ics lab8 perflab性能优化实验

/********************************************************
 * Kernels to be optimized for the CS:APP Performance Lab
 ********************************************************/

#include 
#include 
#include “defs.h“

/* 
 * Please fill in the following team struct 
 */
team_t team = {
    “5120379091“              /* Student ID */

    “Gao Ce“           /* Your Name */
    “gaoce270863799@se.sjtu.edu.cn“  /* 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
 ******************************************************/
/*
 *Add the description of your Rotate implementation here!!!
 *1. Brief Intro of method
 *2. CPE Achieved
 *3. other words
 */

/* 
 * 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;
    dst += （dim-1）*dim;
    for （i = 0; i < dim; i+=32）{ 
            for （j = 0; j < dim; j++）{ 
                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;
                src+=dim;
                dst+=1;

                *dst=*src;