109年 - 109 國立中山大學_碩士班招生考試_電機系(己組)：計算機結構#108262

科目：中山◆電機◆計算機結構 | 年份：109年 | 選擇題數：0 | 申論題數：16

試卷資訊

所屬科目：中山◆電機◆計算機結構

選擇題 (0)

申論題 (16)

1.[20%] (1) (8%) Add one-sentence comments to each of the following MIPS code to describe what it computes. (2) (12%) Write the C code that corresponds to the following MIPS code. Assume that registers Sal and $a2 stores the base address of arrays A and B, and the register St1 stores the value of a variable i.

(a) (4%) Amdahl's law

(b) (4%) Data hazard

(c) (4%) Exceptions

(d) (4%) Flush

(e) (4%) Spatial locality

(a) (4%) Given a single precision IEEE 754 bit stream S: 1011 1110 1110 0000 0000 0000 0000 0000. What does this floating number mean? Write your final answer with decimal expression.

(b) (8%) Draw the flow chart of floating-point addition. Also explain how the floating-point addition is performed by using this flow chart.

(c) (8%) Design a hardware unit that can perform floating-point addition. Please draw the block diagram of your designed hardware unit and explain how this hardware works.

(a) (5%) What are the contained pipeline stages and their tasks to be executed?

(b) (9%) Draw the data path of the five-pipeline-stage MIPS processor

(@) (6%) For the lw instruction, which control signals are needed and how are these control signals transferred with pipeline operations for the data path you drew in (b)?

(a) (4%) What are direct-mapped cache, set-associative cache and fully associative cache?

(b) (6%) Draw the architecture of a four-way set-associative cache that contains totally 16 blocks.

(c)(6%) Draw the flow chart of the cache architecture you drew in (b) and explain how the cache works with this flow chart.

(d) (4%)What is the "write-through" scheme? What is the possible problem with this scheme and how can we solve the problem?