In my previous article “Zephyr RTOS Workspace Setup Cheat Sheet (west, SDK, Python, Build Tools)” I showed how to install a Zephyr workspace and the necessary dependencies. One step was to install the toolchain. And it was a simple ‘west sdk install’. Because the correct toolchain is a crucial part, it deserves a closer lookl
The Zephyr project has recently reached version 4.5, following its 10-year anniversary in March 2026. It has certainly come a long way and continues to improve with each release.
While there are plenty of other excellent resources about Zephyr, new users may be confused by outdated tutorials or simply overwhelmed by the amount of information available. This article serves as a cheat sheet to help you get started quickly.
In this article, I show how to install and get started with Zephyr. For this, I use the tinyK22 (NXP K22FN512) and NXP FRDM-MCXN947 board as an example.
In my earlier article All‑in‑One Docker Image for NXP MCUXpresso, VS Code and CI/CD Pipeline, I introduced a container image for streamlined embedded development; in this update, I have extended that image so it can install VS Code extensions, which makes it much more practical for working in restricted or offline environments. In this post I walk through what has changed, how the extension installation works inside the container, and how you can benefit from these improvements in your own workflows.
Working with different toolchains, different versions and targets can be very challenging. To simplify the setup and distribution of the development environment I’m using the concept of containers.
I’m using different Docker images and containers for development and automated CI/CD integration. In the past weeks I have extended, simplified and consolidated my work flows. For example I have now an image with the MCUXpresso tools, with everything running in a container: both locally on my machine with Docker, with VS Code, and the same integrated in the CI/CD pipeline:
Running MCUXpresso Tools in Container
It not only simplifies my development flow: the same way I can easily deploy the installation and setup for a class room lab environment: basically everything is only one click away.
The latest NXP MCUXpresso VS Code extension and LinkServer software included a small but very helpful feature for me: the ability to identify a connected probe with a blinky:
Setting up the correct tools for a development environment can be complex and time consuming. I’m using DevContainers with Docker for almost all of my projects. Because that way I get an isolated development environment. For this I spend time to craft dedicated docker files, to have everything installed correctly with the the right versions.
Outside a docker or DevContainer environment, I’m using the NXP MCUXpresso installer. It is a standalone application which can be used from VS Code to check and install all the necessary tools.
MCUXpresso Installer
In this article, I show how to have the NXP MCUXpresso Installer installed in a DevContainer so it can be used with VS Code.
At the same time, I’m using the cortex-debug extension for other ARM Cortex-M based MCUs, for example the Raspberry Pi Pico series.
Does it makes sense to use cortex-debug for NXP devices, and what would be the benefit? More precise: can I use the NXP MCU-Link (LinkServer based) debug probes with cortex-debug, to debug another NXP MCU?
NXP MCU-Link (right) debugging a NXP MCU-Link Pro (LPC55S69, right)Continue reading →
In my latest research project I’m using the NXP FRDM-MCXN947 board. As part of that project I have to collect and export application trace data for later analysis. In that project I’m using the the Tracealyzer product from Percepio.
Trace Data in Percepio Tracealyzer
In this article I describe different ways to export trace data from the embedded target: using generic GDB commands or with an trace exporter in VS Code.
For an university research project, I need to write some console output and the same time write a file with data from the embedded board to the host. A logical choice for this is using semihosting.