• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 684 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

AWR2243: Question about porting mmwavelink to Linux based System.

Tim Teuber
Intellectual 390 points
Part Number: AWR2243
Other Parts Discussed in Thread: AWR1243

Hello Team,

i am currently working on porting the mmwavelink API to a Linux based Host (Nvidia Jetson Nano) . I am working along the given examples and the porting guide. From my understanding, i can mainly copy the examples and just have to make major changes in the rls_osi.c and the mmwl_port_ftdi.c files. The first two steps of the porting guide are straight forward. For Communication Interface Callbacks, i am using the I2C library's provided by Linux. Is there anything that would speak against using I2C instead of SPI?

Now i am struggling to understand the the functions rlsEnableDevice and rlsDisableDevice.

From my understanding, i have to mainly handle triggering of the nReset with those functions, to power on and off the AWR2243. In the examples, those functions are defined in the FTDI lib. I looked into the source code of this file, given by the mmWave Studio Reference Code. From those functions, I cant really read out how and when the nReset needs to be pulled low. Could you help me with that, perhabs explain those functions.

Furthermore, i am not quite sure how to implement the HostIRQ Masks. Do I just need to write a function that detects changes on a GPIO pin of my host that is connected to HostIRQ and the mmwavelink API handles the rest of it? Or do I need to implement this in my Communication Interface Callback Functions as well?

Are there more implementation examples available then those provided in the DFP Package?

Best regards,

Tim

  • 0
    TI__Guru*** 146574 points

    Tim,

    We have ported a few years ago the mmwave link to linux for an older nvidia jetson platform

    You can download the release from this link valid 7 days

    https://tidrive.ext.ti.com/u/ws63giBOt-nQzreE/8b3eee8a-3bc8-446f-a991-f508e49d745c?l

    Access: p1TPVk`6

    This may help you with your project.

    thank you

    Cesar

  • 0 in reply to Cesar
    Intellectual 390 points

    Hi Cesar,

    Thanks allot for the provided example. Would this work in case of a direct connection between host and AWR2243 without the UB960EVM? As far as I understand, AWR1243 and AWR2243 use the same communication procedure?

    thank you,

    Tim

  • 0 in reply to Tim Teuber
    TI__Guru*** 146574 points

    Hi,

    AWR1243 firmware supported only SPi communication for control. That's why for this project we had to convert SPI to I2C required for FPD-Link

    The AWR2243 firmware supports both SPI and I2C. For direct connection one would have to use SPI for control.

    So this code will not work as is with direct SPI connection, it would have to be modified

    This is just an example of a linux implementation of the mmwave link

    thank you

    Cesar

  • 0 in reply to Cesar
    Intellectual 390 points

    Hi Cesar,

    thank you for the fast reply. Can you tell me if I understand the given project correctly: The Jetson sends commands via I2C and those get converted to SPI by a I2C-SPI before they get passed to the AWR1243? So shouldn't I be able to make this work for my project with minor adjustments, like the used pins and i2c address.

    Best Regards,

    Tim

  • +1 in reply to Tim Teuber
    TI__Guru*** 146574 points

    Hi,

    Yes, this is correct

    Thank you

    Cesar