Linux Kernel v5.16 released, our contributions
Linux v5.16 marks the first kernel release of 2022. As usual, LWN covered the 5.16 merge window (part one, part two) and Kernel Newbies has a detailed breakdown of the features in the 5.16 release.
Here’s a summary of our contributions, organized by SoC family, and a summary graph of contributions by each developer.
We continued to maintain and improve support for Amlogic SoCs in the kernel. Jerome Brunet fixed an audio sync issue on the AXG SoC family where the FIFO may get out of sync with the TDM decoder. Neil Armstrong made it possible to reduce the kernel image size by building the Amlogic GPIO IRQ controller as a module.
We also continued to support MediaTek SoCs. Markus Schneider-Pargmann contributed a fix for runtime power management of the hardware RNG (random number generator).
Corentin Labbe improved support for Gemini based systems with the addition of device trees for two network attached storage (e.g. NAS) systems: the SSI 1328 with a SL3516 SoC and the Edimax NS2502 with a SL3516 SoC. Device tree improvements were also made for USB, IDE, flash and Ethernet peripherals as well as the addition of 2 new vendor prefixes.
Texas Instruments SoCs
We continue to upstream improvements for TI SoCs. Drew Fustini updated the AM3358-based PocketBeagle device tree to make use of a new feature in the OMAP GPIO driver that allows userspace libraries to set bias (e.g. pull up and pull down) on GPIO lines.
Pumpkin i350 Edge AI EVK now available
The Pumpkin i350 EVK Smart Hub Edition is now available on the new OLogic store! The board supports both Yocto and AOSP, developed in partnership by BayLibre and MediaTek:
The i350 EVK is an Edge AI platform designed for mainstream AI + IoT applications that require vision and voice edge processing, such as facial, object, gesture, motion recognition, LPR, voice activation and speed recognition, sound isolation, bio-tech and biometric measurements, and more. Built on the success of our high-end Pumpkin i500, the new i350 is a lower cost, mid-range performance platform that enables engineering teams to design for wider market adoption use cases such as door entry systems, personal gym trainers, child sleep companion robots, and much more.
Built using an ultra-efficient 14nm process, this highly integrated Edge AI platform incorporates a dedicated APU (AI processor) and DSP to enable vision and voice edge AI, with considerably greater performance and power efficiency in common applications.
With the advent of higher-performing AI-integrated chips, such as the MediaTek i350 SoC and development board designed by OLogic, we accelerate your engineering, shorten your development cycle, helping you to launch your new product faster at a lower development cost and faster ROI.