Computer architecture: a quantitative approach
Computer architecture: a quantitative approach
Communications of the ACM
Generation of software tools from processor descriptions for hardware/software codesign
DAC '97 Proceedings of the 34th annual Design Automation Conference
A methodology for accurate performance evaluation in architecture exploration
Proceedings of the 36th annual ACM/IEEE Design Automation Conference
Cycle and phase accurate DSP modeling and integration for HW/SW co-verification
Proceedings of the 36th annual ACM/IEEE Design Automation Conference
Proceedings of the 14th international symposium on Systems synthesis
An ultra-fast instruction set simulator
IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Incas: a cycle accurate model of UltraSPARC
ICCD '95 Proceedings of the 1995 International Conference on Computer Design: VLSI in Computers and Processors
Techniques for Implementing Fast Processor Simulators
SS '98 Proceedings of the The 31st Annual Simulation Symposium
Automatic Generation of Microarchitecture Simulators
ICCL '98 Proceedings of the 1998 International Conference on Computer Languages
Fast Specification of Cycle-Accurate Processor Models
ICCD '01 Proceedings of the International Conference on Computer Design: VLSI in Computers & Processors
Proceedings of the 42nd annual Design Automation Conference
Designing real-time H.264 decoders with dataflow architectures
CODES+ISSS '05 Proceedings of the 3rd IEEE/ACM/IFIP international conference on Hardware/software codesign and system synthesis
Hi-index | 0.00 |
Modeling and simulating pipelined processors in procedurallanguages such as C/C++ requires lots of cost in handlingconcurrent events, which hinders fast simulation. A number ofresearches on simulation have devised speed-up techniques toreduce the number of events. This paper presents a newsimulation approach developed to enhance the simulation ofpipelined processors. The proposed approach is based on earlypipeline evaluation that all the intermediate values of aninstruction are computed in advance, creating a future state for thenext instructions. The future state allows the next instructions tobe computed without considering data dependencies betweennearby instructions. We apply this concept to building a cycle-accurate simulator for a pipelined RISC processor and achieve almost the same speed as the instruction-level simulator.