A new algorithm for computing IEEE-compliant rounding is presented, called
injection-based rounding. Injection-based rounding is simple and facilitate
s using the same rounding circuitry for different precisions. We demonstrat
e the usefulness of injection-based rounding in a design of an IEEE floatin
g-point multiplier capable of performing either a double-precision multipli
cation or a single-precision multiplication. The multiplier is designed to
minimize hardware cost by using only a half-sized multiplication array and
by sharing the rounding circuitry for both precisions. The latency of the m
ultiplier is in single-precision two clock cycles and in double precision t
he latency is three clock cycles, where each pipeline stage contains roughl
y 15 logic levels. (C) 2000 Elsevier Science B.V. All rights reserved.