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

61 Posts

12-04-2020

Portugal

64 Signins

360 Page Views

  • Posts: 61
  • Registered: ‎12-04-2020
  • Location: Portugal
  • Views: 360
  • Message 551 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-12, 16:54 PM

I would also like to test...new kernel, however I am on vacation, and didn't get the P14s with me (currently not a laptop I can carry out and use on battery).

Anyone testing on Pop Os 21.04?  It would be nice to have a .deb package ;)

I guess with this jump we may start to make or buy stickers saying ... "P14s, T14 and T14s AMD COMMUNITY EDITION!"

will keep following the thread during my holidays...

Big thanks to @Osnix and all the other people testing an helping us out!

Reply
Options

407 Posts

03-06-2021

Germany

227 Signins

2420 Page Views

  • Posts: 407
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2420
  • Message 552 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-12, 20:49 PM

@ 6133126 wrote:

Hey @MarkRHPearson , has the amd team been able to find out what the problem is with external screens (no signal on 4k screen when set to 60hz on 4k screen over usb-c dock gen2)? Could you please give us an update?

 

This is really the wrong thread for this issue.

 

That thread is already very weird and confusing, 

can we please have a thread for each issue?

Reply
Options

407 Posts

03-06-2021

Germany

227 Signins

2420 Page Views

  • Posts: 407
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2420
  • Message 553 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-12, 21:01 PM

Just a small update...

 

Using acpi_osi=Linux seems to have an impact on power consummation too.

 

( I wish BIOS writer would stop using decade(s) old & deprecated code )

 

1h/2h/3h runs are here marginally better with the option.

 

3h -> 0.346666 W/h

2h -> 0.36 W/h

1h -> 0.3689 W/h

 

None of my runs was ever over >0.4 W/h, OK if we scale up the 30min test we are

over but still <0.5 W/h.

 

Tomorrow, I won't have much time for testing suspends, but I'll try to find some time 

to look at some amdgpu patches.

Reply
Options

407 Posts

03-06-2021

Germany

227 Signins

2420 Page Views

  • Posts: 407
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2420
  • Message 554 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-12, 21:07 PM

@ linuxmad wrote:

I would also like to test...new kernel, however I am on vacation, and didn' get the P14s with me (currently not a laptop I can carry out and use on battery).

Anyone testing on Pop Os 21.04?  It would be nice to have a .deb package ;)

I guess with this jump we may start to make or buy stickers saying ... "P14s, T14 and T14s AMD COMMUNITY EDITION!"

will keep following the thread during my holidays...

Big thanks to @Osnix and all the other people testing an helping us out!

 

You are welcome :)

 

I never used Pop Os, and for my testing, I never use distro kernels.

I use kernel.org kernels + patches_for_the_particuar_issue, always.

That way we know 'Linux kernel XYZ has or doesn't have issue Z'.

 

Distro kernels are usually ' Frankenstein kernels', especially RH, Fedora & Ubuntu ones.

You can't be sure what kind of patches they have backported, added, removed, or pushed broken ones in.

 

Reply
Options

18 Posts

09-15-2020

Austria

29 Signins

150 Page Views

  • Posts: 18
  • Registered: ‎09-15-2020
  • Location: Austria
  • Views: 150
  • Message 555 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-12, 21:32 PM

you're right, @osnix .. I'll repost it in the linked thread.

Reply
Options

407 Posts

03-06-2021

Germany

227 Signins

2420 Page Views

  • Posts: 407
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2420
  • Message 556 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-12, 21:48 PM

@ MarkRHPearson wrote:

....

As an aside, I personally have only seen the 0.6W in my testing. I' been doing long sleep testing checking the %age drain over 10hr sleeps. My laptop isn' new and I' running Fedora34 and 1.32 BIOS on it. I found similar to osnix with the disabling devices led to higher power but in my case it was marginal.

 

 

I don't have a T series laptop to test, and they seem to have a bit different drain too.

While you see 0.6W/h, other users see 0.8W/h after some workarounds or emergency reset.

 

Also, P14s even, depending on revision it seems, can have a high drain or marginal one. Also never

P14s have different parts? In any way a different battery unit.

 

My record, with all devices, disable, in BIOS, in S3 was 1.86W/h ;)

Also while S0ix looks promising ( but only bc the kernel workarounds BIOS bugs ),

S3 remains still br0ken.

 

I still cannot make sense of the _PTS method and why is written that way.

 

A small snippet:

 

```

 

If ((Arg0 == 0x03))
            {
                \_SB.PCI0.LPC0.EC0.S3ST = 0x00 <-- Why? What does it workaround? :)
                \_SB.PCI0.LPC0.EC0.S3ST = 0x01
                SLTP () <-- then throw a SMI on 0x05 zerowing everything ?
                ACST = \_SB.PCI0.LPC0.EC0.AC._PSR () <-- then AC status ? 

 

```

 

BTW same on WAK, I may be wrong, but for me, those look like workarounds.

 

 

Another thing to note is the only 2 big companies with this kind of code,

eg: Lenovo & HP ;), have both the biggest problems regarding a lot of things including,

suspend & power draw :), the difference is HP doesn't support Linux on the affected machines while

Lenovo does at least on the TP series ;)

 

Gabriel

Reply
Options

15 Posts

06-22-2021

Germany

34 Signins

135 Page Views

  • Posts: 15
  • Registered: ‎06-22-2021
  • Location: Germany
  • Views: 135
  • Message 557 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-12, 21:57 PM

i did a quick test with @osnix patches on my t14s.

 

Kernel options ubuntu defaults +(as suggested by osnix):

CONFIG_AMD_PMC=y

CONFIG_I2C_HID_ACPI=m

Disable CONFIG_HSA_AMD

 

 

Mon 12 Jul 2021 10:54:48 PM CEST - 56.31 Wh

Mon 12 Jul 2021 11:41:00 PM CEST - 56.25 Wh

 

(numbers from upower before and after sleep - windows 10 mode in bios - and yes, no hibernation but s0ix sleep and yes, disconnected from power)

 

since these numbers are a bit too good to be true i guess its an imprecision in the battery estimates  due to the short 44min sleep time. or is there a more reliable way to do battery measurments then upower ?

 

Ill let it sleep over night and report a more reliable number tomorrow.

Reply
Options

407 Posts

03-06-2021

Germany

227 Signins

2420 Page Views

  • Posts: 407
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2420
  • Message 558 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-12, 22:20 PM
Upower, if I remember correctly, updates the values all X seconds. However see PM ;)
Reply
Options

15 Posts

06-22-2021

Germany

34 Signins

135 Page Views

  • Posts: 15
  • Registered: ‎06-22-2021
  • Location: Germany
  • Views: 135
  • Message 559 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-13, 3:40 AM

So here is a more precise estimate of the power consumption after 5h of s0ix sleep with kernel 5.13.1 and @osnix  patches on a t14s measured with tlp-stat -b :

 

Tue Jul 13 12:10:13 AM CEST 2021
/sys/class/power_supply/BAT0/energy_now                     =  53760 [mWh]

 

Tue Jul 13 05:10:56 AM CEST 2021
/sys/class/power_supply/BAT0/energy_now                     =  51490 [mWh]

 

this amounts to 0.454 W/h which is a further improvement to the ~0.6 W/h which i had previously using only the power drain trick (https://forums.lenovo.com/t5/Other-Linux-Discussions/T14-AMD-battery-drain-in-standby-Linux/m-p/5037674?page=52#5371858)

 

Reply
Options

26 Posts

10-10-2020

Belgium

35 Signins

155 Page Views

  • Posts: 26
  • Registered: ‎10-10-2020
  • Location: Belgium
  • Views: 155
  • Message 560 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-07-13, 5:40 AM

Just some numbers from my T14 Gen 1, Fedora 34, Bios 1.29, Sleep mode Linux :

 

lun 12 jui 2021 22:31:09 CEST
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=186
POWER_SUPPLY_VOLTAGE_MIN_DESIGN=11550000
POWER_SUPPLY_VOLTAGE_NOW=11493000
POWER_SUPPLY_POWER_NOW=6999000
POWER_SUPPLY_ENERGY_FULL_DESIGN=50500000
POWER_SUPPLY_ENERGY_FULL=51480000
POWER_SUPPLY_ENERGY_NOW=29620000
POWER_SUPPLY_CAPACITY=57
POWER_SUPPLY_CAPACITY_LEVEL=Normal
POWER_SUPPLY_MODEL_NAME=5B10W139
POWER_SUPPLY_MANUFACTURER=LGC
POWER_SUPPLY_SERIAL_NUMBER= 4048
mar 13 jui 2021 07:32:57 CEST
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=186
POWER_SUPPLY_VOLTAGE_MIN_DESIGN=11550000
POWER_SUPPLY_VOLTAGE_NOW=11175000
POWER_SUPPLY_POWER_NOW=2424000
POWER_SUPPLY_ENERGY_FULL_DESIGN=50500000
POWER_SUPPLY_ENERGY_FULL=51480000
POWER_SUPPLY_ENERGY_NOW=22100000
POWER_SUPPLY_CAPACITY=42
POWER_SUPPLY_CAPACITY_LEVEL=Normal
POWER_SUPPLY_MODEL_NAME=5B10W139
POWER_SUPPLY_MANUFACTURER=LGC
POWER_SUPPLY_SERIAL_NUMBER= 4048

 

From 10:30pm to 7:30, 9 hours sleep : 29620000−22100000 = 7520000/1000000 = 7,52 W divided by 9 = 0,83W/h.

It went from 57% to 42%.

 

I'd like to test @osnix fixes next.

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