arXiv subjects

Mingjie Xing

Publications and source records attributed to Mingjie Xing.

16 recordsLinked to original sources

arXiv2026

Toward Compiler World Models: Learning Latent Dynamics for Efficient Tensor Program Search

Tensor program optimization is essential for modern machine learning systems, but its search space is enormous. Existing auto-schedulers reduce measurement cost with learned cost models, yet they usually evaluate each candidate as a static code snapshot, ignoring the schedule trajectory that produced it. This makes them insensitive to action dependencies and vulnerable to superficial code variations. We propose a \emph{world-model-inspired} evaluator that models schedule evaluation as action-conditioned latent dynamics over program states. Starting from the initial program, it rolls out scheduling actions in a continuous latent space with a lightweight transition model, avoiding expensive AST mutation and repeated code encoding. The final dynamic representation is combined with action and hardware features to rank candidates. Implemented in TVM AutoScheduler, our method improves representative-subgraph latency over Ansor by 1.37$\times$ on GPU and 1.54$\times$ on CPU under the same 64-trial budget. It also matches Ansor-10K within 2.2% geometric mean using 10$\times$ fewer measurements, and accelerates full-model inference over PyTorch/PyTorch-opt(cuDNN) by 4.61$\times$/3.67$\times$ geometric mean.

cs.LG↗

arXiv2026

A Case for Agentic Tuning: From Documentation to Action in PostgreSQL

Documentation has long guided computer system tuning by distilling expert knowledge into per-parameter recommendations. Yet such guides capture only what experts conclude, discarding how they reason. This fundamental gap manifests in three concrete deficiencies: documentation grows stale as software evolves, fails under heterogeneous workloads, and ignores inter-parameter dependencies. We propose shifting from static documentation to dynamic action for system tuning. We introduce PerfEvolve, which translates expert tuning methodologies into executable skills that equip LLM-based agents to perform version-consistency verification, workload-specific profiling, and multi-parameter joint optimization. Evaluated on PostgreSQL under TPC-C and TPC-H benchmarks, PerfEvolve outperforms state-of-the-art documentation-driven tuning baselines by up to 35.2%. The tool is available at https://github.com/ISCAS-OSLab/PerfEvolve.

cs.SE↗

arXiv2026

ECCO: Evidence-Driven Causal Reasoning for Compiler Optimization

Compiler auto-tuning faces a dichotomy between traditional black-box search methods, which lack semantic guidance, and recent Large Language Model (LLM) approaches, which often suffer from superficial pattern matching and causal opacity. In this paper, we introduce ECCO, a framework that bridges interpretable reasoning with combinatorial search. We first propose a reverse engineering methodology to construct a Chain-of-Thought dataset, explicitly mapping static code features to verifiable performance evidence. This enables the model to learn the causal logic governing optimization decisions rather than merely imitating sequences. Leveraging this interpretable prior, we design a collaborative inference mechanism where the LLM functions as a strategist, defining optimization intents that dynamically guide the mutation operations of a genetic algorithm. Experimental results on seven datasets demonstrate that ECCO significantly outperforms the LLVM opt -O3 baseline, achieving an average 24.44% reduction in cycles.

cs.LG↗

arXiv2025

VecIntrinBench: Benchmarking Cross-Architecture Intrinsic Code Migration for RISC-V Vector

Intrinsic functions are specialized functions provided by the compiler that efficiently operate on architecture-specific hardware, allowing programmers to write optimized code in a high-level language that fully exploits hardware features. Using intrinsics to vectorize core code blocks is a standard optimization method in high-performance libraries, often requiring specific vector optimization implementations for multiple mainstream architectures. The promising RISC-V software ecosystem has a significant demand for algorithm library migration and adaptation. Translating existing intrinsic functions to RISC-V Vector (RVV) intrinsic functions across architectures is currently a mainstream approach. Rule-based intrinsic mapping methods and LLM-based code generation can help developers address the code migration challenge. However, existing intrinsic code benchmarks focus on mainstream SIMD intrinsics and lack support for the emerging RISC-V architecture. There is currently no benchmark that comprehensively evaluates the intrinsic migration capabilities for the RVV extension. To fill this gap, we propose VecIntrinBench, the first intrinsic benchmark encompassing RVV extensions. It includes 50 function-level tasks from open source repositories, implemented as scalars, RVV intrinsics, Arm Neon intrinsics, and x86 intrinsics, along with comprehensive functional and performance test cases. We systematically evaluated various code migration approaches on VecIntrinBench, yielding a series of insightful findings. The results demonstrate that advanced Large Language Models (LLMs) achieve a similar effect as rule-based mapping approaches for RISC-V code migration, while also delivering superior performance. We further analyze the reasons and identify future directions for LLM development in the code migration field. The VecIntrinBench is open-sourced to benefit the broader community and developers.

cs.SE↗

arXiv2025

Exploring the Feasibility of End-to-End Large Language Model as a Compiler

In recent years, end-to-end Large Language Model (LLM) technology has shown substantial advantages across various domains. As critical system software and infrastructure, compilers are responsible for transforming source code into target code. While LLMs have been leveraged to assist in compiler development and maintenance, their potential as an end-to-end compiler remains largely unexplored. This paper explores the feasibility of LLM as a Compiler (LaaC) and its future directions. We designed the CompilerEval dataset and framework specifically to evaluate the capabilities of mainstream LLMs in source code comprehension and assembly code generation. In the evaluation, we analyzed various errors, explored multiple methods to improve LLM-generated code, and evaluated cross-platform compilation capabilities. Experimental results demonstrate that LLMs exhibit basic capabilities as compilers but currently achieve low compilation success rates. By optimizing prompts, scaling up the model, and incorporating reasoning methods, the quality of assembly code generated by LLMs can be significantly enhanced. Based on these findings, we maintain an optimistic outlook for LaaC and propose practical architectural designs and future research directions. We believe that with targeted training, knowledge-rich prompts, and specialized infrastructure, LaaC has the potential to generate high-quality assembly code and drive a paradigm shift in the field of compilation.

cs.LG↗

arXiv2025

A Hybrid, Knowledge-Guided Evolutionary Framework for Personalized Compiler Auto-Tuning

Compiler pass auto-tuning is critical for enhancing software performance, yet finding the optimal pass sequence for a specific program is an NP-hard problem. Traditional, general-purpose optimization flags like -O3 and -Oz adopt a one-size-fits-all approach, often failing to unlock a program's full performance potential. To address this challenge, we propose a novel Hybrid, Knowledge-Guided Evolutionary Framework. This framework intelligently guides online, personalized optimization using knowledge extracted from a large-scale offline analysis phase. During the offline stage, we construct a comprehensive compilation knowledge base composed of four key components: (1) Pass Behavioral Vectors to quantitatively capture the effectiveness of each optimization; (2) Pass Groups derived from clustering these vectors based on behavior similarity; (3) a Synergy Pass Graph to model beneficial sequential interactions; and (4) a library of Prototype Pass Sequences evolved for distinct program types. In the online stage, a bespoke genetic algorithm leverages this rich knowledge base through specially designed, knowledge-infused genetic operators. These operators transform the search by performing semantically-aware recombination and targeted, restorative mutations. On a suite of seven public datasets, our framework achieves an average of 11.0% additional LLVM IR instruction reduction over the highly-optimized opt -Oz baseline, demonstrating its state-of-the-art capability in discovering personalized, high-performance optimization sequences.

cs.SE↗

arXiv2025

Behavioral Embeddings of Programs: A Quasi-Dynamic Approach for Optimization Prediction

Learning effective numerical representations, or embeddings, of programs is a fundamental prerequisite for applying machine learning to automate and enhance compiler optimization. Prevailing paradigms, however, present a dilemma. Static representations, derived from source code or intermediate representation (IR), are efficient and deterministic but offer limited insight into how a program will behave or evolve under complex code transformations. Conversely, dynamic representations, which rely on runtime profiling, provide profound insights into performance bottlenecks but are often impractical for large-scale tasks due to prohibitive overhead and inherent non-determinism. This paper transcends this trade-off by proposing a novel quasi-dynamic framework for program representation. The core insight is to model a program's optimization sensitivity. We introduce the Program Behavior Spectrum, a new representation generated by probing a program's IR with a diverse set of optimization sequences and quantifying the resulting changes in its static features. To effectively encode this high-dimensional, continuous spectrum, we pioneer a compositional learning approach. Product Quantization is employed to discretize the continuous reaction vectors into structured, compositional sub-words. Subsequently, a multi-task Transformer model, termed PQ-BERT, is pre-trained to learn the deep contextual grammar of these behavioral codes. Comprehensive experiments on two representative compiler optimization tasks -- Best Pass Prediction and -Oz Benefit Prediction -- demonstrate that our method outperforms state-of-the-art static baselines. Our code is publicly available at https://github.com/Panhaolin2001/PREP/.

cs.LG↗

arXiv2025

GRACE: Cluster-Specific Sequence Reuse for Compiler Auto-Tuning

Compiler auto-tuning aims to improve optimization quality beyond fixed compiler heuristics, but existing approaches often face a trade-off between effectiveness and deployability. Iterative compilation can discover strong program-specific optimization sequences, yet its search cost is often prohibitive for practical reuse. Learning-based methods reduce tuning overhead, but their effectiveness depends on how well optimization knowledge transfers to unseen programs. Recent coreset-based methods improve this trade-off, but they typically either still rely on relatively large test-time search or assume that a single global coreset can serve all programs well. We present GRACE, a compiler auto-tuning framework based on \emph{cluster-specific sequence reuse}. GRACE constructs a small reusable sequence coreset for each group of similar programs by combining global pass synergy analysis, optimization-response-guided program organization, and cluster-specific evolutionary search. At deployment time, it evaluates a small coreset on the target program and optionally performs lightweight refinement within a restricted search space, yielding bounded overhead. We evaluate GRACE on seven benchmark datasets using LLVM 10.0.0 and LLVM 18.1.6. For code-size optimization, GRACE reduces LLVM IR instruction count by 9.92\% and 10.30\% on average relative to \texttt{opt -Oz}, while requiring less than 1\,s tuning time per program at deployment. Under an execution-oriented objective, GRACE reduces estimated cycle counts by 26.84\% and 27.54\% on average relative to \texttt{opt -O3}, and also yields measurable end-to-end speedups on runnable cBench and polybench programs. These results suggest that offline-constructed, cluster-specific sequence coresets provide a practical balance between optimization quality and cost.

cs.SE↗

arXiv2025

Synergy-Guided Compiler Auto-Tuning of Nested LLVM Pass Pipelines

Compiler optimization relies on sequences of passes to improve program performance. Selecting and ordering these passes automatically, known as compiler auto-tuning, is challenging due to the large and complex search space. Existing approaches generally assume a linear sequence of passes, a model compatible with legacy compilers but fundamentally misaligned with the hierarchical design of the LLVM New Pass Manager. This misalignment prevents them from guaranteeing the generation of syntactically valid optimization pipelines. In this work, we present a new auto-tuning framework built from the ground up for the New Pass Manager. We introduce a formal grammar to define the space of valid nested pipelines and a forest-based data structure for their native representation. Upon this foundation, we develop a structure-aware Genetic Algorithm whose operators manipulate these forests directly, ensuring that all candidate solutions are valid by construction. The framework first mines synergistic pass relationships to guide the search. An optional refinement stage further explores subtle performance variations arising from different valid structural arrangements. We evaluate our approach on seven benchmark datasets using LLVM 18.1.6. The discovered pipelines achieve an average of 13.62% additional instruction count reduction compared to the standard opt -Oz optimization level, showing that our framework is capable of navigating this complex, constrained search space to identify valid and effective pass pipelines.

cs.SE↗

arXiv2025

AwareCompiler: Agentic Context-Aware Compiler Optimization via a Synergistic Knowledge-Data Driven Framework

Compiler optimization is crucial for enhancing program performance by transforming the sequence of optimization passes while maintaining correctness. Despite the promising potential of large language models (LLMs)-based agent for software optimization, automating compiler optimization remains challenging due to: (1) semantic misalignment between abstract program representations and concrete optimization passes, (2) inefficient interaction mechanisms between agents and compiler environments, and (3) reward sparsity from the extensive decision-making process within large optimization spaces. This paper introduces \textbf{AwareCompiler}, an agentic framework for compiler optimization that addresses these challenges through three key innovations: structured knowledge integration and dataset construction, knowledge-driven adaptive pass generation, and data-driven hybrid training pipeline. Experimental results on standard benchmarks demonstrate that AwareCompiler significantly outperforms existing baselines in both performance and efficiency, highlighting the effectiveness of our synergistic knowledge-data-driven approach. Our code is publicly available at https://github.com/LHY-24/AwareCompiler.

cs.PL↗

arXiv2025

IntrinTrans: LLM-based Intrinsic Code Translator for RISC-V Vector

The use of intrinsic functions to leverage hardware-specific capabilities is a crucial approach for optimizing library performance. Many mainstream libraries implement a large number of vectorized algorithms on Arm or x86 SIMD (Single-Instruction, Multiple-Data) intrinsic functions. Translating existing vectorized intrinsic code into the intrinsics of an emerging architecture is a practical and effective approach. However, current cross-architecture translation largely relies on manual rewriting or rule-based mapping methods, which are both time-consuming and prone to errors. We present \texttt{IntrinTrans}, a LLM-based agent that utilizes compile-and-test feedback to translate intrinsic code across architectures automatically, and further optimizes the generated intrinsics using register-usage information derived from liveness analysis. To evaluate the effectiveness of our method, we used \texttt{IntrinTrans} to translate the open-source benchmark from Arm Neon Intrinsic to the emerging RISC-V Vector (RVV) Intrinsic implementation and compared its performance with that of the native RVV implementation. Our experiments show that advanced LLMs can generate semantically correct RVV Intrinsic functions with only a finite number of iterations. Depending on the base LLMs, the pass rate ranges from 47% to 100%, achieving performance similar to the native implementation (0.85x to 1.28x).

cs.SE↗

arXiv2025

TuneAgent: Agentic Operating System Kernel Tuning with Reinforcement Learning

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.

cs.LG↗

arXiv2025

Compiler-R1: Towards Agentic Compiler Auto-tuning with Reinforcement Learning

Compiler auto-tuning optimizes pass sequences to improve performance metrics such as Intermediate Representation (IR) instruction count. Although recent advances leveraging Large Language Models (LLMs) have shown promise in automating compiler tuning, two significant challenges still remain: the absence of high-quality reasoning datasets for agents training, and limited effective interactions with the compilation environment. In this work, we introduce Compiler-R1, the first reinforcement learning (RL)-driven framework specifically augmenting LLM capabilities for compiler auto-tuning. Compiler-R1 features a curated, high-quality reasoning dataset and a novel two-stage end-to-end RL training pipeline, enabling efficient environment exploration and learning through an outcome-based reward. Extensive experiments across seven datasets demonstrate Compiler-R1 achieving an average 8.46% IR instruction count reduction compared to opt -Oz, showcasing the strong potential of RL-trained LLMs for compiler optimization. Our code and datasets are publicly available at https://github.com/Panhaolin2001/Compiler-R1.

cs.LG↗

arXiv2025

BYOS: Knowledge-driven Large Language Models Bring Your Own Operating System More Excellent

Operating system (OS) kernel tuning is a critical yet challenging problem for performance optimization, due to the large configuration space, complex interdependencies among configuration options, and the rapid evolution of kernel versions. Recent work has explored large language models (LLMs) for automated kernel tuning, but existing approaches often suffer from hallucinated configurations, limited interpretability, and poor robustness across workloads and kernel versions. We propose BYOS, a knowledge-driven framework that grounds LLM-based Linux kernel tuning in structured domain knowledge. BYOS incorporates three key components: (1) structured knowledge construction and mapping to bridge the semantic gap, (2) knowledge-driven configuration generation to refine the search space, and (3) continuous knowledge maintenance to adapt to kernel evolution. We evaluate BYOS on diverse workloads across multiple Linux distributions and kernel versions. Experimental results show that BYOS consistently outperforms state-of-the-art tuning baselines, achieving 7.1%-155.4% performance improvement while substantially reducing invalid configurations. These results demonstrate the effectiveness of integrating structured knowledge with LLMs for robust and scalable system optimization. The code of BYOS is available at https://github.com/LHY-24/BYOS.

cs.OS↗

arXiv2025

KGCompiler: Deep Learning Compilation Optimization for Knowledge Graph Complex Logical Query Answering

Complex Logical Query Answering (CLQA) involves intricate multi-hop logical reasoning over large-scale and potentially incomplete Knowledge Graphs (KGs). Although existing CLQA algorithms achieve high accuracy in answering such queries, their reasoning time and memory usage scale significantly with the number of First-Order Logic (FOL) operators involved, creating serious challenges for practical deployment. In addition, current research primarily focuses on algorithm-level optimizations for CLQA tasks, often overlooking compiler-level optimizations, which can offer greater generality and scalability. To address these limitations, we introduce a Knowledge Graph Compiler, namely KGCompiler, the first deep learning compiler specifically designed for CLQA tasks. By incorporating KG-specific optimizations proposed in this paper, KGCompiler enhances the reasoning performance of CLQA algorithms without requiring additional manual modifications to their implementations. At the same time, it significantly reduces memory usage. Extensive experiments demonstrate that KGCompiler accelerates CLQA algorithms by factors ranging from 1.04x to 8.26x, with an average speedup of 3.71x. We also provide an interface to enable hands-on experience with KGCompiler.

cs.AI↗

arXiv2024

A Method for Efficient Heterogeneous Parallel Compilation: A Cryptography Case Study

In the era of diminishing returns from Moores Law, heterogeneous computing systems have emerged as a vital approach to enhance computational efficiency. This paper introduces a novel MLIR-based dialect, named hyper, designed to optimize data management and parallel computation across diverse hardware architectures. The hyper dialect abstracts the complexities of heterogeneous computing by providing a unified compilation framework that efficiently schedules tasks and manages data communication. To demonstrate its capabilities, we present HETOCompiler, a cryptography-focused compiler prototype that implements multiple hash algorithms and enables their execution on heterogeneous systems. The proposed approach achieves performance improvements over existing programming models for heterogeneous computing (OpenCL), offering an average speedup of 1.93x, 1.18x, and 1.12x for SHA-1, MD5, and SM3 algorithms, respectively. Our findings highlight the potential of the hyper dialect in harnessing the full computational power of heterogeneous devices, advancing the field of compiler design for heterogeneous systems.

cs.CR↗