.png){kind=icon}
.png){kind=icon}
Return
P4Muse: Enabling Modular P4 Programming via Compiler-Managed Code Merging Without Syntax Modifications
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2025.3589353.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Domain-specific programming languages such as P4 enable flexible and high-performance packet processing for programming network data planes. However, many P4 programs remain monolithic, limiting the development of modular and reusable protocols and libraries. Introducing modularity to P4 has proven challenging, as existing approaches—such as trans-compilers and virtualization—often sidestep direct integration with the P4 language and compiler, constraining backward compatibility and extensibility. This paper introduces P4Muse (P4 Modularity and Unification for Seamless Extensibility), an open-source P4C compiler extension that enhances the modularity of P4 without requiring new syntax or annotations. P4Muse is developed by integrating new compiler passes for automatic code merging, fostering modular design and reuse. We demonstrate its benefits through three classes of use cases that support P4 modularity, enabling code reusability, data plane pipeline composition, and vendor-customer compatibility. using the V1Model architecture and the BMv2 software switch. Our results show that P4Muse effectively supports modular P4 program development without altering existing P4 syntax, providing a robust solution that significantly improves code reusability, flexibility, and extensibility while maintaining backward compatibility.
Keywords:
Programmable data planes
P4
compiler
intermediate representation
modularity
reusability
extensibility
incremental programming
composability
BMv2
vendor-customer framework
Journal
IF:
3.6
Papers:
9.7W
Citations:
29.4W