published the project (SOF), originally created by Intel, to avoid the supply of closed firmware for DSP chips related to sound processing. Subsequently, the project was transferred under the wing of the Linux Foundation organization and is now being developed with the involvement of the community and with the participation of AMD, Google and NXP companies. The project develops SDK to simplify the development of firmware, a sound driver for the Linux nucleus and a set of finished firmware for various DSP chips for which binary assemblies are also formed, certified by digital signature. The firmware code is written in the SI language with assembler inserts and is distributed under the license BSD. On various DSP architectures and hardware platforms. For example, among the supported platforms stated various chips of Intel (Boadwell, Icelake, Tigerlake, Alderlake, etc. .), MediaTek (MT8195), NXP (I.MX8*) and AMD (Renoir),
component DSP based on architectures xtensa hifi 2, 3 and 4 . In the process of development, a special emulator or Qemu can be used. The use of open firmware for DSP allows you to more quickly correct and diagnose problems in the firmware, and also gives users the opportunity to independently adapt firmware for their needs, make specific optimization and create lightweight firmware options containing only the functionality necessary in the product.
The project is provided with a framework for the development, optimization and testing of solutions related to sound processing, as well as the creation of drivers and programs for interaction with DSP. The composition includes the implementation of firmware, tools for testing firmware, utilities for converting ELF files into flavor images suitable for installation on equipment, debuging tools, an emulator DSP, an emulator host-platform (based on Qemu), tools for tracing firmware, scripts for MATLAB/Octave for fine adjustment of the coefficients for sound components, applications for organizing interaction and exchanging data with firmware, ready -made examples of sound processing topologies.
