Simultaneous multithreading: maximizing on-chip parallelism
ISCA '95 Proceedings of the 22nd annual international symposium on Computer architecture
ISCA '96 Proceedings of the 23rd annual international symposium on Computer architecture
The SimpleScalar tool set, version 2.0
ACM SIGARCH Computer Architecture News
MICRO 31 Proceedings of the 31st annual ACM/IEEE international symposium on Microarchitecture
Software-Directed Register Deallocation for Simultaneous Multithreaded Processors
IEEE Transactions on Parallel and Distributed Systems
Multiple-banked register file architectures
Proceedings of the 27th annual international symposium on Computer architecture
Integrating superscalar processor components to implement register caching
ICS '01 Proceedings of the 15th international conference on Supercomputing
Increasing processor performance by implementing deeper pipelines
ISCA '02 Proceedings of the 29th annual international symposium on Computer architecture
Reducing the complexity of the register file in dynamic superscalar processors
Proceedings of the 34th annual ACM/IEEE international symposium on Microarchitecture
The Alpha 21264 Microprocessor
IEEE Micro
Cherry: checkpointed early resource recycling in out-of-order microprocessors
Proceedings of the 35th annual ACM/IEEE international symposium on Microarchitecture
HPCA '98 Proceedings of the 4th International Symposium on High-Performance Computer Architecture
Proceedings of the 31st annual international symposium on Computer architecture
Hi-index | 0.00 |
In simultaneous multithreaded (SMT) processors, a larger multi-ported rename register file is indispensable for holding more intermediate results of in-flight instructions. However, larger rename register file incurs longer access delay and more power consumption, which are becoming a bottleneck in future SMT processors. To tackle these problems, we propose 2L-MuRR, the abbreviation of Multi-usable Rename Register with 2-Level renaming and allocating, which focuses on more efficient utilization of a fewer number of rename registers. Based on the fact that the effective bit-width of most operands is narrower than the full-bit width of a register entry, 2L-MuRR partitions each rename register into several fields of different widths. Either single field or field combination can hold an operand, thus making each rename register multi-usable. The simulations show that 2L-MuRR improves the efficiency of the rename register file significantly, achieving higher performance with much fewer rename registers.