Register Port Prediction for a Banked Register File

@article{Kawashima2015RegisterPP,
  title={Register Port Prediction for a Banked Register File},
  author={Hiroaki Kawashima and Takahiro Sasaki and Yuki Fukazawa and T. Kondo},
  journal={2015 Third International Symposium on Computing and Networking (CANDAR)},
  year={2015},
  pages={551-555}
}
A large multi-port register file is an indispensable component to achieve higher computing performance, especially in recent processors. However, the number of its ports effects to circuit scale, access latency and power consumption significantly. Bank memory is one solution to implement a multi-port memory effectively. However, performance of the bank memory is lower than that of ideal multi-port memory. In order to reduce performance degradation caused by bank conflict, this paper proposes… Expand

References

SHOWING 1-10 OF 13 REFERENCES
Reducing register ports for higher speed and lower energy
TLDR
This work proposes to reduce the number of register ports through two proposals, one for reads and the other for writes, and uses decoupled rename, a technique which separates dependence and physical tagging of register operands, to perform physical register allocation just before writeback. Expand
Register Cache System Not for Latency Reduction Purpose
TLDR
The proposed Non-Latency-Oriented Register Cache System (NORCS) is a free from register cache miss penalties that the conventional systems suffer from, and can simplify the bypass network to the same complexity as a 1-cycle-latency register file. Expand
Multi-ported memories for FPGAs via XOR
TLDR
This paper presents an alternative approach based on the XOR operation that provides multi-ported memories that use far less logic but more block RAMs than LVT designs, and are often smaller and faster for memories that are more than 512 entries deep. Expand
FPGA modeling of diverse superscalar processors
TLDR
FPGA-Sim is described, a configurable, automatically FGPA-synthesizable, and register-transfer-level (RTL) model of an out-of-order superscalar processor that enables FPGA modeling of diverse superscalars out- of-the-box. Expand
The Alpha 21264 microprocessor
TLDR
A unique combination of high clock speeds and advanced microarchitectural techniques, including many forms of out-of-order and speculative execution, provide exceptional core computational performance in the 21264. Expand
FabScalar: Composing synthesizable RTL designs of arbitrary cores within a canonical superscalar template
TLDR
From this idea, a toolset is developed, called FabScalar, for automatically composing the synthesizable register-transfer-level (RTL) designs of arbitrary cores within a canonical superscalar template, which defines canonical pipeline stages and interfaces among them. Expand
Innovative Architecture for Future Generation High-Performance Processors and Systems
  • A. Veidenbaum, K. Joe
  • Computer Science
  • Innovative Architecture for Future Generation High-Performance Processors and Systems, 2003
  • 1998
The following topics are dealt with: embedded systems; computer architecture; embedded low power; high-performance processors; parallel systems; embedded processor design; memory hierarchy;Expand
Multi-bank Register File in the Register Cache System
  • Multi-bank Register File in the Register Cache System
  • 2014
Propose a Small Register File using Bank Memory and Write-back Port Prediction
  • Propose a Small Register File using Bank Memory and Write-back Port Prediction
  • 2015
...
1
2
...