Post Go back to editing

ADRV9002 custom board: HAL & devicetree

Hi team,

following the user-guide of the adrv9002 I developed my application by generating the files with TES and modifying them according to my needs. Now I need to develop a custom board. I have read the section of the user-guide regarding the development of a custom HAL. I have several questions regarding this point:

  • When I switch to the development of a custom board with the definition of a new HAL is it necessary to make a different microSD image?
  • In particular, if I need to add a new peripheral in the new custom board, is it necessary to modify the devicetree file?
  • If yes, what is your recommended procedure? 

Best Regards.

Parents
  • Hello LF92,

    You've touched on some interesting talking points for this stage of customer development. Luckily, the answer to most of your questions is it's up to you. 

    • There's nothing stopping you from using a different SD card image if you need it. If you're planning on using our FPGA in your final application, you could design the custom daughter card to suit our image. It's not impossible, but it would be a bit of work. Or you could develop your own image for the ZC706 or ZCU102, or even a completely different platform, and go from there. This approach is perfectly valid, and it's probably the one most customers will adopt. We do not provide any support for custom FPGA images, as this is far beyond our scope, however the "compile_on_platform.py" script can be used to target any device connected to the PC via SSH. We used this script to push our custom code to a Raspberry Pi with no difficulty.
    • You shouldn't need to modify any of our non-customer files to suit your custom board. Support for any extra peripherals can be handled in the Customer HAL section of the SDK. All extra functions can be added either to an existing file, or to a dedicated .c and .h file pair (so long as these new files are then linked in the makefile).

    This should see you through this phase of development, however we are aware of how challenging this phase can be. Do let us know if you have any further questions.

    Best Regards,
    Oisín.

Reply
  • Hello LF92,

    You've touched on some interesting talking points for this stage of customer development. Luckily, the answer to most of your questions is it's up to you. 

    • There's nothing stopping you from using a different SD card image if you need it. If you're planning on using our FPGA in your final application, you could design the custom daughter card to suit our image. It's not impossible, but it would be a bit of work. Or you could develop your own image for the ZC706 or ZCU102, or even a completely different platform, and go from there. This approach is perfectly valid, and it's probably the one most customers will adopt. We do not provide any support for custom FPGA images, as this is far beyond our scope, however the "compile_on_platform.py" script can be used to target any device connected to the PC via SSH. We used this script to push our custom code to a Raspberry Pi with no difficulty.
    • You shouldn't need to modify any of our non-customer files to suit your custom board. Support for any extra peripherals can be handled in the Customer HAL section of the SDK. All extra functions can be added either to an existing file, or to a dedicated .c and .h file pair (so long as these new files are then linked in the makefile).

    This should see you through this phase of development, however we are aware of how challenging this phase can be. Do let us know if you have any further questions.

    Best Regards,
    Oisín.

Children
No Data