SubGemini: identifying subcircuits using a fast subgraph isomorphism algorithm
DAC '93 Proceedings of the 30th international Design Automation Conference
Instruction generation for hybrid reconfigurable systems
ACM Transactions on Design Automation of Electronic Systems (TODAES)
Reconfigurable Instruction Set Processors from a Hardware/Software Perspective
IEEE Transactions on Software Engineering
Automatic application-specific instruction-set extensions under microarchitectural constraints
Proceedings of the 40th annual Design Automation Conference
ConCISe: A Compiler-Driven CPLD-Based Instruction Set Accelerator
FCCM '99 Proceedings of the Seventh Annual IEEE Symposium on Field-Programmable Custom Computing Machines
Performance Evaluation of the VF Graph Matching Algorithm
ICIAP '99 Proceedings of the 10th International Conference on Image Analysis and Processing
A dynamic instruction set computer
FCCM '95 Proceedings of the IEEE Symposium on FPGA's for Custom Computing Machines
Computer
Processor Acceleration Through Automated Instruction Set Customization
Proceedings of the 36th annual IEEE/ACM International Symposium on Microarchitecture
Application-specific instruction generation for configurable processor architectures
FPGA '04 Proceedings of the 2004 ACM/SIGDA 12th international symposium on Field programmable gate arrays
Characterizing embedded applications for instruction-set extensible processors
Proceedings of the 41st annual Design Automation Conference
Introduction of local memory elements in instruction set extensions
Proceedings of the 41st annual Design Automation Conference
MiBench: A free, commercially representative embedded benchmark suite
WWC '01 Proceedings of the Workload Characterization, 2001. WWC-4. 2001 IEEE International Workshop
Modern development methods and tools for embedded reconfigurable systems: A survey
Integration, the VLSI Journal
| Hi-index | 0.00 |
This paper presents a novel methodology for instruction set customization of RISPs (Reconfigurable Instruction Set Processors) using morphable structures. A morphable structure consists of a group of hardware operators chained together to implement a restricted set of custom instructions. These structures are implemented on the reconfigurable fabric, and the operators are enabled/disabled on demand. The utilization of a predefined set of morphable structures for instruction set customization dispenses the need for hardware synthesis in design exploration, and avoids run-time reconfiguration while minimizing the reconfigurable area. We will describe the two stages of the methodology for constructing the morphable structures, namely template generation and identification of a maximal unique pattern set from the templates. Our preliminary studies show that 23 predefined morphable structures can sufficiently cater to any application in a set of eight MiBench benchmark applications. In addition, to achieve near-optimal performance, the maximum required number of morphable structures for an application is only 8.