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 331 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-05-25, 12:15 PM

On P14s I can confirm that with 1.32  it is draining even when connected to powerline. This behavior wasn't present on 1.29 and it seems to me that is not normal??!!

Reply
Options

105 Posts

02-14-2021

Ukraine

144 Signins

705 Page Views

  • Posts: 105
  • Registered: ‎02-14-2021
  • Location: Ukraine
  • Views: 705
  • Message 332 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-05-25, 12:22 PM

regarding bios 1.29. i did not pay too much attention on draining when powered on, but i suspect it was present there as well.

 

my not very engineering thought is that the device/firmware/software that causes this drain affects both powered on/switched off states.

 

meaning that probably if they fix the drain when powered on, it will be fixed for switched off state.

 

to conclude, for me, bios 1.32 DO NOT fix switched off/powered on state battery drain.

Reply
Options

61 Posts

12-04-2020

Portugal

64 Signins

360 Page Views

  • Posts: 61
  • Registered: ‎12-04-2020
  • Location: Portugal
  • Views: 360
  • Message 333 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-05-25, 12:30 PM

On 1.29 I had no issues (drain) when powered on. Also I am using tlp 60-80 thresholds since I bought the machine. 
If it was present on 1.29 it was imperceptible. For instances I started the day with 97.7% at 8:30 am (plugged) and at  13:15 I am at 97.4% (plugged)...

so something is going on ... 

Reply
Options

16 Posts

02-23-2021

Ukraine

28 Signins

120 Page Views

  • Posts: 16
  • Registered: ‎02-23-2021
  • Location: Ukraine
  • Views: 120
  • Message 334 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-05-25, 12:46 PM

@MarkRHPearson, since you are responsible for Linux on Thinkpad, and the issue seems to be with the firmware, affecting both Windows and Linux, would it be worth moving this discussion to some other section of the forum or, at least, inviting someone from the firmware team to this thread?

 

 

Reply
Options

26 Posts

08-20-2020

Estonia

27 Signins

180 Page Views

  • Posts: 26
  • Registered: ‎08-20-2020
  • Location: Estonia
  • Views: 180
  • Message 335 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-05-25, 13:26 PM

Could it be so that touchpad is not turned off..

I mean, if I go to UEFI, turn off touchpad and trackpoint, I can still use touchpad in linux(Not on UEFI tho)

Maybe if firmware team uses same code path to turn off touchpad in sleep, it never happens?

Just ideas.

Reply
Options

407 Posts

03-06-2021

Germany

227 Signins

2420 Page Views

  • Posts: 407
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2420
  • Message 336 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-05-25, 13:37 PM

@ mikk150 wrote:

Could it be so that touchpad is not turned off..

I mean, if I go to UEFI, turn off touchpad and trackpoint, I can still use touchpad in linux(Not on UEFI tho)

Maybe if firmware team uses same code path to turn off touchpad in sleep, it never happens?

Just ideas.

 

I noticed other strange things too while testing that.

 

Turning OFF devices in BIOS will result in a higher S3 drain..

Turning ON TSME will result in a higher S3 drain.

Having any USB devices plugged while in S3 will result in a higher drain.

( even with USB turned off in BIOS ) 

 

You can blacklist the Linux module for your touchpad and see

if that changes something.

Reply
Options

105 Posts

02-14-2021

Ukraine

144 Signins

705 Page Views

  • Posts: 105
  • Registered: ‎02-14-2021
  • Location: Ukraine
  • Views: 705
  • Message 337 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-05-25, 13:59 PM

for me the driver rules all: keyboard/trackpoint/touchpad

 

driver: i8042

 

May 25 16:53:15 thinkpad kernel: input: AT Translated Set 2 keyboard as /devices/platform/i8042/serio3/input/input34

May 25 16:53:16 thinkpad kernel: input: SynPS/2 Synaptics TouchPad as /devices/platform/i8042/serio4/input/input36

May 25 16:53:17 thinkpad kernel: input: TPPS/2 Elan TrackPoint as /devices/platform/i8042/serio4/serio5/input/input37

 

but, why not, let's try.

 

Reply
Options

407 Posts

03-06-2021

Germany

227 Signins

2420 Page Views

  • Posts: 407
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2420
  • Message 338 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-05-25, 16:45 PM

@ ikovnatsky wrote:

for me the driver rules all: keyboard/trackpoint/touchpad

 

driver: i8042

 

May 25 16:53:15 thinkpad kernel: input: AT Translated Set 2 keyboard as /devices/platform/i8042/serio3/input/input34

May 25 16:53:16 thinkpad kernel: input: SynPS/2 Synaptics TouchPad as /devices/platform/i8042/serio4/input/input36

May 25 16:53:17 thinkpad kernel: input: TPPS/2 Elan TrackPoint as /devices/platform/i8042/serio4/serio5/input/input37

 

but, why not, let' try.

 

 

On P14s it is the psmouse driver.

Reply
Options

860 Posts

07-22-2019

Canada

532 Signins

6451 Page Views

  • Posts: 860
  • Registered: ‎07-22-2019
  • Location: Canada
  • Views: 6451
  • Message 339 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-05-25, 19:50 PM

- Adding draining when connected to power to the list, but as an aside I thought that was a "good thing". As in batteries don't last well if constantly charged. Admittedly I'm basing this off a study I read related to EV (car) batteries and I'm not saying it's being done deliberately here - but I thought that usually charging bounced between 95% and 100% by default (at least on Windows - I've been digging into battery enhancements as a background project to see what we can do in Linux). Given the tiny amounts I'm treating this as lower priority - let me know if that causes any wailing and gnashing of teeth (not dismissing it - and honestly it's interesting - but I'm just setting it in my list of issues)

- @osnix - those results are super weird. Thanks for the links.

- @dwik1 - you're right, that I focus on Linux but the FW is common and I'm talking to that team. I'll make sure they're aware it's Windows as well. Not sure if there is a T14 discussion group somewhere in our forums - but if there is might be worth posting there?

- @Iluha - Curious that it wasn't signed. My priority is getting the FW on LVFS (though the previous findings with two users bricking their systems makes me very nervous). I don't recommend the USB route unless you have to. I don't want to distract the FW team with dealing with signing images until we've got the other issues resolved I'm afraid so might sit on this one for a bit.

 

Mark

 

Reply
Options

407 Posts

03-06-2021

Germany

227 Signins

2420 Page Views

  • Posts: 407
  • Registered: ‎03-06-2021
  • Location: Germany
  • Views: 2420
  • Message 340 of 841

Re:T14 AMD battery drain in standby (Linux)

2021-05-25, 19:53 PM

@ osnix wrote:

@ MarkRHPearson wrote:

 

@osnix - I' have to ask the FW team about those; afraid I don' know the answer. The AMD platforms don' have the same DPTF/platform profile support that is on the Intel platforms, but I don' think that' what you' referencing (correct me if I' wrong)

 

 

Take your time.

 

I' referring to is this driver:

 

https://github.com/torvalds/linux/blob/master/drivers/platform/x86/amd-pmc.c

 

https://github.com/torvalds/linux/blob/master/drivers/platform/x86/amd-pmc.c#L275

 

https://patchwork.kernel.org/project/platform-driver-x86/patch/20201105140531.2955555-1-Shyam-sundar.S-k@amd.com/

 

Also, the AGESA version to know if the BIOS can have S0ix support at all, and whatever is worth

trying to poke at it & the SMU.

 

I' try to do some more testing this week.

 

 

Mark, I found the answer myself.

 

In Windows 10 power mode ( aka S0ix ), we have AMDI0005 exposed.

 

Now, I managed to have a somewhat working 5.12.6+patches S0ix kernel and did some testing.

 

Here are the issues we face right now:

 

nvme controller ( even with patch ) has some issues with AMD Vi

( I may be able to fix it by adding another workaround )

 

```

nvme 0000:01:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x000c address=0xcc342000 flags=0x0000]

 

```

 

Once in sleep mode, the touchpad, the keyboard is toast ;)

It won't respond on keypress. ( it may respond by luck from time to time )

 

( workaround: have a USB mouse plugged and move it if you want to resume, after that keyboard may work too )

 

LID sensor is toast, may work on close, ( but let the box in a weird state ), and never on open.

 

( workaround: don't use it ;) use something like echo mem | sudo tee /sys/power/state )

 

Resume from sleep takes forever.

 

Now battery drain, believe it or not, is _the_ same as S3 on my machine.

However, in S0ix mode I can literally hear, something is working inside the laptop.

 

If you wish to test all that yourself, there are only 6 patches needed to have

a somewhat working kernel 5.12.6 with S0ix. ( Just let me know ).

 

Here is my conclusion:

 

We, the users/testers have zero chance to really find out the problem, regardless

of S0/S3, the issue is within the firmware itself, and without a full BIOS ( the one with the

Admin/Advanced menu ) only firmware teams have a chance to fix that.

 

( OK, I could measure on the board probably but that will kill the laptop,

which I'm not really willing to do at this point ;) )

 

PS: I even tested with weird acpi_osi= Windoze versions, there is no difference in

drain whatsoever.

 

Gabriel

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