Assignment 1 (minishell, M-pipes, heap) cps210 - Spring 1999 Late Policy: Basically after the first day -5 points.

Grade Breakdown:

Shell 30 pts

M-Pipes 35 pts

Heap 35 pts

----------

100 pts

StandardDeviation 9.02

Mean 72.6

Median 72

Hi 89

Lo 62

Overall: Everyone, with a few exceptions, did ok on this assignment. However no one got everything right. I took off points for testing programs that weren't very thorough, i.e. the burden is on you to prove to me your programs actually work. It was also bad to have programs which didn't work, memory allocators that called new or malloc, and mpipes with no write ups or piper programs which didn't compare sys pipes with m-pipes. Those who did well usually made most of their points on flawless minishells and decent memory allocators. Overall the quality for memory allocators was the highest; everyone seems to have done an ok job on that part of the assignment.

Assignment 2 (sthreads, synch problems)

cps210 - Spring 1999

Grade Breakdown:

threads:sleep,wakeup,term,join 25 pts

threads:wait,signal,broadcast

Barriers: 25

StudProf: 25

PCQueue: 25

----------

100 pts

Overall: I was a lot less harsh with grading this time. Most people seemed to do fine on the thread/condition variable stuff. One or two missed a couple of "big" things, but in general it was ok. Pretty much everyone got something working for barrier and studprof test. But, especially for studprof, there seemed to be quite a few who coded their way into compliance. Where I found it I noted it, but keep in mind that locking can be expensive. You don't want to throw more synch than is necessary at a problem. Also Many used a single lock on the barriers to keep those who were waking from continuing(by blocking when re-aquireing the lock from wait()), if you used and alternated between two condition variables it would allow the slaves to continue immediately with no risk but at some expense to keep track of the polarity each slave is on. Most pcqueue's passed the testing.

Criteria For things which locked up I took off 10pts. For style, 2pts. For other miscellaneous problems in thread/lock code 5pts.

StandardDeviation 5.17

Mean 89

Median 87

Hi 98

Lo 50

Assignment 3 (BufferCache)

cps210 - Spring 1999

breakdown in percentage

Successful test 80% (-5->10 for each failure)

Clear/succinct write up 20% ----------

Overall: Everyone seems to have done a really good job on this assignment. The hardest part seems to have been in reading through the code; most people said that it was easier but this may have been a result of the synchronization practice with the prior assignments. This was succinct. Most write-ups were ok. There were a couple that were outstanding in effort, this is why I weight the write up fairly heavily.

StandardDeviation 9.54

Mean 89.55

Median 92

Hi 100

Lo 68