212
edits
Changes
→Optimizing Problems
We grab the devices properties so that we do not exceed resources.
The original algorithm was split into 2 kernels. The first kernel causing no problems is as follows:
int i = blockIdx.x * blockDim.x + threadIdx.x;
