Random number generators based on linear recurrences modulo 2 are among the fastest long-period generators currently available. The uniformity and independence of the points they produce, by taking vectors of successive output values from all possible initial states, can be measured by theoretical figures of merit that can be computed quickly, and the generators having good values for these figures of merit are statistically reliable in general. Some of these generators can also provide disjoint streams and substreams efficiently. In this chapter, we review the most interesting construction methods for these generators, examine their theoretical and empirical properties, describe the relevant computational tools and algorithms, and make comparisons.