A cell-based datapath synthesizer for ASICs

Ginetti, A.; Mahmood, M.; Sharma, Bhanu

doi:10.1109/asic.1993.410750

Cited by 3 publications

(1 citation statement)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are some other precedents in the general literature for the approach we will be taking here with Verilog. In particular: That the datapath and control sections should be distinctly treated, eg [4]; That a specification for synthesis should be couched in terms of synthesizable parameterizable modules [13], [9], 1111; That the speci-0-8 186-7082-7/95 $04.00 0 1995 lEEE fication and simulation of the zero-delay functionality should be separated from that incorporating tiiming [lo].…”

Section: Introductionmentioning

confidence: 99%

Writing style for architectural synthesis

Smith

Proceedings. 1995 IEEE International Verilog HDL Conference

View full text Add to dashboard Cite

An encapsulated writing style of Verilog is described together with models for a corresponding scheduler and synthesizer. The method provides a middle ground between synthesis f r o m behavioral specijlcations, which implies decisions too dificult for architectural synthesis, and synthesis f r o m pure structural specijkations, which necessitates providing full details of the controller and netlist. The input writing style is designed t o be as concise as possible. This is done by the encapsulation of the functionality of major submodules, and implicit (infered) control. In this way it is possible t o describe complex functional architectures in a n economical and transparent manner. This description can be transformed semi-automatically by steps through intermediate styles and eventually t o a pure structural specification containing full details of timing invocations and interconnections t o appropriate controllers. At each stage after the first, the writing styles can be fully accessible t o native Verilog simulators. The efficiency and economy of this style is demonstrated by a simulation of the well known DLX processor benchmark with full pipeline bypasses.

show abstract

Section: Introductionmentioning

confidence: 99%