97 lines
3.1 KiB
Plaintext
97 lines
3.1 KiB
Plaintext
#include <stdlib.h>
|
||
#include <stdio.h>
|
||
#include "add.cuh"
|
||
#include "clock.cuh"
|
||
|
||
const real a = 1.23;
|
||
const real b = 2.34;
|
||
|
||
|
||
void cuda_clock()
|
||
{
|
||
const int N = 1e6;
|
||
const int M = sizeof(real) * N;
|
||
|
||
// cuda 计时。
|
||
float elapsed_time = 0;
|
||
float curr_time = 0;
|
||
cudaEvent_t start, stop;
|
||
CHECK(cudaEventCreate(&start)); // 创建cuda 事件对象。
|
||
CHECK(cudaEventCreate(&stop));
|
||
CHECK(cudaEventRecord(start)); // 记录代表开始的事件。
|
||
cudaEventQuery(start); // 强制刷新 cuda 执行流。
|
||
|
||
// --------------------------------------------------
|
||
real *h_x, *h_y, *h_z;
|
||
h_x = new real[N];
|
||
h_y = new real[N];
|
||
h_z = new real[N];
|
||
if (!h_x || !h_y || !h_z)
|
||
{
|
||
printf("host memory malloc failed!\n");
|
||
return;
|
||
}
|
||
for (int i = 0; i < N; ++i)
|
||
{
|
||
h_x[i] = a;
|
||
h_y[i] = b;
|
||
}
|
||
// --------------------------------------------------
|
||
|
||
// 主机申请及初始化内存的耗时。
|
||
CHECK(cudaEventRecord(stop));
|
||
CHECK(cudaEventSynchronize(stop)); // 强制同步，让主机等待cuda事件执行完毕。
|
||
CHECK(cudaEventElapsedTime(&curr_time, start, stop)); // 计算 start 和stop间的时间差（ms）。
|
||
printf("host memory malloc and copy: %f ms.\n", curr_time - elapsed_time);
|
||
elapsed_time = curr_time;
|
||
|
||
|
||
// --------------------------------------------------
|
||
real *d_x, *d_y, *d_z;
|
||
CHECK(cudaMalloc(&d_x, M));
|
||
CHECK(cudaMalloc(&d_y, M));
|
||
CHECK(cudaMalloc(&d_z, M));
|
||
CHECK(cudaMemcpy(d_x, h_x, M, cudaMemcpyDefault));
|
||
CHECK(cudaMemcpy(d_y, h_y, M, cudaMemcpyDefault));
|
||
// --------------------------------------------------
|
||
|
||
// 设备内存申请和拷贝耗时。
|
||
CHECK(cudaEventRecord(stop));
|
||
CHECK(cudaEventSynchronize(stop));
|
||
CHECK(cudaEventElapsedTime(&curr_time, start, stop));
|
||
printf("device memory malloc: %f ms.\n", curr_time - elapsed_time);
|
||
elapsed_time = curr_time;
|
||
|
||
// --------------------------------------------------
|
||
const int block_size = 128;
|
||
const int grid_size = N/block_size + 1;
|
||
add<<<grid_size, block_size>>>(d_x, d_y, d_z, N);
|
||
// --------------------------------------------------
|
||
|
||
// 核函数运行耗时。
|
||
CHECK(cudaEventRecord(stop));
|
||
CHECK(cudaEventSynchronize(stop));
|
||
CHECK(cudaEventElapsedTime(&curr_time, start, stop));
|
||
printf("kernel function : %f ms.\n", curr_time - elapsed_time);
|
||
elapsed_time = curr_time;
|
||
|
||
// --------------------------------------------------
|
||
CHECK(cudaGetLastError());
|
||
CHECK(cudaMemcpy(h_z, d_z, M, cudaMemcpyDefault));
|
||
check(h_z, N);
|
||
// --------------------------------------------------
|
||
|
||
// 数据拷贝耗时。
|
||
CHECK(cudaEventRecord(stop));
|
||
CHECK(cudaEventSynchronize(stop));
|
||
CHECK(cudaEventElapsedTime(&curr_time, start, stop));
|
||
printf("copy from device to host: %f ms.\n", curr_time - elapsed_time);
|
||
elapsed_time = curr_time;
|
||
|
||
if (h_x) delete[] h_x;
|
||
if (h_y) delete[] h_y;
|
||
if (h_z) delete[] h_z;
|
||
CHECK(cudaFree(d_x));
|
||
CHECK(cudaFree(d_y));
|
||
CHECK(cudaFree(d_z));
|
||
} |