Abstract

We present an architecture that features dynamic multithreading execution of a single program. Threads are created automatically by hardware at procedure and loop boundaries and executed speculatively on a simultaneous multithreading pipeline. Data prediction is used to alleviate dependency constraints and enable lookahead execution of the threads. A two-level hierarchy significantly enlarges the instruction window. Efficient selective recovery from the second level instruction window takes place after a mispredicted input to a thread is corrected. The second level is slower to access but has the advantage of large storage capacity. We show several advantages of this architecture: (1) it minimizes the impact of ICache misses and branch mispredictions by fetching and dispatching instructions out-of-order, (2) it uses a novel value prediction and recovery mechanism to reduce artificial data dependencies created by the use of a stack to manage run-time storage, and (3) it improves the execution throughput of a superscalar by 15% without increasing the execution resources or cache bandwidth, and by 30% with one additional ICache fetch port. The speedup was measured on the integer SPEC95 benchmarks, without any compiler support, using a detailed performance simulator.

Citations

1461	The SimpleScalar Tool Set, Version 2.0 - Burger, Austin - 1997
544	Combining branch predictors - McFarling - 1993
210	The potential for using thread-level data speculation to facilitate automatic parallelization - Steffan, Mowry - 1998
174	Speculative versioning cache - Vijaykumar, Gopal, et al. - 2001
148	ARB: A Hardware Mechanism for Dynamic Reordering of Memory References - Franklin, Sohi - 1996
113	The Multiscalar Architecture - Franklin - 1993
76	The Microarchitectures of Superscalar Processors - Smith, Sohi - 1995
3	Value Locality and Data Prediction - Lipasti, Wilkerson, et al. - 1996
3	Dynamic Instruction Reuse. The 24 - Sodani, Sohi - 1997