How does Spark SQL normally treat a comparison with NULL using ordinary equality?

A. NULL equals every value B. It yields unknown rather than true, so null-specific functions are needed C. NULL is automatically converted to zero D. The query always fails