Single Cycle Processor Design
ALU Design
For the ALU design, Idid not use the built in aritmethic library. Instead Idesigned each tool(adder, subtractor, multiplier and shifter) meyself.
Register File
Splitter
This unit is responsible for the separation of the opcode, Rs, Rt, Rd, shamt and I[0:15] which determines the type of the function.
Control Unit
Firstly, the Control Unit takes the first two bits of the opcode to decide whether its R-type, I-type, mult or my function. Then the remaining 4 bits determine which function is selected. The selected function changes the control signals as demonstrated in the Control Signals table below. For instance, when we use the addition operation, Memtoread value becomes low. When a function is selected, the control unit changes the ALU opcode for each function. To do so we use the multiplexers just like if/else condition loops. Our control Unit design is a bit long and could be optimized.
CPU Design
Table of Control Signals
| Instruction
Type |
RegDst | ALUSrc | Memto
Reg |
RegWrite | Mem
Read |
Mem
Write |
Branch | ALUOp
(3:0) |
| add | 1 | 0 | 0 | 1 | 0 | 0 | 0 | 0000 |
| sub | 1 | 0 | 0 | 1 | 0 | 0 | 0 | 0001 |
| and | 1 | 0 | 0 | 1 | 0 | 0 | 0 | 0011 |
| or | 1 | 0 | 0 | 1 | 0 | 0 | 0 | 0100 |
| slt | 1 | 0 | 0 | 1 | 0 | 0 | 0 | 0111 |
| load | 0 | 1 | 1 | 1 | 1 | 0 | 0 | 0000 |
| store | 0 | 1 | 0 | 0 | 0 | 1 | 0 | 0000 |
| addi | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 0000 |
| beq | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0001 |
| blez | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 0001 |
| mult | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0010 |
| mflo | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1001 |
| Sll | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0110 |
| mfhi | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1000 |
The Assembly Code
| The Assembly Code | Address of Instruction | Instruction Machine Code |
| addi $0, $0, 0 | 00100000000000000000000000000000 | 20000000 |
| addi $1, $1, 0 | 00100000001000010000000000000000 | 20210000 |
| addi $2, $2, 4 | 00100000010000100000000000000100 | 20420004 |
| addi $3, $3, 0 | 00100000011000110000000000000000 | 20630000 |
| addi $4, $4, 4 | 00100000100001000000000000000100 | 20840004 |
| beq $3, $4, 9 | 00010000100000110000000000001001 | 10830009 |
| lw $5, 0($1) | 10001100001001010000000000000000 | 8C250000 |
| lw $6, 0($2) | 10001100010001100000000000000000 | 8C460000 |
| mult $5, $6 | 01100000101001100000000000000000 | 60A60000 |
| mflo $7 | 01001000000000000011100000000000 | 48003800 |
| add $0, $0, $7 | 10000000000000000011100000000000 | 80003800 |
| addi $1, $1, 1 | 00100000001000010000000000000001 | 20210001 |
| addi $2, $2, 1 | 00100000010000100000000000000001 | 20420001 |
| addi $3, $3, 1 | 00100000011000110000000000000001 | 20630001 |
| j 5 | 00001000000000000000000000000101 | 08000005 |
The output value obtained from the assembly code is stored in the 0th register. You can check and see that it is A000. However, when we I use the built in multiplier tool the result is A082. This small error is probably caused by a defect in my self-designed multiplier.
You can find the Logisim file here : Download
Its like you read my mind! You seem to understand a lot about this, such as you wrote the guide in it or something. I think that you just could do with a few percent to pressure the message house a little bit, however other than that, that is great blog. An excellent read. I’ll certainly be back.