Error at example

[kozinove]

I looked at your example
https://github.com/tpn/cuda-samples/blob/master/v10.1/0_Simple/fp16ScalarProduct/fp16ScalarProduct.cu

The example contains an error
for (int i = threadIdx.x + blockDim.x + blockIdx.x; i < size; i+=stride)
Should be
for (int i = threadIdx.x + blockDim.x * blockIdx.x; i < size; i+=stride)