Wolf Pass power control

Vijay Khemka vijaykhemka at fb.com
Thu Aug 27 15:36:10 AEST 2020


Just see if you are missing gpio line names patch from intel in your kernel device tree. Mainly look for PSPowerOK pin if it is configured properly in device tree.

From: openbmc <openbmc-bounces+vijaykhemka=fb.com at lists.ozlabs.org> on behalf of Patrick Voelker <Patrick_Voelker at phoenix.com>
Date: Wednesday, August 26, 2020 at 6:04 PM
To: "OpenBMC (openbmc at lists.ozlabs.org)" <openbmc at lists.ozlabs.org>
Subject: Wolf Pass power control

I’m working on a Wolf Pass build for Phoenix and am having issues with power control.  I believe I’ve got all the GPIOs setup correctly and the power-control service is running.  It asserts POWER_OUT but it’s timing out waiting for PS_PWROK assertion.

As far as I can tell, those GPIOs are setup exactly like it’s setup on the IntelBMC version.  Is there anything that jumps out at you here that I might look at next?  I’m not sure why asserting POWER_OUT wouldn’t cause a power up sequence to occur.

## Current behavior of my wolf pass build at power on request
Jan 01 00:02:42 wolfpass power-control[224]: powerStateFailedTransitionToOn: power-on request event received
Jan 01 00:02:42 wolfpass power-control[224]: power supply power OK watchdog timer started
Jan 01 00:02:42 wolfpass power-control[224]: Moving to "Wait for Power Supply Power OK" state
Jan 01 00:02:42 wolfpass power-control[224]: POWER_OUT set to 0
Jan 01 00:02:42 wolfpass power-control[224]: Error in finding CurrentHostState property
Jan 01 00:02:42 wolfpass power-control[224]: POWER_OUT released
Jan 01 00:02:50 wolfpass power-control[224]: power supply power OK watchdog timer expired
Jan 01 00:02:50 wolfpass power-control[224]: powerStateWaitForPSPowerOK: power supply power OK watchdog timer expired event received
Jan 01 00:02:50 wolfpass power-control[224]: Moving to "Failed Transition to On" state
Jan 01 00:02:50 wolfpass power-control[224]: PowerControl: power supply power good failed to assert

## Example of good power on sequence from IntelBMC
Aug 26 11:43:34 power-control[236]: powerStateOff: power-on request event received.
Aug 26 11:43:34 power-control[236]: power supply power OK watchdog timer started
Aug 26 11:43:34 power-control[236]: Moving to "Wait for Power Supply Power OK" state.
Aug 26 11:43:34 power-control[236]: POWER_OUT set to 0
Aug 26 11:43:34 power-control[236]: Error in finding CurrentHostState property
Aug 26 11:43:34 power-control[236]: POWER_OUT released
Aug 26 11:43:34 power-control[236]: powerStateWaitForPSPowerOK: SIO S5 de-assert event received.
Aug 26 11:43:34 power-control[236]: No action taken.
Aug 26 11:43:34 power-control[236]: SIO_ONCONTROL value changed: 0
Aug 26 11:43:34 power-control[236]: powerStateWaitForPSPowerOK: power supply power OK assert event received.
Aug 26 11:43:34 power-control[236]: SIO power good watchdog timer started
Aug 26 11:43:34 power-control[236]: Moving to "Wait for SIO Power Good" state.
Aug 26 11:43:34 power-control[236]: power supply power OK watchdog timer canceled
Aug 26 11:43:35 power-control[236]: powerStateWaitForSIOPowerGood: SIO power good assert event received.
Aug 26 11:43:35 power-control[236]: Moving to "On" state.
Aug 26 11:43:35 power-control[236]: Host system DC power is on
Aug 26 11:43:35 power-control[236]: SIO power good watchdog timer canceled
Aug 26 11:43:35 power-control[236]: POH timer started
Aug 26 11:43:35 power-control[236]: powerStateOn: SIO power good de-assert event received.
Aug 26 11:43:35 power-control[236]: No action taken.
Aug 26 11:43:35 power-control[236]: powerStateOn: SIO power good assert event received.
Aug 26 11:43:35 power-control[236]: No action taken.
Aug 26 11:44:32 power-control[236]: powerStateOn: POST Complete assert event received.
Aug 26 11:44:32 power-control[236]: No action taken.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ozlabs.org/pipermail/openbmc/attachments/20200827/b75f8f13/attachment.htm>


More information about the openbmc mailing list