1.1.1 Structure and Function of the Processor Flashcards
1
Q
What is the purpose of the ALU and the CU?
A
- The Arithmetic and Logic unit completes mathematical and logical operations on data.
- The control unit uses signals to direct the operations of the CPU. It manages data flow, decodes instructions and coordinates the processors.
2
Q
What is a register, and what do the main 5 each do?
A
- They are small high speed memory cells, temporarily holding data.
- The PC holds the address of the next instruction to be executed.
- The ACC stores calculation results.
- The MAR holds the address of a location to be read/written to.
- The MDR holds read/to be written data.
- The CIR holds the current instruction (operand and opcode).
3
Q
What is a bus, and what do they each do?
A
- They are sets of parallel wires connecting processor components. The number of wires affects the number of bits that can be simultaneously transmit.
- The Data bus is a bidirectional bus used for transmitting data and instructions.
- The Address bus transmits memory addresses.
- The Control bus is a bidirectional bus used to send control signals between internal and external components. (IE clock, write, requests, interrupts).
4
Q
How do assembly language instructions work?
A
- Mnemonics represent instructions as simplified machine code.
- An instruction contains an operand (the data or address of data) and opcode (the type of instruction to be executed on the data).
5
Q
What is the purpose of pipelining?
A
- Completing the FDE cycle of 3 instructions simultaneously by holding data in a CPU-close buffer.
- Reduces CPU idleness by fetching, decoding and executing simultaneously.
6
Q
What are the steps of the FDE cycle?
A
Fetch:
- PC address copied to MAR, instruction at address copied to MDR by data bus.
- PC incremented, MDR instruction copied to CIR.
Decode:
- CIR contents split into operand and opcode.
Execute:
- Decoded instruction is now executed using whichever registers are needed.
7
Q
What factors affect the performance of the CPU?
A
- A higher clock speed means more FDE cycles can be complete in a second, timed by a clock generating alternating 0/1 signals.
- A greater number of cores (independent processors performing FDE cycles) boosts performance of optimised applications.
- The type of cache memory and its transfer rate. Lower levels of cache hold less but transfer much faster, holding frequently used data/instructions.
8
Q
What are the differences between Von Neumann and Harvard Architecture?
A
- Von Neumann includes a shared memory containing data and instructions with one bus. Allows for cheaper, size optimised programs and a simpler CU.
- Harvard architecture has physically separate instruction and data memories and two buses. Allows for different sizes and properties of memories, and faster execution.
- Contemporary processors are often using combined architecture types.