rpcme / stm32f7-renode

About this project

This projects contains a set of artifacts to demonstrate microcontroller based IoT project CI/CD pipelines on AWS using AWS CodePipeline, AWS CodeCommit and GitHub, AWS CodeBuild, and AWS IoT Device Advisor. The focus of the project is to show how to use AWS IoT Device Advisor for Integration test as the Continuous Delivery gate.

Preparing AWS artifacts

Create AWS IoT credentials for AWS IoT Core. If you're using AWS IoT Device Advisor, you will use the same credentials but point the app to a suite-specific endpoint.

Use the -F flag if you have a lot of objects in your IoT Core registry. Otherwise... things could take quite a long time...

cd scripts
mkdir -p $(pwd)/../credentials
mydevice=$(uuidgen)
./create-iot-credential.sh \
    -F \
    -t ${mydevice} \
    -n awscsdk-mqtt-policy \
    -f $(pwd)/../configuration/awscsdk-iotcore-policy.json \
    -o $(pwd)/../credentials \
    -s da-credential-${mydevice}

when you have access to the secret, you can retrieve essential values for doing builds and tests:

aws secretsmanager get-secret-value \
    --secret-id ${my_secret_name} \
    --query SecretString \
    --output text | jq -r .privatekey | base64 -d

aws secretsmanager get-secret-value
    --secret-id ${my_secret_name} \
    --query SecretString \
    --output text | jq -r .certificate | base64 -d

aws secretsmanager get-secret-value \
    --secret-id ${my_secret_name} \
    --query SecretString \
    --output text | jq -r .iotcore_endpoint

Building and flashing the firmware

The firmware uses STM32Cube SDK, STLINK tools (for verifying on hardware target), Renode (for emulation), Arm GCC cross toolchain, make, and (of course) FreeRTOS and related libraries for RTOS, application logic, and AWS IoT connectivity. Although it is better to run the CI process in the cloud, the physical steps are here to help you first see how to build and flash the physical target so you can envisage how the firmware would run in an emulated environment.

Note the steps expect a Linux environment. The development was done on Ubuntu 20.04.

1. Clone the application repository.
2. Use repo to clone dependent repositories.

 repo init -u https://github.com/rpcme/stm32f7-renode

3. Download and unpack the Arm GCC toolchain.

   curl -v https://armkeil.blob.core.windows.net/developer/Files/downloads/gnu-rm/10.3-2021.10/gcc-arm-none-eabi-10.3-2021.10-x86_64-linux.tar.bz2 | tar xjf -

4. Install STLINK tools. Although the flash step below shows you how to flash, you might want to be familiar with all the available tools in the package. Visit https://github.com/stlink-org/stlink for more information.

   sudo apt-get install stlink-tools

5. Remove the driver header file from the STMicro Cube SDK. FreeRTOS+TCP provides the same. Not removing the file will cause a stellar failure.

   rm STM32CubeF7/Drivers/STM32F7xx_HAL_Driver/Inc/stm32f7xx_hal_eth.h

6. Set PATH to include the Arm GCC toolchain.

7. Change directory to the application repository.

8. Build the application.

   make

9. Flash the application. Start by connecting the STM32746-Discovery board and then invoke the following command.

   st-flash --reset write bin/stm32f746-uart.bin 0x8000000

openocd -f board/stm32f7discovery.cfg

monitor reset halt

target remote localhost:3333