Why each project in the I-CUBE-LRWAN expansion Package are not provided with .ioc file for an easy customisation to a different board !!! ST complicate life and impossible to find the solution to well do the job. please help!!!

[SHATE.1]

In the UM2285 Developpment Guideline for STM32 expansion Packagethey say that each expansion Package must contain the .ioc file but in the i-cube-lrwan not the case!!!!! why?

Thanks for help

[Tesla DeLorean]

It was built around the Semtech libraries, not ST's models, by actual developers, not a code generator.

If you want to integrate stuff, do it at the HAL level.

[SHATE.1]

Thank you for your quick answer

But How i can do it at the HAL level???? You know i'm not specialised in the soft development, i'm a Hardware engineer and i developped a custom board with STM32L476 and a Lora Module with the SX1262 semtech radio. in the i-cube-lrwan provided by ST there's an end node project around the STM32L476 but i want to configure the pins and it's very complicated because there's no .ioc file to do it graphically. i used STM32CubeMX to do that in the normal case and make my program after ( there's a good Hands-on explaining that on youtube) but an example of implementation of the LORAWAN stack in a custom board there's nothing!!

Thank you in advance

[Tesla DeLorean]

I don't know your background, I'm an EE, most of my class mates could code.

The CubeL4 repository should have example code/projects. You should be able to build code to drive your peripherals with CubeMX, and between the examples, and mechanically generated code you'll need to extract the HAL based functionality.

For the radio stuff, you're probably going to want find a close proxy, and then port the pin level specifics to your board. Perhaps clone an existing project, and then modify to match your hardware. Get the radio side working, then merge in the peripheral code.

Not sure there is a simple graphical method, you might need to pull an Embedded SW Engineer into the project.

[SHATE.1]

maybe to well understand the problem, you can make a new project with cubeMX using the STM32L476 for example and configure your pins and then download the i-cube-lrwan from st.com and try to port the lorawan stack into your project made by stm32cubeMX and you tell me the result!!

surely the ideal is to have an embedded sw engineer but i have not and that's for this reason that i'm here to look for help from st employer or a generous community member like you because you already answered me.

thanks

[Imen.D · Accepted solution]

Hello @SHATE.1​ ,

I-CUBE-LRWAN expansion Package was designed and its first release published before the guidelines you mentioned were enforced, and probably even before UM2285 document was written.

Its legacy structure is not compatible with STM32CubeMX and our teams are working on this to make a new version that can be integrated in a pack for STM32CubeMX.

Your request is raised internally and will be taken into account. But, currently, there is no short term planning defined yet to integrate .ioc file in the package.

Hope this answered your question.

When your question is answered, please close this topic by choosing Select as Best.

Imen

[Tesla DeLorean]

I understand the scope and complexity of the problem, but I'm not using CubeMX to drive my development. What you propose is a multi-hour task.

Not sure ST staff here will assist you, you're best hope for that would be via the ST FAE (Support Engineer) assigned to your business account out of the local sales office. I'd suspect the radio guys also aren't using CubeMX either, it is an encumbrance, and gets in the way.

The best way forward is to decompose the project into small parts, where you get your sensor and non-LoRa stuff functional, and back port that into a functioning LoRa project for your hardware. Most of the hardware specific stuff for LoRa is in stm32l4xx_hw_conf.h