit 7, the carry flag is set when the value in memory is less than or equal to the accumulator, reset when it is greater than the accumulator. The accumulator is not affected.
It is a "Group One" instruction and therefore has as its addressing modes: Immediate; Zero Page; Zero Page,X; Absolute;
Absolute,X; Absolute,Y; (Indirect,X); (Indirect),Y.
The purpose of the compare instruction is to allow the user to compare a value in memory to the accumulator without changing the value of the accumulator. An example of where this becomes extremely important is when one is receiving command instructions from an external device. In this case, an input byte may have several values. Each value can cause the program to perform a different operation. The only rapid way to determine the value of the input data is to compare the memory with a series of constants.
It is fairly simple to perform "compare to constant" operations.
By use of the immediate addressing mode which will be developed later,, the following example compares an input to three values and branches to different locations for each:
Data
LDA
ADL
ADH
CMP
COUNT 1
BEQ
OFFSET 1
CMP
COUNT 2
BEQ
OFFSET 2
CMP
COUNT 3
BEQ
OFFSET 3
Next Inst.
Comments
Load Value
Address Low
Address High
Compare COUNT 1 to Accumulator
If Equal, take the branch of OFFSET 1
Compare COUNT 2 to Accumulator
If Equal, take the branch of OFFSET 2
Compare COUNT 3 to Accumulator
If Equal, take the branch of OFFSET 3
Otherwise, go to Next Instruction based on default value (COUNT 4).
This example shows how to use the default option. A value was compared against 3 values and if none were equal a fourth, or default value, is assumed. This is a useful technique for code minimization.
The compare instruction is designed to allow a signed compari son between 2 values assuming one makes appropriate use of the Z and
N and C flags. In order to give maximum flexibility to the instruc tion, the instruction performs an effective subtract between the value in memory and the value in the accumulator. The reason it is an ef fective subtract is that subtraction allows the user to compare equal or less with one instruction.
The results of a compare are:
Accumulator < Memory
Accumulator = Memory
Accumulator > Memory
N
Either
Reset
Either
Reset
Set
Set
Reset
Set
Reset
Unchanged
Unchanged
Unchanged
So, to check if the accumulator is less than memory, the com pare is followed by a BCC; to check if equal to is followed by a BEQ; and to check if greater it is followed by a BEQ followed by a BCS.
Greater than or equal is checked by BCS.
46
4.2.2 Bit Testing
The comparison instruction is designed for cases when byte or multiple bytes of values are being compared; however, in the analysis of logic functions, it is very often necessary to determine the con dition of an individual bit. One of the ways to accomplish this is with the use of the AND instruction as previously discussed. In other words, the user can load a value into the accumulator and AND it with a field that contains a one bit only in the corresponding bit posi tion to the bit under test. By using a Branch on Zero Flag after the AND, the status of the bit in memory is testable by this tech nique. However, the use of this technique involves destroying the accumulator value with the AND instruction. Therefore, searching a table looking for a single bit in a given position would necessitate the reloading of the test value (mask) after each AND instruction.
In order to allow memory sampling without disturbing the accumulator, the BIT instruction is used.
4.2.2.1 BIT - Test Bits in Memory with Accumulator
This instruction performs an AND between a memory location and the accumulator but does not store the result of the AND into the accumulator.
The symbolic notation is M A A.
The bit instruction affects the N flag with N being set to the value of bit 7 of the memory being tested, the V flag with V being set equal to bit 6 of the memory being tested and Z being set by the result of the AND operation between the accumulator and the memory if the result is Zero, Z is reset otherwise. It does not affect the accumulator.
The addressing modes are Zero Page and Absolute.
The BIT instruction actually combines two instructions from the PDP-11 and MC6800, that of TST (Test Memory) and (BIT Test).
This, like the compare test, allows the examination of an individual bit without disturbing the value in the accumulator and is illustra ted by the example below:
47
Data
LDA
MASK
BIT
ADL1
ADH1
BNE
+50
BIT
ADL2
ADH2
BNE
-75 etc.
Comments
Load MASK into Accumulator
Branch if Set
Test Second Memory Value for Mask Bit
Branch if Set
The value "MASK" loaded into the accumulator in this example is actually a descriptive title since, this byte is 8 bits, only one of which is a 1. Using this byte in the AND operation inherent in the
BIT test will effectively mask out all bits in the memory location under test except that bit position corresponding to the 1 residing in the accumulator. In Example 4.8, the MASK byte is AND'ed to the data found in location ADH1, ADL1 and if the bit under test is a 1, the branch will be taken; if not a 1, the second memory location will be tested with the same mask, etc.
In addition to the nondestructive feature of the bit which allows us to isolate an individual bit by use of the branch equal or branch no equal test, two modifications to the PDP-11 version of that instruction have been made in the MCS650X microprocessor. These are to allow a test of bit 7 and bit 6 of the field examined with the BIT test. This feature is particularly useful in serving polled interrupts and particularly in dealing with the MCS6520 (Peripheral Interface
Device).
This device has an interrupt sense bit in bit 6 and bit 7 of the status words. It is a standard of the M6800 bus that whenever possible, bit 7 reflects the interrupt status of an I/O device. This means that under normal circumstances, an analysis of the N flag after a load or BIT instruction should indicate the status of the bit 7 on the I/O device being sampled. To facilitate this test using
48
the Bit instruction, bit 7 from the memory being tested is set into the N flag irrespective of the value in the accumulator.
This is different from the bit instruction in the M6800 which re quires that bit 7 also be set on the accumulator to set N. The advantage to the user is that if he decides to test bit 7 in the memory, it is done directly by sampling the N bit with a Bit fol lowed by branch minus or branch plus instruction. This means that
I/O sampling can be accomplished at any time during the operation of instructions irrespective of the value preloaded in the accumu lator .
Another feature of the BIT test is the setting of bit 6 into the V flag. As indicated previously, the V flag is normally reserved for overflow into the sign position during an add and subtract in struction. In other words, the V flag is not disturbed by normal instructions. When the BIT instruction is used, it is assumed that the user is trying to examine the memory that he is testing with the
BIT instruction. In order to receive maximum value from a BIT in struction, bit 6 from the memory being tested is set into the V flag.
In the case of a normal memory operation, this just means that the user should organize his memory such that both of his flags to be tested are in either bit 6 or bit 7, in which case an appropriate mask does not have to be loaded into the accumulator prior to imple menting the BIT instruction. In the case of the MCS6520, the BIT instruction can be used for sampling interrupt, irrespective of the mask. This allows the programmer to totally interrogate both bit 6 and bit 7 of the MCS6520 without disturbing the accumulator. In the case of the concurrent interrupts, i.e., bit 6 and bit 7 both on, the fact that the V flag is automatically set by the BIT instruction allows the user to postpone testing for the "6th bit on" until after he has totally handled the interrupt "for bit 7 on" unless he performs an arithmetic operation subsequent to the BIT operation.
49
CHAPTER 5
NON-INDEXING ADDRESSING TECHNIQUES
5.0 ADDRESSING TECHNIQUES
The addressing modes of the MCS6500 family can be grouped into two major categories: Indexed and Non-Indexed Addressing. This section deals with the Non-Indexed mode of addressing. Before detailing the various modes available to the user, several concepts will be reviewed. The first of these is the concept of memory field, address bus and data bus. Then a brief introduction to two non-indexed addressing modes and timing will be made with the intent of preparing the reader for a discussion of program sequence and the internal activity of the microprocessor during execution of an instruction. This will be followed by a review of how one treats memory and the assorted allocation of memory space to the elements of RAM,
ROM and I/O.
Subsequent to reading this section the user should have an understand ing of the following fundamentals: a) Memory Field b) Address Bus c) Data Bus d) Cycle Timing e) Program Sequence f) Pipelining
With these tools in hand, the reader will be better prepared to readily comprehend the detailed definitions of the non-indexed addressing modes.
As discussed in Section 1.1 the MCS650X microprocessor family is organized around a 16-bit address function. All locations are accessed by a 16-bit word, even though in the case of the MCS6503, the MCS6504, and the MCS6505, only 11 or 12 bits are actually utilized.
50
Sixteen bits of address allow access to 65,536 memory locations, each of which, in the MCS650X family, consists of 8 bits of data. Figure 5.1 displays the total memory field and incorporates the concept of address bus and data bus. The memory address can be regarded as 256 pages (each page defined by the high order byte) of 256 memory locations (bytes) per page.
It will be seen in the detailed discussion of addressing that the lowest order page, page zero, has special significance in the minimization of pro gram code and execution time.
Much of the uniqueness of the MCS6500 product family has to do with how the 16-bit address is created. The simplest way to create a 16-bit ad dress is for the programmer to indicate to the microprocessor the 16 bits necessary to access a particular operand on which the microprocessor is ex pected to operate. An instruction consists of 1, 2, or 3 bytes. It always takes 1 byte to specify the operation which is to be performed (OP CODE) .
This OP CODE is then followed by 0, 1, or 2 bytes of address depending on the specific operation involved. In the case of the simple instructions such as transfer accumulator to X, operations are performed internally and, therefore, no additional bytes are necessary. This instruction mode is known as "Implied" in the sense that the instruction contains both the OP
CODE and the source and destination for the operation. This is the simplest form of addressing and applies to only a limited number of the instructions available in the MCS6500 family. Another form of addressing, absolute ad dressing, is the case when the programmer specifies directly to the micro processor the address he wants the microprocessor to use in fetching the memory value on which the operation will occur. This form is illustrated by the example below.
Example 5.1: Using absolute addressing
Clock Cycle Address Bus Data Bus
1
2
3
4
0100
0101
0102
ADH, ADL
LDA, Absolute
ADL
ADH
Data
In this example, memory location 0100 contains the OP CODE "LDA Abso lute." The next location, 0101, contains ADL which will be defined as the
51
"low order byte of the address," hence address low (ADL). Location 0102 contains ADH— the "high order byte of the address," hence address high (ADH).
At the next clock cycle, the 16 bits composed of ADH and ADL are put on the address bus with the location defined by ADH, ADL containing the data to be loaded into the accumulator. The effective address of the data is best described in Figure 5.1, where the 16-bit address (AB00 through AB15) is composed of ADH and ADL.
This is the normal form for an absolute memory address. The byte of the instruction which is picked up by the program counter is the operation code. This is interpreted by the microprocessor as "Load A -
Absolute." At the same time that this Load A is being interpreted by the microprocessor, the microprocessor accesses the next memory location by putting the program counter content, which was incremented as the OP CODE was fetched, on the address bus.
5.1 CONCEPTS OF PIPELINING AND PROGRAM SEQUENCE
The overlap of fetching the next memory location while interpreting the current data from memory minimizes the operation time of a normal 2- or 3-byte instruction and is referred to as pipelining. It is this feature that allows a 2-byte instruction to only take 2 clock times and a 3-byte instruction to be interpreted in 3 clock cycles.
In the MCS650X microprocessors, a clock cycle is defined as 1 complete operation of each of the 2 phase clocks. Figure 5.2 is a sketch of the address and data bus timing as it relates to the system clocks.
The major point to be noted is microprocessor is a memory cycle in which memory is either read or written.
Simultaneously with the read or write of memory, an internal operation of the microprocessor is also occurring.
-IV
Q w h J hJ W
CQ M
< cn
CO w Pi ps o
Q S
Q
<J §
\£>
CO
LO
LO
\£> w
/^S
CO
4-1 ts
CQ
--------------------<<-----------hJ <U
TJ y o
M u
U
W U)
3
U)
OJ
J-» w TJ
PS t j c
J-4
<u <u
4-* rO
6 pq 3
£5
0) 0J
00 XI cd E
P-H 3
53 hJ
3
M
U w
Q o
IS
Pi
W
CQ s
53 u
Q)
T3
J-»
O
£ o t-J
C/0
CO f-M
Q
§
Pi
<J
53
M
CQ u
<u
TJ o x :
6C
M s r i-H in i-H
CN i-H
CO i-H
00 o \ o
H i H i—1
<3in vO r-*.
O i—1
CN
CO
O o o
1 1 o o o o o 1—1 o o o o o o o o o o o o o o o o
O
O o o
1 1
O m m
Csl
\D
LO
C^l o o o o o o
1—1 o o o o o o o o o i-H i—1 i—1 i-H o o o o i-l i-H i—1 i—1 o o o o
O
O
O
O
1—1 o o o
O
O
O
O o o o o
4V
--------------------(<--------------pc-1
O o O
1 1 1 w
Jj-I
P h ptH
1 o> r--.
CNJ o
00 00
CN Csl l n
'O m
\D i—i i-H i—1 i—1 o o o o i—i o o o
---------------------► i—1 r—i i—1 o o o o o o o o o
1—1
1—1 i-H i-H 1—1 iH 1—1 i—1 I— i—1 1—1 i—1 i—1 i—1 i—1 iH 1—1 iH i-H iH i—1 i—1 i—1 in
CO in i—1 i—1 i-H i-H i—1 i—1 i—1 i—i i—i i-H i-H i—1 i—1 i—) i—1
53 cn
CQ
<
H
<J
*1 _ J ~
0 2 __________
ADDRESS LINE___ J
DATA LINE________ f ~ \ r
Example o f Timing - MCS650X Family
FIGURE 5.2
The following example will let us analyze this effect:
Example 5.2: Demonstration of "Pipelining" effect
Clock
Cycles External Operation
1 Fetch OP CODE
Fetch first-address half from memory
Address Data Internal Operation
100 ADC Increment P-counter to 101
101 ADL Increment P-counter to 102, Interpret
ADC instruction
102 ADH Increment P-counter to 103; Hold ADL
Fetch second ad dress half from memory
Fetch operand from memory
Fetch next OP CODE from memory
ADH,
ADL
103
Data Load Data
Fetch address from memory
104
STA Increment P-counter to 104, Perform ADC operation:
A + M + C
ADL Increment P-counter to 105, Result of
Add accumulator,
Interpret STA Instruc tion
The above example shows the operation of an ADC, add with carry in struction, using absolute addressing. In the first cycle, the OP CODE is fetched from memory addressed by the P-counter. To implement the
54
look-ahead or pipeline in cycle two, the fetch of ADL address low is done simultaneously with the interpretation of the ADC absolute instruction.
By the end of cycle 2, the microprocessor knows that it should access the next memory location for the address high as a result of interpretation of the absolute addressing mode.
The address low (ADL) is stored in the ALU while the address high (ADH) is being fetched in cycle 3.
On the fourth cycle, no internal operation is necessary while the microprocessor is putting the calculated value onto the address bus. How ever, during this cycle, the operand is loaded into the microprocessor.
The 4 cycles have all been involved with memory access for the ADC, absolute instruction. The first to fetch the instruction, the second to fetch the address low, the third to fetch the address high and the fourth to use the calculated address to fetch the operand. Because that completes the memory operations for this instruction, during the fifth cycle the microprocessor starts to fetch the next instruction from memory while it is completing the add operation from the first instruction. During the sixth cycle, the microprocessor is interpreting the new instruction fetched during cycle 5 while transferring the result of the add operation to the accumulator. This means that even though it really takes 6 cycles for the microprocessor to do the ADC instruction, the programmer only need concern himself with the first 4 cycles as the next 2 are overlapped as shown.
All instructions take at least 2 cycles; one to fetch the OP CODE and
1 to interpret the OP CODE and, with few exceptions, the number of cycles that an instruction takes is equal to the number of times that memory must be addressed.
The details of how each addressing mode is overlapped are described in the individual sections and for specific details of each cycle in vari ous operations, the user is referred to the Hardware Manual, Appendix A.
All instructions take at least 2 cycles; one to fetch the OP CODE and
1 to interpret the OP CODE and, with few exceptions, the number of cycles that an instruction takes is equal to the number of times that memory must be addressed.
5.2 M E M O R Y UTILIZATION
As indicated, the 16-bit address allows the user to access greater than 65,000 separate locations. Most of the locations which will be ac cessed in the course of a control problem will be in program or P-counter referenced locations. A typical program will probably range from 1000 to
8000 bytes and will normally be implemented in fixed ROM or non-volatile alterable ROM.
A second type of memory will be the read-write memory in which the user keeps data such as working values, input and output data. Depending on the type of problem being addressed, this RAM usually ranges from 32 bytes to 8000 bytes, although most applications will be under 2000 bytes of RAM.
It would seem there is significant address space not used in most applications. To get the maximum benefit of the addressing space, 2 con cepts are implemented in the MCS6500 family. These are the use of data addressing as 1/0 control and distributed address connections for minimum control lines. The latter concept utilizes the address bus, which is basic to and therefore pervasive in any microcomputer system, as a con trolling network whenever possible. An example of this is the use of the address bus in selecting devices to interface with the microprocessor.
5.2.1 1/0 Control
The advantages of accessing 1/0 as memory are 1) the use of distributed address space allows for simple 1/0 control lines and 2) all of the power of the instructions is applied to 1/0 operations.
This has the advantage of minimizing 1/0 hardware and allows the pro grammer to be innovative in the application of 1/0 devices in solving his problem.
All MCS6500 product family 1/0 devices contain 8-bit registers which are addressed by the microprocessor as though they were a mem ory byte. In the simplest case, the 8-bit register being read con tains a 1 and 0's pattern which corresponds to the TTL voltage level applied to 8 input pins to the 1/0 device.
56
If the register was a flip-flop register driving 8 output pins with TTL levels, the storing of 8 bits of data with a STA instruction into that I/O register would, in effect, be programming the flip-flop to a specific desired state. Thus, one can use the instructions with the I/O just as any other memory location.
5.2.2 Memory Allocation
Figure 5.1 displays the relationship between memory, address bus and data bus while referencing the address values in hexadecimal notation. The previous section has dealt with utilization of memory address space for not only ROM and RAM but for I/O as well. At this time, the concept of allocation of the memory field of Figure 5.1 to the elements of ROM, RAM and I/O will be considered. The allocation below satisfies most applications requirements and represents an optimum allocation for minimization of programming code and speed.
Hexadecimal Address Suggested Allocation of Memory
RAM
1/0
ROM
It should be noted that the 3 memory blocks address defini tions which, while not mandatory or required for proper system opera tion, do represent a logical assignment of space. The justification for this particular allocation will be presented in Section 9.12.
In the meantime, the reader should retain the concept of the various memory blocks allocated to RAM, I/O and ROM as they are useful in the following discussion. With an understanding of pipelining and the concept of memory allocation, the next subject must be: in what manner can data be accessed from the memory field?
5.3 IMPLIED ADDRESSING
Implied addressing is a single-byte instruction.
The byte contains the OP CODE which stipulates an operation internal to the microprocessor. Instructions utilizing this type of addressing in clude operations which clear and set bits in the P (Processor Status) reg ister, incrementing and decrementing internal registers and transferring
/
57
contents of one internal register to another internal register. Operations of this form take 2 clock cycles to execute. The first cycle is the OP
CODE fetch and during this fetch, the program counter increments.
In the second cycle, the incremented P-counter is now the address of the next byte of the instruction. However, since the OP CODE totally de fines the operation, the second memory fetch is worthless and any P-counter increment in the second cycle is suppressed. During the second cycle, the
OP CODE is decoded with recognition of its single byte operation.
In the third cycle, the microprocessor repeats the same address to fetch the next OP CODE. This is the second time the memory address is fetched; once as the second byte of the first instruction and second, as the correct OP CODE address for the next instruction.
A symbolic representation of a 2-cycle instruction is given below.
"PC" means "Program Counter."
Example 5.3: Illustration of implied addressing
Clock
Cycle
2
1
Address Bus Program Counter Data Bus
PC
PC + 1
PC + 1
PC + 1
OP CODE
New
OP CODE
PC + 1 PC + 2 New
OP CODE
Comments
Fetch OP CODE
Ignore New
OP CODE;
Decode Old
OP CODE
Fetch New
OP CODE;
Execute Old
OP CODE
Instructions which use implied addressing and require only 2 cycles include CLC, CLD, CLI, CLV, DEX, DEY, INX, INY, NOP, SEC, SED, SEI,
TAX, TAY, TSX, TXA, TXS, TYA.
Instructions utilizing implied addressing and which require more than
2 cycles are stack operations which include BRK, PHA, PHP, PLA, PLP, RTI,
RTS.
*
58
5.4 IMMEDIATE ADDRESSING
Immediate addressing is a 2-byte instruction.
The first byte contains the OP CODE specifying the operation and ad dress mode. The second byte contains a constant value known to the pro grammer. It is often necessary to compare load and/or test against cer tain known values. Rather than requiring the user to define and load con stants into some auxiliary RAM, the microprocessor allows the user to specify values which are known to him by the immediate addressing mode.
r
Example 5.4: Illustration of immediate addressing
Clock
Cycle Address Bus Program Counter Data Bus
1
2
PC
PC + 1
PC + 1
PC + 2
OP CODE
Data
3 PC + 2 PC + 3 New
OP CODE
Comments
Fetch OP CODE
Fetch Data,
Decode OP CODE
Fetch
OP CODE,
New
Execute Old
OP CODE
Immediate addressing is the simplest form of constant manipulation available to the programmer. It requires a minimum execution time in the sense that 1 cycle is used in loading the OP CODE and as this CODE is be ing interpreted, the constant is being fetched.
Instructions utilizing immediate addressing are ADC, AND, CMP, CPX,
CPY, EOR, LDA, LDX, LDY, ORA, and SBC.
5.5 ABSOLUTE ADDRESSING
Absolute addressing is a 3-byte instruction.
The first byte contains the OP CODE for specifying the operation and address mode. The second byte contains the low order byte of the effective address (that address which contains the data), while the third byte con tains the high order byte of the effective address. Thus the programmer specifies the full 16-bit address and, since any memory location can be specified, this is considered the most normal mode for addressing. Other modes may be considered special subsets of this 16-bit addressing mode.
59
Example 5.5: Illustration of absolute addressing
Clock
Cycle Address Bus Program Counter Data Bus
1
2
3
4
5
PC
PC + 1
PC + 2
ADH, ADL
PC + 3
PC + 1
PC + 2
PC + 3
PC + 3
PC + 4
OP CODE
ADL
ADH
Data
New
OP CODE
Comments
Fetch OP CODE
Fetch ADL,
Decode OP CODE
Fetch ADH,
Hold ADL
Fetch Data
Fetch New
OP CODE,
Execute Old
OP CODE
The basic operation of the microprocessor in an Absolute address mode is to read the OP CODE in the first cycle while finishing the previous operation. In the second cycle, the microprocessor automatically reads the first byte after the OP CODE (in this case the address low) while interpreting the operation code. At the end of this cycle, the microproces sor knows that it needs a second byte for program sequence; therefore, 1 more byte will be accessed using the program counter while temporarily storing the address low. This occurs during the third cycle. In the fourth cycle, the operation is one of taking the address low and address high that were read during cycles 2 and 3 to address the operand. For ex ample, in load A, the effective address is used to fetch from memory the data which is going to be loaded in the accumulator. In the case of stor ing, data is transferred from the accumulator to the addressed memory.
As was illustrated in the review of pipelining, depending on the in struction, it is possible for the microprocessor to start the next instruc tion fetch cycle after the effective address operation and independent of how many more internal cycles it may take to complete the OP CODE. The only exception to this is the case of "Jump Absolute" in which the address low and address high that are fetched in cycle 2 and cycle 3 are used as the 16-bit address for the next OP CODE. The jump absolute therefore only requires 3 cycles. In all other cases, absolute addressing takes 4 cycles,
3 to fetch the full instruction including the effective address, the fourth to perform the memory transfer called for in the instruction.
60
Absolute addressing always takes 3 bytes of program memory; 1 for the
OP CODE, 1 for the address low, 1 for the address high, plus 1 byte of data memory (such as RAM) that is pointed to by the effective address.
Instructions which have absolute addressing capability include ADC,
AND, ASL, BIT, CMP, CPX, CPY, DEC, EOR, INC, JMP, JSR, LDA, LDX, LDY, LSR.
ORA, ROL, SEC, STA, STX, STY.
5.6 ZERO PAGE ADDRESSING
Zero page addressing is a 2-byte instruction. The first byte con tains the OP CODE, while the second byte contains the effective address in page zero of memory.
As seen in absolute addressing, the ability to address anywhere in the 65K memory space costs 3 bytes of program space, plus a minimum of 4 cycles to perform address operations. In order to allow the user a shorten ing of both memory space and execution time, particularly when dealing with working registers and intermediate values, the MCS650X microprocessor family has a special addressing mode that assumes automatically dress high (ADH) to be in the lowest page of memory. In stand the page concept one should think of each of the various memory ad dresses as comprising a consecutive block of 256 locations which have an independent high order address associated with that block. Each block is called a page. Other than for zero page and for calculating indexed ad dresses which will be covered in the following sections, the microprocessor pays little attention to the page concept.
The microprocessor assumes that the high order byte of the effective address for instructions which contain OP CODES which indicate the zero page addressing option is all O's (ADH = 00, hexadecimal). This allows the following sequence to occur.
61
Example 5.6: Illustration of zero page addressing
Clock
Cycle Address Bus Program Counter Data Bus
1
2
PC
PC + 1
PC + 1
PC + 2
OP CODE
ADL
3
4
00, ADL
PC + 2
PC + 2
PC + 3
Data
New
OP CODE
Comments
Fetch OP CODE
Fetch ADL, De code OP CODE
Fetch Data
Fetch New
OP CODE, Exe cute Old
OP CODE
On the first cycle, the microprocessor puts out the program counter, reads the OP CODE and increments the program counter. On the second cycle, the microprocessor puts out the program counter, reads the effective ad dress low, interprets the OP CODE and increments the program counter. So far, the operations are identical to those described in the absolute ad dressing mode. However, by the end of the second cycle, the microprocessor has decoded the fact that this is a zero page operation and on the next cycle, it outputs address 00, as the effective address high, along with the address low that it just fetched and then either reads or writes mem ory at that location, depending on the OP CODE.
The advantage of zero page addressing is that it takes only 2 bytes,
1 for the OP CODE and 1 for the effective address low; and only 3 cycles,
1 to fetch the OP CODE, 1 to fetch the address low, and 1 to fetch the data, as opposed to absolute addressing which takes 3 bytes and 4 cycles.
In order to make most effective utilization of this concept, the user should organize his memory so that he is keeping his most frequently accessed RAM values in the memory locations between 0 and 255. If one organizes the zero page of memory properly, including moving data into these locations for longer loops, significant shortening of program code and execution time can be obtained.
The concept of zero page is so important that the various cross assemblers have error notations which indicate when improper use of this space is made. If one's coding is organized according to the guidelines shown in Section 5.2.2, one normally will find working storage located in values from 0 to 255. This is an important aspect of the discipline known as "memory management."
62
Once the pattern of coding for the MCS650X, which considers working storage or registers in the zero page, becomes a habit, one finds that in most control applications, all of the working registers will take advantage of this programming and the associated time reduction without any special effort on the user's part.
Instructions which allow zero page addressing include ADC, AND, ASL,
BIT, CMP, CPX, CPY, DEC, EOR, INC, LDA, LDX, LDY, LSR, ORA, ROL, SBC, STA,
STX, STY.
5.7 RELATIVE ADDRESSING
As discussed in Section 4.1, all of the branch operations in the micro processor use the concept of relative addressing. In example 5.7, it is seen that for the case of the straightforward branch in which the branch is not taken, on the first program count cycle, the microprocessor puts out program counter as an address, fetches the OP CODE and finishes the previous operation. During the second cycle, the program counter is put on the address bus, picking up the relative offset. Internally, the micro processor is decoding the OP CODE to determine that it is a branch instruc tion.
Example 5.7: Illustration of relative addressing branch not taken
Cycle
1
Address Bus
0100
0101
0102
Data Bus
OP CODE
Offset
Next
OP CODE
External
Operation
Internal
Operation
Fetch
OP CODE
Fetch
Offset
Finish Previous Operation, Increment Pro gram Counter to 101
Interpret Instruction, Increment Pro gram Counter to 102
Fetch Next Check Flags, Increment
OP CODE Program Counter to 0103
This is only the second cycle of an internal operation; therefore, the microprocessor may be storing a computed value from the previous instruc tion at the same time it is finishing interpreting the present instruction.
It is while doing the store operation that the flags In the machine get physically set; therefore, the microprocessor allows the program counter
63
to go 1 more cycle to allow itself time to determine the value of the flags. For example, if the previous instruction is ADC, the flags will not get set until the in which the offset value is fetched.
During the third the microprocessor puts the incremented PC onto the address bus, fetches the next OP CODE and checks the flag in order to decide whether or not the program counter value that is going out is correct and that the branch is not going to be taken. Therefore, an additional type of pipeline, in this case fetching the next OP CODE in a branch sequence, accomplishes the implementation of a branch relative with no branch being taken. This requires 2 cycles. One cycle fetches the branch OP CODE and 1 cycle fetches the next operation, the relative offset.
The second fetch is effectively ignored by virtue of the fact that the branch is not taken, so the program counter location has already been incre mented and the next OP CODE has already been fetched by the microprocessor.
If in the above example it is assumed that the flag is set such that the branch is taken and the relative offset is +50, the microprocessor takes a third cycle to perform the branch operation.
Example 5.8: Illustration of relative addressing branch positive taken, no crossing of page boundaries
Cycle Address Bus Data Bus
External
Operation
Internal
Operation
1
2
3
4
0100
0101
0102
0152
OP CODE
+50
Next
OP CODE
Next
OP CODE
Fetch
OP CODE
Fetch
Offset
Fetch Next
OP CODE
Fetch Next
OP CODE
Finish Previous Oper ation, Increment Pro gram Counter to 101
Interpret Instruction,
Increment Program
Counter to 102
Check Flags, Add Rela tive to PCL, Increment
Program Counter to 103
Transfer Results to
PCL, Increment Program
Counter to 153
In Example 5.8, on the first cycle, a branch OP CODE is fetched while the previous operation is finished. On the second cycle, the offset is fetched while the branch instruction is being interpreted. On the third cycle, the microprocessor uses the adder to add the program count low to
64
the offset and also checks the flags. Because the program count for the next OP CODE in program sequence is already in the program counter and is being incremented, the microprocessor can allow the incrementation process to continue. If the value for the flag is not set, then the microprocessor loads the the add of the program counter low to the offset value, is ignored as it was in the previous example.
If during the third cycle the flag is found to be the correct value for a branch, the OP CODE that ignored. The microprocessor then updates the program counter with the results from the add operation, puts that value out on the address bus twhich fetches a new OP CODE.
This gives the effect in a case where the branch tive 2-cycle branch, i.e., 2 memory references. In the case when the branch is taken, the branch takes 3 cycles as long as the relative value does not force an update to the program counter high. In other words, 3 cycles are required if the cussion of the "page" concept in Section 5.0). If in the above example the branch was back from address 0102 fifty locations, as opposed to +50 locations, the following result would occur:
Example 5.9: Illustration of relative addressing— branch negative taken, crossing of page boundary
Cycle
External
Operations
Internal
Operations Address Bus
0100
Data Bus
OP CODE
2
1
3
0101
0102
01B2
00B2
-50
Next
OP CODE
Discarded
Data
Next
OP CODE
Fetch
OP CODE
Finish Previous
Instruction
Fetch
Offset
Interpret Instruc tion
Fetch Next Check Flags
OP CODE Add Relative to
PCL
Fetch Dis Store Adder in PCL carded Data and Subtract 1 from PCH
Fetch Next Put Out New PCH
OP CODE and Increment PC to 00B3
65
In this example, the adder is used to perform the
The minus branch crosses back over the page boundary, therefore an intermediate result is developed of 01B2 which has no intrinsic value because of the borrow which now has to be reflected into the program counter high. Since this example displays both a negative offset and the crossing of a page boundary, additional explanation is in order.
The value to which the offset will be added is 0102 (hexadecimal).
The offset itself is -50 (hexadecimal).
Subtract low order byte:
02
H.hX
50 r e x
=
=
0000 0010
0101 0000
Take two's complement of 50:
50 = 1010 1111
Add 1 _________ 1
-50 = 1011 0000
Add 02 0000 0010
-50 1011 0000
Carry = /0/ 1011 0010
B 2
Up to this point, the PCH has not been affected; therefore the value on the address bus is 01B2.
The Carry = 0, indicating a borrow.
Subtract high order byte:
01 _ = 0000 0001 ttlLA
00lT1?v = 0000 0000
Take two's complement of 00: ooHEX = m i m i
Add Carry - _________0
-°°HEX = 1111
Add 01 0000 0001
-00 1111 1111
Carry = jTJ 0000 0000
0 0
The presence of the Carry indicates no borrow, hence a positive result.
66
At this time, after the arithmetic operation on both bytes of the P.C., the address bus will be: 00B2.
The microprocessor does put out on the address line the intermediate results (01B2), thereby reading a location within the page it was currently working in, the value of which is ignored. It then subtracts 1, or if this was a branch forward to the next page, the microprocessor would add 1 to program counter high in this fourth cycle. In the fifth cycle, the micro processor will recognize that it has the correct new program counter high and program counter low and is able to start a new instruction operation, thereby giving an effective length to the branch operation when a page crossing is encountered of 4 cycles.
It should be noted that all of the above operations are automatic; once a branch instruction is encountered, the following relative value is calculated and put into the memory location after the branch instruction.
We can see, however, that it is possible to control the execution time of a branch. This is important for counting or estimating execution times of operations. For counting purposes, the following applies:
If a branch is normally not taken, assume 2 cycles for the branch.
If the branch is normally taken but it is not across the page boundary, assume 3 cycles for the branch.
If the branch is over a page boundary, then assume 4 cycles for the branch.
In loops which are repeated many times, one type of bility of taking the branch versus not taking it.
In order to indicate to the programmer when the 4-cycle branch is taken as opposed to the 3-cycle branch, the various assemblers flag all branch operations which cross page boundaries with a warning message and his in such a way that the branch does not cross page boundary.
boundary crossings can be ignored by the programmer.
To summarize, the relative addressing always takes 2 bytes, 1 for the
OP CODE and 1 for the offset.
67
The execution time is as follows:
— 2 cycles
Branch When the
M ^
No Page Crossing
— 3 cycles
Branch When the Branch Is Taken with
.
,
— 4 cycles a Page Crossing
Only branch instructions have relative addressing. The branch instruc tions are: BCC, BEQ, BIT, BMI, BNE, BPL, BSC, BVC, BVS. For a more detailed explanation of relative offset calculations the reader is referred to
Appendix H.
CHAPTER 6
INDEX REGISTERS AND INDEX ADDRESSING CONCEPTS
6.0 GENERAL CONCEPT OF INDEXING
In previous sections techniques for using the program counter to address memory locations after the operation code to develop the address for a particular operation have been discussed. Other than cases when the programmer directly changes the program memory, it can be considered that the addressing modes discussed up until now are fixed or directed addresses and each has the relative merits discussed under each individual section. However, a more powerful concept of addressing is that of computed addressing. There are basically two types of computed address ing; indexed addressing and indirect addressing.
Indexed addressing uses an address which is computed by means of modifying the address data accessed by the program counter with an internal register called an index register.
Indirect addressing uses a computed and stored address which is accessed by an indirect pointer in the programming sequence.
In the MCS650X product family, both of these modes are used and combinations of them are available.
Before undertaking the more difficult concepts of indirect address ing the concept of indexed instructions will be developed.
69
In order to move five bytes of memory from an address contained in FIELD 1 to another set of addresses, starting with FIELD 2, the following program could be written:
LABEL INSTRUCTION
START LDA
STA
LDA
STA
LDA
STA
LDA
STA
LDA
STA
OPERAND COMMENTS
FIELD 1
FIELD 2
FIELD 1 + 1
------ ^
Move First Value
Move Second Value
FIELD 2 + 1 ______ ^
FIELD 1 + 2 Move Third Value
FIELD 2 + 2 ________________________ ^
FIELD 1 + 3 Move Fourth Value
FIELD 2 + 3 ------------------------ ^
FIELD 1 + 4 s\ > Move Fifth Value
FIELD 2 + 4 ------------------------ ^
In this example, data is fetched from the first memory location in
FIELD 1, as addressed by the next one or two bytes in program memory, stored temporarily in A and then written into the first memory location in FIELD 2, also addressed by the next one or two bytes in program memory.
This sequence is repeated, with only the memory addresses changing, until all the data has been transferred. This type of programming is called straight line programming because each repetitive operation is a sepa rate group of instructions listed in sequence or straight line form in program memory. This is necessary even though the instruction OP CODES are identical for each memory transfer operation because the specific memory addresses are different and require a different code to be writ ten into the program memory for each transfer.
It takes a total of 10 instructions to accomplish the move when it is implemented this way. It should be noted that it is not indicated whether or not FIELD 1 and FIELD 2 are Zero Page addresses or Absolute addresses.
If they were Zero Page addresses, the total number of bytes con sumed in solving the problem would be two bytes for each instruction and thereby requiring 20 bytes of memory; if both FIELD 1 and FIELD 2 were Absolute memory locations, each instruction would take 3 bytes and this program would require 30 bytes of program storage.
70
The Zero Page program would execute in three cycles per in struction or 30 cycles and the Absolute location version would execute in four cycles per instruction or 40 cycles.
A new concept has been introduced in this example, that of symbolic notation rather than actual locations for the instructions.
The form that this short program is written in uses symbolic addressing in which the address of the beginning of the program has a name START. Symbolic representations of addresses such as "START" are referred to as labels. The addresses in the two address field used in this example have also been given names, the first address of the first field is called FIELD 1; the first address of the second field is called FIELD 2. Each additional address in the fields has been given a number which is referenced to the first number; for example, the third byte in FIELD 1 is FIELD 1 + 2 . All of these concepts are im plemented to simplify the ease of writing a program because the user does not have to worry about the locations of FIELD 1 and FIELD 2 until after analyzing the memory needs of the whole program. Symbolic notation also results in a more readable program.
Translation from symbolic form instructions and addresses into actual numerical OP CODES and addresses is done by a program called a symbolic assembler. Several different versions of symbolic assemblers and cross assemblers are available for the MCS650X product family.
Symbolic notation will be used throughout the remainder of this text because of its ease of understanding and because individual byte addresses are unnecessary although for an explanation of a particular mode, the byte representation may be used.
In this example, only direct addresses were used. A program to reduce the number of bytes required to move the five values follows:
71
Done
Flow Chart
-
Moving Five Bytes of Data with Loop
FIGURE 6.1
Example 6.2 is a program listing that corresponds to the flow chart:
Example 6.2: Moving Five Bytes of Data With Loop
LABEL
INITIALIZE
START
OTHER
NOTE:
INSTRUCTION
CLC
LDA
STA
LDA
ADC
STA
LDA
ADC
STA
CMP
BNE
OPERAND COMMENTS
FIELD 1 -------- -
FIELD 2 ---------
START + 1 -------
#1
START + 1
OTHER + 1
#1
OTHER + 1 -------
//FIELD 2 + 5 -----
START
Move Loop
Modify Move Values
Check for End
For ease of reading, labels have been written in the form
"FIELD 1". This is incorrect format for use in the various symbolic assemblers. "FIELD 1" must be written "FIELD1" when coding for assembler formats.
72
Assuming Zero Page, direct addressing, Example 6.3 is written below with one byte per line just as it would appear in program memory.
This will provide a more detailed description of Example 6.2.
Example 6.3: Coded Detail of Moving Fields With Loop
LABEL
START
OTHER
CODE NAMES
CLC
LDA
FIELD 1
STA
FIELD 2
LDA
START + 1
ADC
1
STA
START + 1
LDA
OTHER + 1
ADC
1
STA
OTHER + 1
CMP
ORIGINAL FIELD 2 + 5
BNE
START
COMMENTS
Clear Carry
(FIELD 1) - ► A
A — ► ( F I E L D 2)
From Address ► A
A + 1 — ► A
A — ► F r o m Address
To Address ► A
A + 1 — ► A
A + To Address
A - ORIGINAL FIELD 2 + 5
If not, loop to START
In this example, the program is modifying the addresses of one load instruction and one store instruction rather than writing ten in structions to move five bytes of data and fifty instructions to move twenty-five bytes of data.
The address of the Load A instruction is located in memory at
START + 1 and the Store instruction at OTHER + 1. In order to perform this operation, the address must be modified once for each move opera tion until all of the data is moved.
Checking for the end of the moves is accomplished by checking the results of the address modification to determine if the address exceeds the end of the second field. When it does, the routine is complete.
73
If a hundred values were to be moved this program would remain 20 bytes long, whereas the solution to the first problem would require a program of 200 instructions.
The type of coding used in this example is called a "loop".
Although the program loop in this case requires as many bytes as the original program, more values could be moved without increasing the length of the program. The greater the number of repetitive operations that are to be accomplished, the greater the advantage of the loop type program over straight line programming.
Important Note: The execution time required to move the five values is significantly longer using the loop program than the straight line program. In the straight line program, if a Zero Page operation is assumed, the time to perform the total move is 30 cycles. Using the loop program, the execution time to move five values is five times through the entire loop, which takes 25 cycles. Therefore the time to move five values is 125 cycles.
While loops have an advantage in coding space efficiency, all loops cost time. If the programmer has a problem that is extremely time dependent, taking the loop out and going to straight line programming, even though it is extremely inefficient in terms of its utilization of memory, will often solve the timing problem.
The straight line programming technique becomes very useful in some control applications. However, it is not recommended as a standard technique but should only be used when there are extreme timing problems.
Using loops will normally save a significant number of bytes but they will always take more time.
The technique used in the loop program example has two major problems:
1. The necessity to modify program memory. This should be avoided to take advantage of the ability to put programs into read only memory with the corresponding savings in hardware costs.
74
2. Although this is the simplist form of computed addressing, less program bytes would be necessary than the more sophis ticated form of program shown in the following flow chart:
FINISH
Moving Five Bytes o f Data with Counter
FIGURE 6.2
In the MCS650X microprocessor family, the counter is called an index register. It is an 8-bit register which is loaded from memory and has the ability to have one added to it by an increment instruction (INX,INY) and can be compared directly to memory using the compare index instruction
(CPX,CPY). Example 6.4 shows the program listing for the flow chart of
Figure 6.2.
75
Example 6.4: Moving Five Bytes of Data With Index Registers
BYTES LABEL
2
3 LOOP
3
1
2
2
INSTRUCTION
13 for Absolute
LDX
LDA
STA
INX
CPX
BNE
OPERAND
0
FIELD 1,X
FIELD 2,X
5
LOOP
COMMENTS
Load Index With Zero
Increment Count
Compare For End
In this example, index register X is used as an index and as a counter. It is initialized to zero. Data is fetched from memory at the address "FIELD 1 plus the value of register X", and placed in A.
The data is then written from A to memory at the address "FIELD 2 plus the value of register X". Register X is incremented by one and compared with 5 in order to determine if all five data values have been transferred. If not the program loops back to LOOP. In this example, "FIELD 1" is called the "Base Address" which is; the address to which indexing is referenced.
This only takes 11 or 13 bytes, depending on whether or not the field is in Page Zero or in absolute memory. It still takes 13 or 15 cycles per byte moved, again confirming that loops are excellent for coding space but not. for execution time.
It can be seen from the example that there are basically two criterias for an index register; one, that it be a register which is easily incremented, compared, loaded, and stored, and two, that in a single instruction one can specify both the Base Address and the value of X.
In the MCS650X microprocessor, the way that the indexed in struction is symbolically represented is OP CODE, Address, X. This indicates to the symbolic assembler that an instruction OP CODE should be picked, which should specify either the absolute address modified by the content of index X register or Zero Page address modified by the content of index X register.
76
In performing these operations, the microprocessor fetches the instruction OP CODE as previously defined, and fetches the address, modifies the address from the memory by adding the index register to it prior to loading or storing the value of memory.
The index register is a counter. As discussed previously, one of the advantages of the flags in the microprocessor is that a value can be modified and its results tested. Assume the last example is modified so that instead of moving the first value in FIELD 1 to the first value in FIELD 2, the last value in FIELD 1 is moved first to the last value in FIELD 2, then the next to the last value, etc. and finally the first value. With the index register preloaded with 5 and using fields of data The zero indicates that the number of times through the loop is correct and the loop exited by use of the zero test. The program listing for this modification is shown in Example 6.5:
LABEL
LOOP
Register
INSTRUCTION
LDX
LDA
STA
DEX
BNE
OPERAND
5
FIELD 1-1,X
FIELD 2-1,X
LOOP
In this example, Index Register X is again used as an Address
Counter but it will count backwards. It is initialized to five for this example. Data is fetched from memory at the address "FIELD 1 plus the value of Register X" and placed in A. The data is then written from A to memory at the address "FIELD 2 plus the value of Register X."
Register X is decremented by one. If the decremented value is not zero, as determined by a Branch on Zero instruction, the program loops back to LOOP
The loop has been decreased to 9 or 11 bytes and the execution time per byte has been decreased from 15 cycles to 13 cycles per value
77
which shows the advantage of using the flag setting of the decrement index instruction.
The two index registers, X and Y, can now be added to the system block diagram as in Figure 6.3
Partial Block Diagram ofMCS650X Including Index Registers
FIGURE 6.3
Each of the index registers is 8 bits long and is loaded and stored from memory, using techniques similar to the accumulator. Because of this ability, they can be considered as auxiliary channels to flow data through the microprocessor. However, their primary use is in being added to addresses fetched from memory to form a modified effective address, as described previously. Both index registers have the ability to be compared to memory (CPX,CPY) and to be incremented (INX,INY) and decremented (DEX,DEY).
78
Because of OP CODE limitations, X and Y have slightly different uses. X is a little more flexible because it has Zero Page operations which Y does not have with exception of LDX and STX. Aside from which modes they modify, the registers are autonomous, independent and of equal value.
6.1 ABSOLUTE INDEXED
Absolute indexed address is absolute addressing with an index register added to the absolute address. The sequences that occur for indexed absolute addressing without page crossing are as follows:
1
Address
Cycle Bus____
0100
0101
Data
Bus
OP CODE
BAL
External
Operation
Fetch OP CODE
Fetch BAL
Internal
Operation
Increment PC to 101
Finish Previous
Instruction
Increment PC to 102
Interpret In struction
Increment PC to 103
Calculate BAL + X
0102 BAH
BAH,BAL+X OPERAND
Fetch BAH
Put Out
Effective
Address
Fetch Next
OP CODE
103 Next OP
CODE
Finish Operations
BAL and BAH refer to the low and high order bytes of the base address, respectively. While the index X was used in Example 6.7, the index Y is equally applicable.
If a page is not crossed, the results of the address low + X does not cause a carry. The processor is able to pipeline the addition of the
8-bit index register to the lower byte of the base address (BAL) and not suffer any time degradation for absolute indexed addressing over straight absolute addressing. In other words, while BAH is being fetched, the add of X to BAL occurs. Both addressing modes require four cycles with
the only difference being that X or Y must be set at a known value and the OP CODE must indicate an index X or Y.
The second possibility is that when the index register is added to the address low of the base address that the resultant address is in the next page. This is illustrated in Example 6.7.
Example 6.7: Absolute Indexed; With Page Crossing
Address Data
Cycle Bus Bus
External
Operation
Internal
Operation
0100 OP CODE Fetch OP CODE
0101 BAL Fetch BAL
Finish Previous
Operation Increment
PC to 101
Interpret Instruction
Increment PC to 102
0102 BAH Fetch BAH Add BAL + Index
Increment PC to 103
BAH,BAL
+X
Data
(Ignore)
Fetch Data
(Data is ignored)
Add BAH + Carry
BAH+1,
BAL+X
Data Fetch Data
0103 Next OP
CODE
Fetch Next
OP CODE
Finish Operation
The most substantial difference between the page crossing operation and no page crossing is that during the fourth cycle, the address high and the calculated address low is put out, thereby incorrectly addressing the same page as the base address. This operation is carried on in parallel with the adding of the carry to the address high.
During the fourth cycle the address high plus the carry from the adder is put on the address bus, moving the operation to the next page. Thus there are two effects from the page crossing. 1. The addressing of a false address. This is similar to what happens in a branch relative during a page crossing. 2. The operation takes one additional cycle while the new address high is calculated. As with the branch relative
80
this page crossing occurs independently of programmer action and there is no penalty in memory for having crossed the page boundary.
It is possible for the programmer to predict a page crossing by knowing the value of the base address and the maximum offset value in the index register. If timing is of concern, the base address can be adjusted so that the address field is always in one page.
As with absolute addressing, absolute indexed is the most general form of indexing. It is possible to do absolute indexed modified by X, and absolute indexed modified by Y. Instructions which allow absolute indexed by X are ADC, AND, ASL, CMP, DEC,
EOR, INC. LDA, LDY, LSR, ORA, ROL, SBC, and STA.
The instructions which allow indexed absolute by Y are ADC,
AND, CMP, EOR, LDA, LDX, ORA, SBC, and STA.
As with non-computed addressing, there is a memory use advantage to the short-cut of Zero Page addressing. Except in LDX and STX instructions which can be modified by Y, Zero Page is only available modified by X. If the base address plus X exceeds the value that can be stored in a single byte, no carry is generated, therefore there is no page crossing phenomena. A wrap-around will occur within
Page Zero. The following example illustrates the internal operations of Zero Page indexing.
81
Example 6.8: Illustration of Zero Page Indexing
Cycle
1
Address
Bus____
0100
Data
Bus
OP CODE
External
Operation
Fetch OP CODE
0101 BAL
Internal
Operation
Finish Previous
Operation, 0101 ->■ PC
Interpret Instruct ion, 0102 ->■ PC
00,BAL
00,BAL
+X
0102
Data
(Dis carded
Data
Next OP
CODE
Fetch Base
Address Low
(BAL)
Fetch
Discarded
Data
Fetch Data
Fetch Next OP
CODE
Add: BAL + X
Finish Operation
As can be seen from the example, there is no time savings of Zero
Page indexing over absolute indexing without page crossing. In the case of the indexed absolute during cycle 3 the address high is being fetched at the same time as the addition of the index to address low.
In the case of the Zero Page, there is no opportunity for this type of overlap; therefore, indexed Zero Page instructions take one cycle longer than non-indexed instructions.
In both Zero Page indexed and absolute indexed with a page crossing, there are incorrect addresses calculated. Provisions have been made to make certain that, only a READ operation occurs during this time. Memory modifying operations such as STORE, SHIFT, ROTATE, etc. have all been delayed until the correct address is available, thereby prohibiting any possibility of writing data in an incorrect location and destroying the previous data in that location.
As has been previously stated, there is no carry out of the Zero
Page operation. 00 is forced into address high under all circumstances in cycle A. For example, if the index register containing a value of 10 is to be added to base address containing following operation would occur:
82
Example 6.9: Demonstrating the Wrap-Around
Cycle
3
4
Address Bus
0007
Internal Operation
F7 + 10
This indicated the wrap-around effect that occurs with Zero Page indexing with page crossing. This wrap-around does not increase the cycle time over that shown in the previous example.
Only index X is allowed as a modifier in Zero Page. Instructions which have this feature include ADC, AND, ASL, CMP, DEC, EOR, INC, LDA,
LDY, LSR, ORA, ROL, SBC, STA and STY. Note that index Y is allowed in the instructions LDX and STX.
6.3 INDIRECT ADDRESSING
In solving a certain class of problems, it is sometimes necessary to have an address which is a truly computed value, not just a base address with some type of offset, but a value which is calculated or sometimes obtained as a group of addresses. In order to implement this type of indexing or addressing, the use of indirect addressing has been introduced.
In the MCS650X family indirect operations have a special form.
The basic form of the indirect addressing is that of an instruction consisting of an OP CODE followed by a Zero Page address. The micro processor obtains the effective address by picking up from the Zero
Page address the effective address of the operation. The indirect addressing operation is much the same as absolute addressing except indirect addressing uses a Zero Page addressing operation to in directly access the effective address. In the case of absolute addressing the value in the program counter is used as the address to pick up the effective address low, one is added to the program counter which is used to pick up the effective address high. In the case of indirect addressing, the next value after the OP CODE, as addressed with the program counter, is used as a pointer to address the effective
83
address low in the zero page. The pointer is then incremented by one with the effective address high fetched from the next memory location. The next cycle places the effective address high (ADH) and effective address low (ADL) on the address bus to fetch the data.
An illustration of this is shown in Figure 6.4.
Indirect Addressing—Pictorial Drawing
FIGURE 6.4
DATA
The address following the instruction is really the address of an address, or "indirect" address. The indirect address is represented by IAL in the figure.
A more detailed definition of indirect addressing is included in the appendix.
84
Although the MCS650X microprocessor family has indirect operations, it has no simple indirect addressing such as described above. There are two inodes of indirect addressing in the MCS650X microprocessor family: 1.) indexed indirect and 2.) indirect indexed.
The major use of indexed indirect is in picking up data from a table or list of addresses to perform an operation. Examples where indexed indirect is applicable is in polling I/O devices or performing string or multiple string operations. Indexed indirect addressing uses the index register X. Instead of performing the indirect as shown in the Figure 6.4, the index register X is added to the Zero
Page address, thereby allowing varying address for the indirect pointer. The operation and timing of the indexed indirect addressing is shown in Figure 6.5.
0100 OP CODE
0101 IAL
00,IAL+X
\
\
00,IAL+0 ADL 1
ADH1,
ADL1
DATA 1
ADH 1
/
\
00,IAL+2 ADL 2
00,IAL+4
ADH 2
ADL 3
ADH 3
Indexed Indirect Addressing
FIGURE 6.5
/
\
)> ADL2 DATA 2
ADH3,
ADL3
DATA 3
/
85
Example 6.10: Illustration of Indexed Indirect Addressing
Cycle
Address Data
Bus____ Bus
External
Operation
Internal
Operation
0100 OP CODE Fetch OP CODE Finish Previous
Operation, 0101 P(
0101 BAL Fetch BAL Interpret In struction, 0102 -»■ PC
Add BAL + X 00,BAL DATA (Dis carded)
00,BAL ADL
+ X
00,BAL ADH
+ X + 1
Fetch Discard ed DATA
Fetch ADL
Fetch ADH
Add 1 to BAL + X
Hold ADL
6
7
ADH,ADL DATA
0102 Next OP
Fetch DATA
Fetch Next OP
CODE
Finish Operation
0103 PC
One of the advantages of this type of indexing is that a 16-bit address can be fetched with only two bytes of memory, the byte that contains the OP CODE and the byte that contains the indirect pointer.
It does require, however, that there be a table of addresses kept in a read/write memory which is more expensive than having it in read only memory. Therefore, this approach is normally reserved for appli cations where use of indexed indirect results in significant coding or throughput improvement or where the address being fetched is a variable computed address.
It is also obvious from the example that the user pays a minor time penalty for this form of addressing in that indexed indirect always takes six cycles to fetch a single operand which is 25% more than an absolute address and 50% more than a Zero Page reference to an operand. As in the Zero Page indexed, the operation in cycles three and four are located in Zero Page and there is no ability to carry over into the next page. It is possible to develop a value of the index plus the base address where the result exceeded 255, in this case the address put out is a wrap-around to the low part of the Page Zero.
86
Instructions which allow the use of indexed indirect are ADC, AND,
CMP, EOR, LDA, ORA, SBC, STA.
6.5 INDIRECT ADDRESSING
The indirect indexed instruction combines a feature of indirect addressing and a capability of indexing. The usefulness of this in struction is primarily for those operations in which one of several values could be used as part of a subroutine. By having an indirect pointer to the base operation and by using the index register Y in the normal counter type form, one can have the advantages of an address that points anywhere in memory, combined with the advantages of the counter offset capability of the index register.
Figure 6.6 illustrates the indirect indexed concept in flow form while Example 6.11 indicates the internal operation of a non-page roll over of an indirect index.
Indirect Indexed Addressing
FIGURE 6.6
87
DATA 1
DATA 2
DATA 3
Example 6.11: Indirect Indexed Addressing (No Page Crossing)
Cycle
Address Data
Bus Bus
External
Operation
Internal
Operation
1 0100 OP CODE Fetch OP CODE Finish Previous
Operation, 0101 -* PC
0101 IAL Fetch IAL Interpret In struction, 0102 -* PC
3
4
00,IAL
00,IAL
+ 1
BAL
BAH
BAH,BAL
+ Y
DATA
0102 Next OP
CODE
Fetch BAL
Fetch BAH
Fetch Operand
Fetch Next OP
CODE
Add 1 to IAL
Add BAL + Y
Finish Operation
0103 -* PC
The indirect index still requires two bytes of program storage, one for the OP CODE, one for the indirect pointer. Once beyond the indirect, the indexing of the indirect memory location is just the same as though it was an absolute indexed operation in the sense that if there is no page crossing, pipelining occurs in the adding of the index register Y to address low while fetching address high, and therefore, the non-page crossing solution is one cycle shorter than the indexed indirect. In
Example 6.12 it is seen that the page crossing problem that occurs with absolute indexed page crossing also occurs with indirect indexed address ing.
88
Cycle
1
Address Data
Bus
0100
0101
Bus
OP CODE
IAL
External
Operation
Load OP CODE
Fetch IAL
Fetch DATA
(Discarded)
Fetch Data
Internal
Operation
Finish Previous
Operation, 0101 -* PC
Interpret In struction, 0102 -* PC
Add 1 to IAL
Add BAL to Y
3
4
00,IAL BAL
00,IAL BAH
+ 1
BAH,BAL DATA (Dis-
+ Y carded)
BAH + 1 DATA
BAL + Y
0102 Next OP
CODE
Fetch BAL
Fetch BAH
Add 1 to BAH
Fetch Next OP
CODE
Finish This
Operation,
0103 + PC
When there is a page crossing, the base address high and base address low plus Y are pointing to an incorrect location within a referenced page. However, it should be noted that the programmer has control of this incorrect reference in the sense that it is always pointing to the page of the base address. In one more cycle the correct address is referenced. As was true in the case of absolute indexed, the data at the incorrect address is only read. STA and the various read, modify, write memory commands all operate assuming that there will be a page crossing, take the extra cycle time to perform the add and carry and only perform a write on the sixth cycle rather than taking advantage of the five cycle short-cut which is available to read operations. This added cycle guarantees that a memory location will never be written into with incorrect data.
Instructions which allow the use of indexed indirect are ADC, AND,
CMP, EOR, LDA, ORA, SBC, STA.
89
In the following two examples can be seen a comparison between the use of absolute modified by Y and indirect indexed addressing.
In these examples the same function is performed. Values from two memory locations are added and the result stored in a third memory location, assuming that there are several values to be added.
The first example deals with known field locations. The second example, such as might be traditionally used in subroutines, deals with field locations that vary between routines. A two byte pointer for each routine using the subroutine is stored in Page Zero.
The number of values to be added for each routine is also stored.
Example 6.13: Absolute Indexed Add - Sample Program
//Bytes Cycles Label
2
3
2
4
START
LOOP
3
3
2
1
4
2
4
3
14 19
Instruction Comments
LDY //COUNT -1
LE>A FIELD 1,Y
ADC FIELD 2,Y
STA FIELD 3,Y
D$Y
BfL LOOP
Set Y = End of FIELD
Load Location 1
Add Location 2
Store in Location 3
Check for Less Than Zero
Time for 10 Bytes = 1 7 1 Cycles
Example 6.14: Indirect Indexed Add - Sample Program
#Bytes
2
2
2
2
1
2
U
Cycles Label
2
5
5
5
2
22
3
START
LOOP
+ 6 bytes for pointers
Instruction
LDY //COUNT -1
LDA (PNT1), Y
ADC (PNT2), Y
STA (PNT3), Y
DEY
BPL LOOP
Comments
Set Y = End of FIELD
Load FIELD 1 Value
Add FIELD 2 Value
Store FIELD 3 Value
Time for 10 Bytes = 201 Cycles
90
The "count" term in the number of sets of values to be added and stored. Loading
COUNT-1 will allow a fall through the BPL instruction when computation on all set of values has been completed.
cause it only requires two bytes for each indirect pointer, the OP CODE plus the pointer of the Page Zero location, whereas in the case of the absolute, it takes three bytes, the OP CODE, address low and address high.
It is noted that there are six bytes of Page Zero memory used for pointers, two bytes for each pointer. The number of memory locations allocated to the problem are 17 for the indirect and 14 for the problem where the values are known. The execution time is longer in the in direct loop. Even though the increase in time for a single pass through the loop is only three cycles, if many values are to be trans ferred, it adds up. It is important to note that loops require time for setup but it is only used once. But in the loop itself, additional time is multiplied by the number of times the program goes through the loop; therefore, on problems where execution time is important, the time reduction effort should be placed on the loop.
Even though the loop time is longer and the actual memory expended is greater for the indexed indirect add, it has the advantage of not requiring determination of the locations of FIELD 1, FIELD 2, and FIELD
3 at the time the program was written as is necessary with absolute.
An attempt to define problems to take advantage of this shorter memory and execution time by defining fields should be investigated first. However, in almost every program, the same operation must be performed several times. In those cases, it is sometimes more useful to define a subroutine and set the values that the subroutine will operate on as fields in memory. Pointers to these fields are placed in the Zero Page of memory and then the indexed indirect operation is used to perform the function. This is the primary use of the indexed indirect operation.
91
In the case of all of the indirect operations previously described, the indirect reference was always to a Page Zero location from which is picked up the effective address low and effective address high.
There is an exception in the MCS650X microprocessor family for the jump instruction in which absolute indirect jumps are allowed. The use of the absolute indirect jump is best explained in the discussion on interrupts where the addressing mode and its capabilities are explained.
As has been developed in many of the previous examples, an index register has primary values as a modifier and as a counter. As a modifier to a base address operation, it allows the accessing of contiguous groups of data by simple modification of the index. This is the primary application of indexes and it is for this purpose they were created. By virtue of the fact that all of the MCS650X instructions have the base address in the instruction, or in the case of the in direct, in the pointer, a single index can usually be used to service an entire loop, because each of the many instructions in the loop normally are referring to the same relative value in each of the lists.
An example is adding the third byte of a number to its corresponding third byte of another number, then storing the result in the memory location representing the third byte of the result; therefore, the index register only needs to contain three to accomplish all three of these offset functions.
Some other microprocessors use internal registers as indirect point ers. The single register requirement is a significant advantage of the type of indexing done in the MCS650X. Even though the MCS650X has two indexes, moire often than not, a single index will solve many of the problems because of the fact that the data is normally organized in corresponding fields.
The second feature of the MCS650X type of indexing is that, if used properly, the index register also contains the count of the operations to be performed.
92
The examples have tried to show how to take advantage of that feature. There are two approaches to counting; forward counting and reverse counting. In forward counting, the data in memory can be organized such that the index register starts at zero and is added to on each successive operation. The disadvantage of this type of approach is that the compare index instruction, as used in Example
6.13 must be inserted into the loop in order to determine that the correct number of operations is completed.
The reverse counting approach has been used in the latter examples. The data must be organized for reverse counting operation.
The first value to be operated on is at the end of the FIELD, the next value is one memory location in front of that, etc. The ad vantage of this type of operation is that it takes advantage of the combined decrement and test capability of the There are two ways to use the test. First there is the case where the actual number of operations to be performed is loaded into the index register such as was done in Example 6.13. In this case, the index contains the correct count but if added to the base directly, would be point ing to one value beyond the FIELD because the base address contains the first byte. Therefore, when using the actual count in the index register, one always references to the base address minus one. This is easily accomplished as shown in the examples. The cross assembler accepts symbolic references in the form of base address minus one, and the microprocessor very carefully performs the operation shown.
The advantage of putting the actual count in the register is that the branch if not equal instruction CBNE) can be used because the value of the register goes to zero on the last operation.
The second alternative is to load the counter with the count minus one as done in Example 6.14. In this case, the actual value of the base address is used in the offset. However, the branch back to loop now is a branch plus, remembering that the value in the index register will not go to minus (all ones) until we decrement past zero.
93
Values of count minus one through zero will all take the branch.
It is only when attempting to reference less than the base address that the loop will be completed.
Either approach gives minimum coding and only requires that the user develop a philosophy of always organizing his data with the first value at the end. In many cases, the operations such as
MOVE can be performed even if the data is organized the other way.
Experienced programmers find that this reverse counting form is actually more convenient to use and always results in minimum loop time and space.
Although for most applications, the 8-bit index register allows simple count in offset operations, there are a few operations where the 256 count that is available in the 8-bit register is not enough to perform the indexed operations. There are two solutions to this problem. First, to code the program with two sets of bases, that is duplicating the coding for the loop with two different address highs, each one a page apart. The second, more useful solution, is to go to indirect operations because the indirect pointer can be modified to allow an infinite indexed operation. An example of the move done under 256 and over 256 is shown in the following example:
Example: 6.16: Move N Bytes (N<256)
Number of Program Instruction
Cycles Label Mnemonics
2
4
4
2
3
LOOP
LDX
LDA
STA
DEX
BNE
OPERAND
FIELD
//BLOCK
FROM-1,X
TO —1,X
LOOP
Comments
Setup 2 Cycles
LOOP Time:
13 cycles
Memory Required:
11 Bytes
3
2
6
2
3
5
3
2
2
5
5
2
2
3
2
3
Number of Program Instruction
Cycles Label Mnemonics
MOVE 2
3
2
3
2
LDA
STA
LDA
STA
LDA
LOOP
SPECIAL
OUT
STA
LDA
STA
LDX
LDY
LDA
STA
DEY
BNE
INC
INC
DEX
BMI
BNE
LDY
BNE
--
Operand
FIELD
//FROML
FRPOINT
//FROMH
FRiPOINT + 1
//TOL
TOPOINT
#TDH
TOPOINT + 1
//BLOCKS
#0
(FRPOINT) ,Y
(TOPOINT),Y
LOOP
FRFOINT + 1
TOPOINT + 1
OUT
LOOP
//COUNT
LOOP
Menlory required:
AO bytes
Comments
Move from address to an indirect pointer
Move A to address to an index pointer
Setup # of 256 blocks to move
Loop Time: 16 cycles/ byte. Move 256 bytes
Increase high pointer
Check for last move
Setup last move
95
CHAPTER 7
INDEX REGISTER INSTRUCTIONS
The index registers can be treated as auxiliary-general purpose reg isters, having the added ability of being incremented and decremented because of the normal operations in which they are required to perform.
7.0 L D X - LOAD INDEX Y
Load the index register X from memory.
The symbolic notation is M X.
LDX does not affect the C or V flags; sets Z if the value loaded was zero, otherwise resets it; sets N if the value loaded in bit 7 is a 1; otherwise N is reset, and affects only the X register.
The addressing modes for LDX are Immediate; Absolute; Zero Page;
Absolute Indexed by Y; and Zero Page Indexed by Y.
7.1 L D Y - LOAD INDEX REGISTER Y F R O M Y
Load the index register Y from memory.
The symbolic notation is M -*■ Y.
LDY does not affect the C or V flags, sets the N flag if the value loaded in bit 7 is a 1, otherwise resets N, sets Z flag if the loaded value is zero otherwise resets Z and only affects the Y reg ister. The addressing modes for load Y are Immediate; Absolute;
Zero Page; Zero Indexed by X, Absolute Indexed by X.
96
7.2 STX - STORE INDEX REGISTER X IN MEMORY
Transfers value of X register to addressed memory location.
The symbolic notation is X -*■ M.
No flags or registers in the microprocessor are affected by the store operation. The addressing modes for STX are Absolute,
Zero Page, and Zero Page Indexed by Y.
7.5
STY STORE INDEX REGISTER Y IN MEMORY
Transfer the value of the Y register to the addressed memory location. The symbolic notation is Y -*■ STY does not affect any flags or registers in the microprocessor. The addressing modes for
STY are Absolute; Zero Page; and Zero Page Indexed by X.
7.4 M X - INCREMENT INDEX REGISTER X B Y ONE
Increment X adds 1 to the current value of the X register. This is an 8-bit increment which does not affect the carry operation, there fore, if the value of X before the increment was FF, the resulting value is 00. The symbolic notation is X + 1 -»■ X. INX does not affect the carry or overflow flags; it sets the N flag if the result of the increment has a one in bit 7, otherwise resets N; sets the Z flag if the result of the increment is 0, otherwise it resets the Z flag.
INX does not affect any other register other than the X register. INX is a single byte instruction and the only addressing mode is Implied.
7.5 IN Y - INCREMENT INDEX REGISTER Y B Y ONE
Increment Y increments or adds one to the current value in the
Y register, storing the result in the Y register. As in the case of
INX the primary application is to step thru a set of values using the
Y register. The symbolic notation is Y + 1 -*■ Y. The INY does not affect the carry or overflow flags, sets the N flag if the result of the increment has a one in bit 7, otherwise resets N, sets Z if
97
as a result of the increment the Y register is zero otherwise resets the Z flag. Increment Y is a single byte instruction and the only addressing mode is Implied.
7.
6 DEX - DECREMENT INDEX REGISTER X BY ONE
This instruction subtracts one from the current value of the index register X and stores the result in the index register X.
The symbolic notation is X - 1 ->■ X.
DEX does not affect the carry or overflow flag, it sets the
N flag if it has bit 7 on as a result of the decrement, otherwise it resets the N flag; sets the Z flag if X is a 0 as a result of the decrement, otherwise it resets the Z flag.
DEX is a single byte instruction, the addressing mode is
Implied.
7.7
DEY
-
DECREMENT INDEX REGISTER Y BY ONE
This instruction subtracts one from the current value in the in dex register Y and stores the result into the index register Y. The result does not affect or consider carry so that the value in the index register Y is decremented to 0 and then through 0 to FF.
Symbolic notation is Y - 1 ■+ Y.
Decrement Y does not affect the carry or overflow flags; if the
Y register contains bit 7 on as a result of the decrement the N flag is set, otherwise the N flag is reset. If the Y register is 0 as a result of the decrement, the Z flag is set otherwise the Z flag is reset. This instruction only affects the index register Y.
DEY is a single byte instruction and the addressing mode is
Implied.
NOTE: Decrement of the index registers is the most convenient method of using the index registers as a counter, in that the decre ment involves setting the value N on as a result of having passed through 0 and sets Z on when the results of the decrement are 0.
98
7 .8 CPX - COMPARE
This instruction subtracts the value of the addressed memory location from the content of index register X using the adder but does not store the result; therefore, its only use is to set the
N, Z and C flags to allow for comparison between the index register
X and the value in memory.
The symbolic notation is X - M.
The CPX instruction does not affect any register in the machine; it also does not affect the overflow flag. It causes the carry to be set on if the absolute value of the index register X is equal to or greater than the data from memory. If the value of the memory is greater than the content of the index register X, carry is reset.
If the results of the subtraction contain a bit 7, then the N flag is set, if not, it is reset. If the value in memory is equal to the value in index register X, the Z flag is set, otherwise it is reset.
Page.
The addressing modes for CPX are Immediate, Absolute and Zero
7.9 CPY - COMPARE REGISTER Y TO
This instruction performs a two's complement subtraction between the index register Y and the specified memory location. The results of the subtraction are not stored anywhere. The instruction is strict ly used to set the flags.
The symbolic notation for CPY is Y - M.
CPY affects no registers in the microprocessor and also does not affect the overflow flag. If the value in the index register Y is equal to or greater than the value in the memory, the carry flag will be set, otherwise it will be cleared. If the results of the subtrac tion contain bit 7 on the N bit will be set, otherwise it will be cleared. If the value in the index register Y and the value in the memory are equal, the zero flag will be set, otherwise it will be cleared.
The addressing modes for CPY are Immediate, Absolute and Zero Page.
99
7.10 TRANSFERS BETWEEN THE INDEX REGISTERS A N D ACCUMULATOR
There are four instructions which allow the accumulator and in dex registers to be interchanged. They are TXA, TAX which transfers the contents of the index register X to the accumulator A and back, and TYA, TAY which transfers the contents of the index register Y to the accumulator A and back. The usefulness of this will be discussed after the instructions.
7.11 TAX - TRANSFER ACCUMULATOR TO INDEX X
This instruction takes the value from accumulator A and trans fers or loads it into the index register X without disturbing the content of the accumulator A.
The symbolic notation for this is A -> X.
TAX only affects the index register X, does not affect the carry or overflow flags. The N flag is set if the resultant value in the index register X has bit 7 on, otherwise N is reset. The Z bit is set if the content of the register X is 0 as a result of the opera tion, otherwise it is reset. TAX is a single byte instruction and its addressing mode is Implied.
7.12 TXA - TRANSFER INDEX ACCUMULATOR
This instruction moves the value that is in the index register
X to the accumulator A without disturbing the content of the index register X.
The symbolic notation is X -> A.
TXA does not affect any register other than the accumula tor and does not affect the carry or overflow flag. If the result in
A has bit 7 on, then the N flag is set, otherwise it is reset. If the resultant value in the accumulator is 0, then the Z flag is set, other wise it is reset.
The addressing mode is Implied, it is a single byte instruction.
100
7.13 T A Y - TRANSFER ACCUMULA TOR TO INDEX Y
This instruction moves the value of the accumulator into index register Y without affecting the accumulator.
The symbolic notation is A -> Y.
TAY instruction only affects the Y register and does not affect either the carry or overflow flags. If the index register Y has bit 7 on, then N is set, otherwise it is reset. If the content of the index register %
Y equals 0 as a result of the operation, Z is set on, otherwise it is reset.f;
TAY is a single byte instruction and the addressing mode is Implied.
7.14 TYA - TRANSFER INDEX Y TO ACCUMULA TOR
This instruction moves the value that is in the index register Y to accumulator A without disturbing the content of the register Y.
The symbolic notation is Y •+ A.
TYA does not affect any other register other than the accumula tor and does not affect the carry or overflow flag. If the result in the accumulator A has bit 7 on, the N flag is set, otherwise it is reset. If the resultant value in the accumulator A is 0, then the Z flag is set, otherwise it is reset.
The addressing mode is Implied and it is a single byte instruc tion .
Some of the applications of the transfer instructions between accumulator A and index registers X, Y are those when the user wishes to use the index register to access memory locations where there are multiple byte values between the addresses. In this application a count is loaded into the index register, the index register is trans ferred to the accumulator, a value such as 5, 7, 10, etc. is added immediate to the accumulator and results stored back into the index
101
register using the TAX or TAY instruction. The consequence of this type of operation is that it allows the: microprocessor to address non-consecutive locations in memory. Another application is where the internal transfer instructions allow the index registers to hold intermediate values for the accumulator which allows rapid transfer to and from the accumulator to help solve high speed data shuffling problems.
7.15 SUMMARY OF INDEX REGISTER APPLICATIONS AND MANIPULATIONS
Primary use of index register X and Y is as offset and counters for data manipulation in which the index register is used to compute an address based on the value of the index register plus base address specified by the user, either in a fixed instruction format or in a variable pointer type format. In order to operate as both an offset and counter, index registers may be incremented or decremented by one or compared to values from memory. There are limitations on the applications of each of the index registers which have to do with formats which are unique to certain instruction addressing modes.
Because of the ability of the index registers to be loaded, changed and stored, they are also useful as general purpose registers. They can be used as interim storages for moves between memory locations or for moves between memory and the accumulator.
One of the optimum uses of the indexing concept is the case when the index register is being used both as an offset and a counter.
This type of operation uses the ability of- the microprocessor to perform a decrement function on the index registers and set flags.
Therefore, a single decrement instruction not only changes the value in the counter but can also perform a test on the count value.
102
CHAPTER 8
STACK PROCESSING
In all of the discussions on addressing, it has been assumed that either the exact location or at least a relation to an exact location of a memory address was known.
Although this is true in most of the programming for control applica tions, there are certain types of programming and applications which re quire that the basic program not be working with known memory locations but only with a known order for accessing memory. This type of programming is called re-entrant coding and is often used in servicing interrupts.
To implement this type of addressing, the microprocessor maintains a separate address generator which is used by the program to access memory.
This address generator uses a push down stack concept.
Discussions of push down stacks are usually best stated considering that if one were given 3 cards, an ace, a king and a ten and were told that the order of cards was important and asked to lay them down on the table in the order in which they were given, ace first, the king on top of it and finally the ten, and then if they were retrieved, 1 card at a time, the ten is retrieved first even though it was put on last, the king is retrieved second, the ace retrieved last, even though it was put on first.
The only commands needed to implement this operation are "put next card on stack" and "pull next card from the stack." The stack could be processing clubs and then go to diamonds and back to clubs. However, we know that while we are processing clubs, we will always find ten first, king second, etc.
103
The hardware implementation of the ordered card stack which just described is a 16-bit counter, into which the address of a memory location is stored. This counter is called a "Stack Pointer." Every time data is to be pushed onto the stack, the stack pointer is put out on the address bus, data is written into the memory addressed by the stack pointer, and the stack pointer is decremented by 1 as may be seen in Example 8.1.
by 1. The stack pointer is put out on the address bus, and data is read from the memory location addressed by the stack pointer. This implementa is program independent addressing.
Example 8.1: Basic stack map for 3-deep JMP to subroutine sequence
Stack Address
01FF
01FE
01FD
01FC
01FB
01FA
01F9
Data
PCH 1
PCL 1
PCH 2
PCL 2
PCH 3
PCL 3
In the above example, the stack pointer starts out at 01FF. The stack pointer is used to store the first state of the program counter by storing the content of program counter high at 01FF and the content of program counter low at 01FE. The stack pointer would now be pointed at 01FD. The second time the store program count is performed, the program counter high number is stored on the stack at 01FD and the program counter low is stored at 01FC. The stack pointer would now be pointing at 01FB. The same pro cedure is used to store the third program counter.
When data is taken from the stack, the PCL 3 will come first and the
PCH 3 will come second just by adding I to the stack pointer before each memory read. The example above contains the program count for 3 successive jump and store operations where the jump transfers control to a subroutine and stores the value of the program counter onto the stack in order to re member to which address the program should return after completion of the subroutine.
104
Following is an example of a program that would create the Example 8.1 stack operation.
Example 8.2: Basic stack operation
Program
Counter Label Instruction
PCI Jump to Subroutine 1
PC2
SUB1
Jump to Subroutine 2
PC3 Jump to Subroutine 3
SUB3
105
This is known as subroutine nesting and is often encountered in solv ing complex control equations.
To correctly use the stack for this type of operation requires a jump to subroutine and a return from subroutine instruction.
8.1 JSR - JUMP TO SUBROUTINE
This instruction transfers control of the program counter to a sub routine location but leaves a return pointer on the stack to allow the user to return to perform the next instruction in the main program after the subroutine is complete. To accomplish this, JSR instruction stores the program counter address which points to the last byte of the jump instruc tion onto the stack using the stack pointer. The stack byte contains the program count high first, followed by program count low. The JSR then transfers the addresses following the jump instruction to the program counter low and the program counter high, thereby directing the program to begin at that new address.
The symbolic notation for this is PC + 21, (PC + 1) ->■ PCL,
(PC + 2) ■> PCH.
The JSR instruction affects no flags, causes the stack pointer to be decremented by 2 and substitutes new values into the program counter low and the program counter high. The addressing mode for the JSR is always
Absolute.
Example 8.3 gives the details of a JSR instruction.
Example 8.3: Illustration of JSR Instruction
Program Memory
PC Data
0100 JSR
0101 ADL
0102 ADH Subroutine
Stack Memory
Stack
Pointer Stack
01FD
01FE
01FF
02
01
106
Cycle
1
Address Bus
0100
0101
01FF
01FF
Data Bus
OP CODE
New ADL
PCH
External
Operations
Fetch
Instruction
Fetch
New ADL
Store PCH
Internal
Operations
Finish Previous
Operation; Incre ment PC to 0101
Decode JSR;
Increment PC to 0102
Store ADL
Hold ADL, Decre ment S to 01FE
01FE PCL Store PCL Hold ADL, Decre ment S to 01FD
0102
ADH Fetch ADH Store Stack Pointer
ADH, ADL New
OP CODE
* S denotes "Stack Pointer."
Fetch New
OP CODE
ADL ->■ PCL
ADH PCH
In this example, it can be seen that during the first cycle the micro processor fetches the JSR instruction. During the second cycle, address low for new program counter low is fetched. At the end of cycle 2, the microprocessor has decoded the JSR instruction and holds the address low in the microprocessor until the stack operations are complete.
NOTE: The stack is always stored in Page 1 (Hex address 0100-01FF).
The operation of the stack in the MCS650X microprocessor is such that the stack pointer is always left pointing at the next memory location into which data can be stored. In Example 8.3, the stack pointer is assumed to be at 01FF in the beginning and PC at location 0100. During the third cycle, the microprocessor puts the stack pointer onto the address lines and on the fourth writes the contents of the current value of the program counter high, 01, into the memory location indicated by the stack pointer address.
During the time that the write is being accomplished, the stack pointer is being automatically decremented by 1 to 01FE. During the fifth cycle the
PCL is stored in the next memory location with the stack pointer being auto matically decremented.
It should be noted that the program counter low, which is now stored in the stack, is pointing at the last address in the JSR sequence. This is not what would be expected as a result of a JSR instruction. It would be expected that the stack points at the next instruction. This apparent anomaly in the machine is corrected during the Return from Subroutine in struction.
107
Note: At the end of the JSR instruction, the values on the stack con tain the program counter low and the program counter high which referenced the last address of the JSR instruction. Any subroutine calls which want to use the program counter as an intermediate pointer must consider this fact. It should be noted also that the Return from Subroutine instruction performs an automatic increment at the end of the RTS which means that any program counters which are substituted on the stack must be 1 byte or 1 pointer count less than the program count to which the programmer expects the RTS to return.
The advantage of delaying the accessing of the address high until after the current program counter can be written in the stack is that only the address low has to be stored in the microprocessor. This has the effect of shortening the JSR instruction by 1 byte and also minimizing in ternal storage requirements.
After both program counter low and high have been transferred to the stack, the program counter is used to access the next byte which is the ad dress high for the JSR. During this operation, the sixth cycle, internally the microprocessor is storing the stack pointer which is now pointing at
01FD or the next location at which memory can be loaded.
During the seventh cycle the address high from the data bus and the address low stored in the microprocessor are transferred to the new program counter and are used to access the next OP CODE, thus making JSR a 6-cycle instruction.
At the completion of the subroutine the programmer wants to return to the instruction following the Jump-to-Subroutine instruction. This is accomplished by transferring the last 2 stack bytes to the program counter which allows the microprocessor to resume operations at the instruction fol lowing the JSR, and it is done by means of the RTS instruction.
This instruction loads the program count low and program count high from the stack into the program counter and increments the program counter so that it points to the instruction following the JSR. The stack pointer is adjusted by incrementing it twice.
The symbolic notation for the RTS is PC+, INC PC.
108
The RTS instruction does not affect any flags and affects only PCL and PCH. RTS is a single-byte instruction and its addressing mode is Im plied .
The following Example 8.4 gives the details of the RTS instruction.
It is the complete reverse of the JSR shown in Example 8.3.
Example 8.4: Illustration of RTS instruction
Program Memory
PC
0300
0301
Data
RTS
?
Stack Memory
Stack Pointer Stack
01FD
01FE
01FF
?
02
01
Cycle
2
3
4
5
6
7
Return from Subroutine (Example)
Address Bus
0301
01FD
01FE
Data Bus
OP CODE
Discarded
Data
Discarded
Data
02
External
Operations
Fetch
OP CODE
Fetch Dis carded Data
Fetch Dis carded Data
Fetch PCL
01FF
0102
0103
01
Discarded
Data
Next
OP CODE
Fetch PCH
Put Out PC
Fetch Next
OP CODE
Internal
Operations
Finish Previous
Operation, 0301 -* PC
Decode RTS
Increment Stack
Pointer to 01FE
Increment Stack
Pointer to 01FF
Increment PC by 1 to 0103
As we can see, the RTS instruction effectively unwinds what was done to the stack in the JSR instruction. Because RTS is a single-byte
109
instruction it wastes the second memory access in doing a look-ahead oper ation. During the second cycle the value located at the next program ad dress after the RTS is read but not used in this operation. It should be noted that the stack is always left pointing at the next empty location, which means that to pull off the stack, the microprocessor has to wait 1 cycle while it adds 1 to the stack address. This is done to shorten the interrupt sequence which will be discussed below; therefore, cycle 3 is a dead cycle in which the microprocessor fetches but does not use the current value of the stack and, like the fetch of address low on Indexed and Zero
Page Indexed operations, does nothing other than initialize the micro processor to the proper state. It can be seen that the stack pointer de crements as data is pushed on to the stack and increments as data is pulled from the stack. In the fourth cycle of the RTS, the microprocessor puts out the 01FE address, reads the data stored there which is the pro gram count low which was written in the second write cycle of the JSR.
During the fifth cycle, the microprocessor puts out the incremented stack picking up the program count high which was written in the first write cycle of the JSR.
As is indicated during the discussions of JSR, the program counter stored on the stack really points to the last address of the JSR instruc tion itself; therefore, during the sixth cycle the RTS causes the program count from the stack to be incremented. That is the only purpose of the sixth cycle. Finally, in the seventh cycle, the incremented program counter is used to fetch the next instruction; therefore, RTS takes 6 cycles.
Because every subroutine requires 1 JSR followed by 1 RTS, the time to jump to and return from a subroutine is 12 cycles.
In the previous 2 examples, we have shown the operations of the JSR located in location 100 and the RTS located in location 300. The follow ing pictorial diagram, Example 8.5, illustrates how the memory map for this operation might look:
110
Example 8.5: Memory map for RTS instruction
Address
Bus
100
101
102
Data
JSR
04
02
0204 First Instruction of Subroutine^—
--- 0300 RTS
With this capability of subroutining, the microprocessor allows the programmer to go from the main program to 1 subroutine, to the second sub routine, to a third subroutine, then finally working its way back to the main program. Example 8.6 is an expansion of Example 8.2 with the returns included.
Example 8.6: Expansion of RTS memory map
Main Program
JSR SUB1
► Next Inst,
SUB1 Stack Located at
01FF, 01FE
Test a Value
Stack Located at
01FD, 01FC
Stack Located at
01FB, 01FA
111
This concept is known as nesting of subroutines, and the number of subroutines which can be called and returned from in such a manner is limited by only the length of the stack.
As we have seen, the primary requirement for the stack is that irre spective of where or when a stack operation is called, the microprocessor must have an independent counter or register which contains the current memory location value of the stack address. This register is called the
Stack Pointer, S. The stack becomes an auxiliary field in memory which is basically independent of programmer control. We will discuss later how the stack pointer becomes initialized, but once it is initialized, the pri mary requirement is that it be self-adjusted; in other words, operations which put data on the stack cause the pointer to be decremented automati cally; operations which take data off from the stack cause the pointer to be incremented automatically. Only under rare circumstances should the programmer find it necessary to move his stack from one location to another if he is using the stack as designed.
On this basis, there is no need for a stack to be longer than 256 bytes.
To perform a single subroutine call takes only 2 bytes of stack memory.
To perform an interrupt takes only 3 bytes of stack memory. Therefore, with 256 bytes, one can access 128 subroutines deep or interrupt ourselves
85 times. Therefore the length of the stack is extremely unlikely to be limiting. The MCS6501 through MCS6505 have a 256-byte stack length.
Figure 8.1, which is now the complete block diagram, shows all of the microprocessor registers. The 8-bit stack pointer register, S, has been added. It is initialized by the programmer and thereafter automatically increments or decrements, depending on whether data is being put on to the stack or taken off the stack by the microprocessor under control of the program or the interrupt lines.
112
2___ 11___ li
INDEX
X
INDEX
Y
S
H
DATA BUS
HZ ZH
ALU
C = ^ A
INTERNAL
INTERNAL sz.
<2.
ABL
MEMORY
7%
n
PCL
ADL
IE
ADH
0
ABH
S I
£
PCH
A2_
Partial Block Diagram o f MCS650X Including Stack Pointer, S
FIGURE 8.1
The primary purpose of the stack is to furnish a block of memory loca tions in which the microprocessor can write data such as the program coun ter for use in later processing. In many control systems the requirements for Read/Write memory are very small and the stack just represents another demand on Read/Write memory. Therefore these applications would like the stack to be in the Page Zero location in order that memory allocation for the stack, the Zero Page operations, and the indirect addresses can be performed. Therefore, one of the requirements of a stack is that it be easily locatable into Page Zero.
On the other hand, if more than 1 page of RAM is needed because of the amount of data that must be handled by the user programs, having the stack in Page Zero is an unnecessary waste of Page Zero memory in the sense that the stack can take no real advantage of being located in Zero Page, whereas other operations can.
113
In each of the examples, the stack has been located at high order address 01 followed by a low order address. In the same manner as the microprocessor forces locations 00 on to the high order 8 bits of the address lines for Zero Page operations, the microprocessor automatically puts 01 Hex on to the high order 8-bit address lines during each stack operation. This has the advantage to the user of locating the stack into
Page One of memory which would be the next memory location added if the
Zero Page operation requirements exceed Page Zero memory capacity. This has the advantage of the stack not requiring memory to be added specifi cally for the stack but only requiring the allocation of existing memory locations. It should be noted that the selected addressing concepts of the MCS650X microprocessor support devices would involve connecting the memories such that bit 8, which is the selection bit for the Page One versus Page Zero, is a "don't care" for operations in which the user does not need more than 1 page of Read/Write memory. This gives the user the effect of locating stack in Page Zero for those applications.
The second feature that should be noted from the examples is that the stack was located at the end of Page One and decremented from that point towards the beginning of the page. This is the natural operation of the stack. RAM memory comes in discrete increments, 64, 128, 256 bytes so the normal method of^allocating stack addressing is for the user to calcu late the number of bytes probably needed for stack access. This could be done by analyzing the number of subroutines which might be called and the amount of data which might be put onto the stack in order to communicate between subroutines or the number of interrupts plus subroutines which might occur with the respective data that would be stored on the stack for each of them. By counting 3 bytes for each interrupt, 2 bytes for each jump to subroutine, plus 1 byte for each programmer-controlled stack opera tion, the microprocessor designer can estimate the amount of memory which must be allocated for the stack. This is part of his decision-making pro cess in deciding how much memory is necessary for his whole program.
Once the allocation has been made, it is recommended that the user assign his working storage from the beginning of memory forward and always load his stack at the end of either Page Zero, Page One, or at the end of his physical memory which is located in one of those locations. This will
114
give the effect of having the highest bytes of memory allocated to the stack, lower bytes of memory allocated to user working storage and hope fully the two shall never overlap.
It should be noted that the natural operation of the stack, which often is called by hardware not totally under program control, is such that it will continue to decrement throughout the page to which it is allocated irre spective of the user's desire to have it do so. A normal mistake in allo cation in memory can result in the user writing data into a memory location and later accessing it with another subroutine or another part of his pro gram, only to find that the stack has very carefully written over that area as the result of its performing hardware control operations. This is one of the more difficult problems to diagnose. If this problem is suspected by the programmer, he should analyze memory locations higher than unex plained disturbed locations.
There is a distinctive pattern for stack operations which are unique to the user's program but which are quite predictable. An analysis of the value which has been destroyed will often indicate that it is part of an address which would normally be expected during the execution of the pro gram between the time data was stored and the time it was fetched. This is a very strong indication of the fact that the stack somehow or other did get into the user's program area. This is almost always caused by improper con trol of interrupt lines or unexpected operations of interrupt or subroutine calls and has only 2 solutions: (1) If the operation is normal and predict able, the user must assign more memory to his program and particularly re assign his memory such that the stack has more room to operate; or (2) if the operation of the interrupt lines is not predictable, attention must be given to solving the hardware problem that causes this type of unpredictable operation.
8.3.1 Summary of Stack Implementation
The MCS6501 through MCS6505 microprocessors have a single 8-bit stack register. This register is automatically incremented and decre mented under control of the microprocessor to perform stack manipula tion operations, under direction of the user program or the interrupt lines. Once the programmer has initialized the stack pointer to the
115
end of whatever memory he wants the stack to operate in, the program mer can ignore stack addressing other than in those cases where there is an interference between stack operations and his normal program working space.
In the MCS6501 through MCS6505, the stack is automatically located in Page One. The microprocessor always puts out the address
0100 plus stack register for every stack operation. By selected mem ory techniques, the user can either locate the stack in Page Zero or
Page One, depending on whether or not Page One exists for his hard ware .
8.4 USE OF THE STACK B Y THE PROGRAMMER
Discussed in Section 8.1 was the use of the JSR to call a subroutine.
However, not indicated was the technique by which the subroutine knew which data to operate on. There are 3 classical techniques for communi cating data between subroutines. The first and most straightforward tech nique is that each subroutine has a defined set of working registers located in the Page Zero in which the user has left values to be operated on by the subroutine. The registers can either contain the values directly or can contain indirect pointers to addresses to values which would be operated on. The following example shows the combination of these:
Example 8.7: Call-a-move subroutine using preassigned memory locations
Location 10 Count
Location 11, 12 = Base from Address
Main Line Routine
N o . of
Bytes
2
2
2
2
2
2
2
2
2
2
3
23 bytes
Instruction
LDA //Count -1
STA 10
LDA //FRADH
STA 12
LDA //FRADL
STA 11
LDA //TOADL
STA 13
LDA //T0ADH
STA 14
JSR SUB1
Location 13, 14 = Base to Address
Comment
Load Fixed Value for the Move
Set up "FROM" Pointer
Set up "TO" Pointer
116
Subroutine Coding
No. of
Bytes
1
2
1
2
2
2 total 33 bytes
Label
SUB1
LOOP
Instruction
LDY 10
LDA (11), Y
STA (13), Y
DEY
BNE LOOP
RTS
As has been previously developed, the loop time is the overriding con sideration rather than setup time for a large number of executions.
It can be seen that we have used the techniques developed in previous sections of the indirect referencing, the jump to subroutine and the return from subroutine to perform this type of subroutine value communication.
In this operation, there was no use of the stack except for the program counter value.
A second form of communication is the use of the stack itself as an intermediate storage for data which is going to be communicated to the subroutine. In order for the programmer to use the stack as an intermediate storage, he needs instructions which allow him to put data on the stack and to read from the stack. These instructions are known as push and pull in structions .
8.5 PHA - PUSH STACK
This instruction transfers the current value of the accumulator to the next location on the stack, automatically decrementing the stack to point to the next empty location.
the notation + means push to the stack, + means pull from the stack.
The Push A instruction only affects the stack pointer register which is decremented by 1 as a result of the operation. It affects no flags.
PHA is a single-byte instruction and its addressing mode is Implied.
The following example shows the operations which occur during Push A instruction.
117
Example 8.8: Operation of PHA, assuming stack at 01FF
External
Cycles Address Bus Data Bus Operations
Internal
Operations
1 0100 OP CODE Fetch
Instruction
2
3
4
0101
01FF
0101
Next
OP CODE
(A)
Next
OP CODE
Fetch Next
OP CODE and Discard
Write A on
Stack
Fetch Next
OP CODE
Finish Previous
Operation, Incre ment PC to 0101
Interpret PHA In struction, Hold
P-Counter
Decrement Stack
Pointer to 01FE
As can be seen, the PHA takes 3 cycles and takes advantage of the fact that the stack pointer is pointing to the correct location to write the value of A. As a result of this operation, the stack pointer will be setting at 01FE. The notation (A) implies contents of A. Now that the data is on the stack, later on in the program the programmer will call for the data to be retrieved from the stack with a PLA instruction.
This instruction adds 1 to the current value of the stack pointer and uses it to address the stack and loads the contents of the stack into the
A register.
The symbolic notation for this isAt.
The PLA instruction does not affect the carry or overflow flags. It sets N if the bit 7 is on in accumulator A as a result of instructions, otherwise it is reset. If accumulator A is zero as a result of the PLA, then the Z flag is set, otherwise it is reset. The PLA instruction changes content of the accumulator A to the contents of the memory location at stack register plus 1 and also increments the stack register.
The PLA instruction is a single-byte instruction and the addressing mode is Implied.
In the following example, the data stored on the stack in Example 8.8 is transterred to the accumulator.
118
Example 8.9: Operation of PLA stack from Example 8.8
External
Operations Cycles Address Bus Data Bus
0200
PLA
0201
01FE
01FF
0201
Next
OP CODE
(A)
Next
OP CODE
Internal
Operations
Fetch
Instruction
Fetch Next
OP CODE and
Discard
Read Stack
Fetch A
Fetch Next
OP CODE
Finish Previous Opera tion, Increment PC to
101
Interpret Instruction,
Hold P-Counter
Increment Stack Pointer to 01FF
Save Stack
M A
When taking data off the stack, there is 1 extra cycle during which time the current contents of the stack register are accessed but not used and the stack pointer is incremented by 1 to allow access to the value that was previously stored on the stack. The stack pointer is left point ing at this location because it is now considered to be an empty location to be used by the stack during a subsequent operation.
8 .7 USE OF
OPERATIONS
BETWEEN
In Example 8.10, we perform the same operation as we did in Example 8.7; only here, instead of using fixed locations to pick up the pointers, we are going to use the stack as a communications vehicle:
Example 8.10: Call-a-move subroutine using the stack to communicate
Location 11, 12 = Base "FROM" Address
Bytes
Location 13, 14 = Base "TO" Address
Main Line Routine
Instruction
1
3
18
1
2
1
2
2
1
2
1
2
LDA #Count -1
PHA
LDA #FRADL
PHA
LDA #FRADH
PHA
LDA #T0ADL
PHA
LDA #T0ADH
PHA
JSR SUB1
119
Bytes Label
1
2
2
1
2
1
2
2
1
2
2
1
2
1
1
1
Total 42 Bytes
SUB1
LOOP1
Subroutine
Instruction
LDX 6
PLA
STA 10,X
DEX
BNE LOOP 1
PLA
TAY
LDA (11) ,Y
STA (13) ,Y
DEY
BNE LOOP 2
LDA 15
PHA
LDA 16
PHA
RTS
Comments
Move Stack to Memory
Set up Count
Move Memory Location
Restore PC to Stack
We can see from this example that using the stack as a communication vehicle actually increases the number of bytes in the subroutine and the total bytes overall. However, the only time one should be using subroutine in this case is when the subroutine is fairly long and the number of times the subroutine is used is fairly frequent. This technique does reduce the number of bytesjin the calling sequence. The calling sequence is normally repeated once for every time the instruction is called; therefore the use of the stack to communicate should result in a net reduction in the number of bytes used in the total program.
Up until this time, we have been considering that the stack is at a fixed location and that all stack references use the stack pointer. It has not been explained how the stack pointer in the microprocessor gets loaded and accessed. This is done through communication between the stack pointer and index register X.
This instruction transfers the value in the index register X to the stack pointer.
Symbolic notation is X -»■ S.
TXS changes only the stack pointer, making it equal to the content of the index register X. It does not affect any of the flags.
120
TXS is a single-byte instruction and its addressing mode is Implied.
Another application for TXS is the concept of passing parameters to the subroutine by storing them immediately after the jump to subroutine instruction.
In Example 8.11, the from and to address, plus the count of number of values would be written right after the JSR instruction and its address.
By locating the stack in Page Zero, the address of the last byte of the JSR can be incremented to point at the parameter bytes and then used as an indirect pointer to move the parameter to its memory location.
The key to this approach is transferring the stack pointer to X which allows the program to operate directly on the address while it is in the stack.
It should be noted that this approach automatically leaves the address on the stack, positioned so that the RTS picks up the next OP CODE address.
Example 8.11: Jump to subroutine (JSR) followed by parameters
Address Bus Data
0100
0101
0102
0103
0104
0105
0106
0107
0108
JSR
ADL
ADH
To High
To Low
From High
From Low
Count
Next OP CODE
Before concluding this discussion on subroutines and parameter passing, one should again note the use of subroutines should be limited to those cases where the user expects to duplicate code of significant length sev eral times in the program. In these cases, and only in these cases, is subroutine call warranted rather than the normal mode of knowing the addresses and specifying them in an instruction. In all cases where timing is of significant interest, subroutines should also be avoided. Subroutines add significantly to the setup and execution time of problem solution. How ever, subroutines definitely have their place in microcomputer code and there have been presented 3 alternatives for use in application programs.
The user will find a combination of the above techniques most useful for solving his particular problem.
121
This instruction transfers the value in the stack pointer to the index register X.
Symbolic notation is S -> X.
TSX does not affect the carry or overflow flags. It sets N if bit 7 is on in index X as a result of the instruction, otherwise it is reset. If index X is zero as a result of the TSX, the Z flag is set, other wise it is reset. TSX changes the value of index X, making it equal to the content of the stack pointer.
TSX is a single-byte instruction and the addressing mode is Implied.
During the interrupt sequences, the current contents of the processor status register (P) are saved on the stack automatically. However, there are times in a program where the current contents of the P register must be saved for performing some type of other operation. A particular example of this would be the case of a subroutine which is called independently and which involves decimal arithmetic. It is important that the programmer keeps track of the arithmetic mode the program is in at all times. One way to do this is to establish the convention that the machine will always be in binary or decimal mode, with every subroutine changing its mode being responsible for restoring it back to the known state. This is a superior convention to the one that is about to be described.
A more general convention would be one in which the subroutine that wanted to change modes of operation would push P onto the stack, then set the decimal mode to perform the subroutine and then pull P back from the stack prior to returning from the subroutine.
Instructions which allow the user to accomplish this are as follows:
8.11 PHP - PUSH TUS O N
This instruction transfers the contents of the processor status reg ister unchanged to the stack, as governed by the stack pointer.
Symbolic notation for this is P+.
The PHP instruction affects no registers or flags in the micropro cessor .
PHP is a single-byte instruction and the addressing mode is Implied.
122
This instruction transfers the next value on the stack to the Proces sor Status register, thereby changing all of the flags and setting the mode switches to the values from the stack.
Symbolic notation is +P.
The PLP instruction affects no registers in the processor other than the status register. This instruction could affect all flags in the status register.
PLP is a single-byte instruction and the addressing mode is Implied.
8.13 S U M M A R Y O N THE STACK
The stack in the MCS650X family is a push-down stack implemented by a processor register called the stack pointer which the programmer ini tializes by means of a Load X immediately followed by a TXS instruction and thereafter is controlled by the microprocessor which loads data into mem ory based on an address constructed by adding the contents of the stack pointer to a fixed address, Hex address 0100. Every time the microproces sor loads data into memory using the stack pointer, it automatically decre ments the stack pointer, thereby leaving the stack pointer pointing at the next open memory byte. Every time the microprocessor accesses data from the stack, it adds 1 to the current value of the stack pointer and reads the memory location by putting out the address 0100 plus the stack pointer.
The status register is automatically pointing at the next memory location to which data can now be written. The stack makes an interesting place to store interim data without the programmer having to worry about the actual memory location in which data will be directly stored.
There are 8 instructions which affect the stack. They are: BRK,
JSR, PHA, PHP, PLA, PLP, RTI, and RTS.
BRK and RTI involve the handling of the interrupts.
CHAPTER 9
RESET AND INTERRUPT CONSIDERATIONS
9.0 VECTORS
Before developing the concepts of how the MCS650X Microprocessors handle interrupts and start-up, a brief definition of the concept of vector pointers needs to be developed.
In the sections on Jumps and Branches, it was always assumed that the program counter is changed by the microprocessor under control of the programmer while accessing addresses which were in program sequence. In order to get the microprocessor started and in order to properly handle external control or interrupt, there has been developed a different way of setting the program counter to point at a specific location. This concept is called vectored pointers. A vector pointer consists of a pro gram counter high and program counter low value which, under control of the microprocessor, is loaded in the program counter when certain external events occur. The word vector is developed from the fact that the micro processor directly controls the memory location from which a particular operation will fetch the program counter value and hence the concept of vector.
By allowing the programmer to specify the vector address and then by allowing the programmer to write coding that the address points to, the microprocessor makes available to the programmer all of the control necessary to develop a general purpose control program. The microprocessor has fixed address in memory from which it picks up the vectors. By this
implementation, minimum hardware in the microprocessor is obtained. Loca tions FFFA through FFFF are reserved for vector pointers for the micro processor. Into these locations are stored respectively the interrupt vectors or pointers for: non-maskable interrupt, reset and interrupt request.
9.1 RESET OR RESTART
In the microprocessor, there is a state counter which controls when the microprocessor is going to use the program counter to access memory to pick up an instruction, then after the instruction is loaded, the microprocessor goes through a fixed sequence of interpreting instructions and then develops a series of operations which are based on the OP CODE decoding.
Up to this point, it has been assumed that the program counter was set at some location and that all program counter changes are then directed by the program once the program counter had been initialized.
Instructions exist for the initialization and loading of all other registers in the microprocessor except for the initial setting of the program counter. It is for this initial setting of the program counter to a fixed location in the restart vector location specified by the micro processor programmer that the reset line in the microprocessor is pri marily used.
The reset line is controlled during power on initialization and is a common line which is connected to all devices in the microcomputer sys tem which have to be initialized to a known state. The initialization of most I/O devices is such that they are brought up in a benign state such that with minimum coding in the microcomputer, the programmer can con figure and control the I/O in an orderly fashion.
The concept has important systems implications in systems where damage can be done if peripheral devices came up in unknown states. There fore, in the MCS650X, power on or reset control operates at two levels.
First, by holding of an external line to ground, and having this external line connected to all the devices during power up transient conditions, the entire microcomputer system is initialized to a known disabled state.
Second, the releases of the reset line from the ground or TTL zero condition to a TTL one condition causes the microprocessor to be automat ically initialized, first by the internal hardware vector which causes it to be pointed to a known program location, ana secondly through a software program which is written by the user to control the orderly start-up of the microcomputer system.
All of the MCS650X family parts also obey a discipline that while the reset line is low, the system is in a stop or reset state. The micro processor is guaranteed to be in a Read state and upon release of the re set line from ground to positive, the microprocessor will continue to hold the line in a Read state until it has addressed the specified vectored count location, at which time control of the microprocessor is available to the programmer.
NOTE: The MC6800 family also follows this convention.
While the reset line is in the low state, it can be assumed that internal registers may be initialized to any random condition; therefore, no conditions about the internal state of the microprocessor are assumed other than that the microprocessor will, one cycle after the reset line goes high, implement the following sequence:
5
6
7
8
1
2
3
4
Cycles Address Bus Data Bus
?
? + 1
?
0100 + SP
0100 + SP-1
?
?
0100 + SP-2
FFFC
?
Start PCL
FFFD
PCH PCL
?
Start PCH
First
OP CODE
External Operation
Don't Care
Don't Care
Don't Care
Don't Care
Fetch First Vector
Fetch Second Vector
Load First OP CODE
Internal Operation
Hold During Reset
First Start State
Second Start State
Third Start State
Fourth Start State
Hold PCL
The start cycle actually takes seven cycles from the time the reset line is let go to TTL plus. On the €iighth cycle, the vector fetched from the memory location FFFC and FFFD is used to access the next instruction.
The microprocessor is now in a normal program load sequence, the location where the vector points should be the first OP CODE which the programmer desires to perform.
The second point that should be noted is that the microprocessor actually accesses the stack three times during the start sequence in cycles 3, 4 and 5. This is because the start sequence is in effect a specialized form of interrupt with the exception that the read/write line is disabled so that no writes to stack are accomplished during any of the cycles.
There are two major facts to remember about initialization. One, the only automatic operations of the microprocessor during reset are to turn on the interrupt disable bit and to force the program counter to the vector location specified in locations FFFC and FFFD and to load the first instruc tion from that location. Therefore, the first operation in any normal pro gram will be to initialize the stack. This should be done by having pre viously decided what the stack value should be for initial operations and then doing a LDX immediate of this value followed by a TXS. By this simple operation, the microprocessor is ready for any interrupt or non-maskable interrupt operation which might occur during the rest of the start-up sequence.
127
Once this is accomplished, the two non variable operations of the machine are under control. The program counter is initialized and under programmer control and the stack is initialized and under program control
The next operations during the initialization sequences will consist of configuring and setting up the various control functions necessary to perform the I/O desired for the microprocessor.
Specific discussion for considerations regarding the start-up are covered in Section 11.
The major things which have to be considered include the current state of the I/O device and the non destructive operations that will allow the state to be changed to the active state.
The initialization programs mostly consist of loading accumulator
A immediately with a bit pattern and storing it in the data control regis ter of an I/O device.
Note: The interrupt disable is automatically set by the micro processor during the start sequence. This is to minimize the possibility of a series of interrupts occurring during the start-up sequence because of uncontrolled external values although it is usually possible to control interrupts as part of the configuration.
The programmer should consider two effects. First, that the non maskable interrupt is not blockable by this technique since it would be possible to configure a device that was connected to a non maskable inter rupt and have to service the interrupt immediately. Secondly, the mask must be cleared at the end of the start sequence unless the user has specific reason to inhibit interrupts after he has done the start-up sequence. Therefore, the next to last instruction of the start-up sequence should be CLI.
It should be noted that the start-up routine is a series of sequential operations which should occur only during power on initial ization and is the first step in the programmed logic machine.
128
Because the execution of the routine during powpr on occurs very seldom in the normal operation of the machine, the coding for power on sequence should tend to minimize the use of memory space rather than speed.
The last instruction in the start-up sequence should initialize the decimal mode flag to the normal setting for the program.
The next instruction should be the beginning of the user's normal programming for his device, everything preceding that being known as
"housekeeping."
9.4 RESTART
It should be noted that the basic microprocessor control philosophy allows for a single common reset line which initializes all devices.
This line can be used to clear the microprocessor to a known state and to reset all peripherals to a known state; therefore, it can be used as a result of power interruption, during the power on sequence, or as an external clear by the user to re-initialize the system.
As discussed in the hardware manual, restart is often used as an aid to making sure the microprocessor has been properly interconnected and that programs have been loaded in the correct locations.
9.5 INTERRUPT CONSIDERATIONS
Up until this point, the microprocessor has to proceed under program mer control through a variety of sequences. The only way for the program mer to change the sequence of operations of the microprocessor was to change the program counter location to point at new operations. The microprocessor is in control of fetching the next instruction at the conclusion of the current instruction. The only way that external events could control the microprocessor, if it were not for interrupts, would be for the programmer to periodically interrupt or stop processing data and check to see whether or not an external event which might cause him to change his direction has occurred. The problem with this technique is that
129
I/O events are usually asynchronous, i.e., not timed with the micro processor internal instructions, therefore, it would be possible for the event to occur shortly after the programmer has stopped to look at I/O events which would mean that the event would not be sampled until the programmer took the time to stop his coding and sample again.
Because the sampling of I/O devices normally takes several byte counts or cycles to accomplish, the frequent insertion of checking routines into straight line code results in significant delays to the entire program. In trying to use this technique, there has to be a trade-off between the fact that the program wastes a significant amount of time checking events which have not yet occurred versus delaying checking of an event which has occurred and if not timely serviced the data may be lost.
In order to solve this dichotomy, the concept of interrupt is used to signal the microprocessor that an external event has occurred and the microprocessor should devote attention to it immediately. This technique accomplishes processing in which the microprocessor's program is inter rupted and the event that caused the interrupt is serviced.
Transferring most of data and control to I/O devices in driven environment will usually result in maximum program and/or program mer efficiency. Each event is serviced when it occurs which means there is a minimum amount of delaying in servicing events, also a minimum amount of coding because of elimination of the need to determine occurrence of several events simultaneously; each interrupting event is handled as a unique combination. It is possible to interrupt an interrupt processing routine and, therefore, all the interrupt logic uses the stack which allows processing of successive interrupts without any penalty other than increasing the stack length.
A real world example of an event which should interrupt is when the user is given a panic button indicating to the microcomputer some event has occurred which requires total immediate attention of the microprocessor to solving that problem.
130
The action and events are as follows: The microprocessor user pushes the panic button; the panic switch sensor causes an external device to indicate to the microprocessor an interrupt is desired; the microprocessor checks the status of the internal interrupt inhibit signal; if the internal inhibit is set, then the interrupt is
However, if it is reset or when it becomes reset through some program reaction, the following set of operations occur:
Cycles Address Bus
1
3
4
5
6
7
8
PC
PC
01FF
01FE
01FD
FFFE
FFFF
Vector
1
PCH PCL
Data Bus
OP CODE
OP CODE
PCH
PCL
P
New PCL
New PCH
OP CODE
External Operation Internal Operation
Fetch OP CODE
Fetch OP CODE
Hold Program Counter,
Finish Previous
Operation
Force a BRK
Instruction, Hold
P-Counter
Store PCH on Stack Decrement Stack
Pointer to 01FE
Store PCL on Stack Decrement Stack
Store P on Stack
Fetch Vector Low
Fetch Vector High
Pointer to 01FD
Decrement Stack
Pointer to 01FC
Put Away Stack
Fetch Interrupt
Program
Vector Low ->
PCL and Set I
Increment PC to
PC + 1
As can be seen in Example 9.2, the microprocessor uses the stack to save the reentrant or recovery code and then uses the interrupt vectors
FFFE and FFFF, (or FFFA and FFFB), depending on whether or not an interrupt request or a non maskable interrupt request had occurred. It should be noted that the interrupt disable is turned on at this point by the micro processor automatically.
131
Because the interrupt disable had to be off for an interrupt request to have been honored, the return from interrupt which loads the processor status from before the interrupt occured has the effect of clearing the interrupt disable bit. After the interrupt has been acknowledged by the microprocessor by transferring to the proper vector location, there are a variety of operations which the user can perform to service the interrupt; however, all operations should end with a single instruction which reinitializes the microprocessor back to the point at which the interrupt occurred. This instruction is called the RTI instruction.
This instruction transfers from the stack into the microprocessor the processor status and the program counter location for the instruction which was interrupted. By virtue of the interrupt having stored this data before executing the instruction and thei fact that the RTI reinitializes the microprocessor to the same state as when it was interrupted, the combination of interrupt plus RTI allows truly reentrant coding.
The symbolic notation for RTI is +P +PC.
The RTI instruction reinitializes all flags to the position to the point they were at the time the interrupt was taken and sets the program counter back to its pre-interrupt state. It affects no other registers in the microprocessor.
RTI is a single byte instruction and its addressing mode is Implied.
In the following example, we can see the internal operation of the
RTI which restores the microprocessor:
Example 9.3: Return from Interrupt
Cycles
2
1
3
4
5
6
7
Address Bus
0300
0301
01FC
01FD
01FE
01FF
PCH PCL
Data Bus
RTI
?
?
P
PCL
PCH
External Operation
Fetch OP CODE
Fetch Next OP CODE
Discarded Stack
Fetch
Fetch P Register
Fetch PCL
Fetch PCH
OP CODE Fetch OP CODE
Internal Operation
Finish Previous
Operation,Increment
PC to 0301
Decode RTI
Increment Stack
Pointer to 01FD
Increment Stack
Pointer to 01FE
Increment Stack Point er to 01FF, Hold PCL
M+PCL, Store
Stack Pointer
Increment New PC
Note the effects of the extra cycle (3) necessary to read data from stack which causes the RTI to take six cycles. The RTI has restored the stack, program counter and status register to the point they were at before the interrupt was acknowledged.
There is no automatic save of any of the other registers in the microprocessor. Because the interrupt occurred to allow data to be trans ferred using the microprocessor, the programmer must save the various in ternal registers at the time the interrupt is taken and restore them prior to returning from the interrupt. Saving of the registers is best done on the stack as this allows as many consecutive interrupts as the program ming will allow for. Therefore, the routines which save all registers and restore them are as follows:
Example 9.4: Illustration of Save and Restore for Interrupts
13
4
2
4
2
4
2
3
2
3
3
16
Cycle Bytes
1
1
1
1
1
1
1
1
5
1
1
SAVE
RESTORE
PHA
TXA
PHA
TYA
PHA
PLA
TAY
PLA
TAX
PLA
Save A
Save X
Save Y
Restore Y
Restore X
Restore A
133
The SAVE coding assumes that the programmer wants to save and to restore registers A, X and Y. It should be noted that for many inter rupts , the amount of coding that has to be performed in the interrupt is fairly small.
In this type of operation, it is usually more desirable to shorten the interrupt processing time and not use all of the registers in the machine. Therefore, a more normal interrupt processing routine would consist of just saving registers A and X which means that the restore routine would be just restore registers X and A. This has the effect of shortening the interrupt routine by two bytes, and also shortens the restore routine by two bytes and will cut 5 cycles out of the interrupt routine and 6 cycles out of the restore routine.
This technique combined with automatic features of the interrupt and the RTI allows multiple interrupts to occur with successive inter rupts interrupting the current interrupt. This is one of the advantages of the use of the stack so that as many interrupts can interrupt other interrupts as can be held in the stack. every interrupt if all registers are saved, so 42 sequences of interrupts can be stored in one page. However, in more practical situations, consecu tive interrupts hardly ever get more than about three deep.
The advantage of allowing an interrupt to interrupt an interrupt is that the whole concept behind the interrupt is that asynchronous events can be responded to as rapidly as possible; therefore, it is desirable to allow the processing to service one interrupt to be interrupted to service the second, as long as the first interrupt has been properly serviced.
To review how this is accomplished using the normal interrupt capability of the MCS650X, it is important that we which is inherent in the MCS6500 family
M6800.
As has already been discussed, all I/O operations on this type of microprocessor are accomplished by reading and writing registers which
134
actually represent connections to physical devices or to physical pins which connect to physical devices.
Up until this point, this discussion has addressed itself to transferring of data into and out of the microprocessor. However, there is a concept that is inherent in the bus discipline that says that when ever an interrupt device capable of generating an interrupt desires to accomplish an interrupt, it performs two acts; first, it sets a bit, usually bit 7, in a register whose primary purpose is to communicate to the microprocessor the status of the device. The interrupting device causes one of perhaps many output lines to be brought low. These collector-or'd outputs are connected together to the IRQ pin on the
MCS650X microprocessor.
The interrupt request to the MCS650X is the IRQ pin being at a
TTL zero. In order to minimize the handshaking necessary to accomplish an interrupt, all interrupting devices obey a rule that says that once an interrupt has been requested by setting the bit and pulling interrupt low, the interrupt will be held by the device until the condition that caused the interrupt has been satisfied. This allows several devices to interrupt simultaneously and also allows the microprocessor to ignore an interrupt until it is ready to service it. This ignoring is done by the interrupt disable bit which can be set on by the programmer and is initialized on by the interrupt sequence or by the start sequence.
Once the interrupt line is low and interrupt disable is off, the microprocessor takes an interrupt which sets on the interrupt disable.
The interrupt disable then keeps the input low line from causing more than one interrupt until an interrupt has been serviced. There is no other
* handshaking between the microprocessor and the interrupting device other than the collector-or'd line. This means that the microprocessor must use the normal addressing registers to determine which of several collector- or 'd devices caused the line to go low and to process the interrupt which has been requested.
135
Once the processor has found the interrupting device by means of analyzing status bits which indicates an interrupt has been requested, the microprocessor then clears the status by reading or writing data as indicated by the status register.
It should be noted that a significant difference between status registers and data registers in I/O devices is that status registers are never cleared by being read, only by being written into or by the microprocessor transferring data from a data register which corresponds to some status in the status register. Detailed examples of this interaction are discussed in Chapter 11. The clearing of the status register also releases the collector-or' d output thereby releasing the interrupt pin request.
The basic interaction between the microprocessor and interrupting device is when interrupting device sets the status bit and brings its output IRQ line low. If its output IRQ line is connected to the micro processor interrupt request line, the microprocessor waits until the interrupt disable is cleared, takes th€ interrupt vector, and sets the interrupt disable which inhibits further interrupts in the IRQ line.
The microprocessor determines which interrupting device is causing an interrupt and transfers data from that device.
Transferring of data clears the interrupt status and the IRQ pin. At this point, the programmer could decide that he was ready to accept another interrupt even though the data may have been read but not yet operated on.
Allowing interrupts at this point, gives the most efficient operation of the microprocessor in most applications.
There are also times when a programmer may be working on some coding the timing of which is so important that he cannot afford to allow an interrupt to occur. During these times, he needs to be able to turn on the interrupt disable. To accomplish this, the microprocessor has a set and clear interrupt disable capability.
136
As was indicated above, any one of several devices are collector-or'd to cause an IRQ. The effect of any one of the devices or combination of them having polled the IRQ line low is always the same. The interrupt stores the current status of the program counter and processor on the stack and transfers to a fixed vector address. In servicing the inter rupt, it is important to save those registers which will be used in the analysis of the interrupt and during the interrupt processing, so the normal first steps of the interrupt routine are to do the SAVE pro cedures .
The next operation is to determine which of the various potential interrupting devices caused the interrupt. To accomplish this, the programmer should make use of the fact that all interrupting devices signal the interrupt by a bit in the status register. All currently implemented 6800 and 6500 peripherals always have interrupt indicators; either bit 7 or bit 6 in their status register. Therefore, the basic loop that a user will use to verify the existence of an interrupt on one of five devices is' as follows:
Example 9.5: Interrupt Polling
No.of Bytes
3
2
3
2
3
3
2
3
2
2
Cycles
4
2
4
2
4
2
4
2
4
2
LDA
BMI
LDA
BMI
LDA
BMI
LDA
BMI -
LDA
BMI
RES1 JMP
FIRST LDA
CLI
Process 1 etc.
Status 1
FIRST
Status 2
SECOND
Status 3
THIRD
Status 4
FOURTH
Status 5
FIFTH to RESTORE
DATA 1
137
In this example, the simplest case where the potential interrupts are indicated by bit 7 being on, has been assumed. This allows advantage to be taken of the free N-bit test by following the load of the first status register with a branch on result minus. If the first device has an active interrupt request, the BMI will be taken to FIRST where the data is transferred. This automatically clears the interrupt for the first device.
To allow multiple interrupts, the load A is followed by the CLI instruction which allows the program to accept another interrupt. As a result of the
CLI, one of two things can occur; there is not another interrupt currently active, in which case, the microprocessor will continue to process the first interrupt down to the point where the interrupt is complete and the first subroutine does a jump to RESTORE, which is the routine that unsaves the registers that were used in the process of servicing the interrupt.
If another device has an active interrupt which occurred either prior to the first interrupt or subsequent to it but before the microprocessor has reached the point where the CLI occurs, then the microprocessor will immediately interrupt again following the CLI, go back and save registers as defined before and come back into the polling loop. Therefore, multiple interrupts are serviced in the order in which they are looked at in polling sequence. Polling means that the program is asking each device individu ally whether or not it is the one that requested an interrupt.
It should be noted that polling has the effect of giving perfect priority in the sense that no matter which two interrupts occur before the microprocessor gets to service one, the polling sequence always gives priority to the highest priority device first, then the second, then the third, etc. In light of the fact that this polling sequence requires no additional hardware to implement other than is available in the inter rupting devices themselves, this is the least expensive form of interrupt and the one that should be used whenever possible because of its indepen dence from external hardware.
Although it would appear that the last interrupting device in a sequence pays a significant time penalty based on the amount of instruc tions to be executed before the last device is serviced, the amount of time to perform polls is only six cycles per device and, therefore, the extra penalty that the last device has to pay over the first device is 24 cycles. This is in comparison to a minimum time to cause an inter rupt (eight cycles), plus store time for registers (in the range of another
8
to 13 cycles) which means that the delay to the last devices is roughly twice what it would be for the first device.
This timing just described represents a most interesting part of the analysis of interrupts for a microprocessor. There is a significant amount of fixed overhead which must be paid for the interrupt. This over head includes the fact that the interrupts can only occur at the end of an instruction so, therefore, if an interrupt occurs prior to the end of an instruction, the microprocessor delays until the end of the instruction to service it. Therefore, case analysis, one has to consider the fact that the interrupt might be occurring in the middle of a seven cycle, read/modify/write instruction which means that the worst case time to process the first instruction in an interrupt sequence is
14 cycles (7 cycles plus the 7 cycles for the interrupt).
In light of the fact that saving of additional registers is often required (at least the accumulator A must be saved), at least twice the number of cycles will be required. Consequently the absolute minimum worse case time for an interrupt is 17 cycles plus the time to transfer data which is another 4 cycles. Therefore interrupt driven systems must be capable of handling a delay of at least
2 0
cycles and more realistically,
20 to 50 cycles before the first interrupt is serviced. This means that devices which are running totally interrupt driven must not require succes sive bytes of data to be transferred to the microprocessor in less than 30 or 40 cycles and on a given system, only one device is capable of operating at that rate at one time. This limits the interrupt driven frequency of data transfer to 40KHZ at a one megahertz clock system and 80KHZ on a two megahertz clock system.
139
An even more serious problem is the timing delay when an interrupt has just started to be serviced. The interrupt mask is on and higher priority interrupts are blocked from service. In this case, the delay to the service can easily stretch out to
1 0 0
cycles before the interrupt mask is cleared. This is one of the reasons for clearing up the inter rupt mask as soon as data is transferred. (The non-maskable interrupt which will be discussed later is one solution to this problem.) A second is to only use interrupts for systems that have adequate buffering and/or slower transfer rates. This does not imply that most microprocessor applications should not be primarily interrupt driven. The MCS650X inter rupt system is designed to be very economical and easy to apply. It should be used for almost all control applications, other than when the throughput described is not sufficient to handle the particular problem. It should be remembered that at one megahertz the fast MCS650X is not really capable of handling problems with more than 50KHZ byte throughput for a sustained period of operation. It is also true that in most control applications, many of the signals occur at much slower rates or are bufferable so that the response time to a request for service is significantly longer than the 20 to 50 cycles that can normally be expected with a polling system.
Because of this, it is expected that most applications will be quite satisfied using the polling technique described above.
However, there are occasions where several high speed peripherals can be managed by the microprocessor if the user is willing to make the investment to attain a truly vectored interrupt. There is a second level of interrupt vectoring possible by just putting one high priority device on the non-maskable interrupt line. However, the case when multiple inputs are desired with both priority encoding and true vectoring, the
MCS650X when combined with appropriate hardware has the ability in the first polling instruction to transfer control to appropriate interrupting device service software.
140
The MCS6520 contains, in its two bytes of memory, an indirect pointer to the address of the subroutine in which resides the interrupt processing for the devices, which the priority encoder has selected. This gives an effective service time of approxi mately 24 cycles to a prioritized interrupt and is one of the primary applications of the jump indirect capability.
9.8.1 JMP Indirect
This instruction establishes a new value for the program counter.
It affects only the program counter in the microprocessor and affects no flags in the status register.
JMP Indirect is a three byte instruction.
In the JMP Indirect instruction, the second and third bytes of the instruction represent the indirect low and high bytes respectively of the memory location containing ADL. Once ADL is fetched, the program counter is incremented with the next memory location containing ADH.
Example 9.6: Illustration of JMP Indirect
Cycle
1
Address
Bus____
0100
0101
Data
Bus
OP CODE
IAL
^
External
Operation
Fetch OP CODE
Fetch IAL
Internal
Operation
Finish Previous
Operation.
Increment PC to 0101
Interpret Instructions
Increment PC to 102
6
3
4
5
0 1 0 2
IAH, IAL
IAH.IAL+1
ADH,ADL
ADL
ADH
Next OP
CODE
Fetch IAH
Fetch ADL
Fetch ADH
Fetch Next
OP CODE
Store IAL
Add 1 to IAL
Store ADL
141
There is an interrupt request line (IRQ) that, when low, indicates one of the devices which are connected to the interrupt request line requires service. At the beginning of the interrupt service routine, the user should save, on the stack, whatever registers will be used in his interrupt processing routine. His program then goes through a polling sequence to determine the interrupting device by analyzing the status registers in the order of priority of service for the I/O devices. On finding a device which requires service, the data for that device should be read or written as soon as possible and the interrupt disable cleared so that the microprocessor can interrupt again to service lower priority devices. Devices with over 40KHz byte transfer, etc., and mixed devices with over 20KHz should not normally be run interrupt driven. All others should be run interrupt driven as it minimizes the service time and programming for interrupt I/O operations. :
9.10 NON-MASKABLE INTERR
As is discussed, it is often desirable to have the ability to inter rupt an interrupt with a high priority device which cannot afford to wait during the time interrupts are disabled. For this reason, the MCS650X has a second interrupt line, called a Non-Maskable Interrupt. The input characteristics of this line are different than the interrupt request line which senses it needs service when it remains low. The non-maskable input is an edge sensitive input which means that when the collector-or’d input transitions from high to low, the microprocessor sets an internal flag such that at the beginning of the next instruction, no matter what the status of the interrupt disable, the microprocessor performs the interrupt sequence shown in Example 9.2 except that the vector pointer put out in cycle
6
and 7 is FFFA and FFFB.
This gives two effects of a non-maskable interrupt. First, no matter what the status of the interrupt disable, the non-maskable inter rupt will interrupt at the beginning of the next instruction, therefore, the maximum response time to the vector point is 14 cycles. Secondly, the internal logic of the MCS650X is such that if an interrupt request and non maskable interrupt occur simultaneously or if the non-maskable interrupt occurs prior to the time that the vectors are selected, the microprocessor
142
always assigns highest priority to the non-maskable interrupt. Therefore, the FFFA and FFFB vector are always taken if both interrupts are active at the time the vector is selected. Thus the non-maskable interrupt is always a higher priority fast response line, and can, in any given system be used to give priority to the high speed device.
It is possible to connect multiple devices to the non-maskable inter rupt line except for the fact that the non-maskable interrupt is edge sensitive. Therefore, the same logic that allows the IRQ to stay low until the status has been checked and the data transferred will keep the non-mask able interrupt line in a low state until such time as the first interrupt is serviced. If subsequent to the first interrupt of a non-maskable inter rupt line occuring, a second device which is collector-ored would have turned on its status and collector-orfd output, the clearing of the first interrupt request would not cause the line to re-initialize itself to the high state and the microprocessor would ignore the second interrupt. There fore, multiple lines connected to the non-maskable interrupt must be careful ly serviced.
In any case, NMI is always one free high priority vectored interrupt.
By virtue of the fact that it goes to a different vector pointer, the microprocessor programmer can be guaranteed that in 17 cycles he can trans fer data from the interrupting device on the non-maskable interrupt input.
The IRQ and NMI are lines which, externally to the microprocessor, control the action to the microprocessor through an interrupt sequence.
As is mentioned during the discussion on the start command, the restart cycle is a pseudo interrupt operation with a different vector being selected for reset which has priority over both interrupt and non maskable interrupt. Non-maskable interrupt has priority over interrupt.
There is also a software technique which allows the user to simulate an interrupt with a microprocessor command, BRK. It is primarily used for causing the microprocessor to go to a halt condition or stop condition during program debugging.
143
9.11 BRK - BREAK COMM A N D
The break command causes the microprocessor to go through an inter rupt sequence under program control. This means that the program counter of the second byte after the BRK is automatically stored on the stack along with the processor status at the beginning of the break instruction.
The microprocessor then transfers control to the interrupt vector.
Symbolic notation for break is PC + 24- (FFFE)-^PCL (FFFF}->-PCH.
Other than changing the program counter, the break instruction changes no values in either the registers or the flags.
The BRK is a single byte instruction and its addressing mode is
Implied.
As is indicated, the most typical use for the break instruction is during program debugging. When the user decides that the particular pro gram is not operating correctly, he may decide to patch in the break instruction over some code that already exists and halt the program when it gets to that point. In order to minimize the hardware cost of the break which is applicable only for debugging, the microprocessor makes use of the interrupt vector point to allow the user to trap out that a break has occurred. In order to know whether the vector was fetched in response to an interrupt or in response to a BRK instruction, the B flag is stored on the stack, at stack pointer plus
1
, containing a one in the break bit position, indicating the interrupt was caused by a BRK instruction. The B bit in the stack contains 0 if it was caused by a normal IRQ. Therefore, the coding to analyze for this is as follows in Example 9.6.
Example 9.7: Break-Interrupt Processing
Cycles Bytes
4
3
2
2
11
1
1
2
2
6
Check for A BRK
PLA
PHA
BNE BRKP
Flag
Load status register
Restore onto Stack
Branch to Break Programming
Normal Interrupt Processing
This coding can be inserted any place in the interrupt processing routine. During debugging, if the user can afford the execution time, it should be placed immediately after the save routine. If not, it can be put at the end of the polling routine which gives a priority to the polling devices as far as servicing the interrupts. However, it should be noted that in order not to lose the break, the returns from all inter rupts during debugging should go through an equivalent routine.
Once the user has determined that the break is on, a second analysis and correction must be made. It does not operate in a normal manner of holding the program counter pointing at the next location in memory during the BRK instruction. Because of this, the value on the stack for the program counter is at the break instruction plus two. If the break had been patched over an instruction, this is usually of no significant consequence to the user. However, if it is desired to process the next byte after the break instruction, the use of decrement memory instructions in the stack must be used.
It is recommended that the user normally takes care of patching programs with break by processing a full instruction prior to returning and then use jump returns.
An interesting characteristic about the break instruction is that it's
OP CODE is all zero's (0), therefore, BRK coding can be used to patch fusable link PROMS through a break to an E-ROM routine which inserts patch coding.
An example of using the break to patch with is shown below:
Example 9.8: Patching with a break utilizing PROMs
Old Code
Patched
Code
FC21
FC22
FC23
FC24
FC21
FC22
FC23
FC24
LDA
05
21
Next OP CODE
05
21
Next OP CODE
145
The interrupt vector routine points to:
Patch LDA
06
21
JMP
24
FC
This coding substitutes:
LDA 2106 for the
LDA 2105 coding at
FC21 by use of the BRK and a break processing routine.
A series of requirements were discussed to this point for the memory organization which can be illustrated by the following memory map:
Hex Address
0000-00FF
RAM used for zero page and indirect memory addressing operation.
0100-01FF
RAM used for stack processing and for absolute addressi
0200-3FFF Normally RAM.
4000-7FFF Normally I/O
8000-FFF9 Program storage normally ROM.
FFFA Vector low address for NMI.
FFFB Vector high address for NMI.
FFFC Vector low address for RESET.
FFFD Vector high address for RESET.
FFFE Vector low address for IRQ + BRK.
FFFF
Vector high address for IRQ + BRK.
The addressing schemes for I/O control between locations 4000 and
8000 Hex, have not been fully developed. This is described in detail in the Hardware Manual, Chapter 2. The Zero Page addressing requires that
RAM should be located starting in location 00. If more than one RAM page is necessary, RAM location 0100 through 01FF should be reserved for the stack or at least a portion of parts should be reserved for the stack with the rest of it being available to the user to use as normal RAM.
Locations from 0200 up to 4000 are normally reserved for RAM expansion.
146
In small memory configurations such as are inherent in a MCS6530 class de vice, in order to minimize the addressing lines, page two (02XX) will be normally used for input/output as opposed to using the 40XX page which is used for devices which require significant amount of outboard RAM, ROM and I/O.
Because of the fact that the MCS650X has three very important vector points selected in highest order memory, it is usually more useful to write programs with the memory storage located at a starting address which allows the programmer to make sure that the last address in his
ROM contains the start and interrupt vectors. Because of these alloca tions, the user finds himself working in three directions. RAM is assigned in location 0000 working up. I/O devices are started at location 4000 starting up and ROM starts at location FFFF and works down.
Although this seems like an unusual concept, one must remember that the hardware really only gives performance to either end of memory and, therefore, data located in the middle has no priority one over the other.
So starting at either end is just as useful a technique as starting at one end and working up.
In order to take maximum advantage of the capability of the micro processor, particularly when using a symbolic assembler, working data should be located starting in the location
0
, and stack addresses should be reserved until after analysis of the working storage requirements have been completed. Program storage should start in high order memory with some guess as to the amount of memory required being taken and that being taken as a start address. However, care should be taken to assign the three fixed vectors almost immediately at least symbolically as they are all necessary for correct operation of the microprocessor.
CHAPTER 10
SHIFT AND MEMORY MODIFY INSTRUCTIONS
In many cases operations of the control systems must operate a bit at a time. Data is often available only bit-serial and sometimes sequential bit operations are the only way to solve a particular problem. In addition to that, in order to combine bits into a field, shift and rotate instruc tions are necessary. Multiply and divide routines all require the ability to move bits relative to one another in a full multiple byte field.
The shift instruction is one that takes a register such as the accumulator and moves all of the bits in the accumulator
1
bit to the right or 1 bit to the left. Examples of the shift and rotate instructions in the
MCS650X are shown below:
Example 10.1: General shift and rotate
Shift Right Before
After
B7
0
B
6
B7
B5 B4 B3
B
6
B5 B4
B2 Bl BO
B3 B2 Bl
Carry
BO
Shift Left Before
After
B7
B
6
B
6
B5 B4 B3
B5 B4 B3 B2
B2 Bl BO
Bl BO
0
Rotate Left Before
After
B7
B
6
B
6
B5 B4 B3
B5 B4 B3 B2
B2 Bl BO
Bl BO C B7
147
As you can see from our example, moving data 1 bit to the right is called shift right. The natural consequence of the shift right is that the input bit or high order bit in this case is set to 0. Moving the data in the register 1 bit to the left is called shift left. In this case, the
0 is inserted in the low order position. These are the 2 shift capabilities that exist in the MCS650X microprocessor.
It should be noted that in both cases, the bit that is shifted from the register, the low order bit in shift right, and the high order bit in shift left, is stored in the carry flag. This is to allow the programmer to test the bit by means of the carry branches that are available and also to allow the rotate capability to transfer bits in multiple precision shifts.
The second part of the multiple precision shift instruction is the rotate which is shown in Example 10.1, in which the value of the carry bit becomes the low order bit of the register, and the output bit from the shift is stored in carry.
This instruction shifts either the accumulator or a specified memory location
1
bit to the right, with the higher bit of the result always being set to
0
, and the low bit which is shifted out of the field being stored in the carry' flag. B7 BO
The symbolic notation for LSR is Q - ► --
1
^C~|
The shift right instruction either affects the accumulator by shift ing it right
1
or is a read/modify/write instruction which changes a speci fied memory location but does not affect any internal registers. The shift right does not affect the overflow flag. The N flag is always reset. The
Z flag is set if the result of the shift is 0 and reset otherwise. The carry is set equal to bit
0
of the input.
LSR is a read/write/modify instruction and has the following address ing modes: Accumulator; Zero Page; Zero Page,X; Absolute; Absolute,X.
148
The shift left instruction shifts either the accumulator or the ad dress memory location
1
bit to the left, with the bit
0
always being set to 0 and the bit 7 output always being contained in the carry flag. ASL either shifts the accumulator left
1
bit or is a read/modify/write instruc tion that affects only memory.
3 7 3 0
The symbolic notation for ASL is C
The instruction does not affect the overflow bit, sets N equal to the result bit 7 (bit
6
in the input), sets Z flag if the result is equal to
0, otherwise resets Z and stores the input bit 7 in the carry flag.
ASL is a read/modify/write instruction and has the following address ing modes: Accumulator; Zero Page; Zero Page,X; Absolute; Absolute,X
10.3 ROL - ROTATELEFT
The rotate left instruction shifts either the accumulator or addressed memory left
1
bit, with the input carry being stored in
0
and with the input bit 7 being stored in the carry flags.
The symbolic notation for ROL is ^ ^B/^ ^B0
The ROL instruction either shifts the accumulator left 1 bit and stores the carry in accumulator bit
0
or does not affect the internal reg isters at all. The ROL instruction sets carry equal to the input bit 7, sets N equal to the input bit
6
, sets the Z flag if the result of tate is 0, otherwise it resets Z and does not affect at all.
ROL is a read/modify/write instruction and it has the following address ing modes: Accumulator; Zero Page; Zero Page,X; Absolute; Absolute,X.
149
10.4 ROR - ROTATE RIGHT (Available on Microprocessors after June, 1976)
The rotate right instruction shifts either the accumulator or addressied memory right
1
bit with bit
0
shifted into the carry and carry shifted into bit 7.
The symbolic notation for ROR is
B7 B0
c
The ROR instruction either shifts the accumulator right 1 bit and stores the carry in accumulator bit 7 or does not affect the internal regis ters at all. The ROR instruction sets carry equal to input bit 0, sets N equal to the input carry and sets the Z flag if the result of the rotate is 0 otherwise it resets Z and does not affect the overflow flag at all.
ROR is a read/modify/write instruction and it has the following address ing modes: Accumulator; Zero Page; Absolute; Zero Page,X; Absolute,X.
As indicated, all of the shift instructions can operate on the accumu lator. This is a special addressing mode that is unique to the shift in structions and operates with the following set of operations:
Example 10.2: Rotate accumulator left
External
Operation Cycles Address Bus Data Bus
100 OP CODE Fetch Next
OP CODE
Internal
Operation
101
101
102
Next
OP CODE
Next
OP CODE
Fetch Dis carded OP CODE
Fetch Next
OP CODE
Fetch Second
Byte
Finish Previous
Operation; Increment
PC to 101
Decode Current In struction; Hold P-
Counter
Shift Through the
Adder
Store Results into A;
Interpret Next OP CODE
As we can see, the accumulator instructions have the same effect as the single-byte non-stack instructions in the sense that the instruction con tains both the OP CODE and the register in which the operations are going to be performed; therefore, in cycle
2
, the microprocessor holds the pro-
150
gram counter and in cycle 3, fetches the same program counter location and starts the next instruction operation. At the same time, it is transferring the results from the adder into the accumulator; this is because of the look ahead and pipelining characteristics of the MCS650X. - The accumulator shift and rotate operations take only
2
cycles and
1
byte of memory.
The MCS650X has a series of instructions which allow the user to change the contents of memory directly with a single instruction. These instructions include all of the shift, rotate, increment and decrement mem ory instructions. The operation of each of these instructions is the same in that the addressing mode that is defined for the instruction is imple mented the same way as if for normal instructions. After the address has been calculated, the effective address is used to read the memory location into the microprocessor arithmetic unit (ALU). The ALU performs the opera tion and then the same effective address is used to write the results back into memory. The most difficult operation is the addressing mode Absolute
Indexed which is illustrated in Example 10.3 for the rotate left instruc tion, ROL:
Example 10.3: Rotate memory left Absolute»X
Cycles
1
3
4
Address Bus
100
101
102
ADH, ADL + X
Data Bus
OP CODE
ADL
ADH
External
Operation
Fetch
OP CODE
Fetch ADL
Fetch ADH
False Read
Internal
Operation
Finish Previous
Operation, Incre ment PC to 101
Decode Current In struction, Increment
PC to 102
Add ADL + X, Incre ment PC to 103
Add Carry from
Previous Add to ADH
5 Data Fetch Value
6
7
8
ADH + C,
ADL + X
ADH + C,
ADL + X
ADH + C,
ADL + X
103
Shifted
Data
OP CODE
Destroy
Memory
Store
Results
Fetch Next
OP CODE
Perform Rotate,
Turn on Write
Set Flags
Increment PC to 104
151
Cycle 4 is a wasted cycle because read/modify/write instruction should wait until the carry had been added to the address high in order to avoid writing a false memory location. This is the same logic that is used in the store instruction in which the look-ahead or the short cut addressing mode is not taken advantage of. Cycle 4 is an intermediate read, and cycle 5 is when the actual data that is going to be operated on is read.
The address lines now hold at that address for cycles 5, 6 and 7. The microprocessor signals both itself and the outside world those operations during which it will not recognize the ready line. It does this by pulling the Write line. The Write line is pulifed^in cycle 6 because data is writ ten into the memory location that is going to be written into again in cycle 7 with correct data.
Because data bits read from memory have to be modified and returned, there is no pipelining effect other than the overlap of the adding in the address low and index register. The 7 cycles it takes to perform read/ modify/write Absolute Indexed,X instruction is the worst case in timing for any section of the machine except for interrupt. This unique ability to modify memory directly is perhaps best illustrated by the coding in
Example 10.4 which is used to shift a 4-bit BCD number, which has been accumulated in the high 4 bits of the accumulator as part of the decoding operation, from the accumulator; into a memory field. Figure 10.1 is a flow chart of this example. Examples such as this often occur in pointof-sale terminals and other machines in which BCD data is entered sequen tially. This example assumes that the value is keyboard entered, through which data is entered into the accumulator from left to right but has to be shifted into memory from right to left. The value in the field before the shift is a 1729 which after the shift will be a 17,295.
152
Flow Chart for Moving in a New BCD Number
FIGURE 10.1
153
Example 10.4: Move a new BCD number Into field
Field
Before
00
00
17
29
Accumulator 50
After
00
01
72
95
00
Coding
Bytes
2
2 LOOP-2
3 LOOP-1
1
2
1
_2
14 bytes
Instruction
LDY 4
LDX 4
R O L W i c e -1, X
DEX
BNE LOOP-1
DEY
BNE LOOP-2
Set Up shifts for 4 mOVeS shift the field 1 bit four times.
There are several new concepts introduced in this example; the first is the use of index register Y as just a counter to count the number of times the character has been bit-shifted. It is a common approach to use bit shifts, as is implemented in the MCS650X family, to shift data into memory. The power of being able to communicate directly in memory is shown by shifting bits from one byte to the next byte using a single ROL indexed instruction. This example uses a loop within a loop and it should be noted that LOOP 1 occurs 4 times for every time LOOP 2 occurs. The in ternal loop is very important in the sense that this loop executes 16 times for the problem; therefore, its execution time should be optimized.
In addition to having the ability to shift and rotate memory, the
MCS650X has the ability to increment and decrement memory locations.
154
10
.7
INC - INCREMENT MEMORY BY ONE
This instruction adds 1 to the contents of the addressed memory loca tion.
The symbolic notation is M + 1
-*■
M. .
and does not affect the carry or overflow flags. If bit 7 is on as the result of the increment, N is set, otherwise it is reset; if the increment causes the result to become 0, the Z flag is set on, otherwise it is reset.
The addressing modes for increment are: Zero Page; Zero Page,X; Abso lute; Absolute,X.
10.8 DEC - DECREMENT MEMORY BY ONE
This instruction subtracts 1, in two's complement, from the contents of the addressed memory location.
Symbolic notation for this instruction is M - 1 -* M.
The decrement instruction does not affect any the
# microprocessor. It does not affect the carry or overflow flags. If bit 7 is on as a result of the decrement, then the N flag is set, otherwise it is reset. If the result of the decrement is 0, the Z flag is set, other wise it is reset.
The addressing modes for decrement are: Zero Page; Zero Page,X;
Absolute; Absolute,X.
In many examples through the report, we have used the ability to incre ment and decrement registers in the microprocessors. The advantages of incrementing and decrementing in memory are that it is possible to keep externail counters or to directly influence a bit value by means of these instructions. It is sometimes useful during I/O instructions.
10.9 GENERAL NOTE ON READ/MODIFY/WRITE INSTRUCTIONS
The ability to read, modify and write memory is unique to MCS6500 class microprocessors. The usefulness of the instructions is limited only by the user's approach to organizing memory. Even though the instructions are fairly long in execution, they are significantly shorter than having to load and save other registers to perform the same function. Experience in organizing programs to take advantage of this manipulation of memory will allow the user to fully appreciate the power of these instructions.
155
i
e
CHAPTER 11
PERIPHERAL PROGRAMMING
11.0 REVIEW OFMCS6520 FOR I/O OPERATIONS
It should be noted that in the following discussions, the major difference between the MCS6530 I/O and the main register of the MCS6520 is that the extra bit in the control register need not be used in the
MCS6530. All registers in the HCS6530 are directly addressable.
Example 11.1: The MCS6520 Register Map
A
= Base Address
PI
PI
PIBD
PIBC
* =
* + 1
*
= * + 1
* = * + 1
*
—
* 4 - 1 i
1
-------------- 1
A DATA DIRECT-| l
1
1
A DATA (AD)
A CONTROL (AC)
A SIDE l~B DATA DIRECT-!
| ION (BD) i i
J
1
1
B DATA (BD)
B CONTROL (BC)
B SIDE
In Example 11.1 a programming form to describe the PIA is shown.
The programming form is used in the Cross-Assembler and Resident
Assembler with the MCS650X product family. The notation * = is used to define any location. The notation means that the assembler instruct ion counter is set equal to the value following the equal sign. The expression * = * + 1 causes the assembler to recognize that there is one byte of memory associated with the term; therefore, we can see that the definition of the four registers PIAD, PIAC, PIBD and PIBC are consecutive memory locations starting at some base address, with
156
the first byte addressed as PIAD, the second byte addressed as PIAC, the third byte addressed as PIBD, and the fourth byte as PIBC. This is a normal way a MCS6520 would be organized and this is the way the programming form should be set up. The base address is picked up by an algorithm described in the hardware manual but normally it is a value between 4004 and 4080 Hex. Each MCS6520 is given a base address which works progressively up from 4004 Hex.
In Example 11.1 two registers are shown in dotted lines. This is because each of the A DATA (AD) and B DATA (BD) parts of the MCS6520 are actually two registers having the same address, one which specifies the direction of each of the input/output paths (the Data Direction
Register), the second one which is actually the connection to the in put/output paths (the Data Register). Because of pin limitations on the MCS6520, the microprocessor can only directly address one of the registers at a time. Differentiation as to which register is being con nected to the microprocessor is a function of bit
2
in the respective control register (AC and BC). If bit 2 is off, the Data Direction Reg ister is being addressed; if it is on, the Data Register is being ad dressed .
During the initialization sequence, therefore, the MCS6520 starts out with all registers at zero. This means that the microprocessor is addressing the Data Direction Register. The PIA initialization is done by writing the direction of the pins into the Data Direction Register
CAD, BD) and then setting on the control flag as described below. After that, the program will normally be dealing with the data registers.
Example 11.2: General PIA Initialization
LDA # DIRECT
STA PIAD
LDA # CONTR
STA PIAC
Initialize Direction
Initialize Control
157
Example 11.2 illustrates a general form of initialization and can be completed for as many PIA's as there are in the system.
11.1 MCS6520 INTERRUPT CONTROL
The MCS6520 has a basic interrupt capability which is under control of the programmer. Almost all MCS6500 I/O devices that allow interrupts have an interrupt control register that allows the user to disable the interrupt. This will keep inputs which are not necessarily active from causing spurious interrupts which must be handled by the microprocessor.
Examples of this are open tape loops or other signals which have high impedance noise sensitive inputs except when connected to some kind of media. In this type of application, normally the interrupt is enabled by some physical action from the person using the device such as load ing of; the cassette, pushing the power-on switch, etc. In the case of the MCS6520, there are two interrupt causing conditions for each control register.
Each of these interrupts concern themselves with one input pin.
The Control Register allows the programmer to decide whether or not the pin is sensitive to positive edge signals or negative edge signals and whether or not an interrupt shall occur when the- selected transition has occurred.
It should be noted that, therefore, it is possible for a line to cause a status bit to be set without causing an interrupt. The com prehensive I/O Program in Section 11.5 uses this combination.
158
Example 11.3: Interrupt Mode Setup
Bit 7 Status Bit: Bits 1 0
Set on Negative Edge
Set on Negative Edge
Set on Positive Edge
Set on Positive Edge
0
0
1
1
0
1
0
1
Bit 6 Status Bit:
Set on Negative Edge
Set on Negative Edge
Set on Positive Edge
Set on Positive Edge
Bits A
0
0
1
1
3*
0
1
0
1
*if bit 5 equals zero
Interrupt
No
Yes
No
Yes
Interrupt
No
Yes
No
Yes
The proper combination of bits are usually determined during the design of the MCS6520 interconnection and form the constant which is loaded in the control register. The constant that is loaded in the control register should contain bit 2 on. For example, to allow bit 7 to be set on negative going signals with interrupt enable and bit 6 to be set on positive signals with Interrupt disable, the control value would be Hex 15.
With bit 5 on, the pin that controls bit 6 can be set as an output pin. The output pin is either controllable by the microprocessor directly or acts as a handshake to reflect the status of reads and writes of the data register. The operation of the output pins CA2, CB2 depends on how bits 5, A, and 3 are programmed, as shown in Example 11.A.
Example 11.A ; CA2, CB2 Output Control
CA2 Output With: Bit 5 on
Bit A Bit 3
Low on read or write until bit 7 on
0
0
Low on read or write for one cycle
0
1
Always 0
Always 1
1
1
0
1
159
The decision as to whether or not to use the one cycle low until bit 7 comes on is a hardware decision, depending on the device which is hooked to the pin.
It should be of interest to the programmer to note that bit 6 con trols pins known as CA2 or CB2 which can be considered to be auxiliary outputs which are controlled by bit 3 assuming the processor is initialized so that bit 5 and bit 4 are ones.
Example 11.5 shows the use of controlling bit 3 using AND and
OR instructions; however, it should be noted that this technique applies for any individual bit in the PIA data direction register also: or CB2 Using Bit 3 Control
Set CA2
LDA
ORA
STA
PIAC
#$08
PIAC
Clear CA2
LDA
AND
STA
PIAC
#$F7
PIAC
Note: $ - Direction to Assembler for Hex Notation
# - Direction to Assembler for Production Operator
By similar techniques, every
MCS6520 can be controlled. There
1. In the MCS6520, both bit 6 and bit 7 are cleared on either side by reading of the corresponding data register if bit 6 has been set up as an input. This means that polling sequences for I/O instructions should only read the status registers and then read the data registers after the status has been determined, otherwise false clearing of the status data may occur.
2. Even though the data, a read of B data must be done to clear bit 7.
B
160
THEMCS6520
11.2.1 Shortcut Polling Sequences
In section 9.7, the techniques for using a LOAD A to poll for interrupts was covered; however, the I/O devices on the MCS6520 can either set bit
6
or bit 7 on to cause an interrupt; therefore, a different technique needs to be used to analyze the MCS6520 to , poll a series of 6520's each one of which could have caused the interrupt. It is for this purpose that the BIT instruction senses both bit
6
and bit 7. Coding for a full poll of a PIA is as shown:
Example 11.6: Polling the MCS6520
Interrupt Vector JMP STORE
LDA #C0
BIT PIAAC
BEQ NXT1
BMI SEVEN
Set up Mask for
6
If 7, go to save— otherwise clear
and 7
Check for neither
6
or 7
Process BIT
6
INTERRUPT
NXTI BIT PIABC
BEQ NXTZ etc.
This program takes full advantage of the BIT instruction by checking for both bit 7 and
6
clear. BMI to SEVEN just checks N is on and that N is a higher priority. If bit
6
is one, the overflow bit will also be set, allowing the finish of the process seven routine to test the overflow and jump back to the process bit
6 coding.
161
Bit
6
and bit 7 were sampled by the single BIT instruction.
Speed was accomplished by loading the mask for just bit
6
and 7 into the register which allows the BEQ instruction to determine that neither of the two flags is on.
This routine depends on the fact that in the MCS6520, if
CA2 or CB2 is an output, bit
6
is always zero.
11.2.2 Bit Organization on MCS6520's
In the microprocessor, there is a definite positional pref erence for the testing of single bits. In the MCS6520 Data Direct ion Register, it is possible to select any combinations of input/ output pins by the pattern that is loaded in the Data Direction
Register. A one bit corresponds to an output and a zero bit corresponds to an input. The natural tendency would be to use
MCS6520s with all eight bits organized into a byte. There is relatively little advantage to organizing this way unless the eight bits are to be treated as a single byte by the program. This is often not the case, more often the bits are a collection of switches, coils, lights, etc.
On such combinations, advantage should be taken of the fact that bit 7 is directly testable so that a more useful combination of eight pins on one MCS6520 register would be seven outputs and a single input with the single input on bit 7. This organization allows the programmer to load and branch on that location without ever having to perform a bit or shift instruction to isolate a particular bit.
A similar,- capability -for setting a single bit involves the organization of data with seven inputs and a single output with a single output located in bit 0. This bit may be set or cleared by an INC or DEC instruction without affecting the rest of the bits in the register because the input pins ignore signals written from the microprocessor. Therefore, the more skilled MCS6500 programmer will often mix single outputs on bit 0 and a single input on bit 7 with bits of the corresponding opposite type.
162
11.2.3 Use of READ/MODIFY/WRITE Instruction For Keyboard Encoding
A rather unique use of the memory with a READ/MODIFY/WRITE operation involves setting the data register at all zeros, then using the three state output of the B side to sample a keyboard.
The following Figure 11.1 shows the connection for a 64 key key board organized
8 x
8
:
B Side BD
\ V
\
►
V
1
8 x
8
Switch
Decode
Matrix y * *_ t
AD
A SIDE i_t
Keyboard Encoding Matrix Diagram
FIGURE 11.1
163
The B side is set up to act as a strobe so that each of the output lines will have a ground on it during one scan cycle.
The eight A side data inputs are then sampled and decoded by the microprocessor giving a 64 key keyboard which is directly trans latable into code.
Figure 11.1 and Example 11.7 make use of the capability of the microprocessor to move a bit through the MCS6520 register location. This program also uses the compare instruction and the ability to detect a carry during a shift.
Example 11.7: Coding for Strobing an
8
x
8
Keyboard
Output Strobe is indicated by a one in Data Director Register.
Any connection is indicated by a zero in register bit.
LDX itO
STX PIABD
LDA PIABC
AND //FB
STA PIABC
STX PIABC
SEC Set low end bit on
LOOP ROL PIABD
BCS DONE
LDA PIAAD
CMP //FF
BEQ LOOP
DONE ------
Initialize B Data Register
Initialize Control Register to
Address Data Direction Register
Shift for Strobe if all sampled, exit
Check for no zeros
If any zeros, then process them.
A and PIABD can now be used to find out just what key is depressed.
Done
Then process Accmulator for Zero Bits
Keyboard. Strobe Sequence
FIG URE 11.2
165
Initialize
For
S trobing
Although they have separate addressing, the Data Direction and
Input/Output Registers operate the same as on the MCS6520.
Programming of the Interval Timer has some special problems.
First of all, the time is effectively located in all addresses from
XXX4-XXXF. By picking the proper address, the programmer is able to control the P scale for the timeout. Initialization of the Interval
Timer is done by a LOAD A followed by STORE A into the timing count.
The value stored in the timing counter represents the number of states which the counter will count through. The address used to load will determine how many additional divisions of the basic clock cycle will be counted.
When the counter finally counts to zero, it continues to count past zero at the one cycle clock rate in order to give the user an opportunity to sample the Status Register, then come back read the Count Register to determine how long it has been since an interrupt occurred..
Servicing an interrupt is the same for this Control Register as for any other interrupting register. Bit 7 is set on in the Status
Register to indicate that the Interval Timer is in the interrupt state and bit 7 is reset by the reading of the Counter.
11.3.1 Reading of the Counter Register
Because of the nature of counting past zero, the number in the Count Register is in two's complement form. It can be added directly to and used to correct the next count in a sequential string of counts or for correction for one cycle accuracy.
The specific details of organizing to get coding assembled is a function of the software that is used to implement the coding. Two software programs are currently available for the MCS650X family.
The Cross Assembler is available on various time share systems or for batch use on the user's system. Its documentation is covered in
166
the Cross-Assemble*r Manual, publication number 6500-60. The Resident
Assembler is available in the Microcomputer Development Terminal, as well as for sale in ROMs. The documentation for this is covered in the Resident Assembler Manual, publication number 6500-65.
The major advantages of using an assembler are that the assembler takes mnemonics and labels and calculates the fixed code. Reference to the OP CODE tables in the appendix shows that coding in Hex is quite difficult because there is no ordered pattern to the instruction
Hex codes.
The Cross Assembler or Resident Assembler allows one to specify all inputs and outputs in symbolic form on a documented listing. Symbolic addressing is a technique which has the following advantages over numerical addressing:
1. It allows the user to postpone until the last minute actual memory allocation in a program which is being developed. In a microprocessor that has memory-oriented features such as
Zero Page, memory management is important. It is desirable to have as many as possible of the read/write values in the
Zero Page. However, until the coding is complete, the organi zation of Zero Page may be in doubt. Values which are originally assigned in Zero Page may not be as valuable there after some analysis of the coding either indicates that the applications of these values use indirect references or index ing by Y which does not allow the program to really take advantage of Zero Page locations whereas some other code which may not be as frequently used might still result in a code reduction by use of Zero Page. This allocation, if all the fields are defined symbolically, can be done on the final assembly without any changing in the user's codes.
2. Use of symbolic addresses for programming branches leads to a better documented program and as one soon determines calculation of relative branches is difficult and subject
167
to change any time a coding change is made. For example, if one has organized a program with a loop in which three or four branches all return to the same point and then discovers a programming error which requires a single instruction to be added between the return point and various branches, each branch would have to be edited and recalculated. The symbolic assembler accomplishes this automatically on the next assembly pass.
11.4.1 Label Standards
The MCS650X assemblers have been done on a reserve word basis in which the various mnemonics which have been described are always considered to be OP CODE mnemonics. If any three character fields exactly match a mnemonic then the assembler assumes that the field is an OP CODE and proceeds to evaluate the addressing. Any other label may be located in free form anywhere in the coding. This means that one should organize one's labels such that he never has a three character label which inadvertently might be considered an OP CODE. The easiest way to accomplish this is to always follow a pattern on labels.
Good programming practice requires that the user develop a systems flow chart for his own basic program and individual flow charts for subroutines before starting the coding. From the time the routine is flow charted, it is very easy for the user to then assign a mnemonic label to the basic subroutine.
In this text, notations like LOOP, LOOP 1, etc. are used.
In an ADD, loop would be ADLP.
The MCS650X assembler allows six spaces for labels. It is good practice to use two characters to generally identify the subroutine, two more characters for mnemonic purposes and then a numbering system which allows correlation between various addresses within a LOOP within a subroutine. By strictly numbering such that ADLP1 is different from ADLP3, each can be addresses within the same LOOP.
168
It is assumed that the PIA's are connected in the normal manner of Status Register Address equal to Data Register Address + 1.
The following table and flow chart defines the program implemented in the example.
Table #1 contains the address of all of the MCS6520 Status Registers.
Table #2 contains the address of the put-away location for the respective data.
PIA #1 Table #1
PIA #1 ADL
ADH
Table #2
ADL
ADH
List
Value 1
Value 2
Value 3 etc.
169
Using six character labels, there are a hundred com binations of code which could be used in a given routine or loop without the user having to think through the rest of mnemonic notation. The use of characters plus a numeric for all references is sound programming practice. The advantage of using this technique allows one to use three character mnemonics without ever interfering with the reserve word of the microprocessor OP CODE mnemonics because they never have a numeric in the mnemonic.
Figure 11.3 demonstrates the program flow in support of the Cross-
Assembler listing (Example 11.9) of a time-sharing routine of a program which illustrates the use of the indexed indirect to perform a search of eight devices which have active signals for servicing. The implementa tion of the eight devices is done in MCS6520's where the MCS6520 status is set up to be a flag in bit 7 of a Control Register.
Initialize Index to End of Table
\
Fetch Next Status Register
Decrement X by 2
Program Flow - Polling for Active Signal
FIGURE 11.3
171
10
11
12
(0000)
0002
13 0004
14 0006
15 0008
16 000A
17 OOOC
18 000E
19 0010
20 0012
21 0014
22 0016
23 0018
24 001A
25 ' 001C
26 00
27 0020
28
29
30
0022
0200
31 0250
32 02A0
33 02F0
34 0340
35 0390
36 03E0
37 0430
38
39
40
41
42 0480
CARD = LOC
3
CODE CARD
CA R D S E R IA L N U M B E R
AH SYSTEMS BENCHMARK ■ POLLING 8 PERIPHERALS
05 40
07 40
09 40
OB 40
11 40
13 40
21 40
23 40
00 02
50 02
AO 02
FO 02
40 03
90 03
EO 03
30 04
m e m o r y l o c a t i o n
SET TABLES AND STORAGE AREAS
TABLEI
*=$02
.WORD PIA1AC
.WORD PIA1BC
.WORD PIA2AC
.WORD PIA2BC
.WORD PIA3AC
.WORD PIA3BC
.WORD PIA4AC
.WORD PIA4BC
TABLE2 .WORD STORE1
.WORD STORE2
.WORD STORE3
•WORD STORE4
.WORD STORE5
.WORD ST0RE6
.WORD STORE7
.WORD STORES
INITIALIZE PC_________________ y
(TABLE OF PIA PERIPHERAL CONTROL)
POINTERS TO STORE INPUT DATA FROM PERIPHERALS
STORE1
*=$200
*=*+80
STORE2 *=*+80
STORE3 *=*+80
STORE4 *-*+80
STORE5 *=*+80
STORE6 *=*+80
STORE7 *=*+80
STORES *=*+80
SET SPACE FOR DATA INPUT ON PAGE 2
FOR EACH DEVICE SET BUFFER 80 CHARACTERS LONG
MAIN PROGRAM
(*=$FC00)
•PROGRAM l o c a t i o n
43 FCOO
44 FC02
45 FC04
46 FC06
47
48
FCO?
FC08
49 FCOA
50
51
52
53 FCOC
54 FCOE
55 FClO
56 FC12
57 FC14
58 FC16
59
60
61
62
63 FC18
64 4004
65 4005
66 4006
67 4007
68 4008
69 4008
70 4009
71 400A
72 400B
73 400C
74 4010
75 4011
76 4012
77 4013
78 4014
75 4020
80 4021
81 4022
82 4023
83
^OPCODE
■ eLABEL
A l @
30 06
CA
CA
DO F8
FO F4
PLOP2 LDA (TABLE1-2.X)
INITIALIZE INDEX REGISTER X WITH 16
INDIRECT ADDRESSING OF PERIPHERAL CONTROL
\ VALU E BMI DOIT
\
(PEX)------ \
\
A D O R E SS
O E X _____ \ __
BNEfPL0P2^— \
IF FLAG SET BRANCH AND SERVICE THE DEVICE
IF NOT SEARCH NEXT ONE
m n e m o n ic
BEQ PLOPl START AGAIN TO POLL FROM THE BEGINNING
D6 00
A1 00
81 10
F6 10
F6 00
DO E8
SERVICE ROUTINE
DOIT DEO TABLE1-2.X
LDA (TABLE1-2,X)
STA (TABLE2-2,X)
INC TABLE2-2.X
INC TABLE1-2,X
8NE PLOPl
ASS1CN PIA LOCATION
*-$4004
PIA1AD *-*+1
PIA1AC *-*+1
PIA1BD *-*+1
PIA1BC *-*+1
*-$4008
PIA2AD *-*+1
PIA2AC *-*+1
PIA2BD *-*+1
PIA2BC *=*+1
*-$4010
PIA3AD *-*+1
PIA3AC *-*+L
PIA3BD *-*+1
PIA3BC *-*+1
*-$4020
PIA4AD *-*+1
PIA4AC *-*+1
PIA4BD *-*+1
PIA4BC *-*+1
.END
'^— S Y M B O LIC A D D R E S S
MOVE THE POINTER TO PIA DATA REGISTER
READ DATA IN
STORE THE DATA INTO THE BUFFER
SET BUFFER POINTER TO NEXT LOCATION
WHEN DONE START FROM BEGINNINC ACAIN
FIRST PERIPHERAL
SECOND
THIRD
FOURTH
FIFTH
SIXTH
SEVENTH
EICHTH
END OF PROGRAM
172
APPENDIX A
INSTRUCTION LIST
ALPHABETIC BY MNEMONIC
DEFINITION OF
INSTRUCTION GROUPS
A-1
A .l INTRODUCTION
The microprocessor instruction set is divided into three basic groups.
The first group has the greatest addressing flexibility and consists of the most general purpose instructions silch as Load, Add, Store, etc.
The second group includes the Read, Modijfy, Write instructions such as
Shift, Increment, Decrement and the Register X movement instructions. The third group contains all the remaining instructions, including all stack operations, the register Y, compares for
1
X and Y and instructions which do not fit naturally into Group One or Groi^p Two.
There are eight Group One instructi|ons, eight Group Two instructions, and all of the 39 remaining instructions! are Group Three instructions.
The three groups are obtained by organizing the OP CODE pattern to give maximum addressing flexibility (16 addressing combinations) to Group
One,to give eight combinations to Group Two instructions and the Group
Three instructions are basically individually decoded.
These instructions are: Add With Cjarry (ADC), (AND), Compare (CMP),
Exclusive Or (EOR) , Load A (LDA) , Or (OR|A) , Subtract With Carry (SBC) , and
Store A (STA). Each of these instructions has a potential for 16 addressing modes. However, in the MCS6501 through MCS6505, only eight of the available modes have been used.
Addressing modes for Group One are: Immediate, Zero Page, Zero Page
Indexed by X, Absolute, Absolute Indexed! by X, Absolute Indexed by Y,
Indexed Indirect, Indirect Indexed. Thei unused eight addressing modes are to be used in future versions of the MCS|650X product family to allow addressing of additional on-chip registeirs, of on-chip I/O ports, and to allow two byte word processing.
A. 3 GROUP TWO INSTRUCTIONS
Group Two instructions are primarily Read, Modify, Write instructions.
There are really two subcategories within the Group Two instructions.
The components of the first group are shift and rotate instructions and are: Shift Right (LSR), Shift Left (ASL), Rotate Left (ROL), and Rotate
Right (ROR).
The second subgroup includes the Increment (INC) and Decrement (DEC) instructions and the two index register X instructions, Load X (LDX) and
Store X (STX).
These instructions would normally have eight addressing modes available to them because of the bit pattern. However, to allow for upward expansion, only the following addressing modes have been de fined: Zero Page, Zero Page Indexed by X, Absolute, Absolute Indexed by X, and a special Accumulator (or Register) mode. The four shift in structions all have register A operations; the incremented or decremented
Load X and Store X instructions also have accumulator modes although the
Increment and Decrement Accumulator has been reserved for other purposes.
Load X from A has been assigned its own mnemonic, TAX. Also included in this group are the special functions of Decrement X which is one of the special cases of Store X. Included also in this group in the X decodes are the TXS and TSX instructions.
All Group One instructions have all addressing modes available to each instruction. In the case of Group Two instructions, another address ing mode has been added; that of the accumulator and the other special de codes have also been implemented in this basic group. However, the primary function of Group Two instructions is to perform some memory operation using the appropriate index.
It should be noted for documentation purposes that the X instructions have a special mode of addressing in which register Y is used for all in dexing operations; thus, instead of Zero Page Indexed by X, X instructions have Zero Page Indexed by Y, and instead of having Absolute Indexed by X,
X instructions have Absolute Indexed by Y.
A-4
A.4 GROUP THREE INSTRUCTIONS
There are really two major classifications of Group Three in structions; the modify Y register instructions, Load Y (LDY), Store Y
(STY) , Compare Y (CPY) , and Compare X (cj?X) , instructions actually occupy about half of the OP CODE space ft>r the Group Three instructions.
Increment X (INX) and Increment Y (INY) £re special subsets of the Compare X and Compare Y instructions and all of th£ branch instructions are in the
Group Three instructions.
Instructions in this group consist <t>f all of the branches: BCC, BCS,
BEQ, BMI, BNE, BPL, BPC and BPS. All of the flag operations are also de voted to one addressing mode; they are: CLC, SEC, CLD, SED, CLI, SEI and
CLV. All of the push and pull instructions and stack operation instructions are Group Three instructions. These include: BRK, JSR, PHA, PHP, PLA and
PLP. The JMP and BIT instructions are also included in this group. There is no common addressing mode available t
0
members of this group. Load Y,
Store Y, BIT, Compare X and Compare Y haijre Zero Page and Absolute, and all of the Y and X instructions allow Zero P^ge Indexed operations and Immediate.
APPENDIX B
INSTRUCTION LIST
ALPHABETIC BY MNEMONIC
WITH OP CODES, EXECUTION CYCLES
AND MEMORY REQUIREMENTS
B-1
The following notation applies to this summary:
A
X, Y
M
P
S
J
_
+
A
¥
+
+
-*■
V
PC
PCH
PCL
OPER
//
Accumulator
Index Registers
Memory
Processor Status Register
Stack Pointer
Change
No Change
Add
Logical AND
Subtract
Logical Exclusive Or
Transfer from Stack
Transfer to Stack
Transfer to
Transfer to
Logical OR
Program Counter
Program Counter High
Program Counter Low
OPERAND
IMMEDIATE ADDRESSING MODE
Note: At the top of each table is located in parentheses a reference number (Ref: XX) which directs the user to that Section in the MCS6500 Microcomputer Family
Programming Manual in which the instruction is defined and discussed.
B-2
A D C Add memory to accumulator with carry
Operation: A + H + C + A, C N 2 - C I D V
/ / / ___ /
(Ref: 2.2.1)
ADC
Addressing
Mode
Assembly Language
Form
OP
CODE
No.
Bytes
No.
Cycles
Immediate
Zero Page
Zero Page, X
Absolute
Absolute, X
Absolute, Y
(Indirect, X)
(Indirect), Y
ADC # Oper
ADC Oper
ADC
ADC
ADC
ADC
ADC
ADC
Oper, X
Oper
Oper, X
Oper, Y
(Oper, X)
(Oper), Y
* Add 1 if page boundary is crossed.
79
61
71
69
65
75
6
D
7D
2
2
2
3
3
3
2
2
2
3
4
4
4*
4*
6
5*
AND
"A N D ” m em ory with accumulator
Logical AND to the accumulator
Operation: A A H - > A
(Ref: 2.2.3.0)
N Z C I D V
/ / -------
AND
Addressing
Mode
Assembly Language
Form
OP
CODE
No.
Bytes
No.
Cycles
Immediate
Zero Page
Zero Page, X
Absolute
Absolute, X
Absolute, Y
(Indirect, X)
(Indirect), Y
AND # Oper
AND Oper
AND
AND
AND
Oper, X
Oper
Oper, X
Oper, Y AND
AND
AND
(Oper, X)
(Oper), Y
* Add 1 if page boundary is crossed.
29
25
35
2D
3D
39
2 1
31
2
2
2
3
3
3
2
2
6
5
2
3
4
4
4*
4*
B-3
ASL
Operation: C
ASL
Shift Left One Bit |Memory or Accumulator)
7 6 5 4 3 2 1 0
-0 N * C I D V y / / --------
Addressing
Mode
(Ref: 10.2)
Assembly Lknguage
Forjn
OP
CODE
No.
Bytes
ASL
No.
Cycles
Accumulator
Zero Page
Zero Page, X
Absolute
Absolute, X
ASL A
ASL Oper
ASL Oper,
|x
ASL Oper
ASL Oper, X
— 1
0A
06
16
0E
IE
3
3
2
2
1
5
2
6
7
6
BCC
BCC
Branch on Carry Clear
l : Branch on C = 0
Addressing
Mode
Relative
(Ref: 4.1.1.3)
Assembly Language
FoiV
BCC Oper
* Add 1 if branch occurs to same pag^.
* Add 2 if branch occurs to different page.
N 2 C I D V
BCC
OP No.
No.
CODE Bytes Cycles
90
2 2
*
B-4
BCS
BCS Branch On carry set
BCS
Operation: Branch on C = 1
Addressing
Mode
N Z C I D V
(Ref: 4.I.1.4)
--------------
1
---------
Assembly Language
Forjn
OP
CODE
No.
Bytes
N o .
Cycles
Relative BCS Oper B0
2 2
*
* Add 1 if branch occurs to same page.
* Add 2 if branch occurs to next page.
BEQ
Operation: Branch on Z = 1
Addressing
Mode
B E Q
Branch on result zero
(Ref: 4.1.1.5)
Assembly Language
Foriti
N Z C I D V
BEQ
OP
CODE
N o .
Bytes
No.
Cycles
2
*
BEQ Oper F0
2
Relative
* Add 1 if branch occurs to same page*
* Add 2 if branch occurs to next page.
B-5
BIT
BIT
Test bits in memory with accumulator
Operation: A A M, + N, + V
Bit
6
and 7 are transferred to the statjis register. % C I D V
If the result of A A M is zero then Z = t, otherwise M.,/------ M,
Z = 0
(Ref: 4.2.1.1)
Addressing
Mode
Assembly Language
For^n
OP No.
No.
CODE Bytes Cycles
BIT
Zero Page
Absolute
BIT Oper
BIT Oper
24
2C
2
3
3
4
BMI
Operation: Branch on N = 1
BMI
Branch on result minus
(Ref: 4 jl.1.1)
Addressing
Mode For m
Relative BMI Oper
* Add 1 if branch occurs to same pag^.
* Add 2 if branch occurs to different^ page.
N % C I D V
BMI
OP
CODE
No.
Bytes
No.
Cycles
30 2
2*
BNE
BNE
Branch on result not zero
BNE
N £ C I D V
Addressing
Mode
(Ref: 4.1.1.6)
Assembly Language
Form
Relative BNE Oper
* Add 1 if branch occurs to same page.
* Add 2 if branch occurs to different p£ge.
0
?
CODE
No.
Bytes
No.
Cycles
D0
2 2
*
BPL
Operation: Branch on N = 0
BPL
Branch on result plus
Addressing
Mode
(Ref: 4.1,1.2)
Assembly Language
Form
Relative BPL Oper
BPL
OP
CODE
No.
Bytes
No.
Cycles
2*
1 0
2
* Add 1 if branch occurs to same page.
* Add 2 if branch occurs to different page.
B-7
BRK
B R K
Force \fireak
Operation: Forced Interrupt PC + 2 + p +
(Ref: 9.11)
Addressing
Mode
Assembly Language
Form
Implied BRK
1. A BRK command cannot be masked bji setting I.
N
i
C I D V
BRK
OP No.
No.
CODE Bytes Cycles
0 0
1
7
BVC
Operation: Branch on V = 0
B V C
Branch on\ overflow clear
(Ref: 4 .
1
.
1
.
8
)
Addressing
Mode
Assembly Language
Fc rm
N % C I D V
OP
CODE
No.
BVC
No.
Cycles
50
2
* Relative BVC Opei
* Add 1 if branch occurs to same pa^e.
* Add 2 if branch occurs to different page.
2
BVS
Operation: Branch on V = 1
BVS
Branch on overflow set
Addressing
Mode
(Ref: 4.1.1.7)
Assembly Language
Form
N Z C I D V
BVS
OP
CODE
No.
Bytes
No.
Cycles
Relative BVS Oper 70
2 2
*
* Add 1 if branch occurs to same page.
* Add 2 if branch occurs to different page.
CLC
Operation: 0 -> C
Addressing
Mode
Implied
CLC
Clear carry flag
(Ref: 3.0.2)
Assembly Language
Form
CLC
CLC
N - 2 C I D V
0
OP
CODE
No.
Bytes
No.
Cycles
18
1 2
CLD
Operation: 0 D
Addressing
Mode
Implied
C L D
Clear decimal m ode
(Ref: 3.3.2)
Assembly Language
Form
CLD
N £ C I D V
------------------
0
-
OP
CODE
No.
Bytes
No.
Cycles
CLD D
8 1 2
CLI
Operation: 0 I
Addressing
Mode
CLI
Clear interrupt disable bit
(Ref: 3.2.2)
Assembly Language
Form
CLI
N 2 C I D V
------------
0
-
OP
CODE
No.
Bytes
No.
Cycles
Implied CLI 58
1 2
B-10
CLV
Operation: | + V
Addressing
Mode
Implied
CLV
Clear overflow flag
(Ref: 3 .6 .1 )
Assembly Language
Form
CLV
CLV
N 2 - C I D V
0
OP No.
No.
CODE Bytes Cycles
B
8 1 2
CMP
Operation: A - M
Addressing
Mode
CMP
Compare memory and accumulator
N Z C I D V
/ / / ____
(Ref: 4.2.1)
Assembly Language
Form
CMP
OP
CODE
No.
Bytes
No.
Cycles
Immediate
Zero Page
Zero Page, X
Absolute
Absolute, X
Absolute, Y
(Indirect, X)
(Indirect), Y
CMP #Oper
CMP Oper
CMP Oper, X
CMP Oper
CMP Oper, X
CMP Oper, Y
CMP (Oper, X)
CMP (Oper), Y
* Add 1 if page boundary is crossed.
C9
C5
D5
CD
DD
D9
Cl
D1
2
2
2
2
2
3
3
3
2
3
4
4
4*
4*
6
5*
B-ll
CPX
Operation: X - M
Addressing
Mode
CPX Compare Memory and Index X
N Z C I D V
/ / / --------
(Ref: 7.8)
Assembly Language
Form
CPX
OP
CODE
No.
Bytes
No.
Cycles
Immediate
Zero Page
Absolute
CPX # Oper
CPX Oper
CPX Oper
E0
E4
EC
2
2
3
2
3
4
CPY
Operation: Y - M
Addressing
Mode
CPY
Compare memory and index Y
N Z C I D V
/ / / --------
(Ref: 7.9)
Assembly Language
Form
CPY
OP
CODE
No.
Bytes
No.
Cycles
Immediate
Zero Page
Absolute
CPY #Oper
CPY Oper
CPY Oper
C0
C4
CC
2
2
3
2
3
4
B-12
DEC
Operation: M - 1 ->- M
DEC
Decrement memory by one
Addressing
Mode
(Ref: 10.7)
Assembly Language
Form
N Z C I D V
/ / -----------
DEC
OP
CODE
No.
Bytes
No.
Cycles
Zero Page
Zero Page, X
Absolute
Absolute, X
DEC Oper
DEC Oper, X
DEC Oper
DEC Oper, X
C
6
D
6
CE
DE
2
2
3
3
6
5
6
7
DEX
Operation: X - 1 ->- X
DEX
Decrement index X by one
(Ref: 7.6)
Addressing
Mode
Assembly Language
Form
Implied DEX
DEX
N Z C I D V
^
OP
CODE
No.
Bytes
No.
Cycles
CA
1 2
B-13
DEY
Operation: Y - 1 Y
Addressing
Mode
Implied
DEY
Decrement index Y by one
(Ref: 7.7)
Assembly Language
Form
DEY
DEY
N Z C I D V
/ / -----------
OP
CODE
N o .
Bytes
N o .
Cycles
8 8 1 2
EOR
Operation: A ^ M A
EOR
"Exclusive—Or” memory with accumulator
N Z C I D V
(Ref: 2.2.3.2) ^
Addressing
Mode
Assembly Language
Form
EOR
OP
CODE
N o .
Bytes
No.
Cycles
Immediate
Zero Page
Zero Page, X
Absolute
Absolute, X
Absolute, Y
(Indirect, X)
(Indirect),Y
EOR #Oper
EOR Oper
EOR Oper, X
EOR Oper
EOR Oper, X
EOR Oper, Y
EOR (Oper, X)
EOR (Oper), Y
* Add 1 if page boundary is crossed.
B-14
5D
59
41
51
49
45
55
4D
2
2
2
2
2
3
3
3
2
3
4
4
4*
4*
6
5*
INC
Operation: M + 1 M
Addressing
Mode
Zero Page
Zero Page, X
Absolute
Absolute, X
INC Increment memory by one
(Ref: 10.6)
Assembly Language
Form
INC Oper
INC Oper, X
INC Oper
INC Oper, X
INC
N Z C I D V
/ /
-----------
OP
CODE
No.
Bytes
No.
Cycles
E
6
F
6
EE
FE
2
2
3
3
5
6
6
7
INX
Operation: X + 1 X
INX
Increment Index X by one
Addressing
Mode
(Ref: 7.4)
Assembly Language
Form
Implied INX
N Z C I D V
/ / -----------
INX
OP
CODE
No.
Bytes
No.
Cycles
E
8 1 2
B-15
INY
Operation: Y + l + Y
INY Increment Index Y by one
(Ref: 7.5)
Addressing
Mode
Assembly Language
Form
N Z C I D V
/ / --------
INY
OP
CODE
No.
Bytes
No.
Cycles
Implied INY C
8 1 2
JMP
Operation: (PC + 1) PCL
(PC +
2")
(PC + L)
-*■
JMP
Jump to new location
(Ref.
(Ref: 4.0.2)
9
.
8
.
1
)
Addressing
Mode
Assembly Language
Form
N Z C I D V
___________
JMP
OP
CODE
No.
Bytes
N o .
Cycles
Absolute
Indirect
JMP Oper
JMP (Oper)
4C
6
C
3
3
3
5
JSR
JSR
Jump to new location saving return address
Operation: PC + 2 +, (PC + 1) ->■ PCL N Z C I D V
(PC + 2) ->■ PCH ___________
(Ref: 8.1)
Addressing
Mode
Assembly Language
Form
OP
CODE
No.
Bytes
JSR
No.
Cycles
Absolute JSR Oper
2 0
3
6
LDA
Operation: M ->■ A
Addressing
Mode
LDA
Load accumulator with memory
N Z C I D V
/ / ______
(Ref: 2.1.1)
Assembly Language
Form
LDA
OP
CODE
No.
Bytes
N o .
Cycles
Immediate
Zero Page
Zero Page, X
Absolute
Absolute, X
Absolute, Y
(Indirect, X)
(Indirect), Y
LDA # Oper
LDA Oper
LDA Oper, X
LDA Oper
LDA Oper, X
LDA Oper, Y
LDA (Oper, X)
LDA (Oper), Y
*
Add 1 if page boundary is crossed.
B9
A1
B1
A9
A5
B5
AD
BD
2
2
2
2
2
3
3
3
2
3
4 ,
4
4*
4*
6
5*
B-17
LDX
Operation: M -> X
LDX Load index X with memory
Addressing
Mode
(Ref: 7.0)
Assembly Language
Form
Immediate
Zero Page
Zero Page, Y
Absolute
Absolute, Y
LDX # Oper
LDX Oper
LDX Oper, Y
LDX Oper
LDX Oper, Y
* Add 1 when page boundary is crossed.
N Z C I D V
/ / _
LDX
OP
CODE
N o .
Bytes
N o .
Cycles
A2
A
6
B
6
AE
BE
2
3
3
2
2
2
3
4
4
4*
LDY
Operation: M -> Y
LDY
Load index Y with memory
Addressing
Mode
Immediate
Zero Page
Zero Page, X
Absolute
Absolute, X
(Ref: 7.1)
Assembly Language
Form
LDY #Oper
LDY Oper
LDY Oper, X
LDY Oper
LDY Oper, X
* Add 1 when page boundary is crossed.
N Z C I D V
/ / -----------
OP
CODE
No.
Bytes
No.
Cycles
A0
A4
B4
AC
BC
2
2
2
3
3
LDY
2
3
4
4
4*
B-18
LSR
LSR
Shift right one bit (memory or accumulator)
Operation: 0 —^ 7
6
5 4 3
2 1 0
-+ C N Z C I D V
0
/ / -----
(Ref: 10.1)
Addressing
Mode
Assembly Language
Form
OP
CODE
No.
Bytes
LSR
No.
Cycles
Accumulator
Zero Page
Zero Page, X
Absolute
Absolute, X
LSR A
LSR Oper
LSR Oper, X
LSR Oper
LSR Oper, X
4A
46
56
4E
5E
1
2
2
3
3
6
2
5
6
7
NOP
NOP
No operation
Operation: No Operation (2 cycles)
Addressing
Mode
Implied
NOP
N Z C I D V
Assembly Language
Form
NOP
OP
CODE
No.
Bytes
No.
Cycles
EA
1 2
ORA
Operation: A V M A
Addressing
Mode
ORA “OR" memory with accumulator
N Z C I D V
/ / ______
(Ref: 2.2.3.1)
Assembly Language
Form
ORA
OP
CODE
No.
Bytes
N o .
Cycles
Immediate
Zero Page
Zero Page, X
Absolute
Absolute, X
Absolute, Y
(Indirect, X)
(Indirect), Y
* Add 1 on page crossing
ORA #Oper
ORA Oper
ORA Oper, X
ORA Oper
ORA Oper, X
ORA Oper, Y
ORA (Oper, X)
ORA (Oper), Y
0D
ID
19
09
05
15
0 1
1 1 2
2
2
3
3
3
2
2
6
5
2
3
4
4
4*
4*
PHA
Operation: A +
Addressing
Mode
Implied
PHA
Push accumulator on stack
N Z C I D V
(Ref: 8.5)
Assembly Language
Form
PHA
PHA
OP
CODE
No.
Bytes
No.
Cycles'
48
1
3
B-20
PHP
Operation: P+
Addressing
Mode
Implied
PHP
Push processor status on stack
(Ref: 8.11)
Assembly Language
Form
PHP
N Z C I D V
-----------
OP
CODE
No.
Bytes
No.
Cycles
08
1
3
PLA
Operation: A t
Addressing
Mode
Implied
P L A
Pull accumulator from stack
N Z C I D V
/ / -----------
(Ref:
8
.
6
)
Assembly Language
Form
PLA
PLA
OP
CODE
No.
Bytes
No.
Cycles
6 8 1
4
B-21
PLP
Operation: P +
Addressing
Mode
PLP
Pull processor status from stack
N Z C I D V
From Stack
(Ref: 8.12)
Assembly Language
Form
PLP
OP
CODE
No.
Bytes
N o .
Cycles
Implied PLP 28
1
4
ROL
Operation:
R O L
R otate one bit left (m em ory or accumulator)
M or
7
6
5 4 3 2
1
0
ZS7
N Z C I D V
/ / / --------
(Ref: 10.3)
Addressing
Mode
Assembly Language
Form
OP
CODE
N o .
Bytes
ROL
No.
Cycles
Accumulator
Zero Page
Zero Page, X
Absolute
Absolute, X
ROL A
ROL Oper
ROL Oper, X
ROL Oper
ROL Oper, X
2A
26
36
2E
3E
1
2
2
3
3
2
5
6
6
7
B-22
ROR
ROR
Rotate one bit right (memory or accumulator)
ROR
Operation:
— ► c
► 7
6
5 4 3
2 1
(Ref: 10.4)
0
N Z C I D V
/ / / -------
Addressing
Mode
Accumulator
Zero Page
Zero Page,X
Absolute
Absolute,X
Assembly Language
ROR A
Form
ROR Oper
ROR Oper,X
ROR Oper
ROR Oper,X
6
OP
CODE
A
6 6
76
6
E
7E
N o .
Bytes
1
2
2
3
3
No.
Cycles
6
2
5
6
7
Note: ROR instruction will be available on MCS650X micro processors after June, 1976.
RTI
Operation: P+ PC+
Addressing
Mode
Implied
RTI
Return from interrupt
^ n
(Ref: 9.6)
Assembly Language
Form
RTI
RTI
N Z C I D V
From Stack
OP
CODE
No.
Bytes
No.
Cycles
40
1 6
RTS
Operation: PC+, PC + l-> PC
RTS
Return from subroutine
(Ref: 8.2)
N Z C I D V
Addressing
Mode
Assembly Language
Form
RTS
OP
CODE
No.
Bytes
No.
Cycles
Implied RTS
6 0 1 6
B-23
SBC
SBC
Subtract m em ory from accumulator with borrow
Operation: A - M - C A N Z C I D V
Note: C = Borrow (Ref: 2.2.2) / / / --- /
Addressing
Mode
Assembly Language
Form
SBC
OP
CODE
No.
Bytes
No.
Cycles
Immediate
Zero Page
Zero Page, X
Absolute
Absolute, X
Absolute, Y
(Indirect, X)
(Indirect), Y
SBC # Oper
SBC Oper
, Oper, X
SBC Oper
SBC Oper, X
SBC Oper, Y
SBC (Oper, X)
SBC (Oper), Y
* Add 1 when page boundary is crossed.
E9
E5
F5
ED
FD
F9
El
FI
2
2
2
3
3
3
2
2
2
3
4
4
4*
4*
6
5*
SEC
Operation: 1 + C
Addressing
Mode
Implied
SEC
S et carry flag
(Ref: 3.0.1)
Assembly Language
Form
SEC
SEC
N Z C I D V
---
1
------
OP
CODE
No.
Bytes
N o .
Cycles
38
1 2
B-24
SED
Operation: 1 -> D
Addressing
Mode
Implied
SED
Set decimal mode
SED
N Z C I D V
(Ref: 3.3.1)
Assembly Language
Form
SED
OP
CODE
No.
Bytes
No.
Cycles
F
8 1 2
SEI
Operation: 1 I
Addressing
Mode
SEI
Set interrupt disable status
(Ref: 3.2.1)
Assembly Language
Form
SEI
N Z C I D V
1
OP
CODE
No.
Bytes
No.
Cycles
Implied SEI 78
1 2
B-25
STA
Operation: A M
Addressing
Mode
Zero Page
Zero Page, X
Absolute
Absolute, X
Absolute, Y
(Indirect, X)
(Indirect), Y
STA
Store accumulator in memory
N Z C I D V
(Ref: 2.1.2)
Assembly Language
Form
STA
OP
CODE
No.
Bytes
No.
Cycles
STA Oper
STA Oper, X
STA Oper
STA Oper, X
STA Oper, Y
STA (Oper, X)
STA (Oper), Y
85
95
8
D
9D
99
81
91
2
2
2
3
3
3
2 6
6
3
4
4
5
5
STX
Operation: X + M
Addressing
Mode
Zero Page
Zero Page, Y
Absolute
STX
Store index X in memory
N Z C I D V
(Ref: 7.2)
Assembly Language
Form
STX
OP
CODE
No.
Bytes
No.
Cycles
STX Oper
STX Oper, Y
STX Oper
8 6
96
8
E
2
2
3
3
4
4
STY
Operation: Y ->■ M
Addressing
Mode
Zero Page
Zero Page, X
Absolute
STY Store index Y in memory
N Z C I D V
(Ref: 7.3)
Assembly Language
Form
STY Oper
STY Oper, X
STY Oper
OP
CODE
No.
Bytes
No.
Cycles
84
94
8
C
2
2
3
3
4
4
TAX
Operation: A -»■ X
Addressing
Mode
Implied
-------
TAX
Transfer accumulator to index X
N Z C I D V
/ / -----------
(Ref: 7.11)
Assembly Language
Form
OP
CODE
TAX
No.
Bytes
No.
Cycles
TAX AA
1 2
B-27
m
Operation: A -> Y
Addressing
Mode
TAY
Transfer accumulator to index Y
N Z C I D V
/ / -----------
(Ref: 7.13)
Assembly Language
Form
OP
CODE
No.
Bytes
T A Y
IM I
No.
Cycles
Implied TAY A
8 1 2
TYA
Operation: Y -> A
Addressing
Mode
TYA
Transfer index Y to accumulator
N Z C I D V
/ / -----------
(Ref: 7.14)
Assembly Language
Form
TYA
OP
CODE
No.
Bytes
No.
Cycles
Implied TYA 98
1 2
B-28
TSX
Operation: S -> X
Addressing
Mode
TSX
Transfer stack pointer to index X
N Z C I D V
^ (Ref: 8.9)
Assembly Language
Form
TSX
OP
CODE
No.
Bytes
No.
Cycles
Implied TSX BA
1 2
TXA
Operation: X -> A
Addressing
Mode
TXA
Transfer index X to accumulator
(Ref: 7.12)
N Z C I D V
J J ______ v v
Assembly Language
Form
TXA
OP
CODE
No.
Bytes
No.
Cycles
Implied TXA
8
A
1 2
TXS
Operation: X + S
Addressing
Mode
TXS
Transfer index X to stack pointer
N Z C I D V
(Ref:
8
.
8
)
Assembly Language
Form
TXS
OP
CODE
N o .
Bytes
No.
Cycles
Implied TXS 9A
1 2
B-29
APPENDIX C
INSTRUCTION ADDRESSING
MODES AND
RELATED EXECUTION TIMES
C/5
LU s z o h-
D
O
X
UJ
Q
UJ
J> o
>• o o
_o o
UJ cc
<
00
UJ
Q
O
2
CO
C/5
UJ cc o
Q
< z o h“
O
D
CC h“
C/5 z
)
3 ej;pu| ein|osqv
A'(^eJipui)
(X 'laa-MPUl) aAijeiay p3l|dlU|
A amiosqv
X 'amiosqv amiosqv
A '
0 6 ej ojaz
X '«
6 ej oiaz aBej o
» 2 siejpeuiuui joie|niun33V
).33j;pu| ain|osqv
A '(iMJipul)
(X '»3»i!PU|)
3Ai^e|ay pai|duj|
A 'ajnjosqv
X *a»n|osqv amiosqv
A 'aBej ojez
X 'aBej ojez sited ojaz
3 ieipeiuui| jo}B|niunDov
* *
*
• ^
*
CM •
•
*
5 f
.
CM CM CM
*
...........
<*
.
.
i n
.
* *
^ . . . . ^ . . . i n . .
c o
.
co •
■
-co
. .
.
lo
CM CM CM .
.CM........
- • • • C M ........CM CM
J J J J Z O l L L L l C CCCCCCc
0
C
*
* * c c ^ X £ e r £ < < a . < a . ; J £ - « / > o o a - < x > - X > x < «
2
<
QQQtflOOClI j j O OI-l-ajLUUjLUl-(-l-<<W)XX>
C
0
C
0
C
0
C
0
C
0 t-t-t-t-t-leg c o
>T3 m “
........................................ in -S®
O
<0
*
............. i n . . . .
..................... c o ......
•
* * * * * *
CM CM CM .
* *
<0
.
•
.Q
£3 *n
CM C M ..................... **•
............ .
* * * *
^ ^ ....... .................... * * . . . . o ™
* *
^ ^
*
...... ............. ^ .
*
.
•
”
„
o>—
.
.............. ^ ^ ^ <D ■ .
si
CO X O
................................................................................... .E .o
t o ........................................ . .t o • . T t t o . . . ^ it
- - 2
CN C M ......................CN CN CN .
CN
■
O s>
C M ..................................... C C
O O
-a 73
73 73
< <
® a o
CT c
1 c o c
<0
C-2
APPENDIX D
OPERATION CODE INSTRUCTION LISTING
HEXIDECIMAL SEQUENCE
0 0
- BRK
0 1
- ORA - (Indirect,X)
0 2
- Future Expansion
03 - Future Expansion
04 - Future Expansion
05 - ORA - Zero Page
06 - ASL - Zero Page
07 - Future Expansion
08 - PHP
09 - ORA - Immediate
0A - ASL - Accumulator
0B - Future Expansion
0C - Future Expansion
0D ORA - Absolute
0E - ASL - Absolute
0F - Future Expansion
1 0
- BPL
1 1
- ORA - (Indirect),Y
1 2
- Future Expansion
13 - Future Expansion
14 - Future Expansion
15 - ORA - Zero Page,X
16 - ASL - Zero Page,X
17 - Future Expansion
18 - CLC
19 - ORA - Absolute,Y
1A - Future Expansion
IB - Future Expansion
1C - Future Expansion
ID ORA - Absolute,X
IE - ASL - Absolute,X
IF - Future Expansion
D-2
2(1
- JSR
2
:. - AND - (Indirect,X)
2\\ - Future Expansion
23 - Future Expansion
2H - BIT - Zero Page
25 - AND - Zero Page
2
(> - ROL - Zero Page
2 7 - Future Expansion
21!
- PLP
2
!) - AND - Immediate
2 a - ROL - Accumulator
21!
- Future Expansion
2
<: - BIT - Absolute
2
D - AND - Absolute
2 1
! - ROL - Absolute
21
’ - Future Expansion
30 - BMI
31 - AND - (Indirect),Y
3:> - Future Expansion
33 - Future Expansion
34 - Future Expansion
35 - AND - Zero Page,X
3'3 - ROL - Zero Page,X
37 - Future Expansion
33 - SEC
39 - AND - Absolute,Y
35 - Future Expansion
3: - Future Expansion
33 - AND - Absolute,X
3E - ROL - Absolute,X
3F - Future Expansion
40 - RTI
41 - EOR - (Indirect,X)
42 - Future Expansion
43 - Future Expansion
44 - Future Expansion
45 - EOR - Zero Page
46 - LSR - Zero Page
47 - Future Expansion
48 - PEA
49 - EOR - Immediate
4A - LSR - Accumulator
4B - Future Expansion
4C - JMP - Absolute
4D - EOR - Absolute
4E - LSR - Absolute
4F - Future Expansion
50 - BVC
51 - EOR - (Indirect),Y
52 - Future Expansion
53 - Future Expansion
54 - Future Expansion
55 - EOR - Zero Page,X
56 - LSR - Zero Page,X
57 - Future Expansion
58 - CLI
59 - EOR - Absolute,Y
5A - Future Expansion
5B - Future Expansion
5C - Future Expansion
5D - EOR - Absolute,X
5E - LSR - Absolute,X
5F - Future Expansion
60 - RTS
6
:. - ADC - (Indirect,X)
62 - Future Expansion
62 - Future Expansion
64 - Future Expansion
65 - ADC - Zero Page
6 6
- ROR - Zero Page
67 - Future Expansion
6 8
- PLA
69 - ADC - Immediate
6
A - ROR - Accumulator
6
B - Future Expansion
6
C - JMP - Indirect
6
D - ADC - Absolute
6
E - ROR - Absolute
6
F - Future Expansion
70 - BVS
71 - ADC - (Indirect),Y
72 - Future Expansion
73 - Future Expansion
74 - Future Expansion
75 - ADC - Zero Page,X
76 - ROR - Zero Page,X
77 - Future Expansion
78 - SEI
79 - ADC - Absolute,Y
7A - Future Expansion
7B - Future Expansion
7C - Future Expansion
7D - ADC - Absolute,X
7E - ROR - Absolute,X
7F - Future Expansion
D-3
80 - Future Expansion
81 - STA - (Indirect,X)
82 - Future Expansion
83 - Future Expansion
84 - STY - Zero Page
85 - STA - Zero Page
8 6
- STX - Zero Page
87 - Future Expansion
8 8
- DEY
89 - Future Expansion
8
A - TXA
8
B - Future Expansion
8
C - STY - Absolute
8
D - STA - Absolute
8
E - STX - Absolute
8
F - Future Expansion
90 - BCC
91 - STA - (Indirect),Y
92 - Future Expansion
93 - Future Expansion
94 - STY - Zero Page,X
95 - STA - Zero Page,X
96 - STX - Zero Page.Y
97 - Future Expansion
98 - TYA
99 - STA - Absolute,Y
9A - TXS
9B - Future Expansion
9C - Future Expansion
9D - STA - Absolute,X
9E - Future Expansion
9F - Future Expansion
A? - LDY - Immediate
A1 - LDA - (Indirect,X)
A2 - LDX - Immediate
A3 - Future Expansion
A4 - LDY - Zero Page
A5 - LDA - Zero Page
A
6
- LDX - Zero Page
A7 - Future Expansion
A
8
- TAY
A9 - LDA - Immediate
AA - TAX
AB - Future Expansion
AC - LDY - Absolute
AD - LDA - Absolute
AE - LDX - Absolute
AF - Future Expansion
B0 - BCS
Bl - LDA - (Indirect),Y
B2 - Future Expansion
B3 - Future Expansion
B4 - LDY - Zero Page,X
B5 - LDA - Zero Page,X
B
6
- LDX - Zero Page,Y
B7 - Future Expansion
B
8
- CLV
B9 - LDA - Absolute,Y
BA - TSX
BB - Future Expansion
BC - LDY - Absolute,X
BD - LDA - Absolute,X
BE - LDX - Absolute,Y
BF - Future Expansion
C0 - CPY - Immediate
Cl - CMP - (Indirect,X)
C2 - Future Expansion
C3 - Future Expansion
C4 - CPY - Zero Page
C5 - CMP - Zero Page
C
6
- DEC - Zero Page
C7 - Future Expansion
C
8
- INY
C9 - CMP - Immediate
CA - DEX
CB - Future Expansion
CC - CPY - Absolute
CD - CMP - Absolute
CE - DEC - Absolute
CF - Future Expansion
D0 - BNE
D1 - CMP - (Indirect),Y
D2 - Future Expansion
D3 - Future Expansion
D4 - Future Expansion
D5 - CMP - Zero Page,X
D
6
- DEC - Zero Page,X
D7 - Future Expansion
D
8
- CLD
D9 - CMP - Absolute,Y
DA - Future Expansion
DB - Future Expansion
DC - Future Expansion
DD - CMP - Absolute,X
DE - DEC - Absolute,X
DF - Future Expansion
E0 - CPX - Immediate
El - SBC - (Indirect,X)
E2 - Future Expansion
E3 - Future Expansion
E4 - CPX - Zero Page
E5 - SBC - Zero Page
E
6
- INC - Zero Page
E7 - Future Expansion
E
8
- INX
E9 - SBC - Immediate
EA - NOP
EB - Future Expansion
EC - CPX - Absolute
ED - SBC - Absolute
EE - INC - Absolute
EF - Future Expansion
F0 - BEQ
FI - SBC - (Indirect),Y
F2 - Future Expansion
F3 - Future Expansion
F4 - Future Expansion
F5 - SBC - Zero Page,X
F
6
- INC - Zero Page,X
F7 - Future Expansion
F
8
- SED
F9 - SBC - Absolute,Y
FA - Future Expansion
FB - Future Expansion
FC - Future Expansion
FD - SBC - Absolute,X
FE - INC - Absolute,X
FF - Future Expansion
D-5
APPENDIX E
SUMMARY OF ADDRESSING MODES
E-1
This appendix is to serve the user in providing a reference for the MCS650X addressing modes. Each mode of address is shown with a symbolic illustration of the bus status at each cycle during the instruction fetch and execution. The example number as found in the text is provided for reference purposes.
Example 5.3: Illustration of implied addressing
Clock
Cycle
1
2
Address Bus Program Counter Data Bus
PC
PC + 1
PC + 1
PC + 1 o
:5
CODE
Niw
0? CODE
PC + 1 PC + 2 New
OP CODE
Comments
Fetch OP CODE
Ignore New
OP CODE;
Decode Old
OP CODE
Fetch New
OP CODE;
Execute Old
OP CODE
E-2
E.2 IMMEDIATE ADDRESSING
Example 5.4: Illustration of immediate addressing
Clock
Cycle Address Bus Program Counter Data Bus j.
2
PC
PC + 1
PC + 1
PC + 2
OP CODE
Data
PC + 2 PC + 3 New
OP CODE
Comments
Fetch OP CODE
Fetch Data,
Decode OP CODE
Fetch New
OP CODE,
Execute Old
OP CODE
Example 5.5: Illustration of absolute addressing
Clock
Cycle
1
2
Address Bus Program Counter Data Bus
PC
PC + 1
PC + 1
PC + 2
OP CODE
ADL
PC + 2 PC + 3 ADH
4
5
ADH, ADL
PC + 3
PC + 3
PC + 4
Data
New
OP CODE
Comments
Fetch OP CODE
Fetch ADL,
Decode OP CODE
Fetch ADH,
Retail ADL
Fetch Data
Fetch New
OP CODE,
Execute Old
OP CODE
E-3
E.4 ZERO PAGE ADDRESSING
Example 5.6: Illustration of zero page addressing
Clock
Cycle
1
2
Address Bus Program Counter Data Bus
PC
PC + 1
PC + 1
PC + 2
OP CODE
ADL
3
4
00, ADL
PC + 2
PC + 2
PC + 3
Data
New
OP CODE
Comments
Fetch OP CODE
Fetch ADL, De code OP CODE
Fetch Data
Fetch New
OP CODE, Exe cute Old
OP CODE
Example 5.8: Illustration of relative addressing— branch positive taken, no crossing of page bcundaries
Cycle Data Bus
External
Operation
Internal
Operation
1
Address Bus
0100 OP CODE
0101
0102
0152
+50
Next
OP CODE
Next
OP CODE
Fetch
OP CODE
Fetch
Offset
Finish Previous Operation, Increment Pro gram Counter to 101
Interpret Instruction,
Increment Program
Counter to 102
Fetch Next Check Flags, Add Rela-
OP COD'S tive to PCL, Increment
Program Counter to 103
Fetch text Transfer Results to
OP COD I PCL, Increment Program
Counter to 153
E-4
E.6 ABSOLUTE INDEXED ADDRESSING - (with page crossing)
Step 5 is deleted and the data in step 4 is valid when no page crossing occurs.
Example 6.7: Absolute Indexed; With Page Crossing
Cycle
Address Data
Bus Bus
External
Operation
Internal
Operation
1
0100
OP CODE Fetch OP CODE
2
0101 BAL Fetch BAL
Finish Previous
Operation Increment
PC to 101
Interpret Instruction
Increment PC to 102
3 0102 BAH Fetch BAH
4
5
BAH,BAL
+X
Data
(Ignore)
BAH+1,
BAL+X
Data
Fetch Data
(Data is ignored)
Fetch Data
Add BAL + Index
Increment PC to 103
Add BAH + Carry
0103 Next OP
CODE
Fetch Next
OP CODE
Finish Operation
E-5
E.
7
ZERO PAGE INDEXED ADDRESSING
Example
6
.
8
: Illustration of Zero Page Indexing
Cycle
1
Address
Bus____
0100
Data
Bus
OP CODE
Externa L
Operation
Fetch OP :0DE
0101
BAL
Internal
Operation
Finish Previous
Operation
Interpret Instruct ion
00,BAL
00,BAL
+X
0102
Data
(Dis carded
Data
Next OP
CODE
Fetch Base
Address Low
(BAL)
Fetch
Discarded
Data
Fetch Data
Fetch Next OP
CODE
Add: BAL + X
Finish Operation
E-6
E.8 INDEXED INDIRECT ADDRESSING
Example 6.10: Illustration of Indexed Indirect Addressing
Cycle
1
Address Data
Bus Bus
0100
OP CODE
External
Operation
Fetch OP CODE
Internal
Operation
Finish Previous
Operation
2
0101
BAL Fetch BAL Interpret In struction
3 Add BAL + X
4
5
00,BAL DATA (Dis carded)
00,BAL ADL
+ X
00,BAL
+ X + 1
Fetch Discard ed DATA
Fetch ADL
Fetch ADH
Add 1 to BAL + X
Hold ADL
6
7
ADH,ADL DATA
0102 Next OP
Fetch DATA
Fetch Next 01’
CODE
Finish Operation
E-7
E. 9 INDIRECT INDEXED ADDRESSING (with page crossing)
Step 6 is deleted and the data in step 5 is valid when no page crossing occurs.
Example 6.12: Indirect Indexed Addressing (With Page Crossing)
Cycle
1
Address Data
Bus Bus
0100
OP CODE
External
Operation
Load OP CODE
Internal
Operation
Finish Previous
Operation
010L
IAL Fetch IAL Interpret In struction
00,IAL BAL
00,IAL BAH
+
1
Fetch B a L
Fetch BAH
Add 1 to IAL
Add BAL to Y
Add 1 to BAH BAH,BAL DATA (Dis-
+ Y carded)
BAH + 1 DATA
BAL + Y
0102 Next OP
CODE
Fetch Dl^TA
(Discarded)
Fetch Data
CODE
Finish This
Operation
E-8
APPENDIX F
MCS650X
PROGRAMMING
MODEL
F-l
LU
Q
O
C 3
2
<
CC
(D
O
0
CC
.
o
X
O to
CO cc
LU
I—
M
CJ
LU oc cc o
ID o o
<
X
LU
Q z
DC
LU
I
C
/3 o
LU
CC
DC
LU
I
W
C 3
LU
CC
X
LU
Q
LU
< cc cc u o
QC
Qr- 1 o
; 1
| j
1 1 j I
1 1
1 1 j 1
1 1
1 1 j 1
1. .. J f"
1 1 1
1 1 1
I 1 1
1 1 1
1 1 1
1 1 1
1 1 1
1 1 1
<.
o r~
1
1
1
1
1
1
1
1
|
>o r~
1
1
1
1
1
1
1
1
|
X o i .
_i o
I o
Oo
CO o
---
O
N
CD
O
0 o
<
1 -
C
/3 cc
LU
DC
LU
(“
CO
C3
LU
CC c
/3
D
LU
_1
CD
<
C
/3
LU
^ Q
UJ cc r> i -
<
LU
Li-
CS c
/3 cc o c
C
LU
O o cc
O-
Q -
^ U
O
Q o >
U cc o £ i < l - C C <
CC c c t l O LU CC LU C D
<
LU
5
LU QC o >
LU
O N - Q C Q U -0 2
>
E
CD
3
>+-
>+-
O
_
03 O OJ -Q
E
CD
-O E
CD
C
>
03
E o o
-C
JZ o jO co Q
15 ___
s ___ £ --if?
£ ___
F-2
APPENDIX G
DISCUSSION - INDIRECT ADDRESSING
G-1
The MCS650X microprocessors have a special form of addressing known as Indirect. The writeup on Indirect addressing describes the basic operation of Indirect.
It is the intent of this discussion to acquaint the user with some of the uses and applications of Indirect addressing.
The Indirect address is really an address that would have been coded in line as in the case of Absolute except for the fact that the address is not known at the time the user writes the program. As has been in dicated several times in the basic body of the documentation, it is sig nificantly more efficient with the organization of the MCS650X to assign addresses and implement them if the addressing structure is known. How ever, that is not always possible to do. For instance, in order to mini mize the coding of a subroutine or general purpose set of coding, it is often desirable to work with a range of addressing that is not possible to cover in a normal index, or in the case of subroutine where it is neces sary for the addresses to be variable depending on which part of the whole program called the address.
It is probably this discussion which best amplifies the need for calculated addresses. It should be fairly obvious to the user that a general purpose subroutine cannot contain the; address of the operations.
Therefore, instead of having the instruction LDA followed by the value that the programmer wants to load, in a subroutine it may be desirable to do a Load A from a calculated or specified address.
The use of the Indirect Addressing Mode is to give the user a loca tion in Page Zero in which can be put the calculated address. Then the subroutine instruction can call this calculated address using the form
Load A from an address pointed to by the next byte in program sequence.
The word "indirect" technically comes from the fact that instead of taking the address which is immediately following the instruction, the next value in program sequence is a pointer to the address.
The Indirect pointer will be referred to from now on as IAL, because it is a Zero Page address and, therefore, is a low order byte. The in direct instructions are written in the form "Load A" followed by IAL.
G-2
IAL points to an address which had been previously stored into Page Zero.
This gives the user the flexibility of addressing anywhere in memory with a calculated address. However, the real value of Indirect is not in just having Indirect but having the ability to have Indirect modified.
This is the reason for which indirect indexed Instruction is implemented rather than straight indirect. An example of che indirect indexed in sub- routining is covered in Section 6.5, , but it should be noted that the in direct indexed instruction should be used whenever the user does not know the exact address at time of compilation. Although there may be other in teresting and esoteric uses of the indirect index instruction, this is the most common one.
The second form of indirect is very powerful for certain types of applications. Chapter 11 shows the use of tabJ.es which have pointers and the advantage of running down one table of pointers until a match is found and then using the same index to address a second table to perform an opera tion. This is the classical stack processor type of architecture but it requires a special discipline at the time a program is originally defined.
Both the indirects require a concept of memory management that is not obvious to the novice programmer.
The concept of indexed indirect is that memory has to be viewed as a series of tables, in which access to one set of tables is accomplished by indexing through a list of pointers. One set of tables might be searched to perform some type of testing or operation. Then the same index is then used to process another set of pointers. This concept is only applicable to operations in which a variety of inputs are being serviced. A classical application is when several remote devices are being managed by the same control program. An example might be having three teletypes tied on to a device, each teletype is being manually controlled and can be under con trol of the user program. In this type of message handling environment, the control program for the teletypes does nothing more than collect strings of data from the input device and then performs operations on the string
G-3
upon seeing a control signal, usually a carriage return in this case of the teletype. Because any one of the teletypes can be causing any one of the series of operations, this program does not lend itself well to the concept of absolute addressing. In fact, most: of the subroutines which deal with the individual processing should be written address independent.
This normally allows the addition of more devices without paying any penal ty in terms of programming. Therefore, this is a subroutine or nonabsolute type of operation in which the indirect indexed would not apply because each of the various operations use a function of position. In other words, one can assign a series of tables that point at the teletype itself; an other set that points at an outgoing message jitream and another set that points to a series of tables which keep the status of the device. Each of these pointers is considered to be an individual address at the be ginning of a string. Each string is a variable length. The teletype strings may consist of a three character message followed by a character return or a 40 character message followed by a character return. In the
MCS650X, this system will be implemented by means of developing a series of indirect pointers. Each teletype will have an indirect pointer. Its
I/O port has another indirect pointer that points at the put-away string, another one that points at the teletype message output string, another one that points at its status table. If all of the teletypes work this way, it can be seen that the coding to put data into the input message table is the same for all the teletypes and is totally independent of the teletype in which data is being stored.
The index register X serves as a control for the tables so that if all each operation. A sample operation might be: read teletype three, trans fer the data to teletype three input register, update teletype three counter, check to see that teletype three is still active, and decide whether or not to return to signal teletype three back. The coding to perform each of these operations would be exactly the same as coding for teletype two, if the tables were organized such that X was an index register for the pointers.
G-4
indirect was designed and only when a program can be organized for this technique is the indirect used to its maximum potential. The advantages for organizing for this type of approach when the problem requires string manipulation is significant; the comprehensive I/O program is roughly one half the memory and one fourth the execution time of several other microprocessors which do not have this indexed indirect feature.
APPENDIX H
REVIEW OF BINARY
AND
BINARY CODED DECIMAL
ARITHMETIC
H-1
The number 1789 is assumed by most people to mean one thousand, seven hundred eighty-nine, or 1 x lO
3
+ 7 x 10
2
+
8
x 10
1
+ 9 x 10°. However, until the number base is defined, it might mean
1 x 163 + 7 x 162 + x 161 -h 9 x 16° which is hexadecimal and the form used in the microprocessor.
and
1789^0 for base 16 for hexadecimal. Because very few computers or I/O de vices allow subscripting, all hexadecimal numbers are preceded by a $ notation. Then 1789 means base 10 and $1789 means base 16. Why hexadeci mal? This is a convenient way of representing 2 digits in
8
bits.
The MCS650X is a byte-oriented microprocessor which means most opera tions have
8
-bit operations.
There are 2 ways to look at
8
bits. The first is as
8
individual bits in which 00001000 means that bit 3 (bit 7 to 0 representation) is on and all other bits are off or as an
8
-bit binary number in which case the value is
0 x
2 7
+
0 x
2 6
+
0 x
2 5
+
0 x
2 tt + l x
2 3
+
0
x
2 2
+
0 x
2 1
+
0 x
2
° =
8 or $08.
For logic analysis purposes, each bit is unique, but for arithmetic purposes, the
8
bits are treated as a binary lumber.
H-2
Binary Arithmetic Rules;
0 + 0 = 0
0 + 1
=
1
1
+
0 = 1
1
+
1 = 0
with a carry
Carry occurs when the resulting number is too long for the base. In decimal,
8
+ 4 = 2 + 10.
In hexadecimal,
$ 8
+ $4 = $C (see hexadecimal details), so that 8 + 4 has a carry in base
1 0
but not in base 16.
Using these rules to add
8
+ 2 in
0 0 0 0 1 0 0 0
0 0 0 0 0 0 1 0
0 0 0 0 1 0 1 0
8
+ 2
1 0
1
x
1
x
1
x
2 3
2 1
2 3
+
1
x
2 1
Therefore,
8
bits, and binary addition performed using the basic binary add equation,
R. = (A. ¥ B. V C . ,), where, as defined is notation for
J J J
Exclusive-Or.
In most applications, it is also necessary to subtract. Subtract operations either require a different hardware implementation or a new way of representing numbers.
A combination of this is to implement a simple inverter in each bit.
This would make
00001100
11110011
12
-12
However, when subtracting 12 from 12, the result should also be 0.
00001100
11110011
1 1 1 1 1 1 1 1
+12
-12
0
However, if a carry is added to the complemented number:
1 Carry
00001100 12
11110011 -12
0 0 0 0 0 0 0 0
=
0
If, instead of representing -12 as the complement of 12, it is represented as the complement plus carry, the following is obtained:
H-3
11110011 = 12
1
= Carry
11110100
=
-12
00001100 +12
0 0 0 0 0 0 0 0
=
0
This representation is called two's complement and represents the way that negative numbers are kept in the microcomputer. Below are examples of negative numbers represented in two's complement form.
- 0
=
0 0 0 0 0 0 0 0
- 1
=
1 1 1 1 1 1 1 1
-2
=
11111110
-3 = 11111101
-4 = 11111100
-5 = 11111011
-6
=
11111010
-8
=
11111000
-9 = 11110111
Hexadecimal is the representation of numbers to the base 16. The fol lowing table shows the advantages of Hex:
Hexadecimal
8
9
A
B
C
D
E
4
5
6
7
0
1
2
3
F
Binary
0 0 0 0
0 0 0 1
0 0 1 0
0C11
0100
0101
0 1 1 0
0111
1 0 0 0
1001
1010
1011
1100
1101
1110
1111
Decimal
0 0
0 1
0 2
08
09
10
11
12
13
14
03
04
05
06
07
15
H-4
Because 16 is a multiple of 2, hexadecimal is a convenient shorthand for representation of 4 binary digits or bits. The rules on arithmetic also hold.
Binary Hex
0100 1111 4F
+ Q U O 0010 + 62
1011 0001 B1
To take advantage of this shorthand, all addresses in this manual are shown in hexadecimal notation. It should be nc ted that the reader should learn to operate in Hex as soon as possible. Continual translation back to decimal is both time consuming and error prcne. Working in Hex and binary will quickly force learning of hexadecimal manipulation and the famil iarity with working with this convenient representation.
Although many microcomputer applications cin successfully be accom plished with binary operations, some applications are best performed in decimal. Although the use of 1 decimal character per byte would be a legitimate way to solve this problem, this is an inefficient use of the cap ability of the
8
-bit byte.
The microprocessor allows the use of packed BCD representation. This representation is, in
4
-bit form:
0
=
0 0 0 0
1 = 0001
2
=
0010
3 = 0011
4 = 0100
5 = 0101
6
=
0110
7 = 0111
8
=
1000
9 = 1001
In BCD, the number 79 is represented:
Binary BCD Hex
01111001 = 79 = 79
H-5
The microprocessor automatically takes this into account and corrects for the fact that
Decimal BCD Hex
79 = 01111001 79 = 01111001
+12
=
00010010 12
=
00010010
91 = 10010001
8 8
'
The only difference between Hex and BCD representation is that the microprocessor automatically adjusts for the fact that BCD does not allow for Hex values A - F during add and subtract operations.
The offset which follows a branch instruction is in signed two's complement form which means that
$+50 = +80 = 01010000 and $-50 = -80 = 10110000
Proof = 00000000
The sign for this operation is in bit 7 whore an 0 equals positive and a
1
equals negative.
This bit is correct for the two's complement representation but also flags the microprocessor whether to carry or borrow from the address high byte.
The following 4 examples represent the combinations of offsets which might occur (all notations are in hexadecimal):
Example H.4.1: Forward reference, no page crossing
0105
0106
0107
INE
+55
Next CP CODE
To calculate next instruction if the branch is taken
Offset
Address Low for next
OP CODE
+55 0101C101
07 00000111
5C 01011100 with no carry, giving 015C as the result.
H-6
Example H.4.2: Backward reference, no page crossing
015A
015B
015C
BNE
-55
Next OP CODE
To calculate If branch is taken,
Offset -55 = AB = 10101011
+ Address Low for
Next OP CODE +5C = 5C =
07 07
01011100
00000111 thus giving 0107 as the result.
Example H.4.3: Backward reference if page boundary crossed
0105 BNE
0106 -55
0107 Next OP CODE
To calculate if branch is taken, first calculate a low byte
Offset
Address Low for
-55 = AB = 10101011
_07_ = 07^ = 00000111
B2 = B2 = 10110010
There is no carry from a negative offset; therefore, a carry must be made:
-1 = -1 = FF = 11111111
+ Address High = 01^ = 01 = 00000001
00 00
This gives 00 B2 as a result.
00000000
Example H.4.4: Forward reference across page bcundary
00B0 BNE
00B1 +55
00B2 Next OP COEE
To calculate next instruction if branch is taken,
H^7
Offset
Address Low for Next
OP CODE
55 = 01010101
B2 = 1011(010
07 0000(111 with carry on positive number.
+1 1 = 0000C001
Address High 00 = 0000C000
1
= oooooooT
which gives 0107.
H-8
MOSTECHNOLOGY INC.
___________V A L L E Y FO RGE CORPORATE CENTER
950 RITTENHOUSE ROAD, NORRISTOWN, PA. 19401
TEL: (215) 666-7950 TWX: 510/660/4033
advertisement
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Related manuals
advertisement
Table of contents
- 17 Manual Introduction
- 18 Microprocessor Architecture
- 19 The Data B u s
- 20 The Accumulator
- 20 LDA— Load Accumulator with Memory
- 21 STA— Store Accumulator in Memory
- 22 The Arithmetic Unit
- 23 ADC— Add Memory with Carry to Accumulator
- 24 Multiple Precision Addition
- 26 Signed Arithmetic
- 29 Decimal A d d i t i o n
- 30 .1.3 Add Summary
- 34 Signed Arithmetic
- 35 Decimal Subtract
- 37 ORA— "OR" Memory with Accumulator
- 37 EOR— "Exclusive OR" Memory with Accumulator
- 40 SEC— Set Carry Flag
- 41 CLC— Clear Carry Flag
- 41 Zero Flag (Z)
- 42 Decimal Mode Flag (D)
- 42 SED— Set Decimal Mode
- 43 CLD— Clear Decimal Mode
- 43 Break Command (B)
- 43 Expansion Bit
- 43 Overflow ( V )
- 44 CLV— Clear Overflow F l a g
- 44 Determination of Overflow
- 45 Negative Flag (N)
- 46 Flag S u m m a r y
- 47 Concepts of Program Sequence
- 52 JMP— Jump to New Location
- 53 Branching
- 56 Branch Instructions
- 56 BMI— Branch on Result Minus
- 56 BPL— Branch on Result P l u s
- 56 BCC— Branch on Carry Clear
- 56 BCS— Branch on Carry S e t
- 57 BEQ— Branch on Result Z e r o
- 57 BNE— Branch on Result Not Z e r o
- 57 BVS— Branch on Overflow Set
- 57 BVC— Branch on Overflow Clear
- 58 Branch S u m m a r y
- 58 Solution to Branch Out of Range
- 61 Test Instructions
- 61 CMP— Compare Memory and Accumulator
- 63 Bit Testing
- 63 BIT— Test Bits in Memory with Accumulator
- 66 Addressing Techniques
- 68 Concepts of Pipelining and Program Sequence
- 72 Memory Utilization
- 72 1/0 Control
- 73 Memory Allocation
- 73 Implied Addressing
- 75 Immediate Addressing
- 75 Absolute Addressing
- 77 Zero Page Addressing
- 79 Relative Addressing