• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 104 subscribers
  • Views 110 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

AM6548: Powering up RTI1 watchdog also powers up R5F core 1

Jan Kiszka
Intellectual 410 points
Part Number: AM6548

This is to bring attention for the LKML bug report and its related questions about the affect processor to this forum:

https://lore.kernel.org/lkml/e0bc7eb5-f1b9-4e99-a39f-ca6f31daa9d8@siemens.com/T/#u

Can someone confirm if that powering relationship exist? Following up on LKML would also be appreciated.

Thanks in advance!

  • 0
    TI__Guru*** 135118 points

    Hello Jan Kiszka

    Thank you for the query and suggestions.

    I am reassigning the thread to the expert to support.

    Regards,

    Sreenivasa

  • 0
    TI__Guru* 76425 points

    Hello Jan,

    Thank you for highlighting this to us. I'll check with Hari & Suman to make sure they've seen the bug report.

    To confirm, we have several k3 devices with an R5F subsystem that would not be running anyway: AM65x, AM64x, AM62A (AM62x & AM62A also have R5F cores that will be running device management firmware, but I'm assuming this doesn't apply since that DM firmware must be running regardless). Which devices have you tested & observed this on so far?

    Regards,

    Nick

  • 0 in reply to Nick Saulnier
    Intellectual 410 points

    Only AM6548. I don't have a setup ready for AM64x or AM62x that is similar in this regard (watchdog specifically).

  • 0 in reply to Jan Kiszka
    TI__Guru* 76425 points

    Got it, thanks for confirming. Hari took a look earlier this week, and

    1) There DOES seem to be a power relationship where powering up RTI watchdog powers up the R5F, and

    2) There might be a logic error where the remoteproc driver detects whether a remote core is actually running code, or just powered on.

    As of Wednesday Hari was waiting to discuss with Suman, not sure of the current status.

    Thanks again for reporting this, and then notifying us on the forums. If Hari hasn't responded within the next couple of business days, please ping the thread and I'll check in with him.

    Regards,

    Nick