This paper describes fault-tolerant and error detection features in HaL's memory management unit (MMU). The proposed fault-tolerant features allow recovery from transient errors in the MMU. It is shown that these features were natural choices considering the architectural and implementation constraints in the MMU's design environment. Three concurrent error detection and correction methods employed in address translation and coherence tables in the MMU are described. Virtually-indexed and virtually-tagged cache architecture is exploited to provide an almost fault-secure hardware coherence mechanism in the MMU, with very small performance overhead (less than 0.01% in the instruction throughput). Low overhead linear polynomial codes have been chosen in these designs to minimize both the hardware and software instrumentation impact.>

The major objective of this VLSI module design is to determine how to modify a fast floating-point multiplier so that it can perform division and square root in accordance with IEEE standards. This has been achieved by applying the Newton-Ralphson iteration only on the mantissa and adjusting the iterated result by a rounding algorithm. Using 1.0- mu m CMOS standard cell technology, the total area of this module is approximately 7.0 mm*6.5 mm, which is just 25% larger than the floating-point multiplier. The module can compute multiplication, division, and square root in 3, 31, and 43 cycles, respectively. The cycle time, under nominal conditions, is expected to be 20 ns.>