Enabling RISC-V CI in Open-Source Projects: Challenges and Solutions

The adoption of RISC-V as a viable architecture for open-source software development is gaining traction. However, a major challenge remains: ensuring continuous integration (CI) support for RISC-V in upstream projects. At Samsung R&D Institute Poland, we addressed this issue by enabling RISC-V CI for several Freedesktop.org (FDO) projects, including Pixman and GStreamer Orc, and we are currently extending our work to the Opus codec. This work presents our approach to enabling RISC-V CI in FDO projects, addressing the challenges of testing architecture-specific optimizations without native hardware support. We detail our implementation of Docker-based GitLab runners with QEMU emulation, enabling automated multi-architecture testing while minimizing infrastructure overhead. Our work not only enhances software quality by enabling automated testing for RISC-V but also provides a framework for future contributions to seamlessly integrate RISC-V into open-source CI ecosystems.