Mengda He
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Mengda He has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Shengchao Qin
Xidian University, Xi'an, China
5 shared publications - Cheng Wen
Guangzhou Institute of Technology, Xidian University, Guangzhou, China
2 shared publications - Huibiao Zhu
East China Normal University, Shanghai, China
2 shared publications - Lili Xiao
Donghua University, Shanghai, P. R. China
2 shared publications - Annalisa Occhipinti
Teesside University
1 shared publication - Bin Zhang
College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, China
1 shared publication
Latest Publications
- Open Access
- 7 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 4030-4041
Understanding Real-World Ordering-Related Concurrency Bugs under Weak-Memory and Cross-Architecture Settings
- 7 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 4030-4041
Open AccessUnderstanding Real-World Ordering-Related Concurrency Bugs under Weak-Memory and Cross-Architecture Settings
- 23 June 2025
- Published by World Scientific Pub Co Pte Ltd in Journal of Circuits, Systems and Computers
- Vol. 35 (01)
Specifying and Verifying Programs Over the MCA ARMv8 Architecture with TLA+
Modern hardware architectures and mainstream programming languages employ relaxed memory models for efficiency purposes. However, these memory models may bring in many behaviors which do not adhere to people’s intuition. In addition, different relaxed memory models produce different relaxed behaviors. The existing facts make the verification of the multi-threaded programs running against these models more difficult. In this paper, we are committed to proposing a general framework for modeling and verifying programs over various relaxed memory models, and we take the MCA ARMv8 architecture as an example. This architecture allows out-of-order execution through thread-local out-of-order, speculative execution and thread-local buffering. Above all, through analyzing the dependencies among statements, any program under MCA ARMv8 is translated into a unified form, which has the ability to describe each program under various memory models. Then, we model the translated programs with the formal specification language TLA[ ], and verify three properties, namely Reordering, Read-after-write elimination and Barriers, by the model checker TLC. Our verification results indicate that these properties align with the specification of MCA ARMv8. This not only validates the effectiveness of our method but also provides a more consistent approach for ensuring program correctness across a wide range of memory models.
- 23 June 2025
- Published by World Scientific Pub Co Pte Ltd in Journal of Circuits, Systems and Computers
- Vol. 35 (01)
Specifying and Verifying Programs Over the MCA ARMv8 Architecture with TLA+
Modern hardware architectures and mainstream programming languages employ relaxed memory models for efficiency purposes. However, these memory models may bring in many behaviors which do not adhere to people’s intuition. In addition, different relaxed memory models produce different relaxed behaviors. The existing facts make the verification of the multi-threaded programs running against these models more difficult. In this paper, we are committed to proposing a general framework for modeling and verifying programs over various relaxed memory models, and we take the MCA ARMv8 architecture as an example. This architecture allows out-of-order execution through thread-local out-of-order, speculative execution and thread-local buffering. Above all, through analyzing the dependencies among statements, any program under MCA ARMv8 is translated into a unified form, which has the ability to describe each program under various memory models. Then, we model the translated programs with the formal specification language TLA[ ], and verify three properties, namely Reordering, Read-after-write elimination and Barriers, by the model checker TLC. Our verification results indicate that these properties align with the specification of MCA ARMv8. This not only validates the effectiveness of our method but also provides a more consistent approach for ensuring program correctness across a wide range of memory models.
- Open Access
- 21 December 2024
- Published by Elsevier in Expert Systems with Applications
- Vol. 268
Enhancing deep learning for demand forecasting to address large data gaps
- 21 December 2024
- Published by Elsevier in Expert Systems with Applications
- Vol. 268
Open AccessEnhancing deep learning for demand forecasting to address large data gaps
- Open Access
- 5 September 2024
- Published by Association for Computing Machinery (ACM) in Formal Aspects of Computing
- Vol. 36 (3) , 1-24
Trace Semantics for C++11 Memory Model
The C and C++ languages introduced the relaxed-memory concurrency into the language specification for efficiency purposes in 2011. Trace semantics can provide the mathematical foundation for the proposed C++11 memory model, and there is a lack of investigation of trace semantics for C++11. The Promising Semantics (PS) of Kang et al. provides the standard SC-style operational semantics for the C++11 concurrency model, where “SC” refers to “Sequential Consistency”. Inspired by PS, in this article we first investigate the trace semantics for the relaxed read and write accesses under C++11, acting in the denotational semantics style. In our semantic model, a trace is in the form of a sequence of snapshots, and the snapshots record the modification in the relevant global or local variables, and the thread view. Moreover, the trace semantics for the release/acquire accesses under C++11 is also explored, based on the separated thread views and newly added message views. When considering this trace model, different accesses bring in their unique snapshots, and make distinguished effects on the production of the sequences. For any given program, the proposed trace semantics in this article produces all the valid traces directly. Furthermore, our trace semantics, together with that for TSO and MCA ARMv8, has the possibility to be the foundation of the meta model of the trace semantics for weak memory models.
- 5 September 2024
- Published by Association for Computing Machinery (ACM) in Formal Aspects of Computing
- Vol. 36 (3) , 1-24
Open AccessTrace Semantics for C++11 Memory Model
The C and C++ languages introduced the relaxed-memory concurrency into the language specification for efficiency purposes in 2011. Trace semantics can provide the mathematical foundation for the proposed C++11 memory model, and there is a lack of investigation of trace semantics for C++11. The Promising Semantics (PS) of Kang et al. provides the standard SC-style operational semantics for the C++11 concurrency model, where “SC” refers to “Sequential Consistency”. Inspired by PS, in this article we first investigate the trace semantics for the relaxed read and write accesses under C++11, acting in the denotational semantics style. In our semantic model, a trace is in the form of a sequence of snapshots, and the snapshots record the modification in the relevant global or local variables, and the thread view. Moreover, the trace semantics for the release/acquire accesses under C++11 is also explored, based on the separated thread views and newly added message views. When considering this trace model, different accesses bring in their unique snapshots, and make distinguished effects on the production of the sequences. For any given program, the proposed trace semantics in this article produces all the valid traces directly. Furthermore, our trace semantics, together with that for TSO and MCA ARMv8, has the possibility to be the foundation of the meta model of the trace semantics for weak memory models.
- Open Access
- 12 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-13
RPG: Rust Library Fuzzing with Pool-based Fuzz Target Generation and Generic Support
Rust libraries are ubiquitous in Rust-based software development. Guaranteeing their correctness and reliability requires thorough analysis and testing. Fuzzing is a popular bug-finding solution, yet it requires writing fuzz targets for libraries. Recently, some automatic fuzz target generation methods have been proposed. However, two challenges remain: (1) how to generate diverse API sequences that prioritize unsafe code and interactions to reveal bugs in Rust libraries; (2) how to provide support for the generic APIs and verify both syntactic and semantic validity of the fuzz targets to enable more comprehensive testing of Rust libraries. In this paper, we propose RPG, an automatic fuzz target synthesis technique to support Rust library fuzzing. RPG uses a pool-based search to generate diverse and unsafe API sequences, and synthesizes fuzz targets with generic support and validity check. The experimental results demonstrate that RPG enhances both the quality of the generated fuzz targets and the bug-finding ability through pool-based generation and generic support, substantially outperforming the state-of-the-art. Moreover, RPG has discovered 25 previously unknown bugs from 50 well-known Rust libraries available on Crates.io.
- 12 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-13
Open AccessRPG: Rust Library Fuzzing with Pool-based Fuzz Target Generation and Generic Support
Rust libraries are ubiquitous in Rust-based software development. Guaranteeing their correctness and reliability requires thorough analysis and testing. Fuzzing is a popular bug-finding solution, yet it requires writing fuzz targets for libraries. Recently, some automatic fuzz target generation methods have been proposed. However, two challenges remain: (1) how to generate diverse API sequences that prioritize unsafe code and interactions to reveal bugs in Rust libraries; (2) how to provide support for the generic APIs and verify both syntactic and semantic validity of the fuzz targets to enable more comprehensive testing of Rust libraries. In this paper, we propose RPG, an automatic fuzz target synthesis technique to support Rust library fuzzing. RPG uses a pool-based search to generate diverse and unsafe API sequences, and synthesizes fuzz targets with generic support and validity check. The experimental results demonstrate that RPG enhances both the quality of the generated fuzz targets and the bug-finding ability through pool-based generation and generic support, substantially outperforming the state-of-the-art. Moreover, RPG has discovered 25 previously unknown bugs from 50 well-known Rust libraries available on Crates.io.