#include <stdio.h>
__global__ void childKernel(int i)
{
int tid = blockIdx.x*blockDim.x+threadIdx.x;
printf("parent:%d,child:%d\n",i,tid);
for(int j=i;j<i+10;j++)
{
printf(",%d",j);
}
printf("\n");
}
__global__ void kernel()
{
int tid = blockIdx.x*blockDim.x+threadIdx.x;
childKernel<<<1,2>>>(tid);
}
int main()
{
kernel<<<1,1>>>();
cudaDeviceSynchronize();
return 0;
}
cuda核函数再调用核函数,多层并行
最新推荐文章于 2024-04-25 13:23:46 发布