Building a Zephyr-Native Audio Framework With Sound Open Firmware
summary
Claims that Sound Open Firmware can run standalone on MCUs by replacing the Linux host with a Zephyr-based host and shows a working proof of concept. Implementation uses an RT600 Cortex‑M33 as the host and a Hi‑Fi4 DSP, with IPC implemented as a raw messaging unit with shared memory, pipeline defined as static C structures, audio pushed from a static PCM array, and firmware loaded and started by the Zephyr host. Work validates feasibility and lists next steps (device tree, shell, Kconfig).
AI-written summary · about
chapters (14)
- 0:00Introduction
- 1:01What is Sound Open Firmware
- 3:11SOF’s Linux limits
- 6:20SOF’s Zephyr modularity
- 8:31Problem: SOF’s Linux dependency
- 9:37Proposal: Zephyr for MCUs
- 10:41Zephyr SOF host details
- 11:45IPC mechanisms
- 14:00Topology and pipeline
- 15:06Audio playback flow
- 16:14Proof of concept on RT600
- 17:38Future work
- 18:43Key conclusions
- 20:07Q&A
related talks
27mBuilding a Zephyr-Native Audio Framework With Sound Open Firmware
33mEnabling Sound Open Firmware on Arm® Cortex®-a Based Devices Using Zephyr RTOS
36mSound Open Firmware
39mSoftware Defined Hardware on Heterogeneous SoCs Using OpenAMP
34mZephyr and OpenAMP
36mZephyr on a Hearing Aid
25mChallenging Hardware Contests With Zephyr and Lessons Learned
26mLE Audio – Specification and Zephyr Implementation