Skip to main content

Control unit sub-operations of GPP

Short answer

The control unit executes the instruction cycle: fetch (IR←Mem[PC], PC++), decode, operand fetch, execute (ALU operation), and store result. For ADD R1, [2000H], it fetches the instruction, reads memory address 2000H, adds to R1, and writes back. Pipelining overlaps these steps to raise throughput.

Explain the control unit sub-operations (basic operations carried out by the controller, with instruction cycles) of a general-purpose processor with a suitable example.

The control unit (controller) of a GPP repeatedly executes the instruction cycle, a sequence of sub-operations (micro-operations) that bring an instruction from memory and carry it out. The basic stages are fetch, decode, execute and store.

Sub-operations

  1. Fetch: read the instruction from memory.
    • MAR <- PC (put the address on the bus)
    • MDR <- Memory[MAR] (read)
    • IR <- MDR; PC <- PC + 1
  2. Decode: the control unit decodes the opcode in IR, determining the operation, the operand locations (addressing mode) and the control signals needed.
  3. Fetch operands: read registers or memory (e.g. MAR <- address field; MDR <- Memory[MAR]).
  4. Execute: the ALU performs the operation (add, AND, compare, shift, branch condition test).
  5. Store result: write the result into a register or memory (or update PC for jumps).
    +-------+   +--------+   +---------+   +-------+
 -->| Fetch |-->| Decode |-->| Execute |-->| Store |--+
    +-------+   +--------+   +---------+   +-------+  |
        ^                                             |
        +---------------- next instruction -----------+

Example: ADD R1, [2000H] (R1 <- R1 + memory[2000H])

CycleMicro-operation
FetchMAR <- PC; IR <- Mem[MAR]; PC <- PC + 1
DecodeControl unit finds ADD with a direct memory operand
Operand fetchMAR <- 2000H; MDR <- Mem[MAR]
ExecuteALU: R1 + MDR
StoreR1 <- ALU result; status flags updated

Pipelined processors overlap these sub-operations for several instructions to raise throughput.

✓ Answer checked by Claude Opus 5.5 (high effort)

Related questions

All Embedded System chapter 2 questions →