首页 > 解决方案 > 将一维数组的索引转换为访问二维数组

问题描述

void dgem(int n, double *A, double *B, double *C)
{

    for(int i = 0; i < n; ++i){
        for(int j = 0; j < n; j++){

            double cij = C[i+j*n]; /* cij = C[i][j] */

            for(int k = 0; k < n; k++){
                cij += A[i+k*n] * B[k+j*n]; /*cij = A[i][k]*b[k][j]*/
                C[i+j*n] = cij; /* C[i][j] = cij */
            }
        }
    }

}

此代码来自 Computer_Organization_and_Design_5th

这样对吗?double cij = C[i+j*n];

据我所知,应该是C[i*n + j]

int main(void){

    double A[4][4] = {1,2,3,4,
                      5,6,7,8,
                      9,10,11,12,
                      13,14,15,16};
    double * a = &A[0][0];

    int n = 4;
    printf("%f %f %f %f", *(*(A+1)+3), A[1][3], a[1*n + 3], a[1 + 3*n]); /* 
when i == 1 and j == 3 */

    return 0;

}

输出:

8.000000 8.000000 8.000000 14.000000

当我尝试使用 gcc 时,它没有任何意义......

标签: carrays

解决方案


下面的声明

double cij = C[i+j*n]; 

是正确的。为了理解这一点,让我们假设三个双变量的数组double ptr[3] = {1.5,2.5,3.5]在哪里。ptr现在你将如何访问ptr[0]等等ptr[1]

 ----------------------------------------
|    1.5   |   2.5   |   3.5   |   4.5   |
-----------------------------------------
0x100     0x108    0x116     0x124     0x132 <-- lets say starting address 
LSB                                                 of ptr is 0x100
|
ptr

为了row = 1

ptr[row] == *(ptr + row * sizeof(ptr[row]))
2.5      == *(0x100 + 1*8)
2.5      == *(0x108)
2.5      == 2.5

从上面你不能有*(ptr*8 + row)它应该是*(ptr + row*8)

同样 if ptris 2Darray like double ptr[2][3]then

ptr[row][col] == *( *(ptr + row) + col*8)

同样,在您的情况下,有效的 C[i + j*n]不是C[i*n +j]

编辑: - 你有一个2D array喜欢下面

double A[4][4] = { {1,2,3,4} , {5,6,7,8} , {9,10,11,12} , {13,14,15,16} };

double *a = &A[0][0];

现在看起来像

    A[0]          |   A[1]        |   A[2]           |   A[3]             |
  -------------------------------------------------------------------------
  | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 |  16 |
  -------------------------------------------------------------------------
0x100 0x108...................0x156.......................0x156 (assume that 0x100 is starting address of A)
  A
  a
 LSB -->  

现在,当您在a[1*n + 3])内部进行如何扩展时

a[1*n + 3]) == *(a + (1*n + 3)  /*note that a is double pointer, it increments by 8 bytes
a[1*n + 3]) == *(0x100 + (1*4 + 3) *8)
a[1*n + 3]) == *(0x100 + 56)
            == *(0x156)
            ==  8 /* it prints 8.000000 */

当你在a[1+3*n]) 内部如何扩展时

a[1+3*n]) == *(a + (1+3*n)*8)
a[1+3*n]) == *(0x100 + (1+3*4)*8)
a[1+3*n]) == *(0x100 + 96)
          == *(0x196)
          == 14 /* it prints 14.000000 */

当你这样做时*(*(A+1) +3)),它在内部扩展为

*(*(A+1) +3)) == *( *(0x100 +1*32) + 3*8) /* A+1 means increment by size of A[0] i.e 32 */ 
*(*(A+1) +3)) == *( *(0x132) + 24)
*(*(A+1) +3)) == *( 0x132 + 24 ) == *(156)
*(*(A+1) +3)) == 8 /*it prints 8.000000 */

当你这样做时A[1][3],与上述情况相同。


推荐阅读