猿代码 — 科研/AI模型/高性能计算
登录 注册
快捷登录
猿代码-超算人才智造局 高性能计算|并行计算|人工智能 › 首页 ›科技资讯 › 查看内容
0

prefix sum, bank conflict 访问共享内存代码分析，确实很奇妙

摘要: #include stdio.h#include iostream#include stdlib.h#include ctime#define MAX_BLOCK_SZ 1024#define NUM_BANKS 32#define LOG_NUM_BANKS 5#define OFFSET(n) \ ((n) NUM_BANKS + (n) (2 * LOG_NUM_BANKS))int ...

#include <stdio.h>
#include <iostream>
#include <stdlib.h>
#include <ctime>



#define MAX_BLOCK_SZ 1024
#define NUM_BANKS 32
#define LOG_NUM_BANKS 5

#define OFFSET(n) \
    ((n) >> NUM_BANKS + (n) >> (2 * LOG_NUM_BANKS))



int main()
{
    int bs=512;
    int off[32];

    for(int thid=0; thid< 32; thid++){
        int offset = 1;
        for (int d = bs >> 1; d > 0; d >>= 1){
            if (thid < d){
                int ai = offset * ((thid << 1) + 1) - 1;
                int bi = offset * ((thid << 1) + 2) - 1;
                ai += OFFSET(ai);
                bi += OFFSET(bi);
                //sm[bi] += sm[ai];
                std::cout<< "down, thid, offset, d, ai, bi ="<< thid <<", " << offset <<", " << d <<", " << ai << ", "<< bi <<std::endl;
            }
            offset <<= 1;
            off[thid]=offset;
        }
        if (thid == 0){
            //bsum[blockIdx.x] = sm[bs - 1 + OFFSET(bs - 1)];
            //sm[bs - 1 + OFFSET(bs - 1)] = 0;
            std::cout<< "thid, bsum index ="<< thid <<", " << bs - 1 + OFFSET(bs - 1) <<std::endl;
        }
    }

    for(int i=0; i<32; i++){
        std::cout<< off[i]<<", " ;
    }
    std::cout<< std::endl;

   
    for(int thid=0; thid< 32; thid++){
            int offset=off[thid];
        for (int d = 1; d < bs; d <<= 1){
        //int offset = bs/2;
            offset >>= 1;
            if (thid < d){
                int ai = offset * ((thid << 1) + 1) - 1;
                int bi = offset * ((thid << 1) + 2) - 1;
                ai += OFFSET(ai);
                bi += OFFSET(bi);
                //int temp = sm[ai];
                //sm[ai] = sm[bi];
                //sm[bi] += temp;
               
                std::cout<< "up, thid, offset, d, ai, bi ="<< thid <<", " << offset <<", " << d <<", " << ai << ", "<< bi <<std::endl;
            }
        }
    }

    return 0;


}

down, thid, offset, d, ai, bi =0, 1, 256, 0, 1 down, thid, offset, d, ai, bi =0, 2, 128, 1, 3 down, thid, offset, d, ai, bi =0, 4, 64, 3, 7 down, thid, offset, d, ai, bi =0, 8, 32, 7, 15 down, thid, offset, d, ai, bi =0, 16, 16, 15, 31 down, thid, offset, d, ai, bi =0, 32, 8, 31, 63 down, thid, offset, d, ai, bi =0, 64, 4, 63, 127 down, thid, offset, d, ai, bi =0, 128, 2, 127, 255 down, thid, offset, d, ai, bi =0, 256, 1, 255, 511 thid, bsum index =0, 511 down, thid, offset, d, ai, bi =1, 1, 256, 2, 3 down, thid, offset, d, ai, bi =1, 2, 128, 5, 7 down, thid, offset, d, ai, bi =1, 4, 64, 11, 15 down, thid, offset, d, ai, bi =1, 8, 32, 23, 31 down, thid, offset, d, ai, bi =1, 16, 16, 47, 63 down, thid, offset, d, ai, bi =1, 32, 8, 95, 127 down, thid, offset, d, ai, bi =1, 64, 4, 191, 255 down, thid, offset, d, ai, bi =1, 128, 2, 383, 511 down, thid, offset, d, ai, bi =2, 1, 256, 4, 5 down, thid, offset, d, ai, bi =2, 2, 128, 9, 11 down, thid, offset, d, ai, bi =2, 4, 64, 19, 23 down, thid, offset, d, ai, bi =2, 8, 32, 39, 47 down, thid, offset, d, ai, bi =2, 16, 16, 79, 95 down, thid, offset, d, ai, bi =2, 32, 8, 159, 191 down, thid, offset, d, ai, bi =2, 64, 4, 319, 383 down, thid, offset, d, ai, bi =3, 1, 256, 6, 7 down, thid, offset, d, ai, bi =3, 2, 128, 13, 15 down, thid, offset, d, ai, bi =3, 4, 64, 27, 31 down, thid, offset, d, ai, bi =3, 8, 32, 55, 63 down, thid, offset, d, ai, bi =3, 16, 16, 111, 127 down, thid, offset, d, ai, bi =3, 32, 8, 223, 255 down, thid, offset, d, ai, bi =3, 64, 4, 447, 511 down, thid, offset, d, ai, bi =4, 1, 256, 8, 9 down, thid, offset, d, ai, bi =4, 2, 128, 17, 19 down, thid, offset, d, ai, bi =4, 4, 64, 35, 39 down, thid, offset, d, ai, bi =4, 8, 32, 71, 79 down, thid, offset, d, ai, bi =4, 16, 16, 143, 159 down, thid, offset, d, ai, bi =4, 32, 8, 287, 319 down, thid, offset, d, ai, bi =5, 1, 256, 10, 11 down, thid, offset, d, ai, bi =5, 2, 128, 21, 23 down, thid, offset, d, ai, bi =5, 4, 64, 43, 47 down, thid, offset, d, ai, bi =5, 8, 32, 87, 95 down, thid, offset, d, ai, bi =5, 16, 16, 175, 191 down, thid, offset, d, ai, bi =5, 32, 8, 351, 383 down, thid, offset, d, ai, bi =6, 1, 256, 12, 13 down, thid, offset, d, ai, bi =6, 2, 128, 25, 27 down, thid, offset, d, ai, bi =6, 4, 64, 51, 55 down, thid, offset, d, ai, bi =6, 8, 32, 103, 111 down, thid, offset, d, ai, bi =6, 16, 16, 207, 223 down, thid, offset, d, ai, bi =6, 32, 8, 415, 447 down, thid, offset, d, ai, bi =7, 1, 256, 14, 15 down, thid, offset, d, ai, bi =7, 2, 128, 29, 31 down, thid, offset, d, ai, bi =7, 4, 64, 59, 63 down, thid, offset, d, ai, bi =7, 8, 32, 119, 127 down, thid, offset, d, ai, bi =7, 16, 16, 239, 255 down, thid, offset, d, ai, bi =7, 32, 8, 479, 511 down, thid, offset, d, ai, bi =8, 1, 256, 16, 17 down, thid, offset, d, ai, bi =8, 2, 128, 33, 35 down, thid, offset, d, ai, bi =8, 4, 64, 67, 71 down, thid, offset, d, ai, bi =8, 8, 32, 135, 143 down, thid, offset, d, ai, bi =8, 16, 16, 271, 287 down, thid, offset, d, ai, bi =9, 1, 256, 18, 19 down, thid, offset, d, ai, bi =9, 2, 128, 37, 39 down, thid, offset, d, ai, bi =9, 4, 64, 75, 79 down, thid, offset, d, ai, bi =9, 8, 32, 151, 159 down, thid, offset, d, ai, bi =9, 16, 16, 303, 319 down, thid, offset, d, ai, bi =10, 1, 256, 20, 21 down, thid, offset, d, ai, bi =10, 2, 128, 41, 43 down, thid, offset, d, ai, bi =10, 4, 64, 83, 87 down, thid, offset, d, ai, bi =10, 8, 32, 167, 175 down, thid, offset, d, ai, bi =10, 16, 16, 335, 351 down, thid, offset, d, ai, bi =11, 1, 256, 22, 23 down, thid, offset, d, ai, bi =11, 2, 128, 45, 47 down, thid, offset, d, ai, bi =11, 4, 64, 91, 95 down, thid, offset, d, ai, bi =11, 8, 32, 183, 191 down, thid, offset, d, ai, bi =11, 16, 16, 367, 383 down, thid, offset, d, ai, bi =12, 1, 256, 24, 25 down, thid, offset, d, ai, bi =12, 2, 128, 49, 51 down, thid, offset, d, ai, bi =12, 4, 64, 99, 103 down, thid, offset, d, ai, bi =12, 8, 32, 199, 207 down, thid, offset, d, ai, bi =12, 16, 16, 399, 415 down, thid, offset, d, ai, bi =13, 1, 256, 26, 27 down, thid, offset, d, ai, bi =13, 2, 128, 53, 55 down, thid, offset, d, ai, bi =13, 4, 64, 107, 111 down, thid, offset, d, ai, bi =13, 8, 32, 215, 223 down, thid, offset, d, ai, bi =13, 16, 16, 431, 447 down, thid, offset, d, ai, bi =14, 1, 256, 28, 29 down, thid, offset, d, ai, bi =14, 2, 128, 57, 59 down, thid, offset, d, ai, bi =14, 4, 64, 115, 119 down, thid, offset, d, ai, bi =14, 8, 32, 231, 239 down, thid, offset, d, ai, bi =14, 16, 16, 463, 479 down, thid, offset, d, ai, bi =15, 1, 256, 30, 31 down, thid, offset, d, ai, bi =15, 2, 128, 61, 63 down, thid, offset, d, ai, bi =15, 4, 64, 123, 127 down, thid, offset, d, ai, bi =15, 8, 32, 247, 255 down, thid, offset, d, ai, bi =15, 16, 16, 495, 511 down, thid, offset, d, ai, bi =16, 1, 256, 32, 33 down, thid, offset, d, ai, bi =16, 2, 128, 65, 67 down, thid, offset, d, ai, bi =16, 4, 64, 131, 135 down, thid, offset, d, ai, bi =16, 8, 32, 263, 271 down, thid, offset, d, ai, bi =17, 1, 256, 34, 35 down, thid, offset, d, ai, bi =17, 2, 128, 69, 71 down, thid, offset, d, ai, bi =17, 4, 64, 139, 143 down, thid, offset, d, ai, bi =17, 8, 32, 279, 287 down, thid, offset, d, ai, bi =18, 1, 256, 36, 37 down, thid, offset, d, ai, bi =18, 2, 128, 73, 75 down, thid, offset, d, ai, bi =18, 4, 64, 147, 151 down, thid, offset, d, ai, bi =18, 8, 32, 295, 303 down, thid, offset, d, ai, bi =19, 1, 256, 38, 39 down, thid, offset, d, ai, bi =19, 2, 128, 77, 79 down, thid, offset, d, ai, bi =19, 4, 64, 155, 159 down, thid, offset, d, ai, bi =19, 8, 32, 311, 319 down, thid, offset, d, ai, bi =20, 1, 256, 40, 41 down, thid, offset, d, ai, bi =20, 2, 128, 81, 83 down, thid, offset, d, ai, bi =20, 4, 64, 163, 167 down, thid, offset, d, ai, bi =20, 8, 32, 327, 335 down, thid, offset, d, ai, bi =21, 1, 256, 42, 43 down, thid, offset, d, ai, bi =21, 2, 128, 85, 87 down, thid, offset, d, ai, bi =21, 4, 64, 171, 175 down, thid, offset, d, ai, bi =21, 8, 32, 343, 351 down, thid, offset, d, ai, bi =22, 1, 256, 44, 45 down, thid, offset, d, ai, bi =22, 2, 128, 89, 91 down, thid, offset, d, ai, bi =22, 4, 64, 179, 183 down, thid, offset, d, ai, bi =22, 8, 32, 359, 367 down, thid, offset, d, ai, bi =23, 1, 256, 46, 47 down, thid, offset, d, ai, bi =23, 2, 128, 93, 95 down, thid, offset, d, ai, bi =23, 4, 64, 187, 191 down, thid, offset, d, ai, bi =23, 8, 32, 375, 383 down, thid, offset, d, ai, bi =24, 1, 256, 48, 49 down, thid, offset, d, ai, bi =24, 2, 128, 97, 99 down, thid, offset, d, ai, bi =24, 4, 64, 195, 199 down, thid, offset, d, ai, bi =24, 8, 32, 391, 399 down, thid, offset, d, ai, bi =25, 1, 256, 50, 51 down, thid, offset, d, ai, bi =25, 2, 128, 101, 103 down, thid, offset, d, ai, bi =25, 4, 64, 203, 207 down, thid, offset, d, ai, bi =25, 8, 32, 407, 415 down, thid, offset, d, ai, bi =26, 1, 256, 52, 53 down, thid, offset, d, ai, bi =26, 2, 128, 105, 107 down, thid, offset, d, ai, bi =26, 4, 64, 211, 215 down, thid, offset, d, ai, bi =26, 8, 32, 423, 431 down, thid, offset, d, ai, bi =27, 1, 256, 54, 55 down, thid, offset, d, ai, bi =27, 2, 128, 109, 111 down, thid, offset, d, ai, bi =27, 4, 64, 219, 223 down, thid, offset, d, ai, bi =27, 8, 32, 439, 447 down, thid, offset, d, ai, bi =28, 1, 256, 56, 57 down, thid, offset, d, ai, bi =28, 2, 128, 113, 115 down, thid, offset, d, ai, bi =28, 4, 64, 227, 231 down, thid, offset, d, ai, bi =28, 8, 32, 455, 463 down, thid, offset, d, ai, bi =29, 1, 256, 58, 59 down, thid, offset, d, ai, bi =29, 2, 128, 117, 119 down, thid, offset, d, ai, bi =29, 4, 64, 235, 239 down, thid, offset, d, ai, bi =29, 8, 32, 471, 479 down, thid, offset, d, ai, bi =30, 1, 256, 60, 61 down, thid, offset, d, ai, bi =30, 2, 128, 121, 123 down, thid, offset, d, ai, bi =30, 4, 64, 243, 247 down, thid, offset, d, ai, bi =30, 8, 32, 487, 495 down, thid, offset, d, ai, bi =31, 1, 256, 62, 63 down, thid, offset, d, ai, bi =31, 2, 128, 125, 127 down, thid, offset, d, ai, bi =31, 4, 64, 251, 255 down, thid, offset, d, ai, bi =31, 8, 32, 503, 511 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, 512, up, thid, offset, d, ai, bi =0, 256, 1, 255, 511 up, thid, offset, d, ai, bi =0, 128, 2, 127, 255 up, thid, offset, d, ai, bi =0, 64, 4, 63, 127 up, thid, offset, d, ai, bi =0, 32, 8, 31, 63 up, thid, offset, d, ai, bi =0, 16, 16, 15, 31 up, thid, offset, d, ai, bi =0, 8, 32, 7, 15 up, thid, offset, d, ai, bi =0, 4, 64, 3, 7 up, thid, offset, d, ai, bi =0, 2, 128, 1, 3 up, thid, offset, d, ai, bi =0, 1, 256, 0, 1 up, thid, offset, d, ai, bi =1, 128, 2, 383, 511 up, thid, offset, d, ai, bi =1, 64, 4, 191, 255 up, thid, offset, d, ai, bi =1, 32, 8, 95, 127 up, thid, offset, d, ai, bi =1, 16, 16, 47, 63 up, thid, offset, d, ai, bi =1, 8, 32, 23, 31 up, thid, offset, d, ai, bi =1, 4, 64, 11, 15 up, thid, offset, d, ai, bi =1, 2, 128, 5, 7 up, thid, offset, d, ai, bi =1, 1, 256, 2, 3 up, thid, offset, d, ai, bi =2, 64, 4, 319, 383 up, thid, offset, d, ai, bi =2, 32, 8, 159, 191 up, thid, offset, d, ai, bi =2, 16, 16, 79, 95 up, thid, offset, d, ai, bi =2, 8, 32, 39, 47 up, thid, offset, d, ai, bi =2, 4, 64, 19, 23 up, thid, offset, d, ai, bi =2, 2, 128, 9, 11 up, thid, offset, d, ai, bi =2, 1, 256, 4, 5 up, thid, offset, d, ai, bi =3, 64, 4, 447, 511 up, thid, offset, d, ai, bi =3, 32, 8, 223, 255 up, thid, offset, d, ai, bi =3, 16, 16, 111, 127 up, thid, offset, d, ai, bi =3, 8, 32, 55, 63 up, thid, offset, d, ai, bi =3, 4, 64, 27, 31 up, thid, offset, d, ai, bi =3, 2, 128, 13, 15 up, thid, offset, d, ai, bi =3, 1, 256, 6, 7 up, thid, offset, d, ai, bi =4, 32, 8, 287, 319 up, thid, offset, d, ai, bi =4, 16, 16, 143, 159 up, thid, offset, d, ai, bi =4, 8, 32, 71, 79 up, thid, offset, d, ai, bi =4, 4, 64, 35, 39 up, thid, offset, d, ai, bi =4, 2, 128, 17, 19 up, thid, offset, d, ai, bi =4, 1, 256, 8, 9 up, thid, offset, d, ai, bi =5, 32, 8, 351, 383 up, thid, offset, d, ai, bi =5, 16, 16, 175, 191 up, thid, offset, d, ai, bi =5, 8, 32, 87, 95 up, thid, offset, d, ai, bi =5, 4, 64, 43, 47 up, thid, offset, d, ai, bi =5, 2, 128, 21, 23 up, thid, offset, d, ai, bi =5, 1, 256, 10, 11 up, thid, offset, d, ai, bi =6, 32, 8, 415, 447 up, thid, offset, d, ai, bi =6, 16, 16, 207, 223 up, thid, offset, d, ai, bi =6, 8, 32, 103, 111 up, thid, offset, d, ai, bi =6, 4, 64, 51, 55 up, thid, offset, d, ai, bi =6, 2, 128, 25, 27 up, thid, offset, d, ai, bi =6, 1, 256, 12, 13 up, thid, offset, d, ai, bi =7, 32, 8, 479, 511 up, thid, offset, d, ai, bi =7, 16, 16, 239, 255 up, thid, offset, d, ai, bi =7, 8, 32, 119, 127 up, thid, offset, d, ai, bi =7, 4, 64, 59, 63 up, thid, offset, d, ai, bi =7, 2, 128, 29, 31 up, thid, offset, d, ai, bi =7, 1, 256, 14, 15 up, thid, offset, d, ai, bi =8, 16, 16, 271, 287 up, thid, offset, d, ai, bi =8, 8, 32, 135, 143 up, thid, offset, d, ai, bi =8, 4, 64, 67, 71 up, thid, offset, d, ai, bi =8, 2, 128, 33, 35 up, thid, offset, d, ai, bi =8, 1, 256, 16, 17 up, thid, offset, d, ai, bi =9, 16, 16, 303, 319 up, thid, offset, d, ai, bi =9, 8, 32, 151, 159 up, thid, offset, d, ai, bi =9, 4, 64, 75, 79 up, thid, offset, d, ai, bi =9, 2, 128, 37, 39 up, thid, offset, d, ai, bi =9, 1, 256, 18, 19 up, thid, offset, d, ai, bi =10, 16, 16, 335, 351 up, thid, offset, d, ai, bi =10, 8, 32, 167, 175 up, thid, offset, d, ai, bi =10, 4, 64, 83, 87 up, thid, offset, d, ai, bi =10, 2, 128, 41, 43 up, thid, offset, d, ai, bi =10, 1, 256, 20, 21 up, thid, offset, d, ai, bi =11, 16, 16, 367, 383 up, thid, offset, d, ai, bi =11, 8, 32, 183, 191 up, thid, offset, d, ai, bi =11, 4, 64, 91, 95 up, thid, offset, d, ai, bi =11, 2, 128, 45, 47 up, thid, offset, d, ai, bi =11, 1, 256, 22, 23 up, thid, offset, d, ai, bi =12, 16, 16, 399, 415 up, thid, offset, d, ai, bi =12, 8, 32, 199, 207 up, thid, offset, d, ai, bi =12, 4, 64, 99, 103 up, thid, offset, d, ai, bi =12, 2, 128, 49, 51 up, thid, offset, d, ai, bi =12, 1, 256, 24, 25 up, thid, offset, d, ai, bi =13, 16, 16, 431, 447 up, thid, offset, d, ai, bi =13, 8, 32, 215, 223 up, thid, offset, d, ai, bi =13, 4, 64, 107, 111 up, thid, offset, d, ai, bi =13, 2, 128, 53, 55 up, thid, offset, d, ai, bi =13, 1, 256, 26, 27 up, thid, offset, d, ai, bi =14, 16, 16, 463, 479 up, thid, offset, d, ai, bi =14, 8, 32, 231, 239 up, thid, offset, d, ai, bi =14, 4, 64, 115, 119 up, thid, offset, d, ai, bi =14, 2, 128, 57, 59 up, thid, offset, d, ai, bi =14, 1, 256, 28, 29 up, thid, offset, d, ai, bi =15, 16, 16, 495, 511 up, thid, offset, d, ai, bi =15, 8, 32, 247, 255 up, thid, offset, d, ai, bi =15, 4, 64, 123, 127 up, thid, offset, d, ai, bi =15, 2, 128, 61, 63 up, thid, offset, d, ai, bi =15, 1, 256, 30, 31 up, thid, offset, d, ai, bi =16, 8, 32, 263, 271 up, thid, offset, d, ai, bi =16, 4, 64, 131, 135 up, thid, offset, d, ai, bi =16, 2, 128, 65, 67 up, thid, offset, d, ai, bi =16, 1, 256, 32, 33 up, thid, offset, d, ai, bi =17, 8, 32, 279, 287 up, thid, offset, d, ai, bi =17, 4, 64, 139, 143 up, thid, offset, d, ai, bi =17, 2, 128, 69, 71 up, thid, offset, d, ai, bi =17, 1, 256, 34, 35 up, thid, offset, d, ai, bi =18, 8, 32, 295, 303 up, thid, offset, d, ai, bi =18, 4, 64, 147, 151 up, thid, offset, d, ai, bi =18, 2, 128, 73, 75 up, thid, offset, d, ai, bi =18, 1, 256, 36, 37 up, thid, offset, d, ai, bi =19, 8, 32, 311, 319 up, thid, offset, d, ai, bi =19, 4, 64, 155, 159 up, thid, offset, d, ai, bi =19, 2, 128, 77, 79 up, thid, offset, d, ai, bi =19, 1, 256, 38, 39 up, thid, offset, d, ai, bi =20, 8, 32, 327, 335 up, thid, offset, d, ai, bi =20, 4, 64, 163, 167 up, thid, offset, d, ai, bi =20, 2, 128, 81, 83 up, thid, offset, d, ai, bi =20, 1, 256, 40, 41 up, thid, offset, d, ai, bi =21, 8, 32, 343, 351 up, thid, offset, d, ai, bi =21, 4, 64, 171, 175 up, thid, offset, d, ai, bi =21, 2, 128, 85, 87 up, thid, offset, d, ai, bi =21, 1, 256, 42, 43 up, thid, offset, d, ai, bi =22, 8, 32, 359, 367 up, thid, offset, d, ai, bi =22, 4, 64, 179, 183 up, thid, offset, d, ai, bi =22, 2, 128, 89, 91 up, thid, offset, d, ai, bi =22, 1, 256, 44, 45 up, thid, offset, d, ai, bi =23, 8, 32, 375, 383 up, thid, offset, d, ai, bi =23, 4, 64, 187, 191 up, thid, offset, d, ai, bi =23, 2, 128, 93, 95 up, thid, offset, d, ai, bi =23, 1, 256, 46, 47 up, thid, offset, d, ai, bi =24, 8, 32, 391, 399 up, thid, offset, d, ai, bi =24, 4, 64, 195, 199 up, thid, offset, d, ai, bi =24, 2, 128, 97, 99 up, thid, offset, d, ai, bi =24, 1, 256, 48, 49 up, thid, offset, d, ai, bi =25, 8, 32, 407, 415 up, thid, offset, d, ai, bi =25, 4, 64, 203, 207 up, thid, offset, d, ai, bi =25, 2, 128, 101, 103 up, thid, offset, d, ai, bi =25, 1, 256, 50, 51 up, thid, offset, d, ai, bi =26, 8, 32, 423, 431 up, thid, offset, d, ai, bi =26, 4, 64, 211, 215 up, thid, offset, d, ai, bi =26, 2, 128, 105, 107 up, thid, offset, d, ai, bi =26, 1, 256, 52, 53 up, thid, offset, d, ai, bi =27, 8, 32, 439, 447 up, thid, offset, d, ai, bi =27, 4, 64, 219, 223 up, thid, offset, d, ai, bi =27, 2, 128, 109, 111 up, thid, offset, d, ai, bi =27, 1, 256, 54, 55 up, thid, offset, d, ai, bi =28, 8, 32, 455, 463 up, thid, offset, d, ai, bi =28, 4, 64, 227, 231 up, thid, offset, d, ai, bi =28, 2, 128, 113, 115 up, thid, offset, d, ai, bi =28, 1, 256, 56, 57 up, thid, offset, d, ai, bi =29, 8, 32, 471, 479 up, thid, offset, d, ai, bi =29, 4, 64, 235, 239 up, thid, offset, d, ai, bi =29, 2, 128, 117, 119 up, thid, offset, d, ai, bi =29, 1, 256, 58, 59 up, thid, offset, d, ai, bi =30, 8, 32, 487, 495 up, thid, offset, d, ai, bi =30, 4, 64, 243, 247 up, thid, offset, d, ai, bi =30, 2, 128, 121, 123 up, thid, offset, d, ai, bi =30, 1, 256, 60, 61 up, thid, offset, d, ai, bi =31, 8, 32, 503, 511 up, thid, offset, d, ai, bi =31, 4, 64, 251, 255 up, thid, offset, d, ai, bi =31, 2, 128, 125, 127 up, thid, offset, d, ai, bi =31, 1, 256, 62, 63

可见确实不会发生bank冲突

收藏 分享 邀请
上一篇：CUDA计时，clock不准下一篇：【猿代码】HPC在未来技术趋势中的地位如何？

说点什么...

已有0条评论

最新评论...

神经蛙
本文作者
2023-12-21 18:53
  • 0
    粉丝
  • 273
    阅读
  • 0
    回复
作者其他文章
相关分类
资讯幻灯片
热门评论
热门专题
排行榜
  • 日
  • 周
  • 月

微信公众号

Copyright   ©2015-2023   猿代码-超算人才智造局 高性能计算|并行计算|人工智能      ( 京ICP备2021026424号-2 )