TuneAgent: Agentic Operating System Kernel Tuning with Reinforcement Learning
Hongyu Lin, Yuchen Li, Haoran Luo, Zhenghong Lin, Libo Zhang, Mingjie Xing, Yanjun Wu
Abstract
Linux kernel tuning is essential for optimizing operating system (OS) performance, yet remains challenging due to the complex kernel space, sparse performance feedback, and strong workload sensitivity. We present TuneAgent, an agentic Linux kernel tuning framework powered by rule-based reinforcement learning (RL). TuneAgent formulates the kernel space as a constrained RL environment, enabling large language models (LLMs) to autonomously explore the kernel while enforcing valid and precise configuration modifications. To address sparse performance feedback, we design structured reward functions that jointly promote reasoning standardization, configuration correctness, and performance awareness. Furthermore, we propose a two-phase training strategy that first ensures format and semantic correctness and then transitions to performance-driven exploration, accelerating convergence and reducing overhead. Experimental results show that TuneAgent consistently outperforms existing baselines, achieving up to 5.6% relative overall performance improvement while maintaining high configuration validity. We further demonstrate its robustness across multiple real-world applications, highlighting its practicality and adaptability in diverse deployment environments.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
1 top-tier paper has cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
Cited by top-tier papers1
Builds on4
- Drive-R1: Bridging Reasoning and Planning in VLMs for Autonomous Driving with Reinforcement LearningYue Li, Meng Tian, Dechang Zhu, Jiangtong Zhu et al.AAAI 2026 · 27 citations
- Towards a Machine Learning-Assisted Kernel with LAKEHenrique Fingler, Isha Tarte, Hangchen Yu, Ariel Szekely et al.ASPLOS 2023 · 19 citations
- Compiler-R1: Towards Agentic Compiler Auto-tuning with Reinforcement LearningHaolin Pan, Hongyu Lin, Haoran Luo, Yang Liu et al.NeurIPS 2025 · 14 citations
- AutoOS: Make Your OS More Powerful by Exploiting Large Language ModelsHuilai Chen, Yuanbo Wen, Limin Cheng, Shouxu Kuang et al.ICML 2024 · 8 citations
Related papers
- STARK: Strategic Team of Agents for Refining KernelsJuncheng Dong, Yang Yang, Tao Liu, Yang Wang et al.ICLR 2026 · 26 citations
- PerfDojo: Automated ML Library Generation for Heterogeneous ArchitecturesAndrei Ivanov, Siyuan Shen, Gioele Gottardo, Marcin Chrapek et al.SC 2025 · 2 citations
- Retroformer: Retrospective Large Language Agents with Policy Gradient OptimizationWeiran Yao, Shelby Heinecke, Juan Carlos Niebles, Zhiwei Liu et al.ICLR 2024 · 124 citations
- Towards Better Correctness and Efficiency in Code GenerationYunlong Feng, Yang Xu, Xiao Xu, Binyuan Hui et al.AAAI 2026 · 3 citations
- AuTAgent: A Reinforcement Learning Framework for Tool-Augmented Audio ReasoningSiqian Tong, Xuan Li, Yiwei Wang, Baolong Bi et al.ICML 2026 · 3 citations
