• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 98 subscribers
  • Views 469 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.

Linux/DRA72: Carlife conneting failure by accident

Wesley
Prodigy 250 points
Part Number: DRA72

Tool/software: Linux

now,we have a problem that Carlife conneting failure sometime,we debug found below message did not sent out.

[1959616](824|0xB2FAF450426] g_carplay gadget: bulk_write eap trying to send 8 bytes

[1959999](6468] g_carplay gadget: bulk_write eap sent 8 bytes               //actually this not sent out

 

 

 

Thank you

Qin Xin

Huizhou Desay SV Automotive Co.,Ltd

Tel：(86)0752 2639273


  • 0
    TI__Mastermind 47963 points

    Hi Qin Xin,

    It looks like this new thread is a continuation of a previous thread based on the description, can you refer that thread. The debug message says that it sent out the 8 bytes, but your comment says it did not send out. How was this checked? Do you connect USB analyzer to collect the USB data transfers? 

    Also, can you share more details on the following:

    - TI SDK release in question

    - TI EVM or a Custom Board?

    - Complete kennel log

    - Description of the use case/ Steps used to reproduce

    - Praveen

  • 0 in reply to Praveen Rao
    Prodigy 250 points

    Hi Praveen,

    The attachment is the kernel log and USB data analyzer, other information as below,

    - TI SDK release in question --psdk-linux-03.02.00.03

    - TI EVM or a Custom Board? -- Custom Board

    - Complete kennel log -- refer to attachment

    - Description of the use case/ Steps used to reproduce -- test phone is iphone XS,repeat less than ten percent

    https://e2e.ti.com/cfs-file/__key/communityserver-discussions-components-files/791/log_2600_data.7z

    --Qin Xin

  • 0 in reply to Wesley
    TI__Mastermind 47963 points

    Hi Qin Xin,

    Thanks for the logs.

    Can you point me to the line number in the kernel log where the issue happened? I do see some other errors, but not related to the issue described in the previous post.

    One question, Have you confirmed that the issue due to the g_carplay gadget. My understanding that this is implemented by you and is not part of the TI SDK. Is that correct?

    Also, how is the iPhone connected to the DRA72? Is there a hub between them?  if there is a hub, Is the analyzer log attached is taken by connecting USB analyzer b/n the DRA72 and hub? 

    Thanks,

    Praveen

  • 0 in reply to Praveen Rao
    Prodigy 250 points

    Hi Praveen,

    The log Starting from line 99607 to the end are related to the issue.

    Yes,the g_carplay gadget is our own.

    no,it didn't connected a hub.

    Thanks,

    Qin Xin

  • 0 in reply to Wesley
    TI__Mastermind 47963 points

    Hi Qin Xin,

    Checked the log at the location mention, but we are not able to figure out much. There is nothing that says weather the packet (8 bytes) was not sent or a failure to send  those bytes. For the request sent on line 99607, the confirmation of packet being sent is on line 99645 as printed by your gadget.

    If you would like to know more on packets were handled at the usb driver level, one suggestion is to collect the ftrace of dwc3 events as below  and do the analysis.

    In Kernel enable CONFIG_EVENT_TRACING=y

    cd /sys/kernel/debug/tracing/
    echo 1 > tracing_on
    echo 1 > events/dwc3/enable
    echo 0 > events/dwc3/dwc3_readl/enable
    echo 0 > events/dwc3/dwc3_writel/enable
    echo 1 > free_buffer
    echo 102400 > buffer_size_kb

    ## the trace buffer
    cat /sys/kernel/debug/tracing/trace

    If you see any issue there, please let us know and we can look further.

    For now, making this thread as closed.

    -Praveen