CS224 – Implementing the MIPS Processor with Pipelined Microarchitecture (Solution)

$ 20.99
Category:

Description

Lab Location: EA-Z04

Purpose: In this lab you will implement and test a pipelined MIPS processor using the digital design engineering tools (System Verilog HDL, Xilinx Vivado and FPGA, Digilent BASYS3 development board). To do this, you will need to add pipeline registers, forwarding MUXes, a hazard unit, etc to your datapath, and of course make the control pipelined as well. You will be provided a skeleton System Verilog code for the Pipelined MIPS processor and fill the necessary parts to make it work. Then, you will synthesize them and demonstrate on the BASYS3 board.

Summary

2. Please upload your programs of Part 1 (PRELIMINARY WORK) to the Unilica by
StudentID_FirstName_LastName_SecNo_PRELIM_LabNo.txt Upload only the programs. Only a NOTEPAD FILE (txt file) is accepted. Your paper submission for preliminary work must match MOSS submission.
We use MOSS testing only for code submissions. So, you are not supposed to submit the whole preliminary report online. You will ONLY submit the code online.

No late submission will be accepted.

At the conclusion of the demo for getting your grade, you will upload your lab (only lab) program work to the Unilica Assignment, for similarity testing by MOSS. See Part 3 below for details.

Part 1. Preliminary Work / Preliminary Design Report (60 points)

At the end of this lab, you will have implemented the pipelined MIPS architecture that can be seen in the file that is provided as PipelineDatapath.PNGPipelineDatapath.PNG (Notice that there is no early branch prediction in this pipeline. Hence, the branch resolution is done in the Execute stage. Note also that hardware for jump instruction is not present; therefore, you do NOT need to implement it). Be sure to have a printout of the pipelined processor with you, to use during the lab. Your PDR should contain the following items:

a) You have to provide a neat presentation prepared by Word or a word processor with similar output quality such as Latex as you were asked in Lab 1. At the top of the paper on left provide the following information and staple all papers. In this part provide the program listings with proper identification (please make sure that this info is there for proper grading of your work, otherwise some points will be taken off).
CS224
Your Full Name/Bilkent ID

b) [13 points] The list of all hazards that can occur in this pipeline. For each hazard, give its type (data or control), its specific name (“compute-use” “load-use”, “load-store” “J-type jump”, “branch” etc), the pipeline stages that are affected, the solution (forwarding, stalling, flushing, combination of these), and explanation of what, when, how.
c) [8 points] The logic equations for each signal output by the hazard unit, as a function of the input signals that come to the hazard unit. This hazard unit should handle all the data and control hazards that can occur in your pipeline (listed in b above) so that your pipelined processor computes correctly.
d) [30 points] You are given a skeleton System Verilog code for your pipelined MIPS processor in the file PipelinedMIPSProPipelinedMIPSProcessorToFillNewcessorToFillNew.txt.txt. The places in the code that needs to be modified are shown with comment blocks above them. Fill them and highlight the changes you made in the code in your report. You can use a different text highlight color (do this by hand after getting the printout) for this purpose. You do NOT need to follow the skeleton code point by point. If you think your design is better, you are welcome to try it in your code, as long as your version of the code works, too. Note that this is a design problem and there is no single solution to it but if you feel comfortable with the skeleton code then you can use it.
e) [9 points] Study the four example programs which respectively have no hazard, compute-use hazard, load-use hazard and branch-hazard. For hazardous programs explicitly show the places where hazards occur and the registers causing the hazard.
● No Hazard
addi $t0, $zero, 7 8’h00: 32’h20080007;
addi $t1, $zero, 5 8’h04: 32’h20090005;
addi $t2, $zero, 0 8’h08: 32’h200a0000;
addi $t3, $t0, 15 8’h0c: 32’h210b000f;
add $t2, $t0, $t1 8’h10: 32’h01095020;
or $t2, $t0, $t1 8’h14: 32’h01095025;
and $t2, $t0, $t1 8’h18: 32’h01095024;
sub $t2, $t0, $t1 8’h1c: 32’h01095022;
slt $t2, $t0, $t1 8’h20: 32’h0109502a;
sw $t0, 2($t1) 8’h24: 32’had280002;
lw $t1, 0($t0) 8’h28: 32’h8d090000;
beq $t0, $zero, 1 8’h2c: 32’h1100fff5;
addi $t2, $zero, 10 8’h30: 32’h200a000a;
addi $t1, $zero, 12 8’h34: 32’h2009000c;

● Compute-use hazard
addi $t0, $zero, 58’h00: 32’h20080005;
addi $t0, $zero, 5 8’h00: 32’h20080005;
addi $t1, $t0, 6 8’h04: 32’h21090006;
add $t2, $t1, $t0 8’h08: 32’h01285020;

● Load-use hazard
addi $t0, $zero, 5 8’h00: 32’h20080005;
addi $t1, $zero, 6 8’h04: 32’h20090006;
addi $a0, $zero, 1 8’h08: 32’h20040001;
addi $a1, $zero, 2 8’h0c: 32’h20050002;
sw $t0, 0($t1) 8’h10: 32’had280000;
lw $t1, 1($t0) 8’h14: 32’h8d090001;
add $t2, $t1, $a0 8’h18: 32’h01245020;
sub $t2, $t1, $a1 8’h1c: 32’h01255022;

● Branch hazard
addi $t1, $zero, 2 8’h00: 32’h20090002;
beq $zero, $zero, 2 8’h04: 32’h10000002;
addi $t1, $zero, 5 8’h08: 32’h20090005;
addi $t1, $t1, 6 8’h0c: 32’h21290006;
addi $t1, $zero, 8 8’h10: 32’h20090008;
addi $a0, $zero, 0 8’h14: 32’h20040000;
addi $a1, $zero, 0 8’h18: 32’h20050000;
sw $t1, 0($zero) 8’h1c: 32’hac090000;

Part 2: Simulation and Hardware Testing
Simulation (20 points)

Hardware Testing (20 points)
To implement the modified processor, you will need to assemble the modified System Verilog modules that model the changes you made to implement the pipelined microarchitecture. You should also consider the following: to slow down the execution to an observable rate, the clock signal should be hand-pushed, to be under user control. One clock pulse per push and release means that instructions advance one stage in the pipeline. Once the pipeline is full, it means that each push will cause one instruction to finish unless a flush or stall caused nothing to complete that cycle. Similarly the reset signal should be under hand-pushed user control. So these inputs need to come from push buttons, and to be debounced and synchronized. The memwrite and regwrite outputs (along with any other control signals that you want to bring out for viewing) can go to LEDs, but the low-order bits of writedata (which is RF[rt]) and of dataaddr (which is the ALU result) should go to the 7-segment display, in order to be viewed in a humanunderstandable way. [Consider why it isn’t necessary to see all 32 bits of these busses, just the low-order bits are enough.] So a new top-level System Verilog module is needed, containing top.v as well as 2 instantiations of pulse_controller, and 1 instantiation of display_controller, and some hand-pushed signals coming from push buttons, and some anode and cathode outputs going to the 7-segment display unit on the BASYS3 board, and some signals going to LEDs.

Make the constraint file that maps the inputs and outputs of your top-level System Verilog model to the inputs (100 Mhz clock and pushbutton switches) and outputs (AN and CA signals to the 7segment display, and memwrite (plus others?) going to a LED) of the BASYS3 board and its segment display, and memwrite (plus others?) going to a LED) of the BASYS3 board and its FPGA.

Part 3. Submit your code for MOSS similarity testing

Combine all the new and modified Verilog codes into a file called

Part 4. Cleanup
1. After saving any files that you might want to have in the future to your own storage device, erase all the files you created from the computer in the lab.
2. When applicable put back all the hardware, boards, wires, tools, etc where they came from.
3. Clean up your lab desk, to leave it completely clean and ready for the next group who will come.

Part 5. Lab Policies
1. You can do the lab only in your section. Missing your section time and doing in another day is not allowed.
4. You must be in lab, working on the lab, from the time lab starts until your work is finished and you leave.
5. No cell phone usage during lab.
6. Internet usage is permitted only to lab-related technical sites.
7. For labs that involve hardware for design you will always use the same board provided to you by the lab engineer.

1

Reviews

There are no reviews yet.

Be the first to review “CS224 – Implementing the MIPS Processor with Pipelined Microarchitecture (Solution)”

Your email address will not be published. Required fields are marked *