• State Resolved
  • Locked Locked
  • Replies 21 replies
  • Answers 3 answers
  • Subscribers 29 subscribers
  • Views 1712 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.

AWR1243: Cascaded chips EVM + MMWCAS-DSP-EVM

Badeea Alazem
Intellectual 315 points
Part Number: AWR1243
Other Parts Discussed in Thread: MMWCAS-DSP-EVM, TDA2
I am configuring and capturing data using the Lua shell in mmWave studio. first I run the "Cascade_Configuration_MIMO.lua" script which executes successfully (attached is a log of the output). after which I run the "Cascade_Capture.lua" scrip which also executes successfully (output log attached).  However, the "/mnt/ssd/Cascade_Capture" folder is empty, no bin files exist after the capture script is done executing. 
the board is drawing 1.1A when spikes up to 1.5A during the capture script execution indicating the board is transmitting and hopefully receiving a signal. 
I attached the scripts I used, i modified " Cascade_Capture.lua " according to the instructions provided. 
could you point me to what i might be missing here. 
  • 0
    TI__Guru*** 147894 points

    Hi,

    What is the hardware you are using.

    Do you have one of the cascade imaging kit that was just released?

    thank you

    Cesar

  • 0 in reply to Badeea Alazem
    TI__Mastermind 18470 points

    Hi Badeea, 

    You are running with the default mmwave studio lua scripts which come with mmwave studio 2.1 release, correct? 

    Is this problem repeatable? Does it occur on every boot of the MMWCAS-DSP-EVM?

    I would like you to check out the condition of the SSD mount and file system on the MMWCAS-DSP-EVM board. After boot, can you please do the following at the UART terminal prompt:

    1. login (root, no password)

    2. run "df" command and confirm that the /dev/nvme0n1p1 is mounted to /mnt/ssd - if it is not, then this is the problem

    • If the SSD fails to mount on boot, then the OS will place the /mnt/ssd folder on the SDcard filesystem. Then the capture application will attempt to write out to the SDCard file system which can sometimes not keep up with the radar data writes. 

    3. if it is mounted, I would like you to reformat the disk and then power cycle the board and try the capture again. You can reformat with the following commands: 

    • First unmount the SSD using "umount /mnt/ssd"
    • Then format it with the ext4 filesystem using "mkfs.ext4 /dev/nvme0n1p1
    • Then mount it again with "mount /dev/nvme0n1p1 /mnt/ssd"

    4. If it is not mounted, I would like you visually inspect the SSD mounting to the m.2 socket on the MMWCAS-DSP-EVM. There is a single screw which holds the SSD in place, please verify that the SSD is securely attached and mounted into the m.2 socket. 

    Please let me know if either of those steps resolves the problem. 

    Thank you,

    -Randy

  • 0 in reply to Randy Rosales
    Intellectual 315 points

    Hello Randy,

    Thank you for the reply.  

    Correct, I am using the latest version of mmWave Studio 2.1. it is also repeatable. 

    I will follow you suggestions above and get back to you soon with results.

    Thank you,

    Badeea 

  • 0 in reply to Badeea Alazem
    TI__Mastermind 18470 points

    Hi Badeea, 

    I just noticed a typo I made above. Please re-reference my umount command exerpt. 

    Thanks,

    -Randy

  • 0 in reply to Randy Rosales
    Intellectual 315 points

    Hi Randy, 

    I tried the steps above and attached a log of the process. It did not work, the directory in which to save the bin files get's created but it is empty. I also noticed something weird. when i run the Capture script and list the contents of the directory right away, i see the bin files, but by the time the scripts finishes execution the files will no longer be there (see the last pic in the attached document).

    Thank you, 

    BadeeaUntitled document (1).pdf

  • 0 in reply to Badeea Alazem
    TI__Mastermind 18470 points

    Hi Badeea, 

    I am checking the log files you sent over as I see some MSS boot error messages being sent over. I'll double check that these are occuring on my EVM as well. 

    Thank you,

    -Randy

  • 0 in reply to Randy Rosales
    Intellectual 315 points

    Hello Randy, 

    Any updates?

    Thank you, 

    Badeea

  • 0 in reply to Badeea Alazem
    Intellectual 315 points

    Hello Again,

     one note, the bin files are being created and deleted within the following API call. not sure if that's helpful or not. 

    status = ar1.TDACaptureCard_StartRecord_mult(1, n_files_allocation, data_packaging, capture_directory, num_frames_to_capture)

    Thank you, looking forward to your response.

    Badeea

  • 0 in reply to Badeea Alazem
    TI__Mastermind 18470 points

    Hi Badeea, 

    I am going to need more time to get back into the lab with the kit to look into these problems. The soonest I will be able to do that is middle of next week. I will update this thread as soon as I have more follow-up information. 

    Thank you,

    -Randy

  • 0 in reply to Randy Rosales
    Intellectual 315 points

    Hi Randy, 

    Thank you. would it be possible to let me know when you will debug so i can sit at the kit that i have and try suggestions form you quickly. this way we minimize the time of finding out what's wrong. 

    Thank you,

    Badeea

  • 0 in reply to Badeea Alazem
    Intellectual 315 points

    Hello Randy, 

    Have you had a chance to look into this? 

    Thank you, 

    Badeea

  • 0 in reply to Badeea Alazem
    TI__Mastermind 18470 points

    Hi Badeea, 

    I think I might have overlooked something fundamental here initially. In your logs and scripts zip file, you included the "Cascade_Capture.lua" that you ran. The capture_directory folder is not setup correctly. 

    Please change 

    capture_directory               =   "C:\\ti\\mmwave_studio_02_01_00_00\\mmWaveStudio\\PostProc\\Cascade_Capture"--"Cascade_Capture"

    To something like this

    capture_directory               =   "capture_test1"

    And try the captures again with this edited capture script. Please let me know if that solves your current problem. For sanity checking a new setup, I would recommend running the scripts exactly as they are presented in the mmWave Studio release first.

    The TDA2x captures data to an onboard NVME SSD. This is the SSD mounted to /mnt/ssd in the linux file system that we discussed earlier. The "capture_directory" variable is meant to simply create a new directory on that SSD for the next capture files to be stored into. For example setting

    capture_directory               =   "capture_test1"

    would create the   /mnt/ssd/capture_test1/ on the SSD and then the subsequently triggered capture would store that data into it. 

    After the capture is completed, users can use the "transfer files" button (see below) in the SensorConfig tab to transfer the files over to the connected host PC. These files are transferred over to the "C:\\ti\\mmwave_studio_02_01_00_00\\mmWaveStudio\\PostProc\\" directory you were referencing. But there is no user configuration for determining where the transferred files will be stored. 

    Thank you,

    -Randy

  • 0 in reply to Randy Rosales
    Intellectual 315 points

    Hello Randy, 

    Thank you for the reply. 

    I tried this and it did not work. I don't think that's the issue because i see the folder being created in the /mnt/ssd and i see the bin files appearing for a brief moment then disappearing. I followed the instructions to the letter but whatever is happening does not seem to be typical. Can you please advise on what you believe is the best way to debug this in a more efficient way. 

    Thank you,

    Badeea

  • 0 in reply to Badeea Alazem
    TI__Mastermind 18470 points

    Hi Badeea, 

    Are you using a different capture script from the one you sent me then? Can you send the exact setup and capture script again? Just making sure I am not missing a simple solution here.

    Can you also please send over the UART log from the TDA2 board from initial power on to logon screen?

    Thank you,

    -Randy

  • 0 in reply to Randy Rosales
    Intellectual 315 points

    Hello Randy, 

    Here's the script. 

    Regards, 

    BadeeaScripts.zip

  • 0 in reply to Badeea Alazem
    Intellectual 315 points

    Hello Randy, 

    Any updates? 

    Regards, 

    Badeea

  • 0 in reply to Badeea Alazem
    TI__Mastermind 18470 points

    Hi Badeea, 

    In your capture script, you are setting up: 

    num_frames_to_capture =   1 -- 0: default case; Any positive value - number of frames to capture  

    With this setting I am also not seeing any captured data. But if I change that back to the default: 

    num_frames_to_capture =   0 -- 0: default case; Any positive value - number of frames to capture  

    I am able to get the expected number of frames captured. Can you please make the same change to your environment and let me know if this resolves the issue? That will let me know if your hardware is OK. 

    But I think there may be a problem with the usage of the num_frames_to_capture variable. 

    Thank you,

    -Randy

  • 0 in reply to Randy Rosales
    Intellectual 315 points

    Hello Randy, 

    No luck, I changed back to the default as you suggested but the problem is still there. 

    Thank you, 

    Badeea

  • 0 in reply to Badeea Alazem
    TI__Mastermind 18470 points

    Hi Badeea, 

    Based on feedback and our debug session it looks like there was a problem with your specific kit. Please close this thread for now and open up a new one if the replacement kit gives you any problems. 

    Thank you,

    -Randy

  • 0 in reply to Randy Rosales
    Intellectual 315 points

    Thank you Randy!

    Badeea