As the indexing rules of L1 and L2 caches are different, this optimal coverage is not reachable. Fortunately, the memory locations are usually referenced in sequences or strides. Whenever a reference to the L2 cache is generated, we calculate the location of the next cache line and feed it into the location cache..