1) execution of our testing script
This testing script runs a series of nachos programs to test several aspects
of the assignment. You'll see references to each of these programs in your
webclass grade description. Briefly they are:
* bullet1 - tests if kernel handles address errors
* bullet2 - tests if kernel handles bad exec arg
* bullet3 - tests if kernel handles huge syscall arg
* bullet4 - tries to execute a file that isn't NOFF
* File I/O - creates, writes, and reads a file
* Args - tests passing of command line args
* Exec (forker) - execs and joins several times
- good test of memory reclamation!
For tests that failed we attempted to fix the problems by looking at your
code and modifying a few things (such as the amount of allocated memory).
The bullet proofing tests were only worth a couple of points each. We placed
more worth on the last 3 tests.
If you're interested, the tests and their source are all located in
/usr/class/cs240a/HW2/TestPrograms
2) Manual testing of your shell and cat programs
3) Evaluation of your README and TESTCASE files. We gave a little extra
credit to groups who went above and beyond the call of duty. For
example, in implementing the shell, some groups added the notion of a path.
Other groups allowed file redirection and added backgrounding.
4) Evaluating extra credit efforts. We looked at the process scheduling
and thread fork/yield implementations to determine their value.
A few people didn't have README's so we guessed which group you were
in. The people who we don't have grades for are:
Maria Klemperer-Johnson
Alejandro Steckler
Mark Wihl
If you were in a group, let us know.
Robert and Steve