maowenge的专栏

///****测试在kernel里面交换全局内存的值 #include #include #include #include using namespace std; const int N=300; __global__ void show(int* a) { for(int i=blockIdx.x*blockDim.x+threadIdx.x;i i+=blo

#include #include #include #include using namespace std; const int N=1234567; const int sample=100; const int threads=256; __global__ void judge(int *da,int *data,int sam) { cons

// //#include //#include //#include //#include //using namespace std; // //const int N=1001569; //const int threadPerBlock=256; // // //__global__ void aplusb(int *a,int *b,float *c) //{

#include #include #include #include using namespace std; __global__ void nullKernel() {} int main(void) { const int sum=1000000; cout cudaEvent_t start,stop; float time; c

#include #include #include using namespace std; template __device__ void saxpy_unrolled( float *out, const float *px, const float *py, size_t N, float alpha ) { float x[n],y[n];

#include #include #include #include using namespace std; const int N=500; __global__ void test(int* a,int n) { const int tid=blockIdx.x*blockDim.x+threadIdx.x; int i=blockDim.x*gridDim.x*n+ti

在CUDA中统计运算时间，大致有三种方法：   使用cutil.h中的函数 unsigned int timer=0;                  //创建计时器 cutCreateTimer(&timer);              //开始计时 cutStartTimer(timer); {      //统计的代码段      ………… } //停

#include #include #include #include using namespace std; const int N=123456; const int sample=100; const int threads=256; const int blocks=2048; __global__ void judge(int* da,int* dat