Semester : SEMESTER 1
Subject : DSP System Design
Year : 2017
Term : DECEMBER
Branch : SIGNAL PROCESSING
Scheme : 2015 Full Time
Course Code : 01 EC 6307
Page:3
instructions, A hit takes 1 clock cycle and miss penalty is TOO clock cycles)A load or
store hit takes extra clock cycle ona unified cache if there is only | cache port to satisfy
2 simultaneous requests. What is the average access time in each case? Assume write
through cache with a write buffer and ignore stalls due to write buffer. Misses/1000
instructions for instruction, data & unified cache of different size are given below. The
percentage of instruction references is about 74 9/0 .
Size Instructio Data Unified
n Cache Cache cache
16KB 3.82 409 = | 51.0
32 KB 1.36 38.4 43.3
b. Write an assembly language program to implement y(n) = x(n) * h(n)
(6)