RepoOMP: Repository-Aware Hotspot OpenMP Parallelization via Dependency-Aware Context Reduction

2026-08-06

AI Summary

This work addresses the challenge of OpenMP parallelization in hot loops from mature codebases, where non-local contextual dependencies often lead existing approaches to under-parallelize or produce unstable results. To overcome this, we propose RepoOMP, a hybrid framework that innovatively integrates deterministic rules with large language models. Our approach employs a multi-granularity attribute-performance graph (MAP) for dynamic decision routing and introduces a structured transformation context (STC) to accurately model dependencies, augmented by dependency-aware context compression to reduce interference and inference overhead. Evaluated on 330 real-world hot loops, RepoOMP achieves a median speedup of 2.25×; on NPB and BOTS benchmarks, it delivers average speedups of 8.23× and 8.96×, respectively—outperforming baselines by 18–28% while reducing agent token consumption by 47–68%.

Abstract

OpenMP parallelization of hotspots in mature repositories remains difficult because loop safety and optimization payoff often depend on non-local evidence. Rule-based tools under-parallelize when legality is not locally provable, while agent-based approaches become unstable when retrieval misses decisive dependencies or includes irrelevant code. We present RepoOMP, a hybrid framework that recovers parallelization-relevant evidence before generation. RepoOMP builds a Multi-granularity Attributes Performance graph (MAP), routes hotspots between deterministic rules and an LLM agent, and constructs a Structured Transformation Context (STC) that exposes dependency facts without flooding the model with unrelated repository text. We evaluate RepoOMP on 951 profiled hotspots from NPB, BOTS, FFmpeg, NCNN, and GROMACS. Under compilation, workload-specific checks, and positive speedup, 372 hotspots are accepted, including 330 real-world repository hotspots. RepoOMP achieves average speedups of $8.23\times$ on NPB and $8.96\times$ on BOTS. For the nine detailed real-world kernels used in matched-backbone and robustness analyses, RepoOMP reaches a cross-backbone mean of $5.25\times$, improves speedup by 18--28\%, and reduces agent-side token cost by 47--68\% relative to the unstructured Claude Code baseline. Across 330 accepted real-world hotspots, median speedup is $2.25\times$. Overall, RepoOMP provides an evidence-guided workflow for hotspot parallelization in repository settings. The open-source repository is available at https://github.com/Qlalq/RepoOMP_Simplified.

Problem

• OpenMP parallelization
• hotspot
• repository-aware
• dependency-aware
• code optimization

Innovation

• Repository-Aware Parallelization
• Dependency-Aware Context Reduction
• Structured Transformation Context
• Hybrid Rule-LLM Framework
• OpenMP Hotspot Optimization