Open Source Summit + Embedded Linux Conference + OSPOCon 2021

This panel is intended to discuss the challenges expected to remain with Linux being used in industrial applications in the coming months and those actively improving. The participants will contribute their experiences from working directly with industrial leaders in their fields. The talk will focus on past and current experiences as they shape the current efforts to improve adoption of Linux and Open Source software in general for industrial use cases. While the conversations will primarily be on technical aspects, the compliance and conformance aspects can not be ignored. The panel will look forward to taking questions from the audience and hope for a participative interaction with attendees.

This talk gives an overview of several new features in the networking driver world and how your Ethernet or CAN driver can benefit from it. In most cases there is HW support, but old network drivers implement only a limited amount of functionality: PHY support in Ethernet drivers is hard coded (AX88772) or switches aren't connected to the Kernel's switch framework (AR9331). Today PHYs (gigabit or single pair ETH) need special handling, hard coding PHY support is not an option. So let's convert Ethernet drivers to make use of the phylib as an Abstraction Layer. This brings access to ready to use PHY drivers and new functionalities, like diagnostic, self testing, cable testing, and workarounds for PHY specific errata. The functionality of Ethernet switches perfectly fit under the DSA framework. Which is the kernel's abstraction for different types of switching HW offloading capabilities. Another nifty feature to add is RX/TX hardware timestamping that modern MACs support, to increases diagnostic capabilities. Devices can benefit from lower latencies with Byte Queue Limit support, those connected with slow buses (USB, SPI) can combine outgoing network packet into one transaction. The talk will briefly mention XDP, what it does and if your ETH and CAN driver can benefit from it.

Today, organizations are using robotics to address a host of business challenges, from the self-driving car to autonomous walkers to assist older adults, exploring various environments from deep oceans to other planets like Mars. In the past, the integration of these robots took a significant amount of time and effort, and it required specialized expertise in this field. Still, this scenario has dramatically changed thanks to adopting a real-time production system with Linux and the Robot Operating System (ROS). ROS is an open-source software framework for robot development, including middleware, drivers, libraries, tools, and commonly used algorithms for robotics. In this session, we walk the audience through the steps from design to deployment robots using ROS2 Foxy (new version of ROS) from zero to hero using live demo using Python 3 (rclpy) with DDS (Data Distribution Service) simulating real-world environments with Gazebo (open-source 3D robotics simulator). In a nutshell, I will cover designing, developing, testing, and deploying intelligent robotics applications at scale, including integration with critical components, and discuss models that allow for optimized large fleet management.

Static partitioning is becoming increasingly common in embedded. A static hypervisor, such as Xen dom0less, is employed to split the hardware resources into multiple domains and run a different OS in each domain. For instance, Linux and Zephyr. Only the simplest static partitioning configurations don't involve any data exchanges between the domains. Often, communication and data exchanges between two or more environments are required to complete the data processing pipeline that implements the target application. However, the VM-to-VM communication mechanisms available in static partitioning configurations are typically more limited compared to general-purpose hypervisors. For example, PV drivers are not available to Xen dom0less domains. This presentation will discuss the need for communication in static partitioning setups and it will present the technical challenges involved in getting traditional communication methods to work, including Xen PV drivers and VirtIO. The talk will also provide simpler alternatives based on shared memory and interrupt notifications to set up domain-to-domain data streams: simpler techniques that are easily exploitable both by Linux and by tiny baremetal applications as well.

This talk will explore the future of Linux on RISC-V, an open instruction set (ISA).  I walk through the pieces needed to boot Linux on RISC-V including the Privileged Architecture, SBI, OpenSBI and U-Boot.  I will explore how to run a complete system including binary distributions and how to build from source yourself.  I will look at what Linux-capable "hard" RISC-V SoC's currently exist and those that are on the horizon. I will also talk about how support in Linux for RISC-V is continuing to evolve such as the introduction of KVM RISC-V support and look at RISC-V hardware support that is in the process of being upstreamed. I will describe how to participate in the creation of RISC-V specifications, and how the new RISC-V Platform Specification is trying to standardize boot and runtime requirements.  Google Slides: https://tinyurl.com/rv-linux-21

RISC-V is an open standard instruction set architecture which is becoming a popular choice for new hardware designs, ranging from low-key ones primarily targeting IoT to high performance multicore SoC capable of running data centers. And modern times RISC-V SoCs quite often have QSPI flash onboard which makes them perfect candidates to use XIP technology. XIP stands for eXecute In Place – a technology that allows code to be executed directly from flash without copying the code to RAM first.That allows to optimize memory footprint very tightly and thus opens up to really low-power IoT Linux appliances. XIP support for 64-bit RISC-V targets (developed by the author of this talk) has recently been accepted into the mainline, so as a part of this talk we'll present a demo how to run a mainline kernel configured for XIP on a RISC-V board. Another part of the talk will cover extending XIP support for RISC-V to 32-bit and MMU-less systems and the opportunities this will open up, especially in the area of low-key battery-powered RISC-V systems with RAM shortage.

There are a variety of approaches to leveraging Linux in real-time and safety-critical applications, such as those required for planes, trains, automobiles, and robots. Some approaches utilize tick-less PREEMPT_RT kernels that isolate cores for user-level processes, some introduce user-level or virtualization-based unikernels, some leverage Linux CPU hot-plug features to offload cores to auxiliary runtimes or bare metal applications, and some defer to secondary CPU clusters and run realtime or safety workloads on compute islands as found in heterogeneous SoCs. This technical presentation covers emerging “hypervisor-less virtio” technology and its resultant unifying system architecture for the sharing of resources, such as files, tty, IPC, network interfaces and others, whether between processes, kernels, cores and/or CPU clusters - with Linux as the virtio backend for all the approaches mentioned above. Although this talk goes deep to cover the low-level details and present real-world performance results of important hypervisor-less virtio use cases, it is also broadly relevant to computer scientists and technology leaders who create, deliver, and capture embedded software value using a Linux-first approach to real-time and/or safety-based applications.