Getting Started with HydraBus and STM32CubeIDE

Getting Started with HydraBus and STM32CubeIDE

The aim of this page is to explain How to Build / Flash and Debug HydraBus firmware with STM32CubeIDE on Windows & Linux.

1) Hardware requirements

• HydraBus v1
• STLINK-V3SET (Programmer/Debugger)
• 10x high quality Dupont Wire 2.54mm female to female with min length 8cm (do not exceed 10cm if possible especially to reach maximum speed with SWD 24MHz...)
  • Only 9x "Dupont Wire 2.54mm female to female" are required but it is always better to have more/spare ...

2) Connect STLINK-V3SET (Programmer/Debugger) to HydraBus v1

2-1) Connect STLINK-V3SET SWD to HydraBus v1 SWD (for basic debugging):

2-2) Connect STLINK-V3SET SWD / SWO & USART1(TX/RX) to HydraBus v1 (for advanced debugging/traces):

3) Software requirements

3-1) Install STM32CubeIDE v1.4 or more

• Download and install STM32CubeIDE v1.4 or more for Linux or Windows
  • Note1: STM32CubeIDE is free but it requires a registration on ST website to download it
  • Note2: All steps in this page have been validated with STM32CubeIDE v1.4.2 and using firmware https://github.com/hydrabus/hydrafw/commit/0099a031d02c396ab06968420faf9eb08ecd2348

3-2) Install developer tools and clone project repository

• Official compiler for hydrafw v0.8beta release is GCC ARM 4.9 2015q3(GNU_ARM_4_9_2015q3)

• WARNING: GCC ARM 4.8.x & GCC ARM 5.0/5-2016-q1 shall be avoided as they crash during link of chibios ...

  • This issue shall be fixed with https://launchpad.net/gcc-arm-embedded/+milestone/8-2018-q4-major or more to be confirmed

• Since 8600bd92a5ed083174e0867355f65a743dcfb9df, it is possible to use GCC 8.1 and newer to compile hydrafw

  3-2-1) For Windows

  3-2-1-1) Prerequisites for Windows

  • Install GCC toolchain GNU_ARM_4_9_2015q3 from https://launchpad.net/gcc-arm-embedded/+milestone/4.9-2015-q3-update
    • At end of Installation Install wizard Complete choose only Add path to environment variable
  • Install git from http://msysgit.github.io
  • Install Python 3.x see https://www.python.org/downloads/windows
  • Default Install Example: C:\Python3\
  • Add in environment variable PATH the path to Default Install: C:\Python3\
  • Open a Command Prompt window (cmd.exe) and type following commands:

  git clone https://github.com/hydrabus/hydrafw.git hydrafw --recursive
  cd hydrafw/
  cd ./scripts
  python get-pip.py
  python -m pip install GitPython --upgrade
  python -m pip install intelhex --upgrade
  

  Note: 
  For get-pip.py if you need a proxy for internet access set following variables before to launch
  python get-pip.py:
  set http_proxy=http://proxy.myproxy.com
  set https_proxy=https://proxy.myproxy.com
  

  3-2-1) For Linux

  3-2-1-1) Prerequisites for Linux

  cd ~
  sudo apt-get install git dfu-util python putty
  wget https://launchpad.net/gcc-arm-embedded/4.9/4.9-2015-q3-update/+download/gcc-arm-none-eabi-4_9-2015q3-20150921-linux.tar.bz2
  tar xjf gcc-arm-none-eabi-4_9-2015q3-20150921-linux.tar.bz2
  echo 'PATH=$PATH:~/gcc-arm-none-eabi-4_9-2015q3/bin' >> ~/.bashrc
  source ~/.bashrc
  git clone https://github.com/hydrabus/hydrafw.git hydrafw --recursive
  cd ~/hydrafw
  cd ./scripts
  sudo python get-pip.py
  sudo python -m pip install GitPython --upgrade
  sudo python -m pip install intelhex --allow-external intelhex --allow-unverified intelhex
  cd ..
  

  • For Debian / Ubuntu on 64bits/amd64 system see https://wiki.debian.org/Multiarch/HOWTO in order to enable multi architecture for i386 support.
  • When multiarch is enabled you shall also install libc6-i686:i386 (Debian) or libc6:i386 (Ubuntu)

  Note: 
  For get-pip.py if you need a proxy for internet access set following variables before to launch
  python get-pip.py:
  export http_proxy=http://proxy.myproxy.com
  export https_proxy=https://proxy.myproxy.com
  sudo -E python get-pip.py
  

4) Start STM32CubeIDE v1.4 and Import project

4-1) Start STM32CubeIDE v1.4

4-2) Import the project

• Select menu File -> Import...

  • Select General -> Existing Projects into Workspace

  • Click on Next

  • Unselect Option Search for nested projects

  • Now only hydrafw project is displayed

  • Click on Finish

4-3) Enable HYDRAFW_DEBUG option on the project Debug configuration

• Select hydrafw project (click on project with right mouse button)

  • A contextual menu appears, select Properties

  

  • Click on C/C++ Build => Environment and select configuration Debug
  • Add Variable HYDRAFW_DEBUG with value 1

5) Clean the project

• Select hydrafw project (click on project with right mouse button)

  • A contextual menu appears, select Clean Project

  • Clean shall display in Console Build Finished. 0 errors, 0 warnings.

6) Build the project

• Select hydrafw project (click on project with right mouse button)
  • A contextual menu appears, select Build Project
  • Build shall display in Console Build Finished. 0 errors

7) Flash & Debug the project

• Select hydrafw project (click on project with right mouse button)
  • A contextual menu appears, select Debug As -> Debug Configurations...
  • Debug Configuration view appears, double click on STM32 Cortex-M C/C++ Application
    • The entry hydrafw Debug is automatically created
• Select Debugger Tab and configure it like in picture
  • When all is configured click on Debug button
• Debug session is started and program is stopped in start of main() code (you can click on Resume (F8) to resume execution of code)

8) Advanced debugging with SWO and STM32CubeIDE Serial Wire Viewer(SWV)

• Select menu Window -> Show View -> Other... -> SVW to add all different Serial Wire Viewer features
  • Add SWV TraceLog, SWV ITM Data Console and SWV Data Trace Timeline Graph
  • Then select SWV TraceLog and click on Configure trace icon to configure SWV
  • Interesting things to configure are mainly to enable Timestamps and ITM Sitmulus Ports 0 (for SWO/SWV ITM Data Console especially for debug trace and crash/exception details)
  • Then click on OK button
  • Last step is to start the Trace(using Serial Wire Viewer settings) by clicking on Start Trace button and finally click on Resume (F8) button to see the trace and data in live

We can see in real-time the different trace and console data during execution of code