Welcome to our peer-to-peer forums, where owners help owners. Need help now? Visit eSupport here.

English Community

Linux Operating SystemsOther Linux Discussions
All Forum Topics
Options

104 Posts

02-14-2021

Ukraine

139 Signins

665 Page Views

  • Posts: 104
  • Registered: ‎02-14-2021
  • Location: Ukraine
  • Views: 665
  • Message 571 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-14, 5:25 AM

basically it’s a tweaks from our side. and the firmware (hardware or software) looks faulty anyway. at least how i see it. 

Reply
Options

390 Posts

03-06-2021

Germany

221 Signins

2355 Page Views

  • Posts: 390
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2355
  • Message 572 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-14, 11:02 AM

@ Drishal wrote:

I wonder at this point, what if its more of an AMD renoir issue than a thinkpad issue?

 

No, it is not.

 

If that would be the case, no workaround would ever work. A broken SoC remains broken

and can be only fixed ( workaround ) from the vendor. In this case with uucode & AGESA.

 

Alone the fact that 'disabling' devices in BIOS *increases* the power consummation,

shows the BIOS is doing funny things.

 

 

Reply
Options

22 Posts

08-20-2020

Estonia

22 Signins

165 Page Views

  • Posts: 22
  • Registered: ‎08-20-2020
  • Location: Estonia
  • Views: 165
  • Message 573 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-14, 12:26 PM

@osnix are your patches backport from 5.14 or did you fix something else?

I compiled 5.14-rc1 and sleep power usage is up again to 0.8W

yesterday before I put my laptop to sleep, I also disabled TLP, maybe that is causing issues(I wanted to test that maybe it really does lower power usage even further on sleep)

Reply
Options

390 Posts

03-06-2021

Germany

221 Signins

2355 Page Views

  • Posts: 390
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2355
  • Message 574 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-14, 13:11 PM

@ mikk150 wrote:

@osnix are your patches backport from 5.14 or did you fix something else?

I compiled 5.14-rc1 and sleep power usage is up again to 0.8W

yesterday before I put my laptop to sleep, I also disabled TLP, maybe that is causing issues(I wanted to test that maybe it really does lower power usage even further on sleep)

 

5.14-rc1 still needs some patch(es). 

 

Here 5.14-rc1+git+patch -> 10h sleep -> 0.325 W/h 

 

```

 

$ uname -r
5.14.0-rc1-S0ix-00016-g40226a3d96ef-dirty

crazy@ThinkPad-P14s:~/suspend$ sudo ./bat_test -s
Mi 14. Jul 02:28:27 CEST 2021
POWER_SUPPLY_NAME=BAT0
POWER_SUPPLY_TYPE=Battery
POWER_SUPPLY_STATUS=Discharging
POWER_SUPPLY_PRESENT=1
POWER_SUPPLY_TECHNOLOGY=Li-poly
POWER_SUPPLY_CYCLE_COUNT=16
POWER_SUPPLY_VOLTAGE_MIN_DESIGN=11550000
POWER_SUPPLY_VOLTAGE_NOW=12770000
POWER_SUPPLY_POWER_NOW=8964000
POWER_SUPPLY_ENERGY_FULL_DESIGN=50500000
POWER_SUPPLY_ENERGY_FULL=51480000
POWER_SUPPLY_ENERGY_NOW=49640000
POWER_SUPPLY_CAPACITY=96
POWER_SUPPLY_CAPACITY_LEVEL=Normal
POWER_SUPPLY_MODEL_NAME=5B10W139
POWER_SUPPLY_MANUFACTURER=LGC
POWER_SUPPLY_SERIAL_NUMBER= 8689

crazy@ThinkPad-P14s:~/suspend$ sudo ./bat_test -r
Mi 14. Jul 12:28:50 CEST 2021
POWER_SUPPLY_NAME=BAT0
POWER_SUPPLY_TYPE=Battery
POWER_SUPPLY_STATUS=Discharging
POWER_SUPPLY_PRESENT=1
POWER_SUPPLY_TECHNOLOGY=Li-poly
POWER_SUPPLY_CYCLE_COUNT=16
POWER_SUPPLY_VOLTAGE_MIN_DESIGN=11550000
POWER_SUPPLY_VOLTAGE_NOW=12599000
POWER_SUPPLY_POWER_NOW=3993000
POWER_SUPPLY_ENERGY_FULL_DESIGN=50500000
POWER_SUPPLY_ENERGY_FULL=51480000
POWER_SUPPLY_ENERGY_NOW=46390000
POWER_SUPPLY_CAPACITY=90
POWER_SUPPLY_CAPACITY_LEVEL=Normal
POWER_SUPPLY_MODEL_NAME=5B10W139
POWER_SUPPLY_MANUFACTURER=LGC
POWER_SUPPLY_SERIAL_NUMBER= 8689

Checking smu_fw_info


=== SMU Statistics ===
Table Version: 3
Hint Count: 1
S0i3 Cycle Count: 1
Time (in us) to S0i3: 356015
Time (in us) in S0i3: 35988703308

=== Active time (in us) ===
DISPLAY  : 0
CPU      : 244435
GFX      : 0
VDD      : 244431
ACP      : 0
VCN      : 0
DF       : 40399
USB0     : 8281
USB1     : 3354

Checking s0ix_stats

=== S0ix statistics ===
S0ix Entry Time: 3895811847
S0ix Exit Time: 1731353570634
Residency Time: 35988703308

 

```

 

Also, I know what is wrong with long wakeup on either kernel.

IOMMU+nvme doesn't play nice ( even with workarounds ).

However IOMMU is not something I know well, so I'll forward that to NVME/Block kernel devels.

 

A solution to work around this is to set iommu to pt, eg: append iommu=pt to kernel CMD.

 

 

A will post some updates a bit later, need to finish some work first :)

Reply
Options

390 Posts

03-06-2021

Germany

221 Signins

2355 Page Views

  • Posts: 390
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2355
  • Message 575 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-14, 13:50 PM

@ mikk150 wrote:

@osnix are your patches backport from 5.14 or did you fix something else?

I compiled 5.14-rc1 and sleep power usage is up again to 0.8W

yesterday before I put my laptop to sleep, I also disabled TLP, maybe that is causing issues(I wanted to test that maybe it really does lower power usage even further on sleep)

 

just to be sure do you have at least:

 

CONFIG_AMD_PMC=y

CONFIG_I2C_HID_ACPI=m

And disabled CONFIG_HSA_AMD

 

in your kernel config?

 

Try this patch against 5.14rc1:

 

https://crazy.dev.frugalware.org/S0ix-5.14rc1.patch

 

Be sure you have at least the 3 options set that way.

Boot with tsc=nowatchdog at least, you may want to try acpi_osi=Linux also.

 

Be sure you have all your devices enabled in BIOS, be sure the Always ON USB option is disabled.

 

A note for all, systemD seems to get confused about some KEY events thinking each KEY press is the POWER button.

I don't use GNOME/KDE power management stuff, so I didn't notice first, however, this is not something I have time to address right now.

 

 

 

 

Reply
Options

11 Posts

04-10-2021

India

37 Signins

440 Page Views

  • Posts: 11
  • Registered: ‎04-10-2021
  • Location: India
  • Views: 440
  • Message 576 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-15, 19:27 PM

Hi All,

 

Seems like a new BIOS has been released V1.34. 

 

CHANGES IN THIS RELEASE
 Version 1.34

 
[Important updates]
- Included a security issue update. 
- Change the processing order of the Block SID commands.

[New functions or enhancements]
- Add Support Shutdown Notification for NVMe SSD during BIOS update.
- Add back flash prevention that can't rollback to older BIOS version.

[Problem fixes]
- Fixed an issue that it can't enter to MS when plug headphone.
- Fixed an issue that No action respond after click the option by finger in BIOS/Boot menu with Touch panel SKU.
- Fixed an issue that Linux Fn+H/M/L keep input and touchpad hang problem.

Has anyone tried updating yet ?
Reply
Options

390 Posts

03-06-2021

Germany

221 Signins

2355 Page Views

  • Posts: 390
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2355
  • Message 577 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-15, 20:37 PM

@ cysys wrote:

Hi All,

 

Seems like a new BIOS has been released V1.34. 

 

CHANGES IN THIS RELEASE Version 1.34 [Important updates]- Included a security issue update. - Change the processing order of the Block SID commands.[New functions or enhancements]- Add Support Shutdown Notification for NVMe SSD during BIOS update.- Add back flash prevention that can't rollback to older BIOS version.[Problem fixes]- Fixed an issue that it can't enter to MS when plug headphone.- Fixed an issue that No action respond after click the option by finger in BIOS/Boot menu with Touch panel SKU.- Fixed an issue that Linux Fn+H/M/L keep input and touchpad hang problem.

Has anyone tried updating yet ?

 

Not me, but interesting. Half of the changelog is about Intel machines :)

Fn+H/M/L etc only works on Intel, MS stuff is irrelevant for us, and the first

part is about the FW update issues some users faced in 1.32. 

 

The only thing is that security issue which I have no idea what it may be.

Reply
Options

35 Posts

01-23-2021

Argentina

51 Signins

430 Page Views

  • Posts: 35
  • Registered: ‎01-23-2021
  • Location: Argentina
  • Views: 430
  • Message 578 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-15, 21:44 PM

Concerning the workarounds, why battery disconnect from BIOS would help to minimize the issue?

Is it something that has to be done once? or at every boot?

Reply
Options

390 Posts

03-06-2021

Germany

221 Signins

2355 Page Views

  • Posts: 390
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2355
  • Message 579 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-15, 23:33 PM

@ aproximacion wrote:

Concerning the workarounds, why battery disconnect from BIOS would help to minimize the issue?

 

       

        From what I see something is wrong with the PSU / BAT firmware or a component in between.

        Why? Well, there are some possibilities. As an example a component that is supposed to

        switch on/off some power lines is stuck. FW sends wrong commands, FW between, PSU,

        sensors, etc doing funny things. and so on. Killing the main power line, which that does, will reset

         everything.

 

        If you are a T series owner you may perform the emergency reset.

         

Is it something that has to be done once? or at every boot?

 

      Once, and if your drain increases again, then you'll have to repeat.

Reply
Options

390 Posts

03-06-2021

Germany

221 Signins

2355 Page Views

  • Posts: 390
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2355
  • Message 580 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-15, 23:35 PM

@ osnix wrote:

@ cysys wrote:

Hi All,

 

Seems like a new BIOS has been released V1.34. 

 

CHANGES IN THIS RELEASE Version 1.34 [Important updates]- Included a security issue update. - Change the processing order of the Block SID commands.[New functions or enhancements]- Add Support Shutdown Notification for NVMe SSD during BIOS update.- Add back flash prevention that can't rollback to older BIOS version.[Problem fixes]- Fixed an issue that it can't enter to MS when plug headphone.- Fixed an issue that No action respond after click the option by finger in BIOS/Boot menu with Touch panel SKU.- Fixed an issue that Linux Fn+H/M/L keep input and touchpad hang problem.

Has anyone tried updating yet ?

 

Not me, but interesting. Half of the changelog is about Intel machines :)

Fn+H/M/L etc only works on Intel, MS stuff is irrelevant for us, and the first

part is about the FW update issues some users faced in 1.32. 

 

The only thing is that security issue which I have no idea what it may be.

 

 That is not the full changelog. TSC issue seems to be fixed too, also that ACP sound issue ( but this one is tricky need to test a bit more ).

Reply
Forum Home

Community Guidelines

Please review our Guidelines before posting.

Learn More

Check out current deals!

Go Shop
X

Save

X

Delete

X

No, I don’t want to share ideas Yes, I agree to these terms