HieraSynth: A Parallel Framework for Complete Super-Optimization with Hierarchical Space Decomposition
Year: 2025
Type: article
Abstract: Modern optimizing compilers generate efficient code but rarely achieve theoretical optimality, often necessitating manual fine-tuning. This is especially the case for processors with vector instructions, which can grow the instruction set by an order of magnitude. Super-optimizers can synthesize optimal code, but they face a fundamental scalability constraint: as the size of the instruction set increases, the length of the longest synthesizable program decreases rapidly. To help super-optimizers deal with large instruction sets, we introduce HieraSynth , a parallel framework for super-optimization that decomposes the problem by hierarchically partitioning the space of candidate programs, effectively decreasing the instruction set size. It also prunes search branches when the solver proves unrealizability, and explores independent subspaces in parallel, achieving near-linear speedup. HieraSynth is sufficiently efficient to run to completeness even on many hard problems, which means that it exhaustively explores the program space. This ensures that the synthesized program is optimal according to a cost model. We implement HieraSynth as a library and demonstrate its effectiveness with a RISC-V Vector superoptimizer capable of handling instruction sets with up to 700 instructions while synthesizing 7–8-instruction programs. This is a significant advancement over previous approaches that were limited to 1 − 3 instructions with similar instruction set sizes. Specifically, HieraSynth can handle instruction sets up to 10.66× larger for a given program size, or synthesize up to 4.75× larger programs for a fixed instruction set. Evaluations show that HieraSynth can synthesize code surpassing human-expert optimizations and significantly reduce synthesis time, making super-optimization more practical for modern vector architectures.
DOI: 10.1145/3763162
Authors: Sirui Lu, Rastislav Bodík
Institutions: University of Washington, Google (United States)
Language: English
Subfield: Hardware and Architecture
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
Awards: A1: Knowledge Network Development Infrastructure with Application to COVID-19 Science and Economics, NRI: INT: COLLAB: Program Verification and Synthesis for Collaborative Robots, Convergence Accelerator Phase I (RAISE): Linking the Open Knowledge Network to the Web with End-User Programming, FMitF: Track I: End-User Programming with Synthesis-Guided Interaction Models, G3785592558, RAPID: Collecting Reliable COVID-19 Datasets in Crisis Conditions, G4713059963, FMitF: Track II: Programming by Demonstration for the Browser with Applications in Data Science, CAPA: Collaborative Research: ARION: Taming Heterogeneity with DSLs, Approximation, and Synthesis
OpenAlex ID: W4414988627