Hi there, I faced an issue when tried to define a helper function for my CUDA Kernel like ```CUDA inline __host__ __device__ float4 operator+(float4 a, float4 b) { return make_float4(a.x + b.x, a.y + b.y, a.z + b.z, a.w + b.w); } ``` And then got ``` input_line_11:3:1: error: function definition is not allowed here { ``` Is it a drawback of a runtime CUDA compilation? Btw/ This is an amazing project! Thanks a lot