High-level Transformations using Canonical Dataflow Representation
نویسندگان
چکیده
This paper describes a systematic method and an experimental software system for high-level transformations of designs specified at behavioral level. The goal is to transform the initial design specifications into an optimized data flow graph (DFG) better suited for high-level synthesis. The optimizing transformations are based on a canonical Taylor Expansion Diagram (TED) representation, followed by structural transformations of the resulting DFG network. The system is intended for data-flow and computation-intensive designs used in computer graphics and digital signal processing applications.
منابع مشابه
Composing Source-to-Source Data-Flow Transformations with Rewriting Strategies and Dependent Dynamic Rewrite Rules
Data-flow transformations used in optimizing compilers are also useful in other programming tools such as code generators, aspect weavers, domainspecific optimizers, and refactoring tools. These applications require source-tosource transformations rather than transformations on a low-level intermediate representation. In this paper we describe the composition of source-to-source data-flow trans...
متن کاملHigh Level Synthesis with a Dataflow Architectural Template
In this work, we present a new approach to high level synthesis (HLS), where high level functions are first mapped to an architectural template, before hardware synthesis is performed. As FPGA platforms are especially suitable for implementing streaming processing pipelines, we perform transformations on conventional high level programs where they are turned into multi-stage dataflow engines. T...
متن کاملProfile Guided Dataflow Transformation for FPGAs and CPUs
This paper proposes a new high-level approach for optimising field programmable gate array (FPGA) designs. FPGA designs are commonly implemented in low-level hardware description languages (HDLs), which lack the abstractions necessary for identifying opportunities for significant performance improvements. Using a computer vision case study, we show that modelling computation with dataflow abstr...
متن کاملReengineering Product Lines of Dataflow Programs
Dataflow programs (DfPs) are widely used in computing. They are complex graphs where nodes are computations and edges indicate the flow of data. We reverse engineered a legacy DfP by deriving its graph from an elementary graph using domain-specific transformations. (In MDE-speak, our derivations are PIM to PSM mappings). In this paper, we explain how our tool ReFlO (a) implements transformation...
متن کاملCanonical representation for approximating solution of fuzzy polynomial equations
In this paper, the concept of canonical representation is proposed to find fuzzy roots of fuzzy polynomial equations. We transform fuzzy polynomial equations to system of crisp polynomial equations, this transformation is perform by using canonical representation based on three parameters Value, Ambiguity and Fuzziness.
متن کامل