摘要：We identify that a set of multimedia applications exhibit highly regular read-after-read (RAR)and read-after-write (RAW) memory dependence streams. We exploit this regularity to predictboth RAW and RAR memory dependences. We also study how two previously proposed memorydependence prediction-based memory latency reduction techniques perform for this multimediaworkload. In the first technique, a load can obtain a value by simply naming a preceding load (orstore) with which a RAR (or RAW) dependence is predicted. The second technique speculativelyconverts a series of LOAD1-USE1,...,LOADN-USEN(or DEF-STORE-LOAD-USE) chains into a singleLOAD1-USE1...USEN(or DEF-USE) producer/consumer graph. We show that via memorydependence prediction it is possible to correctly predict 33.3% of all loads on the average.Moreover, the two memory dependence prediction based techniques result on averageperformance improvements of 2.6% over a highly-aggressive, out-of-order, superscalarprocessor. The actual range of performance improvements is 0% to 8.5%. When cache latency isincreased from 2 to 3 cycles, performance improves by 3.75% on average, with the range being0% to 16.35%
