Table of Contents

How to Build, Flash and Debug the STM32 nanoBooter and nanoCLR on Windows using Visual Studio Code

⚠️ NOTE about the need to build .NET nanoFramework firmware ⚠️

You only need to build it if you plan to debug the CLR, interpreter, execution engine, drivers, add new targets or add new features at native level. If your goal is to code in C# you just have to flash your MCU with the appropriate firmware image using nanoff. There are available ready to flash firmware images for several targets, please check the Home repository.

About this document

This document describes how to build the required images for .NET nanoFramework firmware for STM32 targets. The build system is based on CMake tool to ease the development in all major platforms.

Using Dev Container

If you want a simple, efficient way, we can recommend you to use Dev Container to build your image. This has few requirements as well like Docker Desktop and Remote Container extension in VS Code but it is already all setup and ready to run!

If you prefer to install all the tools needed on your Windows machine, you should continue this tutorial.

Prerequisites

You'll need:

  • Visual Studio Code
  • Visual Studio Code Extensions . C/C++ - C/C++ IntelliSense, debugging, and code browsing (by Microsoft) . CMake Tools - Extended CMake support in Visual Studio Code (by Microsoft)
  • CMake (Minimum required version is 3.21)
  • A build system for CMake to generate the build files to. We recommend Ninja.
  • GNU ARM Embedded Toolchain
  • OpenOCD. We suggest Microsoft build which provides a Windows distribution including convenient fixes for working with Azure RTOS.
  • ChibiOS - Technically you do not need to download this, the build scripts will do this automatically if you do not specify a path to ChibiOS in the CHIBIOS_SOURCE_FOLDER build option (more info here).

All the the above can be installed by the Power Shell script .\install-nf-tools.ps1 -TargetSeries STM32 from the install-scripts folder within the nf-interpreter project (cloned or downloaded). If you prefer you can do it manually (NOT RECOMMENDED for obvious reasons).

Overview

To simplify: this guide we will put all our tools and source in easily accessible folders and not at the default install paths (you do not have to do the same).

  1. Create a directory structure such as the following:

    • C:\nftools
    • C:\nanoFramework

  2. Download and install Visual Studio Code.

  3. Clone nf-interpreter repository into C:\nanoFramework\nf-interpreter. See next section for more info.

  4. Run the PowerShell script that's on the install-scripts folder that will download and install all the required tools. .\install-nf-tools.ps1 -TargetSeries STM32 -Path 'C:\nftools' For best results, run in an elevated command prompt, otherwise setting system environment variables will fail.

  5. Review and adjust several JSON files to match your environment (as documented below)

  6. Restart Visual Studio Code (due to json changes)

The setup is a lot easier than it seems. The setup scripts do almost everything.

.NET nanoFramework GitHub repo

If you intend to change the nanoBooter or nanoCLR and create Pull Requests then you will need to fork the nanoFramework/nf-interpreter to your own GitHub repo and clone the forked GitHub repo to your Windows system using an Git client such as the GitHub Desktop application.

The main branch is the default working branch. When working on a fix or experimenting a new feature you should do it on its own branch. See the Contributing guide for specific instructions on the suggested contributing workflow.

If you don't intend to make changes to the nanoBooter and nanoCLR, you can just clone nanoFramework/nf-interpreter directly from GitHub.

Make sure to put this folder high enough on your drive, that you won't trigger long filename issues. CMake does not support filenames in excess of 250 characters.

Setting up the build environment

After cloning the repo, you need to setup the build environment. You can use the power shell script or follow the step-by-step instructions.

Automated Install of the build environment

Run Power Shell as an Administrator and run set-executionpolicy RemoteSigned to enable execution of the signed script.

On Windows, one may use the .\install-nf-tools.ps1 Power Shell script located in the repository install-scripts folder to download/install CMake, the toolchain, OpenOCD (for JTAG debugging) and Ninja. You may need to use Run as Administrator for power shell to permit installing modules to unzip the downloaded archives. The script will download the zips and installers into the repository zips folder and extract them into sub-folders of the nanoFramework tools folder C:\nftools or install the tool.

  1. Open Power Shell in the install-scripts folder of the repository and run the script.

Example Power Shell command line:

.\install-nf-tools.ps1 -TargetSeries STM32

You can force the environment variables to be updated by adding -Force to the command line.

The script will create the following sub-folders (see manual install below):

  • C:\nftools
  • C:\nftools\GNU_Tools_ARM_Embedded\8-2019-q3-update
  • C:\nftools\ninja
  • C:\nftools\hex2dfu
  • C:\nftools\openocd

The following Environment Variables will be created for the current Windows User.

  • NF_TOOLS_PATH = C:\nftools
  • GNU_GCC_TOOLCHAIN_PATH = C:\nftools\GNU_Tools_ARM_Embedded\8-2019-q3-update
  • HEX2DFU_PATH = C:\nftools\hex2dfu
  • NINJA_PATH = C:\nftools\ninja

Set up Visual Studio Code

  • Step 1: Install the extensions:

  • Step 2: Run the PowerShell script Initialize-VSCode.ps1 that's on the install-scripts folder. This will adjust the required settings, build launch configuration for debugging and setup the tasks to ease your developer work.

.\Initialize-VSCode.ps1

You can force the environment variables to be updated by adding -Force to the command line. The PowerShell relies on the environment variables described above to properly setup the various VS Code working files. In case you have not used the automated install and the variable are not available you'll have to manually edit tasks.json, launch.json and settings.json to replace the relevant paths.

  • Step 3: Copy the template file (in nf-interpreter\config folder) user-tools-repos.TEMPLATE.json to a (new) file called user-tools-repos.json. Rename the json section user-tools-repos-local to user-tools-repos and adjust paths for the tools and repositories in the user-tools-repos configuration preset. If you don't have the intention to build for a particular platform you can simply remove the related options from there. If you don't want to use local clones of the various repositories you can simply set those to null. !!mind to always use forward slashes in the paths!!

  • Step 4: Save any open files and exit VS Code.

Build nanoCLR

  • Step 1: Launch Visual Studio Code from the repository folder, or load it from the File menu, select Open Folder and browse to the repo folder. VS Code could prompt you asking "Would you like to configure this project?". Ignore the prompt as you need to select the build variant first.

  • Step 2: Reopen VS Code. It should load the workspace automatically. In the status bar at the bottom left, click on the No Configure Preset Selected and select the target you want to build from the drop-down list that will open at the top. The respective build preset will be automatically selected by VS Code. More details on this on the documentation about the available targets here.

choose-preset

  • Step 3: In the status bar click Build or hit F7.

  • Step 4: Wait for the build to finish with Build finished with exit code 0 output message.

  • Step 5: In the build folder you'll find several files:

    • nanoBooter.bin
    • nanoBooter.elf
    • nanoBooter.hex
    • nanoCLR.bin
    • nanoCLR.elf
    • nanoCLR.hex

Note: If there are errors during the build process it is possible to end up with a partial build in the build folder, and the CMake/Ninja build process declaring a successful build despite the .bin targets not being created, and a CMake clean not helping. In this case deleting the contents of the build folder should allow the build to complete once you resolve the issues that cause the original failure.

Common Build Issues

The above may have some errors if:

  • CMake is not installed properly, not in the PATH or cannot be found for some reason.
  • Ninja is not recognized: check settings.json or your PATH environment variable and restart Visual Studio Code.
  • COMPILATION object file not found: check that your paths don't exceed 140 chars. Put the solution folder high enough on drive.
  • Reopen VS Code if you have made changes on the CMakePresets.json or CMakeUserPresets.json.

A good remedy for most of the build issues is to manually clean the build folder by deleting it's contents and restarting VS Code.

Flash the STM32 target

There are two options to flash the nanoBooter & nanoCLR images to a target. The first one uses the C/C++ tools in VS Code along with OpenOCD. This is the way to do it if you're planning to debug the code. The second uses a stand alone tool from STM that just flashes the images into the target. Useful if you don't plan to do any debugging.

Starting a debug session in VS Code

  1. Assuming that you have a valid launch.json configuration for the target that you've build, you can go to the Run section.
  2. Choose the launch configuration for nanoBooter corresponding to your target.
  3. Click on the Start Debugging (green arrow). (this will flash the nanoBooter into the target's flash memory)
  4. Stop the debug session.
  5. Choose the launch configuration for nanoCLR corresponding to your target.
  6. Click on the Start Debugging (green arrow). (this will flash the nanoBooter into the target's flash memory)
  7. Stop the debug session.

Note: You don't have to re-flash nanoBooter every time you flash nanoCLR as it won't be erased.

Using STM32CubeProgrammer Utility

Follow the instructions here to install STM32CubeProgrammer and flash nanoBooter.hex and nanoCLR.hex images to the board. Also this is not mandatory as all this can be handled by VS Code.

Next Steps

See Getting Started for instructions on creating and running a 'Hello World' managed application on your nanoFramework board.

Edit this page