Welcome to the first (and only) course where you'll learn the truth about computers.

Syllabus »

My office hours: 2:00PM Monday and Friday and 4:00pm on Wednesday or by appointment.

TA hours: Sunday 4:00pm – 6:00pm and Weds 7:00pm-9:00pm in Williams 303. The TA is Joe Cleveland.

Class Server URL: comp210.eastus.cloudapp.azure.com

Exam dates:

  1. Week of 8 March (evening exam)
  2. Week of 9 April (in class)
  3. Final: Tuesday, 7 May, 10:30AM (scheduled final period)

Notes

  • Note about the Practicum 2 writeup. For phase 4 I want the pseudo-code for the recursive function, not for phase 4.
  • Practicum 2: what is "%fs:0x28"? This is a reference to the way that the intel machine language adresses memory, something that we have not (and will not) talk about in class. If you're interested, see this article: X86 memory organization. You can safely ignore this instruction when you see it, but the larger lesson is that you have to deal with some uncertainty when working with the bomb. You won't know everything, but you have to use what you do know to figure out what is happening.
  • Lab 7: why is the instruction "sar 0x3f, %rdx" there?
  • Integer division always rounds it’s result by rounding towards 0. So if we do 5 the result is 2.5 which is rounded to 2. This is the same result that we get if we shift 0101 to the right one position, i.e., 0010. If we do −5, however, the 3 result is −2.5 which should round toward 0, i.e., to −2. The 2’s complement version of −5 however is 1011. If we shift it right we get 1101 or −3 not −2. To fix this, we can add one before we shift, i.e., −5 + 1 = −4 or 1100. When we shift this right we get 1110 or −2, the correct answer. How can we do this without an if statement to test whether the number is positive or negative? If a 2’s complement number is positive, the sign bit is 0. If the 2’s complement number is negative it’s sign bit is 1. So we shift the sign bit all the way to bit position 0 (63 bits if we have a 64 bit number). If the number is negative, it results in the value ”1”; if the number if positive it results in the value ”0”. We then add this to the number that we are to divide, i.e., we add ”1” to a negative number and ”0” to a positive number. Then when we shift right, the number will be rounded in the correct way.

  • Problem set 2: it's ok to use the time.h library to convert seconds into a date formate. Check out the strftime function in particular.

Coming up...

Problem Set 2 is now due on Friday, 1 February Monday, 11 Feb at the beginning of class.

Past Assignments »

Thursday, 23 January, read the Unix tutorial Intro, tutorial 1, tutorial 2. No quiz on Monday, 28 January

Problem Set 1 is due on Monday, 28 Jan at the beginning of class.

Comp 210 in the news

Fun

  • Interested in what happened to the engineers from Soul of a New Machine? Check out this article from Wired magazine in 2001. And this page has a photo of the development group in 1980.
  • vi or emacs? See this article.
  • Describe computer science in 5 words: BI
  • Bits in real life. Ariane 5, 1996. See here