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.

LP-AM263: ESM Error Signaling Module / Number of interrupt sources in ESM

Part Number: LP-AM263

Hello,

I am trying to understand ESM in combination with the CCM of AM263x. For this purpose I have studied the example of CCM usage in the SDK (version 9.00.35).

My question is now, how many interrupt signals/interrupt input lines has the ESM on AM263x. 

In the forum I have found some forum posts, in which there is spoken from 86 interrupt sources for the ESM. SDK uses also a maximum number of 86 events.

But in the TRM of AM263x (SPRUJ17C) I have found 64 level events (ID 0 - 63) and 23 pulse events (ID0-22).

--> see chapter 10.4.6 ESM0 Interrupt Map


What is now correct 86 or 87 events. I think the SDK function to enable ESM does not activate event 87, is this correct?

Thank you,
Jo Scho

  • Hi Jo,

    I noticed the difference between SDK and the TRM. The CCM1 compare error event is not included in SDK.

    The device has two CCM modules. The CCM compares the outputs of two Cortex-R5 cores running in 1oo1D lockstep configuration. Both CCM modules have the same selftest capability to allow for checking the HW faults with the CCM-R5F itself. Both CCM modules support run-time detection of faults, self-test capability, and error forcing capability.

    The CCM-R5F can run in one of the following four operating modes:
    1. 1oo1D lock step
    2. self-test
    3. error forcing
    4. self-test error forcing

    There are two error flags: CCM self-test error or/and CCM compare error. The error forcing should set both ESM error flags (Self Test, and Compare).

    So I think there are 87 ESM events. 

  • Hello QJ,

    thanks for your answer.

    Is a bugfixed planned for this topic? With what SDK-version would the bugfix be available?

    Thanks and best regards,

    Jo Scho
     

  • Hi Jo,

    I have submitted a bug ticket, and will follow up with this issue.