35template <MemoryOrder memory_order>
42 return a +
adim[0] * (b +
adim[1] * c);
70template <MemoryOrder memory_order>
91 return (i * (i + 1)) / 2 +
j;
Vector implementation and trait types for boost qvm interoperability.
cudaStream_t stream[1]
CUDA streams for parallel computing on CPU and GPU.
DEVICE_QUALIFIER T lower_triangular(T i, T j)
Linear index into a lower triangular matrix.
DEVICE_QUALIFIER int get_linear_index(int a, int b, int c, Vector3i const &adim)