581365-8 Computer Organization II, final exam 1.6.2004         [Suomeksi Kääntöpuolella suomeksi]

Write in each sheet of paper: your name and signature, student id, course name, and page number.

  1. [15 p] IA-64 and Itanium.
    1. [3 p] Explain concept "predicated execution". What is the advantage in it?
    2. [4 p] Explain concept "speculative loading (I.e., control speculation)". What is the advantage in it? When is it disadvantageous to use it?
    3. [4 p] Explain concept "data speculation". What is the advantage in it? When is it disadvantageous to use it?
    4. [4 p] Explain concept "software pipelining". What is the advantage in it?

      5.  
  2.  [15 p] Booth's Algorithm for integer multiplication.
    1. [6 p] What is the basic idea in Booth's algorithm?
    2. [5 p] Give as an example how Booth's algorithm works with multiplication 14*14.
    3. [2 p] When is Booth's algorithm faster than the conventional approach? Give an example.
    4. [2 p] When is Booth's algorithm slower than the conventional approach? Give an example.
       

  3. [15 p] Cache memory. Assume that (with an unrealistically small cache memory solution) the 16-bit byteaddress of a 4-byte word is 0xA0B1 (hexadecimal A0B1). The cache line size (block size) 16 bytes and cache size is 256 bytes.
    1. [4 p] Give an example with a picture and show how the referenced data is found in the cache when the mapping function is direct mapping?
    2. [4 p] Give an example with a picture and show how the referenced data is found in the cache when the mapping function is fully associative mapping?
    3. [4 p]Give an example with a picture and show how the referenced data is found in the cache when the mapping function is 2-way set associative mapping?
    4. [3 p] What are the names/lengths of each field of memory address with each mapping function?
    In your example, you may assume that the referenced data is found in the cache.
     
  4. [15 p] Microprogrammed control
    1. [2 p] What problem is solved by processor control in general?
    2. [5 p] How is microprogrammed control implemented?
    3. [3 p] What advantages/disadvantages does microprogrammed control have as compared to hardwired control?
    4. [3 p] In which three ways can one produce the next micro-program control address ("micro-program PC") and under which circumstances is each of them used?
    5. [2 p] How are interrupts implemented with microprogrammed control?