Final Practice Problems

You’ll be given the following handout for the final exam.

Ch 4: Processor Architecture

• Given a Y86-64 program, assemble the program to Y86-64 object code
• Given Y86-64 object code, disassemble the object code to the Y86-64 assembly program
• Write simple Y86-64 functions
• Write simple HCL expressions
• Be able to describe what happens in each stage for a given Y86-64 instruction
• Understand how pipelining speeds up processing and the limitations of pipelining
• Understand all five stages of a pipelined processor
• Be able to read a pipeline timing diagram
• Data, control, and load/use pipeline hazards
• Avoiding data hazards by forwarding
• Exception handling

Practice Problems:

• 4.1 (p360)
• 4.2 (p360)
• 4.9 (p374)
• 4.12 (p380)
• 4.13 (p387)
• 4.14 (p390)
• 4.18 (p394)
• 4.19 (p406)
• 4.28 (p417)
• 4.31 (p449)
• 4.44 (p468) Note: The solution in the book is wrong. See http://csapp.cs.cmu.edu/3e/errata.html for details.

Ch 5: Optimizing Program Performance

1. Understand the following terms:
   • Memory Aliasing
   • Code Motion
2. Be able to loop unroll a function (in either C or Assembly)

Ch 6: The memory Hierarchy

• The differences between SRAM and DRAM
• The principle of locality and what is the difference between temporal and spatial locality
• The role of caches in the memory hierarchy
• The three types of cache misses: cold (compulsory), conflict, and capacity misses
• The LRU line replacement algorithm
• For a given cache, perform a cache lookup
• Cache terminology: Lines, Sets, Blocks, Tags, and the Valid bit
• The Differences between a fully associative, n-way set associative, and direct-mapped cache
• The difference between write-back and write-through caches

Practice Problems:

• 6.7 (p608)
• 6.8 (p609)
• 6.9 (p616)
• 6.12 (p628)
• 6.13 (p628)
• 6.14 (p629)
• 6.15 (p629)
• 6.16 (p630)