• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 2 answers
  • Subscribers 98 subscribers
  • Views 374 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.

RTOS/TDA2PXEVM: CSI2 capture problem

TG2718
Intellectual 1000 points
Part Number: TDA2PXEVM
Other Parts Discussed in Thread: AWR1243

Tool/software: TI-RTOS

I am reading data from cascaded AWR1243 radar chips into a TDA2PX, using Processor SDK 3.5.

In chains_common_ar12xx.c ChainsCommon_ar12xxSetIssCaptureParams(), I can set pPrm->outParams[i].width = pPrm->outParams[i].height = 0, and collect a single frame of data over CSI2, which I can read from the memory location specified by the CAL DMA register and it looks fine. I thought it may be reasonable to set them to 0, based on comments in struct vpsissCalCfg and struct isshalCalCfg. If I do not set these parameters to zero, the second half of each chirp is filled with 0s.

I have also implemented a separate use case: IssCapture -> NetworkTx (IPU1_1). When I set width and height to 0 as above, the use case times out. If I do not set them to 0, the use case runs fine, but the second half of each chirp is filled with 0s in the data streamed over the network.

Any suggestions what may be happening and how I should configure things? This has proven very time-consuming to debug.

Thanks.

  • 0
    TI__Mastermind 30245 points
    Hi Todd,

    1. Can you please check if you have the fix for the below in your code base:

    pPrm->outParams[i].outBpp = SYSTEM_BPP_BITS16; in the function ChainsCommon_ar12xxSetIssCaptureParams in the file.

    2. What is the size you are expecting and setting in the AWR1243 and the ISS parameters?

    Thanks and Regards,
    Piyali
  • 0 in reply to Piyali Goswami
    Intellectual 1000 points
    Hi Piyali,

    1. I have been using Processor SDK 3.5, and have not yet studied Processor SDK 3.6. Version 3.5 does not have the .outBpp assignment you mentioned above so I had not tried it (Version 3.6 does). I can add it. What does it do? Any other changes in SDK 3.6 that may relate? I was planning to upgrade after fixing this issue.

    2. In terms of size, I have the following:
    pCfgOut->radarParams[i].csi2OutWidth = 2200. Actually, I think this should be 2208 to include Chirp Parameters.
    pCfgOut->radarParams[i].csi2OutHeight = 168
    The radar is set to complex1X, 12 bit (automatically converted to 16 bit fields by pixel processing context). 4 Rx/chip enabled. 275 ADC samples.
    Any other specific parameters you are interested in?

    Anything else I should look at?

    Thanks!
  • 0 in reply to TG2718
    Intellectual 1000 points
    Setting "pPrm->outParams[i].outBpp = SYSTEM_BPP_BITS16;" as suggested resolved our main problem, and we no longer get 0s in the second half of the chirps.

    The last number returned in each chirp is still always 0, for some reason. I am not sure what is causing that.
  • 0 in reply to TG2718
    TI__Mastermind 30245 points
    Glad your main problem is resolved.

    For the last chirp being 0 are you observing the end of the line having 0s or each chirp for each Rx antenna being 0?

    Thanks and Regards,
    Piyali