Module gpu_reduction 

GPU reduction and scan operations (CPU mock implementation).

Provides parallel-style reductions (sum, max, min, dot product), exclusive prefix scan, radix sort (counting sort mock), and histogram.

Functions

gpu_dot

Compute the dot product of two equal-length slices (parallel mock).

gpu_histogram

Compute a histogram of the input data over num_bins equal-width bins spanning [min_val, max_val).

gpu_max

Compute the maximum element in a slice (parallel reduction mock).

gpu_min

Compute the minimum element in a slice (parallel reduction mock).

gpu_prefix_sum

Compute the exclusive prefix scan (prefix sum) of a slice.

gpu_sort_radix

Sort a slice of u64 values using a radix sort (counting sort) mock.

gpu_sum

Compute the sum of all elements in a slice (parallel reduction mock).