---
title: cmprod
description: "复数矩阵的点积运算。"
url: https://www.hikunpeng.com/document/detail/zh/kunpengaccel/system-lib/ug-kvsip/kunpengaccel_kvsip_06_0088.html
sourcePath: /source/zh/kunpengaccel/system-lib/ug-kvsip/kunpengaccel_kvsip_06_0088.html
indexId: d8c99005e3d5d1bc714ba8430585405041e6cc879928dcf045ef62a7ca8976e875
---
# cmprod

复数矩阵的点积运算。

函数接口声明如下：

float型函数操作：

void (vsip_cmprod_f)(const vsip_cmview_f *a, const vsip_cmview_f *b, const vsip_cmview_f *c);

#### 参数

| 参数名 | 描述 | 取值范围 | 输入/输出 |
| --- | --- | --- | --- |
| a | 指向源复数矩阵的指针。 | 非空 | 输入 |
| b | 指向源复数矩阵的指针。 | 非空 | 输入 |
| c | 指向目标复数矩阵的指针。 | 非空 | 输出 |


#### 异常输入

空指针输入时，函数直接返回。


#### 示例

```
#include <stdio.h>
#include <stdlib.h>
#include <stdint.h>
#include "kvsip.h"
#include "vsip.h"
#include "vsip_type.h"

void CmprodExample()
{
    int32_t row = 4;
    int32_t col = 3;
    int64_t rowStride = 1;
    int64_t colStride = row * rowStride;
    uint64_t offset = 0;
    int32_t len = 2 * (col * colStride + offset);
    float *src1 = (float *)malloc(len * sizeof(float));
    float *src2 = (float *)malloc(len * sizeof(float));
    float *dst = (float *)malloc(len * sizeof(float));
    for (int32_t i = 0; i < len; ++i) {
        src1[i] = 1 + i / 100.0f;
        src2[i] = 2 + i / 100.0f;
    }

    vsip_cblock_f *block_a = vsip_cblockbind_f(src1, NULL, len, VSIP_MEM_NONE);
    vsip_cblock_f *block_b = vsip_cblockbind_f(src2, NULL, len, VSIP_MEM_NONE);
    vsip_cblock_f *block_c = vsip_cblockbind_f(dst, NULL, len, VSIP_MEM_NONE);

    vsip_cmview_f *a = vsip_cmbind_f(block_a, offset, colStride, col, rowStride, row);
    vsip_cblockadmit_f(block_a, VSIP_TRUE);
    vsip_cmview_f *b = vsip_cmbind_f(block_b, offset, colStride, col, rowStride, row);
    vsip_cblockadmit_f(block_b, VSIP_TRUE);
    vsip_cmview_f *c = vsip_cmbind_f(block_c, offset, colStride, col, rowStride, row);
    vsip_cblockadmit_f(block_c, VSIP_TRUE);

    vsip_cmprod_f(a, b, c);

    vsip_cmalldestroy_f(a);
    vsip_cmalldestroy_f(b);
    vsip_cmalldestroy_f(c);

    printf("src1:\n");
    for (int32_t i = 0; i < col; ++i) {
        for (int32_t j = 0; j < row; ++j) {
            printf("%.5f ", src1[i * row * 2 + j * 2]);
            printf("%.5f    ", src1[i * row * 2 + j * 2 + 1]);
        }
        printf("\n");
    }
    printf("\nsrc2:\n");
    for (int32_t i = 0; i < col; ++i) {
        for (int32_t j = 0; j < row; ++j) {
            printf("%.5f ", src2[i * row * 2 + j * 2]);
            printf("%.5f    ", src2[i * row * 2 + j * 2 + 1]);
        }
        printf("\n");
    }
    printf("\ndst:\n");
    for (int32_t i = 0; i < col; ++i) {
        for (int32_t j = 0; j < row; ++j) {
            printf("%.5f ", dst[i * row * 2 + j * 2]);
            printf("%.5f    ", dst[i * row * 2 + j * 2 + 1]);
        }
        printf("\n");
    }
    free(src1);
    free(src2);
    free(dst);
}

int main(void) {
    CmprodExample();
    return 0;
}
```

运行结果：

```
src1:
1.00000 1.01000    1.02000 1.03000    1.04000 1.05000    1.06000 1.07000
1.08000 1.09000    1.10000 1.11000    1.12000 1.13000    1.14000 1.15000
1.16000 1.17000    1.18000 1.19000    1.20000 1.21000    1.22000 1.23000

src2:
2.00000 2.01000    2.02000 2.03000    2.04000 2.05000    2.06000 2.07000
2.08000 2.09000    2.10000 2.11000    2.12000 2.13000    2.14000 2.15000
2.16000 2.17000    2.18000 2.19000    2.20000 2.21000    2.22000 2.23000

dst:
-0.09330 12.82900    -0.09390 12.95200    -0.09450 13.07500    -0.09510 13.19800
-0.09570 13.82980    -0.09630 13.96240    -0.09690 14.09500    -0.09750 14.22760
-0.09810 14.83060    -0.09870 14.97280    -0.09930 15.11500    -0.09990 15.25720
```