DSP-MLIR: A MLIR Dialect for Digital Signal Processing

TL;DR

This paper introduces DSP-MLIR, a high-level MLIR dialect for digital signal processing that enables domain-specific optimizations, leading to up to 10x performance improvements in DSP applications.

Contribution

The paper presents a novel MLIR dialect for DSP, along with a compiler and DSL, facilitating high-level optimizations and significant performance gains.

Findings

Up to 10x performance improvement in sample DSP applications

Effective high-level domain-specific optimizations using MLIR dialect

Development of a dedicated DSP compiler and DSL

Abstract

Traditional Digital Signal Processing ( DSP ) compilers work at low level ( C-level / assembly level ) and hence lose much of the optimization opportunities present at high-level ( domain-level ). The emerging multi-level compiler infrastructure MLIR ( Multi-level Intermediate Representation ) allows to specify optimizations at higher level. In this paper, we utilize MLIR framework to introduce a DSP Dialect and perform domain-specific optimizations at dialect -level ( high-level ) and show the usefulness of these optimizations on sample DSP apps. In particular, we develop a compiler for DSP and a DSL (Domain Specific Language) to ease the development of apps. We show the performance improvement in execution time for these sample apps by upto 10x which would have been difficult if the IR were at C/ affine level.