提升CS:APP性能实验室的优化实践与参考

/********************************************************
 * 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 = {
    "09300720017",              /* Team name */

    "09300720017",     /* First member full name */
    "stu09300720017@catlab-fs",  /* 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
 */
char rotate_descr[] = "rotate: Current working version";  
void rotate(int dim, pixel *src, pixel *dst)   
{  
    int i, j;
    int temp;
    int it,jt;
    int im,jm;
    for(jt=0; jt<dim; jt+=32)  {
                                jm=jt+32;
                                     for(it=0; it<dim; it+=32) {
                                                        im=it+32;
                                        for(j=jt; j<jm; j++) {
temp=dim-1-j; 
                                              for(i=it; i<it+32; i++) {