THE QUESTION:

High-level languages provide a "switch" statement that looks as follows. switch (k) \{ case 0: case 1: case 2: case 3: The compiler writer knows that " k " can take contiguous integer values from 0 to 9 during execution, with any other value going to the default case. She decides to use a jump table data structure (implemented as an array indexed by the value contained in k ) to hold the start address for the code for each of the case value as shown below: Jump tadie Assuming we are using the LC-2200 instruction set architecture, which can be found in the appendix: (i) Assume the base address of the jump table is stored in the register $t0, the value of k is stored in register $t1. Help her out by writing a series of instructions that reaches the code for case k. (4 points) (ii) Instead of using a jump table, her friend suggests using a series of conditional branch instructions to compile the switch statement. Which one is faster? Explain your answer, clearly stating your assumptions. ( 6 points) High-level languages provide a "switch" statement that looks as follows. switch (k) \{ case 0: case 1: case 2: case 3: The compiler writer knows that " k " can take contiguous integer values from 0 to 9 during execution, with any other value going to the default case. She decides to use a jump table data structure (implemented as an array indexed by the value contained in k ) to hold the start address for the code for each of the case value as shown below: Jump tadie Assuming we are using the LC-2200 instruction set architecture, which can be found in the appendix: (i) Assume the base address of the jump table is stored in the register $t0, the value of k is stored in register $t1. Help her out by writing a series of instructions that reaches the code for case k. (4 points) (ii) Instead of using a jump table, her friend suggests using a series of conditional branch instructions to compile the switch statement. Which one is faster? Explain your answer, clearly stating your assumptions. ( 6 points)