CSCC69 Operating systems

Prerequisite

This course requires a good understanding of the x86 computer architecture (CSCB58), good C/UNIX system programming skills (CSCB09), and good coding practices (CSCB07).

Course Staff

We encourage you to post questions regarding course materials and projects on Piazza. However, if you need extended support, the course staff will hold office hours.

Course Timing

Course Information

• The course website and its Github repository

  One of the nice things about using Github for the course website is that you can contribute to the course website. If you see something on the course website that should be fixed, or want to improve the UI, please feel free to submit a pull request.

• The Piazza discussion Board

  The discussion board is the best place to ask technical questions, and general questions about the course, projects and labs. For personal issues, please use private posts. I try to respond by the end of the next day. However, due to volume, it may take longer, especially on weekends.

• The anonymous feedback form

  If you have feedback about the course, you can send an anonymous feedback to the course instructor (you also have the option of including your name). Since the sender cannot be determined, comments sent through the feedback form are considered public, and they may receive a response at the beginning of class or on the discussion board.

Marking Scheme

The numeric marks of the projects and final exam will be used to compute a composite numeric score that will determine your final letter grade for the course. The weighting of course work is set as:

A mark of at least 50% on the final exam is required to pass the course. If you receive less than 50% on the final exam, your overall course grade will be capped at 49.

Submission and Grading Policy

For each project, the student or the team will be required to submit the code to the Github repository through Github classroom. Only the final commit submitted to Gradescope will be graded.

The instructor reserves the right to assign different grades to each of the team members based on their individual contributions made to Github repository.

For your work to be graded, it must meet the minimum standards of a professional computer scientist. All files required to build the program must be committed to the repository, and the program must work. Last minute difficulties with git can easily be avoided by ensuring all files are added to the repository well before the deadline, and that you know how to commit them. Your submission may receive a grade of 0, if the code does not compile or does not execute properly.

Each team will have 4 late days which may be spent in units of one day and that can spread into 4 projects. Beyond those grace days, no late submissions will be accepted for any course work, and no make-up assignments will be provided for missed/poorly completed work. It is your responsibility to ensure that all work is completed on time and to the best of your ability.

If an emergency arises that prevents you from being able to complete any piece of work, or attend an exam, contact the instructor immediately.

If a piece of work has been mis-marked or if you believe the rubric used to evaluate the work is not appropriate, you may request a re-mark. For a re-mark to succeed, you must clearly and concisely express what you believe was mis-marked. To request a re-mark, please contact your TA. Requests must be submitted within 1 week of the marks being returned.

AI Policy

Operating systems are learned by designing, implementing, debugging, and reasoning about low-level code. In this course, the Pintos assignments are intended to develop your understanding of kernel threads, synchronization, system calls, user programs, virtual memory, file systems, and debugging in a realistic operating-system environment.

AI tools can be useful learning aids, but they can also complete large parts of the assignments without requiring you to understand the underlying ideas. For that reason, all code submitted for Pintos assignments must be written and understood by you and your team directly.

You may use AI tools for:

• Asking high-level conceptual questions about operating systems.
• Reviewing lecture concepts such as scheduling, synchronization, virtual memory, system calls, page tables, file systems, or race conditions.
• Explaining compiler errors, linker errors, runtime crashes, kernel panics, assertion failures, or test output.
• Understanding documentation, APIs, C syntax, GDB commands, Makefiles, or Pintos tooling.
• Asking for debugging strategies, sanity checks, invariants, or test ideas.
• Asking for feedback on code that you have already written, provided the AI does not rewrite or complete the implementation for you.

You may not use AI tools to:

• Write, generate, or complete any part of an assignment implementation.
• Fill in TODO sections or implement required Pintos functions.
• Generate full functions, patches, diffs, or design-level solutions.
• Convert assignment requirements directly into working code.
• Refactor large parts of your code into a finished solution.
• Debug by asking the AI to identify and fix the bug directly.
• Use coding agents, automated editing tools, or AI autocomplete to modify assignment code.
• Use AI to produce design-document answers that describe code or design decisions you did not make yourself.

This applies to all Pintos assignments, tests, experiments, leaderboards, and design documents.

Each assignment repository will include an AGENTS.md file that describes how AI tools are allowed to interact with the repository. Students must not modify, remove, weaken, or bypass this file. When using an online chat interface such as ChatGPT, Claude, Gemini, or similar tools, students should paste the contents of AGENTS.md at the beginning of the conversation. The purpose of this file is to ensure that AI tools act like teaching assistants rather than solution generators.

AI autocomplete must be disabled while working on assignment code. Non-AI autocomplete, such as completing variable names, function names, or editor snippets, is permitted.

Coding agents such as Cursor Agent, Claude Code, Codex, Copilot Workspace, or similar tools must not be used to edit, generate, or complete Pintos assignment code. They may only be used in a read-only teaching mode consistent with the AGENTS.md file.

Ask yourself whether a course instructor or teaching assistant would provide the same help in office hours.

Allowed:

• “Can you explain what a race condition is?”
• “What does this assertion failure mean?”
• “What should I inspect when debugging a priority-scheduling bug?”
• “Can you explain how condition variables differ from semaphores?”
• “Here is my code. Can you ask me questions that help me debug it?”

Not allowed:

• “Implement priority donation for me.”
• “Write process_wait().”
• “Fix my system call handler.”
• “Complete the page fault handler.”
• “Generate code that passes the Pintos tests.”
• “Here is the assignment handout. Give me the solution.”

You are responsible for understanding every line of code and every design decision in your submission. If you use AI in a permitted way, you should still be able to explain your implementation, justify your design, and debug your code independently.

Violations of this policy will be treated as academic misconduct under the course and university academic-integrity policies. If you are unsure whether a use of AI is permitted, ask the teaching staff before using it.

Academic Integrity

You are expected to comply with the Code of Behaviour on Academic Matters.

Assignment solutions must be prepared individually, except where an assignment handout allows working with a partner. Note that working with a partner may be restricted to just part of an assignment, such as programming task, whereas the rest of the assignment must be solved by an individual.

You are fully responsible for the piece of work you submit to Github repository as your contribution to the project.

When the assignment handout allows you to use snippets of code from the web, you should cite the source in the source code. As a rule of thumb, any piece of code larger than 5 lines that has been copied and re-used as is or even slightly modified must be clearly referenced. However, any piece of code larger than 25 lines should not be re-used.

The use of AI tools is permitted only to the extent allowed by the AI policy described above.

You may discuss projects with other students, for example to clarify the requirements of an project, to work through examples that help you understand the technology used for an project, or to learn how to configure your system to run a supporting piece of software used in an project. However, collaboration at the level of answering written questions or designing and writing code, is strictly forbidden. Written problems and programming projects must be answered, designed and coded by you alone, using the text, your own notes, and other texts and Web sources as aids.

Do not let other students look at your project solutions, since this can lead to copying. Remember you are in violation of the UTSC Academic Code whether you copy someone else’s work or allow someone else to copy your work. These rules are meant to ensure that all students understand their solutions well enough to prepare the solutions themselves. If challenged you must be able to reproduce and explain your work.

You are not allowed to look at solutions available online and you are not allowed to make your solution publicly available online as well, even after the class term.

You are not allowed to ask for help outside of the course Piazza. Asking for help anywhere else online or in private chat groups (unless the private group chat was setup between the group members of the group project) will be considered as unauthorized help.

The course staff reserves the right to use code and text analysis tools to compare your submission with others (including past/present submissions and others available online) to verify that no improper behavior has occurred.

Failure to comply with these guidelines is a serious academic offense. In past academic offense cases, the Associate Dean has imposed penalties for code violations that range from a mark of zero on plagiarized projects to academic suspension from the University.

Accessibility Needs

The University of Toronto is committed to accessibility. If you require accommodations for a disability, or have any accessibility concerns about the course, the classroom or course materials, please contact Accessibility Services as soon as possible: https://www.utsc.utoronto.ca/ability/welcome-accessability-services