Hui Zeng
ABSTRACT
We propose an energy-efficient implementation of register renaming mechanism for high-performance superscalar microprocessors. We use version numbers to identify various instances of each architectural register. This enables the use of an unified register file to hold the current instances of an architectural register and its committed value in a shadow bitcells and to simplify register renaming and register management. Across the SPEC 2000 benchmarks, the proposed mechanism comes within 6% of the performance of a traditional out-of-order design. An average savings of 20% on the energy spent in renaming, register management and instruction commitment is realized compared to a traditional design.
- T. Austin, E. Larson, and D. Ernst, "Simplescalar: an infrastructure for computer system modeling," Computer, vol. 35, no. 2, pp. 59--67, 2002. Google Scholar
Digital Library
- A. Gonzalez, J. Gonzalez, and M. Valero, "Virtual-physical registers," in Proc. HPCA-4, pp. 175--184, 1998. Google ScholarDigital Library
- I. Park, M. Powell, and T. Vijaykumar, "Reducing register ports for higher speed and lower energy," in Proc. MICRO-35, pp. 171--182, 2002. Google ScholarDigital Library
- J. Martinez, J. Renau, M. Huang, and M. Prvulovic, "CHERRY: Checkpointed early resource recycling in out-of-order microprocessors," in Proc. MICRO-35, pp. 3--14, 2002. Google ScholarDigital Library
- S. Balakrishnan and G. Sohi, "Exploiting value locality in physical register files," in Proc. MICRO-36, pp. 265--276, 2003. Google ScholarDigital Library
- M. Martin, A. Roth, and C. Fischer, "Exploiting dead value information," in Proc. MICRO-30, pp. 125--135, 1997. Google ScholarDigital Library
- T. Jones, M. O'Boyle, J. Abella, A. Gonzalez, and O. Ergin, "Compiler directed early register release," in Proc. PACT 2005, pp. 110--119, 2005. Google ScholarDigital Library
- O. Ergin, D. Balkan, D. Ponomarev, and K. Ghose, "Increasing processor performance through early register release," in Proc.ICCD 2004, pp. 480--487, 2004. Google ScholarDigital Library
- M. Kondo and H. Nakamura, "A small, fast and low-power register file by bit-partitioning," in Proc. HPCA-11, pp. 40--49, 2005. Google ScholarDigital Library
- F. Tseng and Y. N. Patt, "Achieving out-of-order performance with almost in-order complexity," in Proc. 35th ISCA '08, pp. 3--12, 2008. Google ScholarDigital Library
- I. Gonzalez, M. Galluzzi, A. Veidenbaum, M. A. Ramirez, A. Cristal, and M. Valero, "A distributed processor state management architecture for large-window processors," in Proc. MICRO-41, pp. 11--22, 2008. Google ScholarDigital Library
Index Terms
- Energy-efficient renaming with register versioning
Recommendations
Virtual register renaming
ARCS'13: Proceedings of the 26th international conference on Architecture of Computing SystemsThis paper presents a novel high performance substrate for building energy-efficient out-of-order superscalar cores. The architecture does not require a reorder buffer or physical registers for register renaming and instruction retirement. Instead, it ...
Register Versioning: A Low-Complexity Implementation of Register Renaming in Out-of-Order Microarchitectures
ICPP '09: Proceedings of the 2009 International Conference on Parallel ProcessingRegister renaming and associated register management mechanisms represent a significant source of complexity in out-of-order micro architectures. We propose the use of register versioning to simplify this logic. Hardware-supported register versioning ...
Tuning the continual flow pipeline architecture with virtual register renaming
Continual Flow Pipelines (CFPs) allow a processor core to process hundreds of in-flight instructions without increasing cycle-critical pipeline resources. When a load misses the data cache, CFP checkpoints the processor register state and then moves all ...
Comments