Qihe: A General-Purpose Static Analysis Framework for Verilog
Qinlin Chen, Nairen Zhang, Jinpeng Wang, Jiacai Cui, Tian Tan, Xiaoxing Ma, Chang Xu, Jian Lu, Yue Li · 2026
Read Qihe: A General-Purpose Static Analysis Framework for Verilog by Qinlin Chen, Nairen Zhang, Jinpeng Wang, Jiacai Cui, Tian Tan, Xiaoxing Ma, Chang Xu, Jian Lu, Yue Li (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-11408.txt.
Introduction
Qinlin Chen and eight coauthors build a static analysis frame for Verilog. Their 2026 preprint names the framework Qihe. Verilog is the tongue used to describe hardware designs. Static analysis reads code without running it at all.
In software, static analysis has thrived for decades. It finds bugs, checks security, and aids understanding of code. General frameworks supply the base that such tools share. Hardware code has lacked a like base for Verilog work. Qihe aims to fill that gap for chip design teams. A shared frame cuts the cost of each new check. It could bring software-grade tooling to hardware design.
Catalog record
| CREATOR | Qinlin Chen, Nairen Zhang, Jinpeng Wang, Jiacai Cui, Tian Tan, Xiaoxing Ma, Chang Xu, Jian Lu, Yue Li |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 122 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-11408
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Qihe: A General-Purpose Static Analysis Framework for Verilog Qinlin Chen, Nairen Zhang, Jinpeng Wang, Jiacai Cui, Tian Tan, Xiaoxing Ma, Chang Xu, Jian Lu, Yue Li Abstract . In the past decades, static analysis has thrived in software, facilitating applications in bug detection, security, and program understanding. These advanced analyses are largely underpinned by general-purpose static analysis frameworks, which offer essential infrastructure to streamline their development. The generality of these frameworks is often founded on components such as a front end for program conversion, an intermediate representation (IR), and a suite of fundamental analyses providing key program information. Soot exemplifies such a framework in Java, sparking thousands of analyses and impacting both academia and industry. Conversely, hardware lacks such a framework, which may limit the exploration of diverse static analysis applications in this field. This gap could overshadow the promising opportunities for sophisticated static analysis in hardware, hindering achievements akin to those witnessed in software. To advance static analysis in hardware, we introduce Qihe, the first general-purpose static analysis framework for Verilog—a highly challenging endeavor given the absence of precedents in hardware. Qihe features an analysis-oriented front end, a Verilog-specific IR, and a suite of diverse fundamental analyses that capture essential hardware-specific characteristics—such as bit-vector arithmetic, register synchronization, and digital component concurrency—and enable the examination of intricate hardware data and control flows. These fundamental analyses are specifically designed to support a wide array of analysis clients for hardware. To validate Qihe’s utility, we further developed a series of analysis clients spanning bug detection, security, and program understanding, and applied them to real-world hardware projects. Unlike software, real-world hardware projects tend to contain fewer but harder-to-detect bugs, as they typically undergo extensive simulation and rigorous verification to prevent the prohibitive costs of hardware defects. Despite this, our preliminary experimental results are highly promising; for example, Qihe uncovered 9 previously unknown bugs in popular real-world hardware projects (averaging 1.5K+ GitHub stars), all of which were subsequently confirmed by developers; moreover, Qihe successfully identified 18 bugs beyond the capabilities of existing static analyses for Verilog bug detection (i.e., linters), and detected 16 vulnerabilities in real-world hardware programs. These results underscore the transformative potential of static analysis in hardware design. By open-sourcing Qihe 1 1 1 https://qihe.pascal-lab.net , which comprises over 100K lines of code, together with its carefully curated and comprehensive documentation 2 2 2 https://qihe-docs.pascal-lab.net , we aim to inspire further innovation and applications of sophisticated static analysis for hardware, aspiring to foster a similarly vibrant ecosystem that software analysis enjoys. 1. Introduction In the software domain, static analysis is widely recognized as an effective technique for enhancing software quality. Over the years, numerous sophisticated static analyses have been developed to achieve various application tasks, such as detecting bugs ( 54 ) , identifying security vulnerabilities ( 3 ) , and aiding in program understanding ( 45 ) . Despite its proven successes in software, static analysis remains underutilized in hardware, particularly with Verilog, the predominant hardware description language (HDL) ( 25 ) . The current situation mirrors the early days of software static analysis, where its applications were limited to compiler-integrated optimizations and basic syntax checks (e.g., early linting tools), with more sophisticated software analyses, such as those for bug detection and security, yet to emerge. In hardware, static analysis techniques are primarily adopted by hardware synthesizers (e.g., Yosys ( 78 ) ) for compiler optimization purposes; linting tools ( 56 ; 12 ; 28 ; 65 ) are popularly used to enforce code-style or syntactic checks but lack the sophistication needed to detect deeper design flaws like improper register resets and hardware Trojans. We argue that one of the main reasons for the stagnation of hardware static analysis is the lack of a general-purpose framework that provides the necessary infrastructure to enable the development of sophisticated analyses. In the software domain, such frameworks have proven transformative. For example, Soot ( 73 ) , a general-purpose static analysis framework for Java, has catalyzed thousands of analysis works in both academia and industry. Research efforts like FlowDroid ( 3 ) , widely used for detecting security vulnerabilities in Android apps, rely on Soot’s fundamental analyses for identifying control and data flows statically. Some companies also leverage FlowDroid and Soot to develop their own analyses, which allows them to identify bugs early in the development cycle, leading to potentially significant cost savings. Drawing from the designs of existing frameworks for software analysis ( 73 ; 20 ; 69 ) , a general-purpose framework typically requires the close cooperation of the following key components: (1) fundamental analyses that provide basic program information used across various application-specific analyses, (2) a front end and an IR that derive an analysis-oriented program abstraction from source code, and (3) an analysis manager that facilitates the reuse of results from existing analyses and the integration of new ones. However, given the absence of a general-purpose framework in hardware, the development of hardware static analysis typically relies on one of the following three kinds of options, each with its own limitations. Linters such as Slang ( 56 ) , Verible ( 12 ) , SVLint ( 28 ) , and Verilator-Lint ( 65 ) are confined to syntax- or pattern-based analyses on abstract syntax trees (ASTs). Tools like Pyverilog ( 68 ) and VeriPy ( 57 ) provide hardware abstractions beyond ASTs but lack the infrastructure to address critical hardware features like synchronization and concurrency, highly limiting their utility for diverse analysis needs. While CIRCT ( 42 ) , a comprehensive hardware compiler framework primarily used for synthesis, can also be adapted for static analyses, it lacks the fundamental facilities necessary for building sophisticated analyses that require intricate value flows and deep semantic reasoning, resulting in substantial development costs. In this paper, we present Qihe, the first general-purpose static analysis framework for Verilog, comprising three key components: • A Set of Fundamental Analyses (Section 3 ): We have developed 22 fundamental analyses to statically construct the hardware program information necessary to support a wide array of analysis clients. This unique contribution stems from our decade-long experience in frontline research and development in foundational static software analysis, sustained communication with industrial hardware practitioners, and an in-depth understanding of Verilog semantics. • An Analysis-Oriented IR and Front End (Section 4 ): We have designed the IR and front end to work collaboratively, meeting specific analysis needs such as supporting analysis for incomplete hardware programs and retaining additional source information, while still ensuring the IR remains simple for ease of analysis. • An Analysis Manager (Section 5 ): We have designed an engineering-focused mechanism that enables new developers to conveniently integrate and execute new analyses within Qihe. To validate Qihe as a general-purpose framework capable of supporting a variety of application-specific analyses, we developed 20 client analyses, which also serve as examples to guide future developers in creating their own. These analyses target common but non-trivial needs identified through surveys on hardware bugs and vulnerabilities ( 48 ; 51 ; 14 ; 15 ; 58 ; 63 ) along with insights from industrial hardware practitioners, spanning diverse application domains, including bug detection, security analysis, and program understanding. We conducted experiments by applying these