Final Examination 97

Final Examination 97 - Name Computer Architecture EE 4720...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Name Computer Architecture EE 4720 Final Examination 10 May 1997, 12:3014:30 CDT Alias Problem 1 (25 pts) Problem 2 (25 pts) Problem 3 (25 pts) Problem 4 (25 pts) Exam Total (100 pts) Good Luck! Problem 1: DLXs immediate instructions use 16-bit immediates. Because in many cases larger immediates are needed, new larger-immediate instructions are to be added to DLX. Larger-immediate instructions specify an integer arithmetic operation and an immediate, but no registers. The operation is performed using the immediate as one source and the most-recent destination register as both the other source and the destination. For example, consider larger-immediate instruction addli in the code fragment below: lw r6, 0(r7) sub r1, r2, r3 sw 0(r4), r5 addli #0x1ffff ! Operation: r1 = r1 + 0x1ffff The addli instruction uses r1 because it was the most-recent destination register used. (The sw does not modify r4 or r5 , so they arent used. The lw does modify r6 , but it is executed before the sub .) The larger-immediate instructions use the J-type format: a 6-bit opcode followed by a 26-bit immediate. ( a ) Let register r10 hold the memory address of an integer. The sum of that integer and 0x981234 is to be written to r11 . Write two code fragments to perform this operation, one with and one without a larger-immediate instruction. (5 pts) 2 Problem 1 continued: ( b ) Show the modifications needed to implement larger-immediate in-structions on the pipeline illustrated below. (Read the next part before solving.) (10 pts) ( c ) Show how bypassing can be implemented for the instruction. (Hint: This is easier than regular bypassing since one source is the most-recent destination.) (10 pts) Addresses must be presented to the register file at the beginning of a cycle and the data wont be available until the end of the cycle. Be sure to label the function of each connection to registers and other devices. ( E.g. , address, write, data.) Avoid magic boxes and clouds. Explain your modifications using an annotated timing diagram showing sample code executing. A detailed logic and timing diagram are preferred over a lengthy verbal description....
View Full Document

Page1 / 9

Final Examination 97 - Name Computer Architecture EE 4720...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online