RE: 回覆: [x86-power-control]: press the power button for a long time that can't force turn off system power

[Chris Chen (TPI)]

Hi Jason and others,

I think I figured out the problem is the GPIOP0 and GPIOP1 passthrough was not set after system booting up. However, as I mentioned when rising the question, I have already set GPIOP0 and P1 passthrough in u-boot, it for now looks like was been turned off during Kernel or OpenBMC application running up. Can you please give me a clue why the GPIO passthrough would be turned off or where should I need to add passthrough setting again?

Thanks.

Regards,
Chris Chen

-----Original Message-----
From: openbmc <openbmc-bounces+chris.chen3=flex.com at lists.ozlabs.org> On Behalf Of Bills, Jason M
Sent: Tuesday, July 27, 2021 12:47 AM
To: openbmc at lists.ozlabs.org
Subject: Re: 回覆: [x86-power-control]: press the power button for a long time that can't force turn off system power



On 7/23/2021 9:04 PM, Chris Chen (TPI) wrote:
> Hi Jason,
> 
> Thanks for your prompt reply.
> 
> The system stays on, because I can hear the fan in PSU is still 
> running and CPU heatsink is hot, besides, the log would show "Graceful 
> power-off timer completed" and "Host0: Moving to "On" state" after 5 
> minutes. Here are all logs of the power-control for your reference. 
> Hope this can provide more details.
The BMC depends on PS_PWROK de-asserting to know that the system has shut down.  If the Graceful power-off timer is expiring, it means that the BMC didn't see PS_PWROK de-assert.

Holding the power button should cause the power button override forced shutdown in hardware.  It sounds like this isn't happening correctly on your system.  I'd suggest reaching out to your Intel representative to get help investigating what is happening in the hardware to prevent the power button override from fully shutting down the system.

> 
> =====
> root at hudsonbay-obmc:~# journalctl | grep -e "power-control" -e "power 
> control" -e "Power Control"
> Jul 24 10:36:06 hudsonbay-obmc systemd[1]: Starting Intel Power Control...
> Jul 24 10:36:09 hudsonbay-obmc power-control[198]: Start Chassis power 
> control service...
> Jul 24 10:36:09 hudsonbay-obmc systemd[1]: Started Intel Power Control.
> Jul 24 10:36:09 hudsonbay-obmc power-control[198]: NMI_OUT set to 0 
> Jul 24 10:36:09 hudsonbay-obmc power-control[198]: POWER_OUT set to 1 
> Jul 24 10:36:09 hudsonbay-obmc power-control[198]: RESET_OUT set to 1 
> Jul 24 10:36:09 hudsonbay-obmc power-control[198]: NMI Source Property 
> Monitor Jul 24 10:36:09 hudsonbay-obmc power-control[198]: 
> Initializing power state.
> Jul 24 10:36:09 hudsonbay-obmc power-control[198]: Host0: Moving to "On" 
> state
> Jul 24 10:36:09 hudsonbay-obmc power-control[198]: POH timer started 
> Jul 24 10:36:38 hudsonbay-obmc power-control[198]: powerStateOn: POST 
> Complete assert event received Jul 24 10:36:38 hudsonbay-obmc 
> power-control[198]: No action taken.
> Jul 24 10:36:41 hudsonbay-obmc power-control[198]: powerStateOn: POST 
> Complete de-assert event received Jul 24 10:36:41 hudsonbay-obmc 
> power-control[198]: Host0: Moving to "Check for Warm Reset" state Jul 
> 24 10:36:41 hudsonbay-obmc power-control[198]: Warm reset check timer 
> started Jul 24 10:36:41 hudsonbay-obmc power-control[198]: 
> RestartCause set to 
> xyz.openbmc_project.State.Host.RestartCause.SoftReset
> Jul 24 10:36:41 hudsonbay-obmc power-control[198]: Host system DC 
> power is off Jul 24 10:36:41 hudsonbay-obmc power-control[198]: POH 
> timer canceled Jul 24 10:36:41 hudsonbay-obmc power-control[198]: 
> failed to reset ACBoot property Jul 24 10:36:42 hudsonbay-obmc 
> power-control[198]: Warm reset check timer completed Jul 24 10:36:42 
> hudsonbay-obmc power-control[198]:
> powerStateCheckForWarmReset: warm reset detected event received Jul 24 
> 10:36:42 hudsonbay-obmc power-control[198]: Host0: Moving to "On"
> state
> Jul 24 10:36:42 hudsonbay-obmc power-control[198]: POH timer started 
> Jul 24 10:36:42 hudsonbay-obmc power-control[198]: Host system DC 
> power is on Jul 24 10:38:16 hudsonbay-obmc power-control[198]: 
> powerStateOn: POST Complete assert event received Jul 24 10:38:16 
> hudsonbay-obmc power-control[198]: No action taken.
> Jul 24 10:38:34 hudsonbay-obmc power-control[198]: PowerControl: power 
> button pressed Jul 24 10:38:34 hudsonbay-obmc power-control[198]: 
> powerStateOn: power button pressed event received Jul 24 10:38:34 
> hudsonbay-obmc power-control[198]: Host0: Moving to "Graceful 
> Transition to Off" state Jul 24 10:38:34 hudsonbay-obmc 
> power-control[198]: Graceful power-off timer started Jul 24 10:43:34 
> hudsonbay-obmc power-control[198]: Graceful power-off timer completed 
> Jul 24 10:43:34 hudsonbay-obmc power-control[198]:
> powerStateGracefulTransitionToOff: graceful power-off timer expired 
> event received Jul 24 10:43:34 hudsonbay-obmc power-control[198]: 
> Host0: Moving to "On"
> state
> =====
> 
> Thanks.
> 
> Regards,
> Chris Chen
> 
> ----------------------------------------------------------------------
> --
> *寄件者:* openbmc <openbmc-
> bounces+chris.chen3=flex.com at lists.ozlabs.org> 代表 Bills, Jason M
> <jason.m.bills at linux.intel.com>
> *寄件日期:* 2021年7月24日 上午 04:36
> *收件者:* openbmc at lists.ozlabs.org <openbmc at lists.ozlabs.org>
> *主旨:* Re: [x86-power-control]: press the power button for a long time 
> that can't force turn off system power
> 
> 
> On 7/23/2021 4:28 AM, Chris Chen (TPI) wrote:
>> Hi all,
>> 
>> I am working on an ast2600 with Intel CPU.
>> The system power currently is able to turn on during Phosphor OpenBMC 
>> boot up after I completed works following:
>> 
>>  1. enable ACPI in u-boot
>>  2. set GPIOP0 ~ P3 pass-through in u-boot
>>  3. porting ESPI driver from AST SDK v6.01 to linux-aspeed repository
>>  4. add "&gpio0" with gpio-line-names which has POWER_BUTTON, 
>>POWER_OUT,
>>     SIO_S3, SIO_S5, etc. defintion in the dts, I think the
>>     "x86-power-control" repository required these.
>>  5. append "x86-power-control" and "intel-ipmi-oem" repositories to 
>>image
>> 
>> However, I always only got the following logs when I pressed the 
>> power button for a long time (> 4s).
>> =====
>> power-control[263]: PowerControl: power button pressed
>> power-control[263]: powerStateOn: power button pressed event received
>> power-control[263]: Host0: Moving to "Graceful Transition to Off" 
>> state
>> power-control[263]: Graceful power-off timer started =====
>> 
>> It doesn't occur "SIO_ONCONTROL value changed: 1 -> SIO power good 
>> de-assert event received", etc. operations and then to turn off the power.
>> 
>> Can anyone do me a favor to give me some clues for what I was wrong?
> Hi Chris,
> 
> Holding the power button should force a hardware shutdown.  The BMC 
> only monitors this flow and doesn't participate.
> 
> Did your system shut down correctly and you just not see the logs in 
> BMC?  Or, did the system stay on?
> 
> Thanks,
> -Jason
>> 
>> 
>> Thank you in advance for your help.
>> 
>> Regards,
>> Chris Chen
>> 
>> Legal Disclaimer :
>> The information contained in this message may be privileged and 
>> confidential.
>> It is intended to be read only by the individual or entity to whom it 
>> is addressed or by their designee. If the reader of this message is 
>> not the intended recipient, you are on notice that any distribution 
>> of this message, in any form, is strictly prohibited. If you have 
>> received this message in error, please immediately notify the sender 
>> and delete or destroy any copy of this message!