https://bugzilla.kernel.org/show_bug.cgi?id=221979
Bug ID: 221979
Summary: Lenovo T14 Gen1 AMD: a phantom ACPI unit corrupts
hardware side on resuming from S3. Endless LMB loop,
USB mouse malfunction. thinkpad_acpi. Sleep Power
Button (SLPB)
Product: ACPI
Version: 2.5
Hardware: AMD
OS: Linux
Status: NEW
Severity: high
Priority: P3
Component: Power-Sleep-Wake
Assignee: [email protected]
Reporter: [email protected]
Regression: No
Severity justified as High because the hardware failure persists across warm
reboots (`sudo reboot` does not clear the corrupted EC/hardware state; only a
full hard power-off restores functionality).
- Thinkpad T14 g1
- Void 6.18.49_1 x86_64 AuthenticAMD uptodate rrrm
- Cinnamon 6.6.7-X11
- AMD Ryzen 5 PRO 4650U with Radeon Graphics
- BIOS 1BET87W(1.56 )
- RAM: 24GiB
On the OS starting all is working fine after it boots, but some time I've
noticed that the resuming from the RAM-sleep just killing my LBM just after I
touch the sensor-mouse.
So then, I've restarted the OS, seen that the touchpad and all it functionality
were ok, and:
```
sudo modprobe -r psmouse
sudo modprobe psmouse
```
What I noticed first? I've noticed that it's doesn't help at all and just
instantly broke the touchpad with the same behavior like after resuming; you
touched the sensor, and then you're in the LMB loop, all works except the LMB.
(The "the touchpad and all it functionality were ok" is not fully true; I've
noticed it later that even `sudo reboot` can't revive touchpad's click button
that is in touchpad's bottom, after the sensor-pad)
Then all combinations with sudo modprobe psmouse never fixed the issue, -
synaptics_intertouch=0, =1, whatever.. and putting them into /etc/default/grub
config never change anything.
I was able only to gut my touchpad with `sudo modprobe psmouse proto=imps` so
the loop is gone, and lost at least the half of the touchpad can do.
A bit later I've found the `xinput disable "SynPS/2 Synaptics TouchPad" &&
xinput enable "SynPS/2 Synaptics TouchPad"` line that only breaks the LMB, you
can use the sensor and forget about the issue before it resumes again, but the
fact that hardware side still will be corrupted doesn't gone.
So, then I've noticed that some day, `sudo modprobe -r psmouse && sudo modprobe
psmouse` doesn't corrupt the touchpad or any other behavior. And other day I've
noticed that resuming doesn't corrupt anything for once. So I've assumed that I
need about 5-20 minutes of a RAM sleep so something will gone wrong, - and that
was wrong assumption.
The truth:
My mind tell me trying close-open the lid while suspended, and I've noticed
that it instantly cause the touchpad's sensor failure, the behaviour with the
loop.
So: You can fully shutdown all the power, then power on, and seen in the OS
that touchpad functional (all) is full and maintained. So then: you're
suspending the OS into the RAM-sleep, or just closing the lid. And instantly
after resuming the hardware initialisation will be corrupted before you fully
shutdown.
The fix:
```
$ cat /proc/acpi/wakeup
Device S-state Status Sysfs node
GPP0 S3 *enabled pci:0000:00:02.7
RESA S3 *enabled pci:0000:06:00.0
GPP2 S3 *enabled pci:0000:00:02.6
GPP4 S4 *enabled pci:0000:00:02.4
GPP5 S3 *disabled
L850 S3 *disabled
GPP6 S3 *enabled pci:0000:00:02.2
GPP7 S3 *disabled pci:0000:00:02.3
GP17 S3 *enabled pci:0000:00:08.1
XHC0 S3 *enabled pci:0000:07:00.3
XHC1 S3 *enabled pci:0000:07:00.4
LID S4 *enabled platform:PNP0C0D:00
SLPB S3 *enabled platform:PNP0C0E:00
$ sudo sh -c "echo SLPB > /proc/acpi/wakeup"
```
And then, when the SLPB is disabled - resuming on lid-open doesn't corrupts the
hardware behaviour, touchpad, its click button and other - working fine. All
working fine.
And I don't know why Kernel enabling the SLPB for me, it's a phantom, I don't
have the Sleep Power Button physically.
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are watching the assignee of the bug.
_______________________________________________
acpi-bugzilla mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/acpi-bugzilla
