• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 97 subscribers
  • Views 525 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.

PROCESSOR-SDK-AM62P: Errata workarounds in BSP and Drivers

Andrea Cecconelli
Prodigy 45 points

Part Number: PROCESSOR-SDK-AM62P
Other Parts Discussed in Thread: AM62P

HI,

I've noticed that most of the proposed fixes for the errata listed in the sprz574_AM62Px Silicon Errata.pdf file seem to need to be implemented at a low level, in the BSP or peripheral drivers. It's correct? In the available SDKs, have they been implemented?
Thank you

  • 0
    TI__Guru**** 181596 points

    Hi Andrea,

    Andrea Cecconelli said:
    need to be implemented at a low level, in the BSP or peripheral drivers. It's correct?

    Yes.

    Andrea Cecconelli said:
    In the available SDKs, have they been implemented?

    Some are implemented, but not all.

    Do you have any specific advisories in question?

  • 0 in reply to Bin Liu
    Prodigy 45 points

    Hi Bin,

    my main concerns are:

    1) errata relating to CAN, MCAN i2278 and MCAN i2279.
    I'm not sure if we will use the "easy" solution #2 based on using the TX FIFO and it would be better if the verification of pending Tx requests was handled by the driver, with adequate notification at the application level.

    2) USART i2310 — USART: Incorrect activation of timeout interrupt

    We will probably use DMA (and the effect is mitigated), but I expect that, in case of CPU usage, register management as requested in workaround#1 will be masked at the application level.

    By the way, does USART i2310 also affect MCU_UART and WAKEUP_UART?
    3) USB i2409 — USB: USB2 PHY hangs due to short sleep

    I expect the workaround to be handled by the BSP during controller initialization.

    4) For the following errata, I haven't analyzed the effect yet. Can you tell me which ones are managed at BSP level?

     DSS i2097

    Interrupt Aggregator i2196

    ECC_AGGR i2049

    MMCHS i2312

    PRG i2253

    PSIL i2137

    RAT i2062

    CPSW i2208, CPSW i2401 

  • 0 in reply to Andrea Cecconelli
    TI__Guru**** 181596 points

    Hi Andrea,

    I will comment on some of the Advisories which I am familiar with. Please create separate E2E threads for the rest of them individually, so that the threads can be assigned to the corresponding expert for comments.

    - 2) USART i2310

    This issue is only observed internally on another device which uses the same USART module as that on AM62P devices, it has not been seen on AM62P.

    The workaround mentioned in the Errata will be implemented in kernel in future SDK release.

    However, the kernel patch mentioned in the link below should be applied to fix a bug related to i2310. This patch won't be needed when the workaround in i2310 is implemented.

    https://e2e.ti.com/support/processors-group/processors/f/processors-forum/1327167/am623-uart-enable-pktdma-increase-total-transmission-time/5050761#5050761

    Andrea Cecconelli said:
    By the way, does USART i2310 also affect MCU_UART and WAKEUP_UART?

    It might.

    - 3) USB i2409

    The workaround is implemented in the coming SDK9.2 release.

    - MMCHS i2312

    The issue is not observed on AM62P. But it has been seen on other device which uses the same MMC module as that on AM62P devices.

    There is no plan to implement the workaround in Linux.