HI folks,
KDE (and I think my other desktop environments - I’ll re-test shortly) recognizes and interprets touchscreen inputs on a cold boot of the machine but not after a restart.
Here’s a link the output of the debugging script with the touchscreen in a non-working state: debug-logs-Lars-2025-02-09_13h43m.tar.gz - Google Drive
I’ll update this post with the output of the script with the touchscreen in a working state shortly.
This is a continuation of my original thread but focused on a single issue.
EDIT: I’ve confirmed that the touchscreen works on XFCE, Cinnamon, and KDE Plasma, and that I can switch desktop environments and maintain touchscreen functionality. Here’s the output of the debugging script under KDE with the touchscreen in a working state: debug-logs-Lars-2025-02-09_13h52m.tar.gz - Google Drive
I was lazy and did not properly run the debugging script.
Here is a full file of the machine with the touchscreen in a working state after booting fresh: debug-logs-Lars-2025-02-12_07h03m touchscreen working.tar.gz - Google Drive
And here is the output of the script after a reboot while the touchscreen is not working: debug-logs-Lars-2025-02-12_07h08m trouchscreen not working.tar.gz - Google Drive
[ 9.235056] elants_i2c i2c-ELAN0001:00: supply vcc33 not found, using dummy regulator
[ 9.235130] elants_i2c i2c-ELAN0001:00: supply vccio not found, using dummy regulator
[ 9.317309] elants_i2c i2c-ELAN0001:00: invalid 'hello' packet: ff ff ff ff
[ 9.338051] elan_i2c i2c-ELAN0000:00: Elan Touchpad: Module ID: 0x008c, Firmware: 0x0003, Sample: 0x0002, IAP: 0x000e
[ 9.338328] input: Elan Touchpad as /devices/pci0000:00/0000:00:15.1/i2c_designware.1/i2c-1/i2c-ELAN0000:00/input/input4
[ 9.400152] elants_i2c i2c-ELAN0001:00: invalid 'hello' packet: ff ff ff ff
[ 9.482220] elants_i2c i2c-ELAN0001:00: invalid 'hello' packet: ff ff ff ff
[ 9.482572] elants_i2c i2c-ELAN0001:00: (read fw id) unexpected response: ff ff
[ 9.482678] input: Elan Touchscreen as /devices/pci0000:00/0000:00:15.0/i2c_designware.0/i2c-0/i2c-ELAN0001:00/input/input5
The touchscreen is not returning any data it seems. Not sure what can be tweaked other than the timing a little bit - can increase the enable/reset delays and see if that helps
1 Like
Thanks for looking into this!
I got as far as searching both versions of dmesg.log for instances of “elan,” looking at them, and scratching my head before I had to go to work this morning.
I just cold booted, tested, and rebooted twice just to make sure I haven’t fooling myself over the last few weeks, but it seems absolutely consistent.
When the touchscreen is in a non-working state, touching it will crash keyd (which is running thanks to WeirdTreeThing’s cros-keyboard-map script. The key remapping will stop working after the first or second touch. Touches before logging in and starting a session in a desktop environment don’t seem to do anything. If I restart the keyd it will stop immediately with a floating point error if I touch the screen again:
kiddobox@fedora:~$ sudo keyd
[sudo] password for kiddobox:
CONFIG: parsing /etc/keyd/cros.conf
Starting keyd v2.5.0 (581b4e5)
DEVICE: match 0001:0001:a38e6885 /etc/keyd/cros.conf (AT Translated Set 2 keyboard)
DEVICE: ignoring 0000:0006:bdb72f48 (Video Bus)
DEVICE: ignoring 04f3:008c:2111b95b (Elan Touchpad)
DEVICE: ignoring 0000:0000:8f67ac85 (Elan Touchscreen)
DEVICE: match 0000:0000:ac8af39d /etc/keyd/cros.conf (avs_nau8825 Headset Jack)
Floating point exception
Seeing “avs_nau8825 Headset Jack” listed is making me suspicious that maybe my other issues aren’t so unrelated to this one.