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.

AM6526: Does Diagnostic TEST code support USB3.0 ?

Part Number: AM6526


Hi,

My customer is using the PDK Diagnostic Test code, but some USB3.0 memories were identified as USB2.0.

We found the following description in the 6.2.4 Diagnostic Applications Support in the PDK-AM65xx User's Guide 07_01_00.

https://software-dl.ti.com/processor-sdk-rtos/esd/AM65X/latest/exports/docs/pdk_am65xx_07_01_00_55/docs/userguide/am65xx/board/board_diag.html#diagnostic-applications-support

 

"Verifies the USB host mode operation of boardf under test. USB modules operates at high speed (2.0) during the test."

The Diagnostic Test code "usbHost_TEST" doesn't support USB3.0 ? or has any problem on USB3.0 mode ?

Regards,
Hideaki

  • Hideaki:

    I was on this discussion with Darren earlier, and you can find the a little more details about usbHost_TESP diag in Section 6.3.2.48:

    https://software-dl.ti.com/processor-sdk-rtos/esd/AM65X/latest/exports/docs/pdk_am65xx_07_01_00_55/docs/userguide/am65xx/board/board_diag_execution.html

    it only runs at USB2.0 speed. 

    I verified linux driver test on the same board and USB3 is supported. 

    I am not aware of any known issues in USB3 Host mode. I think it may not be initialized in the diag test. 

    Jian 

  • Hi Jian,

     

    Thank you for the answer on this thread and off-line as well. The customer needs a diagnostic test code for USB3. Could you prepare it ? I think TI should prepare it if AM65x supports USB3.

     

    Have you heard ever that USB3 on AM65x could work on TI-RTOS ?

     

    Could you tell us why there is not USB in the supported drivers list of PDK 7.00 and 7.01 even though PDK in PSDK6.03 supported USB.?

     

    4.1. Release notes - 07_01_00

     

    4.1.1.2.1. Platform Development Kit (PDK)

      • RTOS device drivers running on R5F, A53
      • Supported drivers and modules
        • Board diagnostic library
        • CPSW2G, CPTS, SGMII/QSGMII
        • CSL-FL (Chip support library)
        • GPIO
        • I2C
        • IPC (inter processor communication)
        • McASP
        • MMCSD with FATFS (A53 with RTOS)
        • OSAL
        • SBL secondary bootloader
        • SCI Client (DMSC interface)
        • SPI driver (SPI, QSPI, OSPI, Hyperflash)
        • UART
        • UDMA and DRU
        • CAL
        • DSS

     

    Regards,

    Hideaki

  • Jian,

    Can we update the latest on this thread?

    Regards

    Karthik

  • update from email correspondence with Darren:

    Two USB3 test binaries from the PDK:

    • USB_Baremetal_HostMsc_usb30 (baremetal)
    • USB_HostMsc_usb30 (RTOS-based)

    has been shared with Darren and brought up at Darren's setup, with the following test results:

    1. on Jian's setup, one compliant USB3-B to A adapter shows consistent USB3 enumeration. 

    2. On Darren's setup, we see the same hardware enumerated to USB2 with these test programs, but enum to USB3 under Linux 

    SERDES config between Linux and RTOS has been captured and compared, it was found PDK tests are using old SERDES config. TI to update SERDES config and retest to see if we can see consistent result across linux and RTOS. 

    Customer may proceed to 1). test the PDK binaries on their hardware (they use Type A connector), 2). update their SERDES config for their diagnostic code, pending completion of their SDK upgrade. 

    Jian

  • Update of debug activities 3/8/2021:

    With the correct adapter, we consistently see

  • update 3/8/2021:

    As we verified the TI EVM can consistently enumerate to USB3 with the thumb drives we tested. We are getting the exact drives that fell back to USB2 on the customer board. We can debug and improve the driver robustness if these drives also fails on the EVM. 

    In parallel, four work items are studied and may be implemented to improve the driver robustness, so we can test the drives when available. 

    Jian

  • Hideaki, 

    I will close this ticket as we identified the gap between RTOS and Linux driver and debug has shifted to software development. 

    regards

    jian