IP Address in Hardware Implementation (External Mode in Configuration Parameters) is not used for code generation

Hello community,

we developed our own toolchain for an Intel Cyclone V FPGA (MitySOM-5CSX) using this tutorial (https://de.mathworks.com/help/ecoder/target-sdk.html), to generate code from Simulink models using the Embedded Coder Support Package for ARM Cortex-A Processors. In the 5th step of this tutorial (Activate the Application Deployment Feature) the IP address of the target board is specified, which then also appears correct under Configuration Parameters -> Hardware Implementation -> External Mode.

Now, if you want to change this IP address at this point (Configuration Parameters -> Hardware Implementation -> External Mode), the embedded coder will not use this new IP address, but only the one that was specified when the target was created. That means, no matter which IP address the user later sets under Configuration Parameters -> Hardware Implementation -> External Mode, only the one specified when creating the target will be used for code generation, which is unsuitable.

Is this a bug or a feature? Where can the user later set the IP address of his target board, which then will be used by the embedded coder?

Thank you guys and with best wishes,

Lars Lindner

Answer 1

[Githin George]

Hi Lars,

It seems like there is no option to change the IP address of the target board from the Configuration Parameters Dialog Box for custom targets as the settings are never reflected for code generation. I am assuming that this behaviour is by design.

I would suggest reaching out to the MathWorks Technical Support team and raising a feature request for the same:

https://www.mathworks.com/support/contact_us.html

The alternative currently is to create multiple custom hardware targets, one for each IP. But the limitation would be that you need to create a custom target for each board.

Finally, you can use the "sefpref" command to change the default IP address property of the target group similar to how it is set in the 5th step of the tutorial:

https://www.mathworks.com/help/ecoder/armcortexa/ug/implement-a-target.html

I hope this helps.

[Lars Lindner]

He Githin, thank you for your answer.

Indeed, we are already using the "setpref" function as described in the 5th step of the tutorial (https://de.mathworks.com/help/ecoder/armcortexa/ug/activate-the-application-deployment-feature.html)!

As I wrote in my question, using this function, the IP address of the target board is specified, which then also appears correct under "Configuration Parameters -> Hardware Implementation -> External Mode". Now, if you want to change this IP address at this point ("Configuration Parameters -> Hardware Implementation -> External Mode"), the embedded coder will NOT use this new IP address, but only the one that was specified when the target was created!! This means, that once the target has been created, the user can no longer set the IP address. To change the IP address, the user would have to create the target again, which is not user-friendly.

With regards,

Lars