Types of Instructions

Understanding instruction types is fundamental to processor design. Instructions define the operation, source operands, and destination where results are stored.

Instruction Register (IR)

The Instruction Register (IR) holds the instruction currently being executed by the processor. It contains binary data that defines:

Operation type (e.g., add, move)
Source operands
Destination register

Each clock cycle, the processor fetches an instruction into the IR and executes it.

Instruction Types

Instructions are classified based on how many operand addresses are explicitly defined.

Three-Address Instructions

These instructions explicitly define two source operands and one destination.

add r2, r1, r0

r2 = r1 + r0

All operands explicitly defined
Fewer instructions required
Higher hardware complexity

Two-Address Instructions

One operand acts as both source and destination.

add r0, r1

r0 = r0 + r1

One implicit operand
Moderate complexity
More instructions than three-address

Single-Address Instructions

Uses an implicit accumulator register along with one explicit operand.

add r0

A = A + r0

Example sequence for r0 + r1:

move A, r1
add r0
move r0, A

Uses accumulator
Simpler hardware
Requires multiple instructions

Zero-Address Instructions

No operands specified. Uses a stack-based architecture.

PUSH r0
PUSH r1
ADD

Operands taken from stack
Simplest hardware design
Highest number of instructions required

Trade-Off Summary

Type	Hardware Complexity	Instruction Count
Three-address	High	Low
Two-address	Medium	Medium
Single-address	Low	High
Zero-address	Very Low	Very High

Design Choice

For this processor project, we will use three-address instructions.

Clear operand definition
Efficient execution
Better for debugging and verification

Interview Questions

Why are three-address instructions efficient?
What is the role of the accumulator?
How do stack-based architectures work?
What are the trade-offs between instruction types?

Introduction

UVMArena