/********************************************************
* Kernels to be optimized for the CS:APP Performance Lab
********************************************************/
#include <stdio.h>
#include <stdlib.h>
#include "defs.h"
/*
* Please fill in the following team struct
*/
team_t team = {
"aaa", /* Team name */
"Habs", /* First member full name */
"abs@nowhdasdaedu", /* First member email address */
"", /* Second member full name (leave blank if none) */
"" /* Second member email addr (leave blank if none) */
};
/***************
* ROTATE KERNEL
***************/
/******************************************************
* Your different versions of the rotate kernel go here
******************************************************/
/*
* naive_rotate - The naive baseline version of rotate
*/
char naive_rotate_descr[] = "naive_rotate: Naive baseline implementation";
void naive_rotate(int dim, pixel *src, pixel *dst)
{
int i, j;
for (i = 0; i < dim; i++)
for (j = 0; j < dim; j++)
dst[RIDX(dim-1-j, i, dim)] = src[RIDX(i, j, dim)];
}
/*
* rotate - Your current working version of rotate
* IMPORTANT: This is the version you will be graded on
* rotate: 16-way unroll, write spatial locality(from bottom)
* didn't use transpose + switch roll
*/
char rotate_descr[] = "rotate: Current working version";
void rotate(int dim, pixel *src, pixel *dst)
{
int i,j;
int offset = (dim<<4);
int dst_offset = dim*dim-dim;
dst += dst_offset;//from bottom
for(i=0; i<dim; i += 16){
for(j=0; j< dim ; j++ ){
*dst++ = *src;
src += dim;
