Embedded, Mobile and Automotive
Over-the-air (OTA) software updates with NVIDIA Jetson
<p>With over half a million developers worldwide, NVIDIA® JetsonTM is one of the most popular embedded computing platforms that brings AI performance to the Edge in a power-efficient and compact form factor</p>
<p>In this talk, we will cover in detail a unique way of integrating a robust and secure over-the-air software (OTA) updates with the NVIDIA JetPack 4.6 production release that works for all Jetson System-on-Modules (SoMs) with support for the official L4T (Ubuntu-based) operating system, running JetPack 4.6 or later. As it is built on top of the latest JetPack SDK release, there is no need to have separate integrations when building and deploying applications on different Jetson SoMs. This has the benefit of enormously cutting down the bringup time of an OTA solution to do all types of OTA updates. It has low maintenance costs since it does not involve modifying the boot process and rather relying on officially supported tools by NVIDIA which should not “break” with new L4T releases or board changes. We will cover the details of the integration with the high-level user flow as outline below, which “clones” the running image of a golden device and deploys it in a robust manner to the entire fleet of devices:</p>
<p>-Download L4T image from NVIDIA
-Use NVIDIA tools to flash A/B setup with L4T image
-Install Mender as deb package, plus integration layer
-Run Mender snapshot to create Artifact
-Deploy Artifact snapshot to fleet</p>
With over half a million developers worldwide, NVIDIA® JetsonTM is one of the most popular embedded computing platforms that brings AI performance to the Edge in a power-efficient and compact form factor
In this talk, we will cover in detail a unique way of integrating a robust and secure over-the-air software (OTA) updates with the NVIDIA JetPack 4.6 production release that works for all Jetson System-on-Modules (SoMs) with support for the official L4T (Ubuntu-based) operating system, running JetPack 4.6 or later. As it is built on top of the latest JetPack SDK release, there is no need to have separate integrations when building and deploying applications on different Jetson SoMs. This has the benefit of enormously cutting down the bringup time of an OTA solution to do all types of OTA updates. It has low maintenance costs since it does not involve modifying the boot process and rather relying on officially supported tools by NVIDIA which should not “break” with new L4T releases or board changes.
We will cover the details of the integration with the high-level user flow as outline below, which “clones” the running image of a golden device and deploys it in a robust manner to the entire fleet of devices:
-Download L4T image from NVIDIA
-Use NVIDIA tools to flash A/B setup with L4T image
-Install Mender as deb package, plus integration layer
-Run Mender snapshot to create Artifact
-Deploy Artifact snapshot to fleet
Additional information
| Type | devroom |
|---|
More sessions
| 2/5/22 |
<p>Once you start out in embedded linux, there is a lot to do. Some things are obvious, some less so. First and foremost, you can put your knowledge of using Linux on the desktop to good use, right? This approach feels like a natural progression, but it has its pitfalls - some of which this talk aims to help you understand and ultimately avoid.</p>
|
| 2/5/22 |
<p>DirectFB2 is a fork of DirectFB, a graphics library designed with embedded systems in mind that was widely used in the GNU/Linux embedded world. DirectFB2 comes with changes such as a Meson build system, a pure C implementation and a modularization of the source code.</p> <p>Access to the low-level display is based on a DRM/KMS system module (or possibly on a legacy Framebuffer system module), and depending on the platform, hardware-accelerated graphics rendering can be achieved using the ...
|
| 2/5/22 |
<p>I would like to share my experience bringing up various Automotive Ethernet Gigabit PHYs on an iMX8 platform.</p> <p>Agenda:</p> <ul> <li>PHY configuration CheckList (= What I need to know about my PHY, my schematic before starting the bring-up)</li> <li>SW implementation (= step-by-step SW integration + common pitfalls to avoid, mainly focused on Linux, but I talk also about U-Boot)</li> <li>Debug tips (= SW and HW tips)</li> </ul>
|
| 2/5/22 |
<p>RAUC is a safe and secure open source software solution for A/B updates of embedded Linux devices. RAUC supports industry-leading build system: the Yocto Project and OpenEmbedded, Buildroot and PTXdist. Porting RAUC to a new device requires several advanced technical steps. Layer meta-rauc-community exists to speed up and simplify the integration process for Yocto and OpenEmbedded by providing examples for popular devices such as Rasperry Pi, Allwinner (Sunxi), NVIDIA Tegra and QEMU.</p>
|
| 2/5/22 |
<p>In this talk, I will show you how to build and deploy ply (a dynamic tracing language for eBPF) to a BeagleBone Black then write ply scripts to attach probes and tracepoints to a running kernel and application on that same target. eBPF has rapidly eclipsed all previous tracers for Linux. While eBPF has taken the cloud native community by storm, the technology has yet to make significant inroads within the embedded Linux ecosystem. I will explain the reasons for this current situation and ...
|
| 2/5/22 |
<p>In this talk, pidge will take a critical look at the places where meta-zephyr succeeds and fails in its original goals, the reasons behind that and the steps being taken to fix those issues.</p>
|
| 2/5/22 |
<p>FPGAs are increasingly being used in today's embedded systems. But they are notoriously complex for having a difficult programming model. In order to counter this complexity, there has been a growing focus to design FPGA hardware at a higher level of abstraction with new languages and compilers. This talk will serve as "one stop shop" for topics related to these developments.</p>
|
