I tried to test following code
include <stdio.h>
global void helloCUDA()
{
printf(“Hello, CUDA!\n”);
}
int main()
{
helloCUDA<<<1, 1>>>();
cudaDeviceSynchronize();
return 0;
}
and I got this error “nvcc error : ‘cudafe++’ died with status 0xC0000005 (ACCESS_VIOLATION)”
I indeed successfully compiled this code a few days ago, but failed today. I have tried re-install cuda and changing the path that locates “cl.exe”, but both do not work.