CMPUT 229 - Computer Organization and Architecture I

Lab 4: Packet Forward - Handle Packet

Prerequisite

This lab assignment builds on top of the previous lab assignment, Lab 3: Packet Forward - Checksum. To understand this lab, it is recommended that you review the Background and Input sections of the webpage for Lab 3. A correct solution to Lab 3 is not required to complete this lab. However, having a correct checksum function will facilitate testing. Read more on how to use your solution to Lab 3 while testing your solution to this lab in the Checksum subsection of the Testing section.

Additional Background

Due to the simplicity of the routing logic, an IP packet can get caught in a loop. For example, Router A forwards the packet to Router B, which forwards it back to Router A through some series of mother routers. This commonly occurs when trying to send packets to a nonexistent destination. To avoid packets remaining on the internet forever, unable to reach a destination, every packet carries an unsigned 1-byte counter, called a Time To Live (TTL). Every router, as part of receiving and forwarding a packet, decrements this counter before passing the packet on. If the TTL ever reaches zero, the packet is simply dropped, and an Internet Control Message Protocol (ICMP) timeout packet is sent back to the device identified by the source IP indicating the failure to send.

Task

Write a RISC-V program that processes an IP packet. Your program will decide whether the packet should be dropped or forwarded. If the packet should be dropped, your program must report the reason. Otherwise, your program must prepare the packet to be forwarded.

Information

In this assignment, to determine if an IP packet can be forwarded we must inspect the following fields:

• IP Version: This field specifies the version of the IP packet. This assignment will only handle IPv4 packets. Therefore, the IP Version must always be four. Your solution must check this condition first because non-IPv4 packets may have different field layouts that the code required by this assignment cannot handle.
• Time To Live (TTL): This field is an unsigned byte that indicates the remaining number of times that the packet can be forwarded before it is dropped. If TTL is less than or equal to one, then the packet should be dropped. This is the second condition that your solution must check. If the TTL is greater than one and the packet is otherwise valid, then the TTL should be decreased before the packet is forwarded.
• Header Checksum: This field contains a halfword value used for error-checking the header. When a packet arrives at a router, the router calculates the checksum of the header and compares it to this field. If the values do not match, the packet should be dropped. If the values do match, and the packet is otherwise valid, then the header checksum should be recalculated after the TTL was decreased and before forwarding. Your solution should assume that checksum.s contains a correct implementation of the checksum function specified in Lab 3 that you can use to calculate the checksum.

Specifications

Write RISC-V assembly code for the following functions in handlePacket.s:

• handlePacket:
• Arguments:
• a0: starting address of an IP packet in memory.
• Return Values:
• If the packet should be forwarded:
• a0: 1
• a1: the starting address of the IP packet
• If the packet should be dropped:
• a0: 0
• a1: a value identifying the reason to drop the packet according to the following convention,
• 0, if the calculated checksum does not match the Header Checksum field.
• 1, if the TTL is less than or equal to one.
• 2, if the IP Version is not four.
• Effect:
• Verify that the packet is IPv4, the header checksum is valid, and the TTL is greater than one. If any of these conditions fail, the packet should be dropped. Otherwise, decrement the TTL and recalculate the header checksum so that the packet can be forwarded.
• Note:
• If the packet should be forwarded, only the TTL and the Header Checksum fields should be modified, the rest of the packet should be unaltered.
• validateIP:
• Arguments:
• a0: starting address of an IP packet in memory.
• Return Values:
• a0:
  • 1: if the IP version of the packet is four.
  • 0: otherwise
• Effect:
• Verify that the packet is IPv4.
• validateTTL:
• Arguments:
• a0: starting address of an IP packet in memory.
• Return Values:
• a0:
  • 1: if the packet's TTL is greater than one.
  • 0: otherwise
• Effect:
• Verify that the packet's TTL is not less than or equal to one.
• validateChecksum:
• Arguments:
• a0: starting address of an IP packet in memory.
• Return Values:
• a0:
  • 1: if the calculated header checksum matches the Header Checksum field.
  • 0: otherwise
• Effect:
• Verify that the packet's Header Checksum is valid.

Testing

Download the code to get started!

Checksum

validateChecksum and handlePacket need the calculated header checksum of an IP packet to determine if its Header Checksum field is valid and to recalculate the checksum if the packet should be forwarded.

At the top of Code/handlePacket.s there is the line: .include "checksum.s". With this include, the functions in handlePacket.s can use the functions that are in checksum.s.

For testing, you should paste the code from your solution to Lab 3 into the file Code/checksum.s with the line .include "common.s" removed. This allows your program to use your checksum function from Lab 3.

The checksum.s file that is in the Code directory when you submit will not be used to grade your solution. While grading, we will replace the checksum.s file with one that contains a correct implementation of the checksum function specified in Lab 3. Therefore, you do not need a correct checksum function to write a solution to this lab, you should write your solution with the assumption that checksum.s contains a correct implementation of the checksum function.

Assume that the calculation performed in the checksum function skips the Header Checksum field.

Input Guarantees

As in Lab 3, the input to this assignment is an IP packet. You can generate packets to use as input with the C program packetGenerator.c provided in Lab 3.

The following is guaranteed for all test cases:

• An IP packet is at most 128-byte large.

Resources

• Slides used for in-class introduction of the lab (.pptx) (.pdf)
• Video for in-class introduction to the lab (YouTube)

Marking Guide

Assignments too short to be adequately judged for code quality will be given a zero for that portion of the evaluation.

• 15% For code cleanliness, readability, and comments
• 15% For correct validateIP, validateTTL, and validateChecksum functions
• 20% For correctly dropping to invalid packets
• 20% For reporting the correct reason to drop invalid packets
• 30% For correctly modifying and forwarding valid packets
• Here is the mark sheet used for grading

Submission

There is a single file to be submitted for this lab:

• handlePacket.s should contain the code for the functions handlePacket, validateIP, validateTTL, and validateChecksum.
• Do not add a main label to this file
• Do not modify the line .include "common.s"
• Do not modify the line .include "checksum.s"
• Submit only your handlePacket.s