✝ JMJ ✝

ad majorem Dei gloriam

CPS 356: Operating Systems

Department of Computer Science, University of Dayton

Spring 2020

Table of Contents (TOC)

Course Description

CPS 356 (3 hours) is a course that introduces the theoretical and practical issues underlying an operating system's structure and operation. Topics include process and thread creation and management, scheduling, concurrent, multi-threaded programming and synchronization, deadlock, memory management, virtual memory, and computer security. Concepts are demonstrated using the C, Go, and Elixir programming languages in a Linux environment. This course assumes no prior experience with Linux, C, Go, Lua, or Elixir.

Course Details

CPS 250 (Introduction to Computer Organization) or ECE 314 (Fundamentals of Computer Architecture) & CPS 350 (Data Structures & Algorithms).
Meeting times:
Mon Wed 3:35pm-4:50pm, Miriam Hall Room 203.
Dr. Perugini, e-mail id: sperugini1, tel: 229-4079, office: AN 145
    Th 9:45am-12:15pm (only when classes are in session); & by appointment.
Student Helpers:
Kyle Mielke (e-mail id: mielkek1) & Zachary Rowland (e-mail id: rowlandz1); Miriam Hall 16
    T11:00am-noon & 5:00pm-6:00pm;
    F 9:00am-11:00am (only when classes are in session); & by appointment.

Anna Duricy (e-mail id: duricya1); Anderson Hall 131
    Th2:00pm-3:15pm; (only when classes are in session); & by appointment.


Required Textbooks

    [LCP] Linux and C Programming by S. Perugini. 2020. Draft (Available as a Resource in Isidore).

    [PLCI] Programming Languages: Concepts and Implementation by S. Perugini. 2020. Draft (Available as a Resource in Isidore; only ``Chapter 14: Concurrency and Synchronization'' needed for this course.).

    [SCMSW] Seven Concurrency Models in Seven Weeks: When Threads Unravel by P. Butcher. Pragmatic Programmers, Dallas, TX, 2014. ISBN-13:978-1-937785-65-9 (textbook webpage contains links to the source code of all programs in the text). An eBook of [SCMSW] is available free to all UD students in the library's eContent collection. To access it conduct a search for the title in the library's catalog at library.udayton.edu.

Steps to access [SCMSW] off-campus:
  1. Go to https://login.libproxy.udayton.edu/menu.
  2. Login using your standard UD credentials.
  3. Click on the link labeled: EBSCO Ebook Collection (formerly NetLibrary).
  4. This link will will take you to a page were you can perform a search. Search for "Seven concurrency models in seven weeks: When threads unravel" Click on "eBook Full Text".

    Note that there is a limit on how many people can view the book and once someone checks it out, no one else can use it. Also, if you incorrectly break your session (closing the page), the book remains checked out for a period of time and you cannot use the book in until the session ends.
    [SMLSW] Seven More Languages in Seven Weeks: Languages that are Shaping the Future by B.A. Tate, F. Daoud, I. Dees, & J. Moffit. Pragmatic Programmers, Dallas, TX, 2014. ISBN-13:978-1-941222-15-7 (textbook webpage contains links to the source code of all programs in the text). An eBook of [SMLSW] is available free to all UD students in the library's eContent collection. To access it conduct a search for the title in the library's catalog at library.udayton.edu. Steps to access [SMLSW] off-campus: follow same steps above for accessing [SCMSW] (with the title of this book).


Course Wordle

CPS 356: Operating Systems/Spring 2020


Learning Outcomes


Course Outline

(required reading assignments and course notes)
  1. Introduction to operating systems & the UNIX/Linux & C programming environment ([OSC10] Ch 1-2; [LCP] Ch 1-5; [USP] Ch 1-2, 4)
    1. introduction to operating systems (review of computer organization, C exercises): Jan 13 15 22 Feb 3
    2. the UNIX philosophy (class UNIX/Linux page, Linux quick reference sheet, vi quick reference sheet, vi editor, UW vi reference): Jan 13 15 22 29
    3. files & directories (manipulation & management): self-study
    4. system libraries & I/O (C quick reference sheet): Jan 15 22 27 29 Feb 3 5
    5. compiling C in UNIX (static vs. dynamic linking, macros, conditional compilation, error handling, & debugging; RMS's gdb tutorial, valgrind): Feb 10

  2. Processes & threads ([OSC10] Ch 3-4; [LCP] Ch 6-7; [USP] Ch 2-6)
    1. processes (identification; getpid, creation; fork, & termination): Jan 29 Feb 3 5 19
    2. memory allocation/deallocation: Jan 22 27 Feb 3 5 10
    3. storage & linkage classes in C: Feb 10
    4. threads & thread-safe functions (strtok): Feb 24
    5. process manipulation (wait): Feb 5
    6. process manipulation (exec & building an argument vector): Feb 24 26 Mar 2
    7. the Linux shell (Learning the Shell) & process environment (variables, configuration, customization; Advanced Linux quick reference sheet): Feb 26
    8. low-level I/O (open & close, & read & write): Feb 24
    9. context switching: Feb 24
    10. (shell) job control (through signals) & terminals: Mar 2 9
    11. files & directories (data structures, inodes, & hard & symbolic links): value added

    12. Exam I (closed book, closed notes) : Feb 12

  3. Scheduling ([OSC10] Ch 5)
    1. types, evaluation criteria, & algorithms (FCFS, SJF, SRTF, RR): Feb 26 Mar .
    2. multi-level queues & multi-level feedback queues: Mar .

  4. Concurrency and Synchronization ([PLCI] Ch 14; [PG] Ch 7, [SCMSW] Ch 5, & [SMLSW] Ch 4)
    1. mutual exclusion & the critical section problem (§§ 6.1-6.4): Mar 30 Apr 1
    2. : Apr 1 6
    3. classical problems of synchronization (§ 6.6): Apr 8 20 27
    4. introduction to Go programming (Go by example, a tour of Go, command-line programming with Go, effective Go): Apr 1 6 .
    5. communicating sequential processes (CSP) (Go/CSP quick reference sheet) in Go ([PG] Ch 7): Apr 1 6 . . 20 27 29

    6. Exam II (closed book, closed notes) : Mar 11 23
      Exam III (closed book, closed notes) : Apr 15

  5. Deadlock ([OSC10] Ch 7): Apr 6 8 20 27
    1. deadlock prevention (§§ 7.1-7.4):
    2. deadlock avoidance & Banker's algorithm (§ 7.5) (Banker's examples):
    3. deadlock detection & recovery (§§ 7.6-7.8):

  6. Memory Management ([OSC10] Ch 9)
    1. fundamentals (overview of hardware, logical vs. physical address space, dynamic loading & linking) (§§ 9.1): Apr .
    2. contiguous allocation (§ 9.2): Apr .
    3. paging (§§ 9.3-9.5): Apr . . .
    4. segmentation ([OSC8] §§ 8.6-8.8): Apr . .
    5. segmentation faults & buffer overflows ([OSC8] §§ 8.6-8.8): Apr 29

  7. Virtual Memory ([OSC10] Ch 10)
    1. demand paging (§§ 10.1-10.3):
    2. page replacement algorithms (FIFO, optimal, LRU; § 10.4):
    3. allocation (§ 10.5) & thrashing (§ 10.6):
    4. course reflection (terms): Apr 29

  8. Appendices ([LCP] Appendices)
    1. Vim quick reference sheet
    2. Linux quick reference sheet
    3. C quick reference sheet
    4. Advanced Linux quick reference sheet
    5. Go quick reference sheet

  9. Final Exam (comprehensive, closed book, closed notes) : Wed May 6, 10:10am-noon, Miriam Hall 203 online on Isidore.


Evaluation Criteria

(all figures below are approximate, and subject to change within the first few weeks of the course)
Component Quantity Points per Total points
Labs (and Lab Quizzes) ~10 varies 245 (+33 EC)
Midterm Project 1 80 80
Exams (and quizzes) 3 2 ~125 375 250
Final exam (comprehensive) 1 300 300
Total:1,000 875

Labs involves analytical, theoretical, and programming exercises. The programming requires a fair amount of critical thought and design, and approximately 500-1000 lines of code. To prepare students for the realities of computer science problems in industry and graduate school (and beyond) this course encourages (and rewards) self-reliance and independent, self-directed work. Handwritten assignments are not accepted. Assignments are due at 3:35pm in class. Late assignments are not accepted. No exceptions. All exams are in-class, closed-book, and closed-notes. Attendance is mandatory at all examinations; make-ups are not given. Any missed examination will result in a zero. Make no assumptions about anything; always consult the instructor first. Final letter grades of A, A-, B+, B, B-, C+, C, C-, and D start approximately at 93, 90, 87, 83, 80, 77, 73, 70, and 60 percent, respectively.



CPS 356 is a challenging course and moves at a very fast pace. Spending a minimum of nine hours outside of class each week reading, studying, and programming is required. I advise you to see me to discuss any problems you may have before you are evaluated. Having said this, CPS 356 is exciting, fun, and essential. The advent of multi-core processors on the desktop makes mastery of core operating system concepts and concurrent programming more necessary than ever for the modern computer scientist.


Classroom & Course Policies

Students are expected to conduct themselves with respect, integrity, and virtue. Keep phones and similar devices in a silent mode and put away during class (i.e., out of sight). The use of laptop computers and similar devices is not permitted in class. Audio or video recording of any kind in class is strictly prohibited.


Academic Integrity

To achieve the course objectives, homework/lab assignments must be a sole result of your work, not be shared with other students, and prepared in accordance with the University Honor Pledge (see below). Moreover, you may not plagiarize code from any, cited or uncited, textbooks, online resources, or other authors. There is no team, group-work assignments in this course. Discussions among classmates must never include pending assignments. No exemptions. All questions and comments about a pending assignment must only be directed to the instructor and teaching assistants. Evidence indicating a violation of this policy will be handled according to the University Academic Honor Code and result in a doubly-weighted zero which will not be dropped (e.g., if the assignment is worth 100 points, you receive a 0/200) or a zero on the next exam. Make no assumptions about this policy; always consult the instructor first. No student should ever feel that they must resort to academic dishonesty. You are encouraged to consult the instructor if you are struggling with the course or an assignment. No grade is worth your integrity. Honesty in your academic work will develop into professional integrity. The faculty and students of the University of Dayton will not tolerate any form of academic dishonesty.

The Honor Pledge as listed in the Academic Honor Code section of the Undergraduate Catalog applies in full to this course.

Honor code FAQ



Lab Assigned Due Total points
Lab #1 Jan 22 Jan 29 20
Lab #2 Jan 29 Feb 5 30
Lab #3 Feb 5 Feb 19 45
Lab #4 Feb 24 Mar 2 20
Lab #5 Mar 2 Mar 4 15
Lab #6 Mar 4 Mar 9 20
Lab #7 Apr 6 Apr 20 15
Lab #8 Apr 20 Apr 27 45
Lab #24 Mar 30 Mar 30 15
Lab #25 Apr 1 Apr 2 20
Lab #27 Apr 8 Apr 20 10
Total Lab Points: 245 (+10 EC)


Lab #1

Assigned: January 22
Due: January 29, 3:35pm

Total points: 20 points

Style guide | Academic Integrity | Evaluation Criteria

[LCP] Programming Exercise 4.31.33. (Use filename /home/<logname>/labs/lab1/countsubsstdin.c)


Lab #2

Assigned: January 29
Due: February 5, 3:35pm

Total points: 30 points

Style guide | Academic Integrity | Evaluation Criteria

  1. (20 points) [LCP] Programming Exercise 4.31.36. (Use filename /home/<logname>/labs/lab2/removesubsargs.c)

  2. (10 points) [LCP] Programming Exercise 7.3.19. (Use filename /home/<logname>/labs/lab2/alternate.c)


Lab #3

Assigned: February 5
Due: February 19, 3:35pm (all equipment issues with RPIs must be resolved before February 12)

Total points: 45 points

Style guide | Academic Integrity | Evaluation Criteria

  1. (20 points) [LCP] Programming Exercise 4.31.38. (Use filename /home/<logname>/labs/lab3/allsubsargs.c.)

  2. (15 points) [LCP] Programming Exercise 7.5.5. (Use filename /home/<logname>/labs/lab3/mathfib.c.)

  3. (10 points) Setup your Raspberry Pi according to the instructions here (or see the Lab #2 in the Lab Manual posted in Isidore). Then, demo the following to one of the TAs in office hours by or before the deadline:
    1. scp'ing a file from your Raspberry Pi to your SUSE account without using the @ symbol in the command line (this means that the account ids must match).
    2. scp'ing a file from your SUSE account to your Raspberry Pi without using the @ symbol in the command line (this means that the account ids must match).
    3. Running Firefox on your Raspberry Pi but displaying the window on your laptop.


Lab #4

Assigned: February 24
Due: March 2, 3:35pm

Total points: 20 points

Style guide | Academic Integrity | Evaluation Criteria

(20 points) [LCP] Programming Exercise 4.31.40. (Use filename /home/<logname>/labs/lab4/parsestring.c)


Lab #5

Assigned: March 2
Due: March 4, 3:35pm

Total points: 15 points

Style guide | Academic Integrity | Evaluation Criteria

(15 points) [LCP] Programming Exercise 7.5.7. (Use filename /home/<logname>/labs/lab5/interface.c.)


Lab #6

Assigned: March 4
Due: March 9, 3:35pm

Total points: 20 points

Style guide | Academic Integrity | Evaluation Criteria

Complete Lab 14 in the course Lab Manual in Isidore first before starting this lab.

(5+6+6+3=20 points) [OSCJ8] Exercise 5.12 (parts a, b, c, and d) on pp. 235-236, or exercise 5.4 (parts a, b, c, and d) on p. 205 of the 7th ed. Consider the following set of processes, with the length of the CPU burst given in milliseconds:

Process Burst Time Priority
P1 10 3
P2 1 1
P3 2 3
P4 1 4
P5 5 2

The processes are assumed to have arrived in the order: P1, P2, P3, P4, P5, all at time 0.

  1. Draw four Gantt charts that illustrate the execution of these processes using the following scheduling algorithms: First-come, first serve (FCFS), Shortest Job First (SJF), non-preemptive priority (a smaller priority number implies a higher priority), and Round Robin (RR) (with quantum = 1).

  2. Give the turnaround time of each process for each of the scheduling algorithms in part a.

  3. Give the waiting time of each process for each of these scheduling algorithms.

  4. Which of these scheduling algorithms results in the minimum average waiting time (over all processes)?


Lab #7

Assigned: April 6
Due: April 15, 3:35pm

Total points: 15 points

Style guide | Academic Integrity | Evaluation Criteria

(15 points) Complete Lab 26: Learning Channels & Communicating Sequential Processes (in Go) in the OS Lab Manual posted in Isidore. It is repeated here for convenience.
  1. (7 points) [PLCI] Programming Exercise 14.4.18 as /home/<logname>/labs/lab7/PingPongChain.go. In this exercise, you do not need a channel to synchronize the processes, so do not use one. This exercise is more about learning how to spawn threads than it is about synchronization. However, you must use a channel to keep the main thread alive.

  2. (8 points) [PLCI] Programming Exercise 14.4.20 as /home/<logname>/labs/lab7/PingPongFan.go. This time you must use a wait group to keep the main thread alive.


Lab #8

Assigned: April 20
Due: April 27, 3:35pm

Total points: 45 points

Style guide | Academic Integrity | Evaluation Criteria

  1. (15 points) [PLCI] Programming Exercise 14.4.24 as /home/<logname>/labs/lab8/pingpongonlytwo.go.
  2. (15 points) [PLCI] Programming Exercise 14.4.32 as /home/<logname>/labs/lab8/AirTrafficControl.go.
  3. (15 points) [PLCI] Programming Exercise 14.4.34 as /home/<logname>/labs/lab8/Bakery.go.


Midterm Project

Assigned: March 4
+20 points Extra Credit Due: March 23 March 30, 5:05pm EST
Due: March 30 April 6, 5:05pm EST

Total points: 80 points

Isidore | Style guide | Academic Integrity | Evaluation Criteria


Design and implement a program (in any language) that simulates some of the job and CPU scheduling of a time-shared operating system.

Detailed Description and Requirements

When jobs initially arrive in the system, they are put on the job scheduling queue which is maintained in FIFO order. The job scheduling algorithm is run when a job arrives or terminates. Job scheduling allows as many jobs to enter the ready state as possible given the following restriction: a job cannot enter the ready state if there is not enough free memory to accommodate that job's memory requirement. Do not start a job unless it is the first job on the job scheduling queue. When a job terminates, its memory is released, which may allow one or more waiting jobs to enter the ready state.

A job can only run if it requires less than or equal to the system's main memory capacity. The system has a total of 512 blocks of usable memory. If a new job arrives needing more than 512 blocks, it is rejected by the system with an appropriate error message. Rejected jobs do not factor into the final statistics (described below).

Note that all jobs in the ready state must fit into available main memory.

Process scheduling is managed as a multilevel feedback queue. The queue has two levels, each queue is organized as a FIFO, and both use a round robin scheduling technique. New jobs are put on the first level when arriving in the ready state. When a job from the first level is given access to the CPU, it is allowed a quantum of 100 time units. If it exceeds that time quantum, it is preempted and moves to the second level.

The jobs on the second level may only be allocated the CPU if there are no jobs on the first level. When a job on the second level is given access to the CPU, it is allowed a quantum of 300 time units. If it exceeds that, it is preempted and put back on the second level of the ready queue.

Process scheduling decisions are made whenever any process leaves the CPU for any reason (e.g., expiration of a quantum or job termination). When a job terminates, do job scheduling first, then process scheduling. Also, give preference to first level jobs (i.e., if a job from the second level of the ready queue is running, and a new job enters the first level, the running job is preempted to the second level in favor of the first level job).

While executing on the CPU, a job may require I/O, which preempts it to the I/O wait queue for the duration of its I/O burst.

When a job completes, put it on a finished list for later processing.

The simulator is driven by the events read from standard input. Examples of possible events are given below. The first field will be the first character of the line, and subsequent fields will be separated by one of more spaces or tabs. The header of each field in the following examples does not appear in the input stream.

A new job arrives:

Event Time Job Memory Run Time
A     140  12  24     2720

Interpretation: job 12 arrives at time 140, requires 24 blocks of memory and uses the CPU for a total of 2720 time units.

A job needs to perform I/O:

Event Time I/O Burst Time
I     214  85

Interpretation: the job currently running on the CPU will not finish its quantum because at time 214 it needs to perform I/O for a duration of 85 time units.

Display the status of the simulator:

Event Time
D     214

Interpretation: display the status of the simulator at time 214.

You may assume that events appear on the input stream in ascending time order. However, realize that the events given in the input stream are not only events which your simulator must handle. For instance, a time quantum expiration is not an event given in the input stream, but it is an event which your simulator must handle. Furthermore, an internal event, such as a time quantum expiration, not in the input stream, may occur at the same time as an event in the input stream (e.g., a new job arrival). Events in the input stream are external events.

The following is a list of internal events (i.e., not given on the input stream) which your simulator must handle:

Assume that context switching and displays take no simulator time (an unrealistic assumption in a real operating system).

When a display is requested, print the contents of all queues as well as the job currently running on the CPU to standard output using only the format used in the sample output given below.

After processing all jobs, write the following to standard output (in this order, as shown on the sample output given below):

  1. completion time for each job (in order of completion),
  2. average turnaround time (where turnaround time is defined as completion time minus arrival time), and
  3. average job scheduling wait time (where wait time is defined as the number of time units spent in the job scheduling queue).

Event Collisions

Often more than one event happen at the same time. Use the following rules to determine which events to process first:

Architectural View of the Simulator

Additional Requirements

  1. Your system must be developed and run on your Raspberry Pi. Multiple programming languages are available on the Pi (C, C++, Java, Python, Perl, Go, etc.) through sudo apt-get install.
  2. Your implementation must be distributed across more than one source code file, in some sensible manner which reflects the logical purpose of the various components of your design, to encourage problem decomposition and modular design.
  3. Include a README file in your submission which describes (i) the language you used to develop your program and (ii) the name of the compiler or interpreter which you used to compile or interpret your program.

Hints and notes

Test data: sample input and output streams

  1. (only events A & D, & E & T) p2stdin_a and p2stdout_a [image]
  2. (all events A, I, & D, & E, C, & T) p2stdin_b and p2stdout_b [image]

This test data is available at /home/perugini_cps356/share/project/. At first, simply try to get only one job through your system; see the test file /home/perugini_cps356/share/project/testdata/one.txt. Once you are confident that your system processes only one job properly, try to get two jobs through the system; see the test file /home/perugini_cps356/share/project/testdata/two.txt

While developing your simulator, you are encouraged to get it to work on the simple test input first (p2d.dat) and progressively enhance and refine your system to the point where it works on the most complex test input (p2a.dat).

Use the Linux diff utility to compare your output to the correct output. For full credit, the output produced by your program must have zero differences, as defined by diff, with the output posted here.

There is also a reference executable of a solution for this project available at /home/perugini_cps356/share/project/OSsim


How to submit

Note: All directory and filenames below are case-sensitive. You must use the directory and filenames exactly as shown below, (i.e., all lower case).

Prepare your submission file as /home/<logname>/project/project.tar. This archive must contain only the most minimal set of files necessary to build your simulator from scratch. Only the file /home/<logname>/project/project.tar will be electronically collected from your account on the deadline.

Failure to follow these submission requirements will result in a 10% penalty.


  1. How do a create my tar file?
      The requirements require you to make a .tar and place it in /home//project/project.tar. A tar file is a single file that contains a packaged up files and/or directories. The command to make a tar file on UNIX systems (Mac OS X and Linux) is as follows:
      tar cvf project.tar <directory_with_code>
      tar cvf project.tar <list of source code files>
      Be very careful to ensure the file name after the cvf is the name of the tar file. If you run the command as follows, you will overwrite one of your source files!
      tar cvf Main.java Scheduler.java Types.java
      The command to untar a tar file is as follows:
      tar xvf project.tar 
      This will create the directory structure with all the files indicated by project.tar.

  2. I developed my project on Windows using and IDE. What should I do?

      Some of you may work locally on your personal computer in some IDE. To upload your code from your computer to the suse server you can either copy and paste the contents of a file into a file on the server or you can use a secure copy program.

      To transfer files to and from your account, use a secure file transfer program such as FileZilla. You can also use PSCP or PSFTP available for free download from the PuTTY Download Page. You can use pscp.exe or psfpt.exe.

      For Windows users you will need to download pscp or psfpt.exe. You will need to have the pscp.exe or psftp.exe in your CMD path (system environment variable PATH) or in your current working directory for this command to work or you will have to specify the path to pscp in the command (C:\putty\pscp.exe for example). Below is an example of the command:

      pscp.exe -r <path_to_local_directory_with_code> <logname>@cpssuse04.cps.udayton.edu:/home/<logname>/project/

      For Mac OS X and Linux users, there is a command built into your terminal called scp (for secure copy. The format of this command is:

      scp -r <path_to_local_directory_with_code> 

      Lastly, it is strongly recommended that you make multiple copies of your project. This will allow you to revert changes if something accidentally gets overwritten.


Ninety percent of your score will come from correctness and 10% of your score will come from following our programming style guide. Applicable submission penalties will then be applied.

In an effort to award partial credit to students who are unable to complete certain parts of this project, students earn up to two different portions of the 70 possible points for correctness:

Note: Depending on how you order the jobs on your I/O wait queue, your dump of the jobs waiting for I/O may not match our output exactly, and for just that queue, that is acceptable. For instance, you might organize your I/O wait queue as a priority queue where the job which comes off first is at the head, or you might maintain jobs on the I/O wait queue in the order in which they are put on and then search for the job to take off when the I/O is complete.



  1. Exam I (closed book, closed notes) : Feb 12

  2. Exam II (closed book, closed notes) : Mar 11 23

  3. Exam III (closed book, closed notes) : Apr 15

  4. Final Exam (comprehensive, closed book, closed notes) : Wed May 6, 10:10am-noon, Miriam Hall 203 online on Isidore.


Quick Reference Sheets


Other Helpful ...

Programming style guide

Practice problems:
see conceptual and programming exercises in [LCP] and [PLCI] (Chapter 14); (outdated, but still relevant) practice problems.

available in Isidore

Computer accounts:
Linux account access | UDit | A beginner's guide to effective e-mail

Helpful links:
academic calendar (PDF) | student handbook | UDit FERPA policies

Dr. Perugini welcomes any feedback you may have on the course motif and approach, style of the lectures, the concepts presented in class, the course webpage, labs, projects, deadlines, exams, course and grading policies, or your general experience in the course.


This material is based upon work supported by the National Science Foundation under Grant Number (NSF Grant Number 1712406). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

Last modified:
ad majorem Dei gloriam

✝ JMJ ✝