#include "common.cuh" #define CUDA_COUNT_EQUAL_CHUNK_SIZE 128 void ggml_cuda_count_equal(ggml_backend_cuda_context & ctx, ggml_tensor * dst);