1. Operands and Branches

What are the standard sizes for floating-point operands?

Floating-Point Operands:

• Single precision = 4B

• Double precision = 8B

What are the components of a floating-point number and its representation formula?

A floating-point number consists of:

• Mantissa (also called significand)
• Exponent

Floating-point representation: $Floating\ Point\ Number = Mantissa \times 2^{Exponent}$

Example: $0.625 = 5 \times 2^{-3}$

What are the four common types of control flow changes?

Types of Control Flow Changes:

• Conditional branches

• Jumps

• Procedure calls

• Procedure returns

How is the target address calculated in PC-Relative Addressing?

PC-Relative Addressing calculation: $Target\ Address = PC + Offset$

What are Register-Indirect Jumps and when are they useful?

Register-Indirect Jumps:

• Target address is stored in a register.

Useful when:

• Target address is unknown at compile time (will be loaded in later)

• Return functions, case statements

• Target too far for PC displacement

What are the three methods for implementing branches?

Methods for branches:

• Compare and branch

• Condition register

• Condition code

How does the Compare and Branch method function?

Compare and Branch:

• Performs a comparison.

• Branch decision depends on comparison result.

• No need for register use but more work per instruction.

How does the Condition Register method function?

Condition Register:

• Performs a comparison.

• Result saved in register.

• Branch occurs if register contains: $0 \quad \text{or} \quad 1$

• Simple but wasteful.

What is the Condition Code (CC) method?

Condition Code (CC):

• Uses bits stored in a special purpose Condition Code (CC) register.

• Has special properti