A design using tag bits presupposes that adjacent memory locations hold dissimilar things: a line of code, a piece of data, a line of code, two pieces of data, and so forth. Most programs do not look like that. How can tag bits be appropriate in a situation in which programs have the more conventional arrangement of code and data?

What will be an ideal response?

First, with small, self-­?contained modules, a module may consist of a small number of lines of code and a few local data storage values. Thus, although code and data do not alternate, relatively small patches of code and data may be spread throughout memory.

Second, there is no predetermined size of code or data areas. Arrangementssuch as paging require the use of an entire page (which may be 2K, 4K, 8K bytes or more) for storage of a separate data area that may be only one byte long. With tag bits, only as many locations are assigned one form of protection as needed.