Standby issue

[jonas_karlsson]

Hi,

We are currently using MX8QXP and PF8100 on a custom board and we are having some problems transitioning to standby mode (KS1). It appears that the SCU does not assert the SCU_PMIC_STANDBY pin, causing the A35s to get stuck in an endless loop waiting for power to be cut.

The power-down sequence is done in two steps. First, we put SW4 in SW_RUN_OFF/SW_STBY_OFF. Secondly, we assert SCU_PMIC_STANDBY. About half of the time none of these two steps happen and we end up in the previously described endless loop. In some rare cases, only the first step is performed (verified with logic analyzer) and the standby pin is never asserted.

Digging through the SCU firmware, it seems that the first step is done through the board_set_power_mode function. But we can't find how and when the second step is performed. Are there any conditions that need to be met for the SCU to assert the standby pin? Is there a good way to debug this further? Attaching an image of the normal behavior when suspending.

SCU firmware version 1.2.7.1

Linux kernel version 4.19.35

[igorpadykov]

Hi Jonas

low power mode is handled by System Controller Unit (SCU) and one can

look at Chapter 5 Power Management sc_fw_api_qx_b0.pdf

included in documentation in SCFW Porting Kit

Behaviour of SCU_PMIC_STANDBY can be checked on i.MX8QXP MEK board using

procedures described in i.MX 8QuadXPlus Power Consumption Measurement

Reasons that SCU does not assert the SCU_PMIC_STANDBY may be that some resource

has enabled interrupt which causes exiting from low power mode.

Best regards
igor
-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

[jonas_karlsson]

Hi Igor,

We tried disabling interrupts and it seems to have resolved the problem. Our question still remains though, what are the conditions that need to be met in order for SCU_PMIC_STANDY to be asserted? If we run into this problem again we want to be able to debug it as fast as possible.

BR,

Jonas

[igorpadykov]

Hi Jonas

>what are the conditions that need to be met in order for SCU_PMIC_STANDY to be asserted?
>If we run into this problem again we want to be able to debug it as fast as possible.

It may be debugged comparing behaviour of SCU_PMIC_STANDBY on i.MX8QXP MEK board using

procedures described in i.MX 8QuadXPlus Power Consumption Measurement

Best regards
igor