OK, so the problem is just a small subset of equipment.

Regarding keyboards, most software nowadays uses unicode instead of the old ASCII sequences with the added kludge of code pages.

So when someone depresses a function key, does a unicode binary sequence get sent?  Is it the correct one?

Gnome has been shifting away from using Xorg protocols, and the latest version of Gnome, 3.3, is purported to only recognize Wayland protocols.  At least this is what Wikipedia says about it:

Wayland or X Window System

Main article: Wayland (display server protocol)

GNOME runs on Wayland and the X Window System.[91] Wayland support was introduced in GNOME 3.10[82] and deemed "for the majority of users […] a usable day to day experience" by 3.20,[92] at which point Wayland became the default user session.[93] GNOME 3.24 will extend Wayland compatibility to Nvidia drivers.[94] With GNOME 3.30, it is possible to run GNOME without X.Org Server, using only Wayland.

I think my latest update of Fedora is using Gnome 3.3

My naive interpretation is trying the fix the keyboard problem using XOrg libraries will become less and less successful.  My login screen does have an option to select a version of Gnome desktop that recognizes XOrg protocols.  I have never tried it.  At the page where I type in my password is a small gear icon.  Selecting that allows me to select one of several desktop environments.

Also Wikipedea has this graphic:  https://upload.wikimedia.org/wikipedia/commons/a/a7/Wayland_display_server_protocol.svg

The graphic suggest keyboard and mouse inputs from the evdev linux kernel module are what Wayland monitors for input.  Forcing keyboard sequences to go first to the evdev module might help response.

I hope this is helpful.

Usedtoberich

PS>  I use an Ideapad 330-17IKB

I would be pleased to provide output of log files or other demonstration of how this is successfully implemented on my machine.  Let me know! 

I think this is not a wayland or X.org issue as it is working seamlessly after suspending and waking the laptop up again. The ACPI signals are only sent or recognized after the suspension of the laptop. acpi_listen recognizes function keys only after the suspension. It really seems to be a bios issue which will hopefully get fixed soon.

Hello,

Could it be a user profile issue?

  My locale is set to en[underscore]US.UTF8  /*  Please excuse me this editing window interprets _ character as an editing command, \ does not escape it.      

When I run acpi-listen

Pressing F1 shows   "button/mute  MUTE  0000080  0000000  K

Pressing Fn+F1 shows   ^[OP

Pressing F2 shows   "button/volumedown  VOLDN  0000080  0000000  K

Pressing Fn+F2 shows   ^[OQ

and so forth across the function buttons.

  Would the "C" locale show the same?  I don't know.

Sincerely,

Usedtoberich

Thank you for your efforts, but it is most certainly not a config issue. No key is detected when presseing FN+F1-8 prior to suspend. Not with xev, skowkey or in the logs. This has to be fixed by Lenovo or maybe some clever kernel update.

Yes. You can suspend as often as you want after the first one. I suspend immediately after boot as a workaround. But still annoying.

Hi @DaveOber,

are there any updates on  this issue. It is probably really a simple BIOS fix, as the function keys are working seamlessly after suspending the system.

Hi All,

Same here, E15 AMD G2.

Installed fedora 32 KDE, upgrade to kernel 5.7.17-200 & latest bios update (1.07).

Function keys working well after sleep.

Was not able to get same result with *ubuntu even after installing 5.8 kernel

Waiting on updates too ;-)

Cheers

EDIT: I installed Kubuntu and same result with 5.8 kernel.

Same here on kernel 5.8.4. Reloading the thinkpad_acpi module doesn't help. Suspending once as a workaround is better than nothing.

I updated my Bios to 1.08 and it is still broken. No communication from Lenovo, I am quite disappointed.