Fast out-of-order processor simulation using memoization

被引:42
作者
Schnarr, E [1 ]
Larus, JR [1 ]
机构
[1] Univ Wisconsin, Madison, WI 53706 USA
关键词
out-of-order processor simulation; direct-execution; memoization;
D O I
10.1145/291006.291063
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
Our new out-of-order processor simulator; FastSim, uses two innovations to speed up simulation 8-15 times (vs. Wisconsin SimpleScalar) with no loss in simulation accuracy. First, FastSim uses speculative direct-execution to accelerate the functional emulation of speculatively executed program code. Second, it uses a variation on memoization - a well-known technique in programming language implementation - to cache microarchitecture states and the resulting simulator actions, and then "fast forwards" the simulation the next time a cached state is reached. Fast-forwarding accelerates simulation by an order of magnitude, while producing exactly the same, cycle-accurate result as conventional simulation.
引用
收藏
页码:283 / 294
页数:12
相关论文
共 10 条
[1]  
Burger D, 1997, 1342 U WISC MAD COMP
[2]  
CMELIK RF, 1994, P 1994 ACM SIGMETRIC
[3]  
CONTE TM, 1996, P 1996 INT C COMP DE
[4]  
HERROD S, 1996, SIMOS SIMULATION ENV
[5]  
LARUS J, 1995, P ACM SIGPLAN 95 C P
[6]  
MUKHERJEE SS, 1997, WORKSH PERF AN ITS I
[7]  
PAI VS, 1997, P 3 INT S HIGH PERF
[8]  
Pai VS, 1997, P 3 WORKSH COMP ARCH
[9]  
*SUN MICR, 1990, SPARC ARCH MAN VERS
[10]  
YEAGER, 1996, IEEE MICRO APR