Toward Performance Portable Programming for Heterogeneous Systems on a Chip: A Case Study with Qualcomm Snapdragon SoC

Future heterogeneous domain-specific systems on a chip (DSSoCs) will be extraordinarily complex in terms of processors, memory hierarchies, and interconnection networks. To manage this complexity, architects, system software designers, and application developers need programming technologies that are flexible, accurate, efficient, and productive. These technologies must be as independent of any one specific architecture as is practical because the sheer dimensionality and scale of the complexity will not allow porting and optimizing applications for each given DSSoC. To address these issues, the authors are developing Cosmic Castle, a performance portable programming toolchain for streaming applications on heterogeneous architectures. The primary focus of Cosmic Castle is on enabling efficient and performant code generation through the smart compiler and intelligent runtime system. This paper presents the preliminary evaluation of the authors’ ongoing work toward Cosmic Castle. Specifically, this paper details the code-porting efforts and evaluates various benchmarks on the Qualcomm Snapdragon SoC using tools developed through Cosmic Castle.