Arch Linux

miroslec

Member

From: Lisbon

Registered: 2013-03-05

Posts: 10

Touchpad Random Fails

Hi.

I've been having a problem with my laptop's Touch-pad, when sometimes it doesn't work, forcing me to reboot until it works.
It seems random and I can't find a way to reproduce the bug.

This is the relevant output:

# dmesg | grep SYNA32A5:00

With the problem:

[   11.932432] i2c_hid_acpi i2c-SYNA32A5:00: HID over i2c has not been provided an Int IRQ
[   11.932461] i2c_hid_acpi: probe of i2c-SYNA32A5:00 failed with error -22

When it runs OK:

[   10.776368] input: SYNA32A5:00 06CB:CE17 Mouse as /devices/platform/AMDI0010:03/i2c-0/i2c-SYNA32A5:00/0018:06CB:CE17.0005/input/input18
[   10.776422] input: SYNA32A5:00 06CB:CE17 Touchpad as /devices/platform/AMDI0010:03/i2c-0/i2c-SYNA32A5:00/0018:06CB:CE17.0005/input/input19
[   10.776458] hid-generic 0018:06CB:CE17.0005: input,hidraw2: I2C HID v1.00 Mouse [SYNA32A5:00 06CB:CE17] on i2c-SYNA32A5:00
[   10.887597] input: SYNA32A5:00 06CB:CE17 Mouse as /devices/platform/AMDI0010:03/i2c-0/i2c-SYNA32A5:00/0018:06CB:CE17.0005/input/input29
[   10.888042] input: SYNA32A5:00 06CB:CE17 Touchpad as /devices/platform/AMDI0010:03/i2c-0/i2c-SYNA32A5:00/0018:06CB:CE17.0005/input/input30
[   10.888131] hid-multitouch 0018:06CB:CE17.0005: input,hidraw2: I2C HID v1.00 Mouse [SYNA32A5:00 06CB:CE17] on i2c-SYNA32A5:00

I've tried some things, like reloading the module, but nothing seems to work.

Any help would be appreciated. Thanks in advance.

Last edited by miroslec (2021-06-18 16:41:44)

magillos

Member

Registered: 2010-07-03

Posts: 34

Re: Touchpad Random Fails

I've got similar issue:

i2c_hid_acpi i2c-DELL0A78:00: HID over i2c has not been provided an Int IRQ

What restores touch pad function for me is:

rmmod i2c_hid_acpi i2c_hid
modprobe i2c-hid-acpi
modprobe i2c-hid

But I haven't yet figured out how to bypass the issue. It doesn't help that it happens quite rarely and randomly.

miroslec

Member

From: Lisbon

Registered: 2013-03-05

Posts: 10

Re: Touchpad Random Fails

The problem seems to have sort itself out (kernel update maybe?).

The bug has only manifested itself once since I opened this thread, but I'll try your suggestion if it happens again.

I have found people with a similar problem using Elantech Touchpads. The common element seems to be the AMD processor. Does your laptop have one by chance?

icar

Member

From: Catalunya

Registered: 2020-07-31

Posts: 450

Re: Touchpad Random Fails

If the mentioned workaround works, one can automate it.

[Unit]
Description=I hope this works
Conflicts=getty@tty1.service
After=systemd-user-sessions.service getty@tty1.service systemd-logind.service

[Service]
ExecStart=/usr/bin/bash -c '\
   /usr/bin/modprobe -r i2c_hid_acpi; \
   ...etc....'

[Install]
WantedBy=multi-user.target

I had this saved from another touchpad issue.

magillos

Member

Registered: 2010-07-03

Posts: 34

Re: Touchpad Random Fails

I am actually testing something very alike. So far it has worked, it seems.

magillos

Member

Registered: 2010-07-03

Posts: 34

Re: Touchpad Random Fails

Does your laptop have one by chance?

Yes, it does. I bought Ryzen laptop recently as it supposed to work nicely with Linux but there are some issues with it that hopefully just need some time to get sorted. Touch pad (not only that it turns off but also switches randomly into mouse mode) and suspend not working being the most annoying.

zjeffer

Member

Registered: 2018-11-08

Posts: 13

Re: Touchpad Random Fails

I have the same issue, the fix by @magillos seems to make it work again. I'm on arch 5.15.2 on a Lenovo Legion 5 Pro with Ryzen 7 5800H.

EDIT: for a 'permanent' fix:

create a file etc/modprobe.d/i2c-touchpad.conf with the following content:

softdep i2c_hid pre: pinctrl_amd

(by https://github.com/Syniurge/i2c-amd-mp2 … -874590455)

It will make sure the pinctrl_amd module will be loaded before loading the i2c_hid touchpad module.

Last edited by zjeffer (2021-11-20 22:32:40)

acheronfail

Member

Registered: 2022-01-03

Posts: 3

Re: Touchpad Random Fails

I can also confirm that on a Metabox (Clevo) Alpha-XR NH58EP I face the same issue, and that @magillos' fix does work. I didn't have the `*_acpi` variant though

rmmod i2c_hid
modprobe i2c-hid

causes it to work again.