How to efficently copy the non-zero elements in an array to another array

[Robert_Crovella]

How to put specific elements from one array to another array use CUDA?

void Filter(int *array,int *part_array,int array_size) { int index=0; for(int i=0;i<array_size;i++) { if(array[i]>M)//M is a rule { part_array[index] = array[i]; index++; } } } This is my host function. I want to change it to a cuda kernel function. I don’t know how to create the kernel function and i don’t want to use thrust. thanks for reply!