Description
Background
The /proc file system is a logical view of internal data that is maintained by the Linux kernel
about a process. For a given process the kernel exposes a logical file called
/proc/[pid]/stat (where [pid ]is the process ID, e.g., /proc/235/stat). This stats
file contains a variety of statistics about the process that can be used to report runtime statistics
about the process. In this homework, you will be using the following 3 values as described in the
following URL: https://man7.org/linux/man-pages/man5/proc.5.html
Word number
in stats file
Logical name
for Value
Notes
14th word utime (float) Cumulative user time. This value must be divided by divide
by sysconf(_SC_CLK_TCK) to get value in seconds
15th word stime (float) Cumulative system time. This value must be divided by
divide by sysconf(_SC_CLK_TCK) to get value in
seconds
23rd word vsize (long) Current virtual memory size in bytes. This value must be
divided by 1000 to get memory size in KB.
Starter code
The starter code for this homework is just the solution for Exercise #9. Invest time to refresh
your memory by reviewing the starter code.
Requirements
Suitably extend (how you modify is up to you) the starter source code to perform the following
additional functionality:
1. Base case — Run a program & record runtime statistics [22 points]: Run a specified
command with suitable arguments and return the output and runtime statistics in the
HTML format shown further below. Recording of runtime statistics is:
a. Outputs are sent line-by-line to the client using chunked encoding
b. Record and print runtime statistics every second by opening & reading
/proc/[pid]/stat
c. Reading and printing statistics each second for the duration of a child process can
be accomplished as suggested below:
int exitCode = 0;
while (waitpid(pid, &exitCode, WNOHANG) == 0) {
sleep(1); // sleep for 1 second.
// Record current statistics about the process.
}
d. It is best to run the above loop in a separate thread to record the necessary
statistics.
e. The recorded statistics is finally sent to the client as a big chunk at the end of
program execution. No special formatting was needed to get the
statistics/numbers to be printed as shown in expected outputs.
Due before: 11:59 PM (before Midnight) on Tue Nov 27 2018
Page 3 of 6
! The base case is relatively straightforward with plenty of time to finish.
Consequently, if the base case does not operate as expected, as per the course policy,
the whole program will be assigned zero score.
2. Setup statistics to plot a chart [8 points]: If the genChart parameter (3rd parameter) to
the serveClient method is true, then the program should also generate the runtime
statistics as a JSON array in the following format:
[ second, utime+stime, vsize ]
See annotated HTML in base case output below for details.
Note: This output should use exactly the same statistics values from base case. Do not reread the stats file to generate the statistics in JSON output.
3. Code quality [5 points]: Five points in this homework are reserved for good code quality,
effective code reuse, and documentation (in the form of comments). These 5 points
would be the hardest to earn in this homework.
Functional testing
The supplied starter code already includes a simple test harness that will also be used for testing
and grading purposes. Several test inputs and expected output files are also supplied to
streamline your testing as shown below. The true/false flag (i.e., last command-line
argument) is used to enable/disable generation of JSON format used to plot charts.
$ ./hw7 base_case1_inputs.txt my_base_case1_output.txt false
$ diff my_base_case1_output.txt base_case1_expected_outputs.txt
Testing for graph case:
$ ./hw7 base_case1_inputs.txt my_graph_case1_output.txt true
$ diff my_graph_case1_output.txt graph_case1_expected_outputs.txt
Needless to add, the above diff commands will not produce any output if the outputs are
identical.
! Due to non-deterministic scheduling and multiple users on the machine, the CPU
time and virtual memory use could be a bit different in some cases. Minor
differences in values are acceptable.
Ensure you test with input files and corresponding expected outputs prior to submission.
Browser testing
You may test operation of your webserver running on os1.csi.miamioh.edu. A partial
screenshot is shown further below for your reference (you may have to use a different port
number instead of 4000).
Due before: 11:59 PM (before Midnight) on Tue Nov 27 2018
Page 4 of 6
Submit to Canvas
This homework assignment must be turned-in electronically via Canvas. Ensure your C++ source
files are named appropriately. Ensure your program compiles (without any warnings or style
errors) successfully. Ensure you have tested operations of your program as indicated. Once you
have tested your implementation, upload the following onto Canvas:
ο The 1 C++ source file you developed for this homework
Upload all the necessary C++ source files to onto Canvas independently. Do not submit
zip/7zip/tar/gzip files. Upload each file independently.
Screenshot of browser test:
Due before: 11:59 PM (before Midnight) on Tue Nov 27 2018
Page 5 of 6
Base case #2 Annotated output HTML
Parts of the HTML highlighted in gray are fixed/constant strings that can be hardcoded in your
program. Values highlighted in red are chunk sizes in hexadecimal notation.
HTTP/1.1 200 OK
Content-Type: text/html
Transfer-Encoding: chunked
Connection: Close
156
Output from program
Runtime statistics
| Time (sec) | User time | System time | Memory (KB) |
|---|---|---|---|
| 1 | 0.99 | 0 | 14336 |
| 2 | 1.99 | 0 | 14336 |
| 3 | 2.99 | 0 | 14336 |
| 4 | 3.36 | 0 | 22728 |
| 5 | 3.36 | 0 | 22728 |
| 6 | 3.36 | 0 | 31117 |
| 7 | 3.36 | 0 | 31117 |
| 8 | 3.36 | 0 | 47894 |
| 9 | 3.36 | 0 | 47894 |
| 10 | 3.45 | 0.54 | 14336 |
| 11 | 4.25 | 0.74 | 14336 |
| 12 | 5.25 | 0.74 | 14336 |
| 13 | 6.25 | 0.74 | 14336 |
| 14 | 6.56 | 0.74 | 22810 |
| 15 | 6.56 | 0.74 | 22810 |
| 16 | 6.56 | 0.74 | 39587 |
| 17 | 6.56 | 0.74 | 39587 |
| 18 | 6.56 | 0.74 | 73146 |
| 19 | 6.56 | 0.74 | 73146 |
| 20 | 6.69 | 1.29 | 39587 |
| 21 | 6.72 | 1.51 | 0 |
0


