=========================preview======================
(COMP180)COMP180-final03_ans.pdf
Back to COMP180 Login to download
======================================================
Computer Organization (COMP 180)
Spring Semester, 2003


Final Examination
May 31, 2003
Name: Student ID:
Email: Lab Section Number:
Instructions:
1.
This examination paper consists of 22 pages, with 10 questions and 1 appendix reference page.

2.
Please write your name, student ID, email and lab section number on this page.

3.
For each subsequent page, please write your student ID at the top of the page in the space provided.

4.
Please answer all the questions in the spaces provided on the examination paper. You may use the empty spaces on the pages for your rough work and afterwards draw a diagonal line through it to show that it is not part of your answer.

5.
Please read each question very carefully and answer clearly and to the point. Make sure that your answers are neatly written.

6.
Keep all pages stapled together.

7.
The examination period will last for 2 hours and 30 minutes.

8.
Stop writing immediately when the time is up.


Question Percentage % Scores
1 10
2 10
3 10
4 10
5 10
6 10
7 10
8 10
9 10
10 10
TOTAL 100

(a) When two twos complement numbers are added, having a carry out from the most significant bit does not necessarily imply an overflow. Ans: T
(b) Nintendo 64 is a commercial product that uses MIPS microprocessors. Ans: T
(c) We need six select lines in a 64-to-1 MUX. Ans: T
(d) An increase of the block size of a cache always decreases the miss rate. Ans: F
(e) In the MIPS architecture, a 32-bit register can represent 232 memory words. Ans: F
(f) In the MIPS addressing modes, the PC-relative addressing uses number of words to specify the branch offset. Ans: T
(g) In a 1-bit ALU, if a subtraction operation is performed, then the values of Binvertand CarryIn should be 1 and 0 respectively. Ans: F
(h) For the final version of the multiplication hardware, if two 32 bit numbers are multiplied, then it requires 32 iterations to finish the calculation (excluding the initialization step).
Ans: T
(i) In the IEEE 754 double precision representation, the value of bias is 1024. Ans: F
(j) One disadvantage of increasing the degree of associativity is that the miss rate also increases. Ans: F

The following table shows the characteristics of running instructions in a processor P.
Instruction Type Cycles
A 2
B 4
C 8
D 10

Assume that P has a clock rate of 600 MHz and the following two benchmarks B1 and B2 are used for testing the processor.
Instruction Type Frequency (%)
A 35
B 25
C 15
D 25

B1:

Instruction Type Frequency (%)
A 35
B 25
C 25
D 15

B2:

Note: Correct your answers to 2 decimal places for Parts (a), (b) and (c). Show all steps.
(a) What are the CPI and MIPS (million instructions per second) ratings of P when using the benchmarks B1 and B2? (You only need to fill in the followi