CompArch 02 CPU

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 105

CHAPTER 2: CPU

1
Computer
Architecture and Organization
Computer Architecture and Organization
LEARNING OBJECTIVES

 Present an overview of essential characteristics of machine

instructions.

 Describe the types of operands used in typical machine

instruction sets.

 Describe the various types of addressing modes common in

instruction sets.

Computer Architecture and Organization 2


WHAT IS AN INSTRUCTION
SET?
 The operation of the processor is determined by the instructions it
executes, referred to as machine instructions or computer instructions.
 The collection of different instructions that the processor can execute is
referred to as the processor’s instruction set.
 Elements of a Machine Instruction
 Operation code: Specifies the operation to be performed (e.g., ADD, I/O).
 Source operand reference: operands that are inputs for the operation.
 Result operand reference: The operation may produce a result.
 Next instruction reference: address of the next instruction.

 Source and result operands can be in one of four areas: Main or virtual
memory, Processor register, immediate and I/O Device

Computer Architecture and Organization 3


INSTRUCTION
REPRESENTATION
 Within the computer, each instruction is represented by a sequence of
bits.
 The instruction is divided into fields

 It’s difficult to deal with binary representations of machine instructions.


 Common practice: use a symbolic representation of machine instructions.
E.g. ADD, SUB, MUL…etc.

Computer Architecture and Organization 4


INSTRUCTION TYPES
 Any program written in a high-level language must be translated into
machine language to be executed.
 Thus, the set of machine instructions must be sufficient to express any
of the instructions from a high-level language.

 With this in mind we can categorize instruction types as follows:


 Data processing: Arithmetic and logic instructions.
 Data storage: Movement of data into or out of register and or memory
locations.
 Data movement: I/O instructions.
 Control: Test and branch instructions.

Computer Architecture and Organization 5


NUMBER OF ADDRESSES
 What is the maximum number of addresses one might need in an
instruction?
 In most architectures, many instructions have one, two, or three operand
addresses, with the address of the next instruction being implicit
(obtained from the PC).

 With 3 addresses, 2 source operand locations and a destination operand


location.
 ADD A, B, C //A = B+C
 Not common:- require a relatively long instruction format to hold the three
address references.

Computer Architecture and Organization 6


NUMBER OF ADDRESSES
 With two-address instructions, and for binary operations, one address
must do double duty as both an operand and a result.
 SUB Y, B //Y = Y – B
 The two-address format reduces the space requirement but also alters the
value of an operand,
 MOVE instruction is used to move one of the values to a result or temporary
location before performing the operation.
 MOVE A, Y , SUB A, B , answer in A
 Reduces length of instruction but requires some extra work
 Temporary storage to hold some results

 Simpler yet is the one-address instruction. For this to work, a second


address must be implicit (AC).
 C = A – B => LOAD A, SUB B, STOR C
 Common on early machines

Computer Architecture and Organization 7


NUMBER OF ADDRESSES
 Write a three, two and one address instructions that could be used to
compute

Y = (A - B)/[C + (D * E)]

 Note that there are four instructions and that the original expression had
five operands.
 Try not to alter the value of any of the operand locations.
 Try to compute the expression with smallest no of instructions possible.

Computer Architecture and Organization 8


Computer Architecture and Organization 9
CON…
 It is, in fact, possible to make do with zero addresses for some instructions.

 Zero-address instructions are applicable to a special memory organization called


a stack.
 The stack is in a known location and, often, at least the top two elements are in
processor registers.
 zero-address instructions would reference the top two stack elements.

10
SO HOW MANY ADDRESSES?
 The number of addresses per instruction is a basic design decision.

 More addresses
 More complex instructions
 Fewer instructions per program

 Fewer addresses
 Less complex instructions
 More instructions per program
 Require less complex processor
 It also results in instructions of shorter length.
 Results in longer execution times

Computer Architecture and Organization 11


INSTRUCTION SET DESIGN
The design of an instruction set is very complex because it affects so many
aspects of the computer system.
The most important of these fundamental design issues include the following:
 Operation list: How many and which operations to provide, and how
complex operations should be.
 Data types: The various types of data upon which operations are performed.
 Instruction format: Instruction length (in bits), number of addresses, size of
various fields, and so on.
 Registers: Number of processor registers that can be referenced by
instructions, and their use.
 Addressing: The mode or modes by which the address of an operand is
specified.

Computer Architecture and Organization 12


TYPES OF OPERAND
 Addresses
 Numbers
 Integer/floating point

 Characters
 ASCII etc.

 Logical Data
 Bits or flags

Computer Architecture and Organization 13


TYPES OF OPERATION

 Data Transfer
 Arithmetic
 Logical
 Conversion
 I/O
 System Control
 Transfer of Control

Computer Architecture and Organization 14


DATA TRANSFER
 Specify
 Source
 Destination
 Amount of data
 Mode of addressing

 May be different instructions for different movements


 e.g. IBM 390……………….(L, LH, LD,ST,STH…..etc)

 Or one instruction and different addresses


 e.g. VAX……….(MOV)

Computer Architecture and Organization 15


ARITHMETIC
 Add, Subtract, Multiply, Divide
 Signed Integer
 Floating point
 May include
 Increment (a++)
 Decrement (a--)
 Negate (-a)
 Absolute

Computer Architecture and Organization 16


LOGICAL
 Bitwise operations based upon Boolean operations
 AND, OR, NOT, EQUAL

Computer Architecture and Organization 17


SHIFT AND ROTATE
OPERATIONS

Examples of Shift and Rotate Operations


Computer Architecture and Organization 18
10100110
CONVERSION
 Change the format or operate on the format of data.
 E.g. Binary to Decimal

INPUT/OUTPUT
 For i/o operation

SYSTEM CONTROL
 Executed only while the processor is in a certain privileged state or is
executing a program in a special privileged area of memory.
 Typically, these instructions are reserved for the use of the operating
system.

Computer Architecture and Organization 19


TRANSFER OF CONTROL
 Branch …..AKA Jump
 Can be conditional or unconditional
 e.g. BRZ,BRP,BRN,BRO

Computer Architecture and Organization 20


TRANSFER OF CONTROL
 Skip
 Implies that one instruction be skipped
 e.g. increment and skip if zero

Computer Architecture and Organization 21


TRANSFER OF CONTROL
 Subroutine call
 Procedure call

Computer Architecture and Organization 22


Use of Stack to Implement Nested Subroutines of Figure 12.8
Computer Architecture and Organization 23
ADDRESSING MODES

 Immediate
 Direct how the processor can determine which address
mode is being used in a particular instruction?
 Indirect
 Register
 Register Indirect
 Displacement (Indexed)
 Stack

Computer Architecture and Organization 24


IMMEDIATE ADDRESSING
Instruction
 The simplest form of addressing Opcode Operand

 Operand is part of instruction


 Used to define & use constants or set initial values of variables
 e.g. ADD 0101
 Add 5 to contents of accumulator
 5 is operand

 Pros
 Fast:- No memory reference to fetch data

 Cons
 Limited range: - the size of the number is restricted to the size of the address
field

Computer Architecture and Organization 25


DIRECT ADDRESSING
 The address field contains the effective address of the operand

EA = A
 e.g. ADD A
 Add contents of memory address A to accumulator
 Look in memory at address A for operand

 Was common in earlier generations of computers


 Single memory reference to access data
 No additional calculations to work out effective address

 Limited address range


 restricted to the size of the address field

Computer Architecture and Organization 26


INDIRECT ADDRESSING
 The address field refer to the address of a word in memory, which in
turn contains a full-length address of the operand.
EA = (A)
 Look in A, find address (A) and look there for operand

 e.g. ADD (A)


 Add contents of cell pointed to by contents of A to accumulator

 Pros
 Increased address space

 Cons
 Instruction execution requires two memory references to fetch the operand:
 one to get its address and a second to get its value.

Computer Architecture and Organization 27


REGISTER ADDRESSING
 Address field refers to a register rather than a main memory address

EA = R
 Pros
 Only a small address field is needed in the instruction
 Very fast: - No time-consuming memory references are required.

 Cons
 Address space is very limited
 Limited number of registers

Computer Architecture and Organization 28


REGISTER INDIRECT
ADDRESSING
 C.f. indirect addressing

EA = (R)
 Operand is in memory cell pointed to by contents of register R
 Pros
 Large address space (2n)
 One fewer memory access than indirect addressing

 Cons
 Extra memory access

Computer Architecture and Organization 29


DISPLACEMENT
ADDRESSING
 Powerful mode of addressing combines the capabilities of direct
addressing and register indirect addressing.
EA = A + (R)
 Address field hold two address fields
 A = base value
 R = register that holds displacement
 or vice versa

Computer Architecture and Organization 30


RELATIVE ADDRESSING
 A version of displacement addressing
 R = Program counter, PC
 EA = A + (PC)
 i.e. the next instruction address is added to the address field to produce
the EA.
 locality of reference

Computer Architecture and Organization 31


BASE-REGISTER
ADDRESSING
 A version of displacement addressing
 EA = A + (R)
 A holds displacement
 R holds pointer to base address
 R may be explicit or implicit
 e.g. segment registers in 80x86

Computer Architecture and Organization 32


INDEXED ADDRESSING
 A version of displacement addressing
 A = base
 R = displacement
 EA = A + (R)
 Good for accessing arrays
 EA = A + (R)
 (R)++

Computer Architecture and Organization 33


STACK ADDRESSING
 Top of stack pointer is maintained in a register.
 In some processors, the top two elements stored in the processor’s registers

 Operand is (implicitly) on top of stack


 e.g.
 ADD Pop top two items from stack and add
 Pros
 No memory reference

 Cons
 Limited applicability

Computer Architecture and Organization 34


SUMMARY OF ADDRESSING

Computer Architecture and Organization 35


36
PART 2

Chapter 2: CPU

Computer Architecture and Organization


LEARNING OBJECTIVES
 Distinguish between user-visible and control/status registers, and
discuss the purposes of registers in each category.
 Summarize the instruction cycle.
 Discuss the principle behind instruction pipelining and how it works in
practice.
 Compare and contrast the various forms of pipeline hazards.

Computer Architecture and Organization 37


INSTRUCTION FORMAT

Reading Assignments

Page 469
Computer Architecture and Organization 38
PROCESSOR
ORGANIZATION
What is the requirements placed on the processor, (the things that it must
do) ?

 Fetch instruction
To do these things, it should be clear that the
processor needs to store some
 Interpret instruction data temporarily.

 Fetch data

 Process data In other words, the processor needs


a small internal memory.
 Write data.
Registers

Computer Architecture and Organization 39


Computer Architecture and Organization 40
REGISTER ORGANIZATION
Registers: - are a very fast, small, and most expensive storage inside the
CPU.
The registers in the processor perform two roles:
 User-visible registers: Enable the machine- or assembly language
programmer to minimize main memory references by optimizing use of
registers.
 Control and status registers:
 Used by the control unit to control the operation of the processor and by
privileged, operating system programs to control the execution of programs.
There is no clean separation of registers into these two categories.

Computer Architecture and Organization 41


USER-VISIBLE REGISTERS
Is one that may be referenced by means of the machine language that the
processor executes. (4 categories )
 General purpose: can be assigned to a variety of functions by the
programmer (with small restrictions)
 Data: used only to hold data
 Address: somewhat general purpose, or they may be devoted to a
particular addressing mode (Segment pointers, Index registers, Stack
pointer).
 Condition codes: are bits set by the processor hardware as the result of
operations.
 partially visible to the user
 read by implicit reference, but the programmer cannot alter them.
 Condition Codes pros and cons ? Check your reference book.

Computer Architecture and Organization 42


REGISTERS DESIGN ISSUES….
1. Use completely general-purpose registers or to specialize their use?
 Make them general purpose
 Increase flexibility and programmer options
 Increase instruction size & complexity

 Make them specialized


 Smaller (faster) instructions format (register is given implicitly)
 limits the programmer’s flexibility

2. How Many Registers?


 Between 8 – 32 registers appears optimum
 Fewer = more memory references
 More does not reduce memory references and takes up processor real
estate
 Some RISC systems use 100 or more registers.

Computer Architecture and Organization 43


REGISTERS DESIGN
ISSUES….
3. How big? (register length)
 Large enough to hold full address (for addresses registers)
 Large enough to hold full word (for data registers
 Some machines allow two contiguous registers to be used as one for
holding double-length values
 C programming
 double int a;
 long int a;

Computer Architecture and Organization 44


CONTROL AND STATUS
REGISTERS
 Employed to control the operation of the processor.
 Most of these, on most machines, are not visible to the user.
 Some of them may be visible to machine instructions executed in a
control or operating system mode.

 Program Counter (PC)


 Instruction Register (IR)
what do these all do?
 Memory Address Register (MAR)
 Memory Buffer Register (MBR)

Computer Architecture and Organization 45


PROGRAM STATUS WORD
(PSW),
PSW : a register or set of registers, which contains condition codes plus other
status information.
 Sign: Contains the sign bit of the result of the last arithmetic operation.
 Zero: Set when the result is 0.
 Carry: Set if an operation resulted in a carry (addition) into or borrow
(subtraction) out of a high-order bit.
 Equal: Set if a logical compare result is equality.
 Overflow: Used to indicate arithmetic overflow.
 Interrupt Enable/Disable: Used to enable or disable interrupts.
 Supervisor: Indicates whether the processor is executing in supervisor or
user mode.
Computer Architecture and Organization 46
CONT…
 Other registers (related to status  The allocation of control
and control) information between registers
 Process control blocks and memory.
 Interrupt vector register  must decide how much control
 System stack pointer
information should be in
registers and how much in
 Page table pointer memory.
Control and status register
organization design issues
 OS support
 N.B. CPU design and operating
system design are closely linked

Computer Architecture and Organization 47


EXAMPLE
MICROPROCESSOR
REGISTER ORGANIZATIONS

Computer Architecture and Organization 48


INDIRECT CYCLE
 Processor’s instruction cycle, Fetch – Execute – Interrupt
 If indirect addressing is used, then additional memory accesses are
required.
 We can think of the fetching of indirect addresses as one more
instruction stages.

Computer Architecture and Organization 49


Computer Architecture and Organization 50
DATA FLOW
 The exact sequence of
events during an
instruction cycle
depends on the design
of the processor.
 But, what happens to
MAR,MBR,PC,IR
 ?????

Computer Architecture and Organization 51


CONT…

Computer Architecture and Organization 52


CONT…

Computer Architecture and Organization 53


INSTRUCTION PIPELINING
 Is similar to an assembly line in a manufacturing plant.
 products at various stages can be worked on simultaneously.
 new inputs are accepted at one end before previously accepted inputs appear
as outputs at the other end.
 Instruction pre-fetch / fetch overlap :- to fetch the next instruction in
parallel with the execution of the current one.
 Pipelining requires registers to store data between stages. (involves
instructions buffering)
 This process will speed up instruction execution. (if equal duration
stages)

Computer Architecture and Organization 54


CON…
In reality, doubling of execution rate is unlikely for two reasons
 The execution time will generally be longer than the fetch time.
 wait for some time before it can empty its buffer

 A conditional branch instruction makes the address of the next


instruction to be fetched unknown.
 wait until it receives the next instruction address from the execute stage.
Then wait until its fetched.
 Any Solutions?

Computer Architecture and Organization 55


CON…
 To gain further speedup, the pipeline must have more stages.
 Consider the following decomposition of the instruction processing.
 Fetch instruction (FI)
 Decode instruction (DI)
 Calculate operands (CO)
 Fetch operands (FO)
 Execute instruction (EI)
 Write operand (WO)

 let us assume equal duration for each stages

Computer Architecture and Organization 56


This six- stage pipeline can reduce the execution time for
9 instructions from 54 time units to 14 time units.
Computer Architecture and Organization 57
CONT…
Keep in mind that in the pervious diagram we assume that
 Each instruction goes through all six stages of the pipeline.
 All of the stages can be performed in parallel.
 There are no memory conflicts. (FI, FO, and WO can occur
simultaneously)
 Another difficulty is the conditional branch instruction, which can
invalidate several instruction fetches.

Computer Architecture and Organization 58


Computer Architecture and Organization 59
Computer Architecture and Organization 60
CONT..
 The greater the number of stages in the pipeline, the faster the execution
rate. But what is the burden
 Movement overhead
 There is some overhead involved in moving data from buffer to buffer and in
performing various preparation and delivery functions.
 This can lengthen the total execution time of a single instruction. This is
significant when sequential instructions are logically dependent
 Very complex:
 The amount of control logic required increases enormously with the number
of stages.

Computer Architecture and Organization 61


PIPELINE HAZARDS
A pipeline hazard occurs when the pipeline, or some portion of the
pipeline, must stall because conditions do not permit continued
execution.
 Resource hazards/structural hazard:- occurs hardware resource can’t
support parallel execution of instructions.
 So instructions must be executed in serial
 Eg. main memory has a single read/write port, source operand for instruction
I1 is in memory

Computer Architecture and Organization 62


CONT…
Data hazards: occurs when there is a conflict in the access of an operand
location.
 Two instructions in a program are to be executed in sequence and both
access a particular memory or register operand.
ADD R1, R2 => R1 = R1 + R2

SUB R3, R1 => R3 = R3 - R1

ADD R1, R2

SUB R3, R1

Computer Architecture and Organization 63


I1 ADD R1, R2
I2 SUB R3, R1

DATA HAZARDS TYPES: RAW


I3 MOV R4, R1
I4 MUL R7,R8

There are three types of data hazards: RAW, WAR and WAW
 Read after write (RAW), or true dependency: An instruction modifies a register
or memory location and a succeeding instruction reads the data in that memory
or register location.
 A hazard occurs if the read takes place before the write operation is complete.
 This is the most common type of hazard and the kind that we use forwarding to
overcome

Clock cycle
Clock cycle
1 2 3 4 5 6 7 8 9
1 2 3 4 5 6 7 8 9
I1 FI DI FO EI WO
I1 FI DI FO EI WO
I2 FI DI FO EI WO
I2 FI DI Idle FO EI WO
I3 FI DI FO EI WO
I3 FI DI Idle FO EI WO
I4 FI DI FO EI WO
I4 FI DI FO EI WO

without forwarding with forwarding


Computer Architecture and Organization 64
DATA HAZARDS TYPES: WAR
 Write after read (WAR), or anti-dependency: An instruction reads a
register or memory location and a succeeding instruction writes to the
location.
 A hazard occurs if the write operation completes before the read operation
takes place.
 This can not happen in our example pipeline because all reads are
early (in FO) and all writes are late (in WO).
 This hazard occurs when there are some instructions that write results
early in the instruction pipeline, and other instructions that read a
source late in the pipeline.
 WAR hazards are rare

Computer Architecture and Organization 65


DATA HAZARDS TYPES: WAW
 Write after write (WAW), or output dependency: Two instructions
both write to the same location.
 A hazard occurs if the write operations take place in the reverse order of the
intended sequence.
 This data hazard only occurs if there are more than on write stages in the
pipeline.

Computer Architecture and Organization 66


CONT…
Control hazards
 A control hazard, also known as a branch hazard
 Occurs when the pipeline makes the wrong decision on a branch
prediction
 therefore brings instructions into the pipeline that must subsequently be
discarded.

Computer Architecture and Organization 67


Memory Address Instruction
1 ADD R1, R2
2 BRZ 7

DEALING WITH BRANCHES


3 SUB R4, R1
⁞ ⁞
7 MOV R4, 10010

 Multiple streams 8 MUL R3,R2

 A brute-force approach
 Allow the pipeline to fetch both instructions, making use of two streams
 Disadvantage:- contention delays for access to the registers and to memory &
additional branch instructions
 IBM 370/168 and the IBM 3033

Clock cycle Clock cycle


1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9
I1 FI DI FO EI WO I1 FI DI FO EI WO

I2 FI DI FO EI WO I2 FI DI FO EI WO

I3 FI DI FO EI WO I3 FI DI FO EI WO

I4 FI DI FO EI WO I7 FI DI FO EI WO

I5 FI DI FO EI WO I8 FI DI FO EI WO

I6 FI DI FO EI WO

I7 FI DI FO EI WO
Pipeline stream 2
I8 FI DI FO EI WO (by assuming the branch is taken)

68
Pipeline stream 1 (by assuming the branch not taken)
DEALING WITH BRANCHES

 Pre-fetch branch target

 The target of the branch is pre-fetched in addition to the instruction following

the branch
 then saved until the branch instruction is executed.

 If the branch is taken, the target has already been pre-fetched.

 Used in IBM 360/91

Computer Architecture and Organization 69


CONT…
 Loop buffer
 A loop buffer is a small, very-high-speed memory
 maintained by the instruction fetch stage of the pipeline and
 containing the n most recently fetched instructions, in sequence.
 If a branch is to be taken, the hardware first checks whether the branch target
is within the buffer.
 If so, the next instruction is fetched from the buffer.

 The loop buffer has three benefits:


 No memory reference required
 Useful for IF–THEN and IF–THEN–ELSE sequences
 This strategy is particularly well suited to dealing with loops, or iterations

Computer Architecture and Organization 70


CONT…
 Branch prediction:- Various techniques can be used to predict whether
a branch will be taken.
 Predict never taken
 Predict always taken Static
 Predict by opcode
 Taken/not taken switch
 Branch history table Dynamic

Static: they do not depend on on the execution history


dynamic: They depend on the execution history

 Delayed branch: automatically rearranging instructions within a


program, so that branch instructions occur later than actually desired.

Computer Architecture and Organization 71


72
PART 3

Computer Architecture and Organization


LEARNING OBJECTIVES

 To know the difference between RISC and CISC architecture.

 To understand the micro-operations involved in instruction cycle.

 Discuss the inputs and outputs of Control Unit and how it works.

 To understand the two implementations of control unit.

Computer Architecture and Organization 73


CPU IS ARCHITECTURE: CISC
vs RISC
 Two Instruction set architectures
 Complex Instruction Set Computers (CISC)
 Reduced Instruction Set Computers (RISC)

Example: Add the content of two memory locations and store the result to another
memory location?
 The CISC Approach
 The primary goal of CISC architecture is to complete a task in as few lines of assembly
as possible.
 This is achieved by building processor hardware that is capable of understanding and
executing a complex instruction.
MULTI [loc 3] , [loc 1], [loc 2]
 When executed, this instruction loads the two values from memory into separate
registers, multiplies the operands in the execution unit, and then stores the product in
the appropriate register (R1).
A single instruction takes
Computer Architecture and Organization 74
more than 1 clock cycle to finish
CONT…THE CISC APPROACH
 Instructions for CISC arc closely resembles a command in a higher level language.

 Potential advantage ? the compiler has to do very little work to translate a high-level

language statement into assembly.

 Because the length of the code is relatively short, very little RAM is required to

store instructions.

 The emphasis is put on building complex instructions directly into the hardware.

 The CISC approach attempts to minimize the number of instructions per program,

sacrificing the number of cycles per instruction.

Computer Architecture and Organization 75


THE RISC APPROACH
 RISC processors only use simple instructions that can be executed
within one clock cycle.
 LOAD A, [loc 1]
 LOAD B, [loc 2]
 PROD A, B
 STORE [loc 3] , A

 Because there are more lines of code, more RAM is needed to store the
assembly level instructions.
 The compiler must also perform more work to convert a high-level
language statement into code of this form.
 The RISC approach reduces the cycles per instruction at the cost of the
number of instructions per program.

Computer Architecture and Organization 76


 Because each instruction requires only one clock cycle to execute, the entire

program in RISC will execute in approximately the same amount of time as of


CISC.
 Potential advantage ? RISC "reduced instructions" require less transistors of hardware

space than the complex instructions, leaving more room for general purpose registers.

 Because all of the instructions execute in a uniform amount of time (i.e. one clock),

pipelining is possible.

 What happen to the registers after we finish the current program?

 In CISC:- processor automatically erases the registers. If one of the operands needs to

be used for another computation, the processor must re-load the data from the
memory bank into a register.

 In RISC:- the operand will remain in the register until another value is loaded in its

place.

Computer Architecture and Organization 77


CISC RISC
The original ISA Redesigned on the original ISA
Instruction can take several clock cycle Instruction take one clock cycle

Hardware-centric design Software-centric design


More efficient use of RAM than RISC Heavy use of RAM

Complex and variable length instructions Simple standardized instructions

Sequential instruction execution Parallel instruction execution

Complexity in CPU (micro-programs) Complexity in compiler

Small code size Large code size


Pipeline is difficult Pipeline is easy
More addressing modes Fewer addressing mode
Have less registers Have more registers

Computer Architecture and Organization 78


Micro-operation and Control
Unit
 What are micro-operations

 Grouping rule

 Examples

 Input and output of CU

 Implementation of CU

Computer Architecture and Organization 79


MICRO-OPERATIONS
 The execution of a program consists of the sequential execution of
instructions.
 Each instruction is executed during an instruction cycle made up of
shorter subcycles
 (fetch, indirect, execute, interrupt).

 The execution of each subcycle involves one or more shorter operations,


that is, micro-operations.
 Micro-operations are the functional, or atomic, operations of a
processor.
 The prefix micro refers to the fact that each step is very simple and
accomplishes very little.

Computer Architecture and Organization 80


Computer Architecture and Organization 81
OP’S IN FETCH CYCLE
 Four registers are involved: MAR, MBR, PC, IR
 Simple fetch cycle actually consists of 3 steps and 4 OP
 Each 𝝁OP involves the movement of data into or out of a register.

Computer Architecture and Organization 82


RULES OF 𝝁OP GROUPING
1. The proper sequence of events must be followed.
 Thus (MAR (PC)) must precede (MBR Memory) because the memory read operation makes

use of the address in the MAR.

2. Conflicts must be avoided.


 One should not attempt to read to and write from the same register in one time unit, because the

results would be unpredictable.

 For example, the micro- operations (MBR Memory) and (IR  MBR) should not occur during

the same time unit.

Note: As long as each 𝝁OP follows the above rules and do not interfere with one
another, several of them can take place during one step, to saving time.

Computer Architecture and Organization 83


OP’S IN INDIRECT CYCLE
 let us assume a one-address instruction format, with direct and indirect
addressing allowed.
 The address field of the instruction is transferred to the MAR.
 This is then used to fetch the address of the operand.
 Finally, the address field of the IR is updated from the MBR, so that it
now contains a direct rather than an indirect address.

Computer Architecture and Organization 84


OP’S IN INTERRUPT CYCLE
 The nature of this cycle varies greatly from one machine to another.

Computer Architecture and Organization 85


OP’S IN EXECUTE CYCLE
 This cycle is different, because of the variety of opcodes, there are a
number of different sequences of micro-operations
 ADD R1, X BSA X
 t1: MAR  (IR(address))
t1: MAR  (IR(address))
 t2: MBR  Memory MBR  (PC)
 t3: R1  (R1) + (MBR) t2: PC  (IR(address))
Memory  (MBR)
 ISZ X
t3: PC  (PC) + I
 t1: MAR  (IR(address))
 t2: MBR  Memory
 t3: MBR  (MBR) + 1
 t4: Memory  (MBR)
 If ((MBR) = 0) then (PC  (PC) + I)

Computer Architecture and Organization 86


THE INSTRUCTION CYCLE
 Defines the complete sequence of micro-operations
 We assume a new 2-bit register called the instruction cycle code (ICC).
 The ICC designates the state of the processor in terms of which portion
of the cycle it is in:
 00: Fetch
 01: Indirect
 10: Execute
 11: Interrupt

 At the end of each of the four cycles, the ICC is set appropriately.

Computer Architecture and Organization 87


Computer Architecture and Organization 88
CONTROL OF THE
PROCESSOR
Define basic elements of processor ?
Describe micro-operations processor performs ?
Determine functions control unit must perform ?
 The functional requirements of the control unit: (those functions that the
control unit must perform)
 Sequencing: The control unit causes the processor to step through a series of
micro-operations in the proper sequence, based on the program being
executed.

 Execution: The control unit causes each micro-operation to be performed


with the use of control signals

Computer Architecture and Organization 89


EXTERNAL SPECIFICATIONS OF
THE CONTROL UNIT
 Clock ---This is how the control unit “keeps time.”
 One micro-instruction (or set of parallel micro-instructions) per clock cycle
IN
 Instruction register
PU
 Op-code and addressing mode for current instruction TS
 Determines which micro-instructions to perform

 Flags
 State of processor
 Results of previous ALU operations

 Control signals from control bus


 Interrupts
 Acknowledgements

Computer Architecture and Organization 90


CONT…
The outputs are as follows:
 Control signals within the processor:
 These are two types: Cause data movement & Activate specific ALU
functions
 Control signals to control bus:
 These are also of two types: control signals to memory, and control signals to
the I/O modules.

Computer Architecture and Organization 91


Computer Architecture and Organization 92
CONTROL SIGNALS
 Three types of control signals are used:
 Those that activate an ALU function;
 Those that activate a data path; and
 Those that are signals on the external system bus or other external interface.

 All of these signals are ultimately applied directly as binary inputs to


individual logic gates
 Let us consider again the fetch cycle to see how the control unit
maintains control.

Computer Architecture and Organization 93


CONTROL SIGNALS IN FETCH
CYCLE
 The first step is to transfer the contents of the PC to the MAR.
 CU activate the control signal that opens the gates between the bits of the PC and the
bits of the MAR.
 The next step is to read a word from memory into the MBR and increment the
PC.
 A control signal that opens gates, allowing the contents of the MAR onto the address
bus;
 A memory read control signal on the control bus;
 A control signal that opens the gates, allowing the contents of the data bus to be
stored in the MBR;
 Control signals to logic that add 1 to the contents of the PC and store the result back
to the PC.
 Following this, the control unit sends a control signal that opens gates between
the MBR and the IR.
Computer Architecture and Organization 94
CONTROL SIGNALS IN FETCH
CYCLE

Computer Architecture and Organization 95


A CONTROL SIGNALS
EXAMPLE

Computer Architecture and Organization 96


For simplicity, the data and control paths for incrementing the PC and for
loading the fixed addresses into the PC and MAR are not shown.

Computer Architecture and Organization 97


INTERNAL PROCESSOR
ORGANIZATION

Computer Architecture and Organization 98


CU IMPLEMENTATION
Control unit can be implemented using
 Hardwired implementation
 Microprogrammed implementation

Hardwired Control Unit


 When the control signals are generated by hardware using conventional logic
design techniques, the CU is said to be hardwired.
 Can be viewed as a state machine that changes from one state to another in
every clock cycle, depending on the inputs of CU.
 The outputs of the state machine are the control signals.
 The sequence of the operation carried out by this machine is determined by
the wiring of the logic elements and hence named as “hardwired”.

Computer Architecture and Organization 99


Computer Architecture and Organization 100
SUMMARY OF HARDWIRED
CU
 Fixed logic circuits that correspond directly to the Boolean expressions are used
to generate the control signals.
 Hardwired control is faster than micro-programmed control.
 A controller that uses this approach can operate at high speed.
 RISC architecture is based on hardwired control unit

Limitation of this implementation


 Complex sequencing & micro-operation logic
 Difficult to design and test
 Inflexible design
 Difficult to add new instructions

Computer Architecture and Organization 101


MICRO-PROGRAMMED
IMPLEMENTATION OF CU
 The control signals associated with operations are stored in special memory

units inaccessible by the programmer as Control Words.

 Control signals are generated by a program are similar to machine language

programs.

 Micro-programmed CU is slower in speed because of the time it takes to

fetch microinstructions from the control memory.

 Is the dominant technique for implementing control units in pure CISC

architectures processors
Computer Architecture and Organization 102
SOME IMPORTANT TERMS IN
MICRO-PROGRAMMED CU
 Control Word : is a word whose individual bits represent various control
signals.
 Micro-instruction : Individual control words in this micro-routine are
referred to as microinstructions.
 Micro-program : A sequence of micro-instructions is called a micro-
program, which is stored in a ROM.
 Micro-routine : Group of micro-instructions each stored in control
word that belongs to same routine.
 Control Memory : the micro-routines for all instructions in the
instruction set of a computer are stored in a special memory called the
Control Memory.

Computer Architecture and Organization 103


Computer Architecture and Organization 104
ADVANTAGES AND DISADVANTAGES
OF
Micro-programmed implementation of CU
 Advantage

 Cheaper and less error prone to implement.

 Simple to design and test.

 Flexible design….. easy to add new features and

 Disadvantage

 Somewhat slower than a hardwired implementation

To know more about this implementation refer to Chapter 16 Micro-programmed


Control (page 586) in your textbook.

Computer Architecture and Organization 105

You might also like