122
edits
Changes
→Assignment 2
=== Assignment 2 ===
[All test from this point on use 100 degrees as rotate input]
Original CPU Implementation:
{|
|'''Function'''|'''CPU-Only'''|'''GPU-CPU'''|'''speedup(%)'''
|-
|'''Cuda Memory Allocation'''|sn/a
| 1164 ms
|%n/a
|-
|'''Copy Image to Device memory'''|sn/a
| 6 ms
|%n/a
|-
|'''Kernel'''|sn/a
| 0 ms
|%n/a
|-
|'''Copy device image to host temp variable'''|sn/a
| 6 ms
|%n/a
|-
|'''copy temp image to original image variable'''|sn/a
| 43 ms
|%n/a
|-
|'''Total Rotation Time (no allocation, with memcpy)'''
|1717ms
| 55ms
|3021.82%
|-
|'''Total Run Time'''
|1775 ms
|1294 ms
|37.17%
|}
