heck-schneider@tutamail.com:
If the Fn keys genuinely no longer work, it would probably be best to
replace the laptop. A refurbished one costs between $50 and $150.
Otherwise, the problem is most likely due to an incorrect installation
or to files having been modified, because in my humble opinion it has
nothing to do with what Google?s AI is saying.
I would add one thing that you may not like: if you don?t have the
passwords, is it actually your PC?
Michel Verdier <listes@verdier.net>:
On 2026-09-11, Steinar Bang wrote:
1. Am I on the right track with suspecting an NVRAM overwrite done by a
kernel upgrade for both the BIOS password and the missing Fn keys?
I have never seen a kernel upgrade breaking the bios. Do you use EFI ?
3. Or does the Fn issue have nothing to do with NVRAM update? Was it
caused by an X/wayland update at around the same time? Is there a
way to get Fn back without messing with the BIOS at all=
You can use xev to check your keyboard. Note that Fn do nothing but
Fn+key give a keycode different from key alone.
<tomas@tuxteam.de>:
... and (sometimes?) BIOSes have a setting to change whether you
get function keys with Fn and the "fancy" things (backlight, etc)
without or the other way around.
Michel Verdier <listes@verdier.net>:
On 2026-09-12, Steinar Bang wrote:
I have never seen a kernel upgrade breaking the bios. Do you use EFI ?
I think the BIOS on the Acer Aspire actually is an UEFI thingy, if
that's what you mean?
Yes. Perhaps an upgrade failed but I don't see how it could break the
bios. Could you provide
dpkg -l firmw\* efi\*
KeymapNotify event, serial 38, synthetic NO, window 0x0,
keys: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
This is generate when you change focus. Could you also check ESC key with
xev ?
David Wright <deblis@lionunicorn.co.uk>:
Acting as hot-keys (usually configured to require holding Fn, though
not in your case), they would typically produce a variety of outputs
in xev, varying from KeyPress/Release, through Focus*/Notify* etc,
to nothing. Nothing is where the action is dealt with entirely by the firmware, with no need to inform the OS. (My keyboard backlight
hot-key, Fn-F5, behaves like that.)
Again typically, the BIOS /swaps/ whether holding Fn yields F1 ? F12
or their hot-key equivalents, but doesn't turn either mode off.
(Of course, the OS could ignore F1 ? F12 altogether if it so wished.)
As for resetting the BIOS to eliminate mobo password(s), the "easiest"
way I know of is to completely discharge the mobo power. With a
laptop, the "ease" depends on how straightforward it is to remove the battery, and coin cell where fitted. (That could involve finding a
competent person to do it for you.)?
Having done that (with AC power disconnected of course), it remains to discharge the so-called flea-power by holding down the power key for
as long as you can. A couple of minutes should suffice, but YMMV.
When you reconnect the battery, passwords stored on the mobo should
have been reset (along with things like the RTC).
? There may be reset pads: jumper pins have disappeared AFAICT as
devices get ever thinner. Ditto coin cells. Insert disclaimers
re loss of warranties, risk of damage, etc.
Michel Verdier <listes@verdier.net>:
On 2026-09-12, Steinar Bang wrote:
[...]
All seems good. I am afraid you have to enter bios.
You could try to reset bios password, perhaps removing cmos battery (I
used this trick but it was a long time ago), perhaps using a jumper on
the motherboard. Search internet to find more infos about your laptop.
Michel Verdier <listes@verdier.net>:
On 2026-09-12, Steinar Bang wrote:
I have never seen a kernel upgrade breaking the bios. Do you use EFI ?
I think the BIOS on the Acer Aspire actually is an UEFI thingy, if
that's what you mean?
Yes. Perhaps an upgrade failed but I don't see how it could break the
bios.
Stefan Monnier <monnier@iro.umontreal.ca>:
Hmm... the fact that you get the exact same result regardless if the
Fn key is pressed or not suggests that maybe it's just that the Fn key
is malfunctioning.
Michel Verdier <listes@verdier.net>:
Fn key doesn't generate event but it changes the keycode of the associate key. It is done somewhere before xev (bios I suppose ? perhaps firmware ?).
David Wright <deblis@lionunicorn.co.uk>:
ScrollLock, similar to NumLock and CapsLock.
Also Fn has no events in xev at all. Its neighbours Ctrl, Windows and
Alt has full event activity.
I would expect Fn alone to generate no activity. (I don't have a
device that does.)
Hm... could it be as simple as the Fn key being physically broken...?
The key doesn't feel stuck in any way...?
That's always possible, but I've only experienced keys that require
extra pressure, like the left Ctrl and Delete keys on one of my
laptops. However, those two keys are at the diagonal extremes of the keyboard, which might be contributory in some way. Or a SpaceBar,
where you had to press hard at one end because its switch underneath
was near the other end.
Keyboards attract dirt, dust and food. Vacuuming can help.
I also recall your laptop case flexes slightly?is it a cheap build?
I saw your post mentioning INSYDEH2O NVRAM passwords. I've never had
one; do you? Does the password have to be clear for this bug to strike?
The firmware can act on a key and tell the OS nothing, as in my
own keyboard backlight example. (I'm assuming the kernel is honest
with xev.) The BIOS swap is all or nothing: it reverses the meaning
of pressed/unpressed for the entire set of Fn-keys.
I don't know the search terms by which you reached [4], but it seems
natural not to self-advertise how to circumvent your own security,
even when googling gives people recipes for doing just that.
Notwithstanding that, similar googling suggests that authorised
resellers can sometimes supply software upon proof of ownership
for unlocking individual machines.
I've no experience of that? I've always set my venerable BIOS password
as my first action upon getting a new PC; a password I've used in
various contexts for over fifty years.
Not for security, but to make sure I can get to the BIOS, change it,
and see that all the passwords are clear? particularly the most
"senior" one, which I then set. (All my computers are hand-me-downs.)
CGS <etphonehomefrance@gmail.com>:
If you search 'unknown BIOS password Acer 5 laptop' you are presented
with a number of promising possibilities.
[snip!]Max Nikulin <manikulin@gmail.com>:
Even if you are not going to flush firmware, read changelog.
I recall I have seen some workarounds related to NVRAM in articles on
UEFI secure boot troubleshooting. Perhaps some commands may be adapted
to inspect if there is no room on the NVRAM storage.
Likely just a coincidence, but the date is not so far from expiration
of MS secure boot keys. Firmware might be buggy. <https://lwn.net/Articles/1029767/>
Jake Edge. Linux and Secure Boot certificate expiration.
20 years ago I fixed repeating corruption of BIOS settings on a
*desktop* by replacing its RTC battery that was discharged. I have no
idea if something similar may happen with modern laptops.
I am skeptical concerning xev. I would start from reading logs for /dev/input* devices detected during boot (sudo journalctl
-b). Libinput tools might be better for diagnostics,
help if the issue with system firmware settings.
Have you tried to boot a live image to confirm that the issue is not
due to some OS setting?
nwe <nwe@gitcoding.net>:
On 9/13/26 11:48 AM, Steinar Bang wrote:
(However, I have never had a key fail for me while it still felt
mechanically ok when pressed...?)
I wouldn't rule it out completely. I've had several (on other
makes/models) over the years that didn't have a noticeable difference
in mechanical response when they quit working.
Upon disassembly (usually hard to do without breaking it) I found a
tiny bit of dust or corrosion. Just enough to prevent the electrical connection from working.
Kamil Jo?ca <kjonca@o2.pl>:
IIRC in acer aspire bios is swith to change this behavior.
Taking into account that "suddenly" you have password on bios I suspcect
you have problem with bios chip.
Vincent Lefevre <vincent@vinc17.net>:
I've tested the Fn key of 2 keyboards with evtest and xev.
With one of them, the Fn key produces lines like
Event: time 1789491077.465131, type 4 (EV_MSC), code 4 (MSC_SCAN), value ff0003
Event: time 1789491077.465131, type 1 (EV_KEY), code 464 (KEY_FN), value 1 Event: time 1789491077.465131, -------------- SYN_REPORT ------------
Event: time 1789491077.553106, type 4 (EV_MSC), code 4 (MSC_SCAN), value ff0003
Event: time 1789491077.553106, type 1 (EV_KEY), code 464 (KEY_FN), value 0 Event: time 1789491077.553106, -------------- SYN_REPORT ------------
with evtest. But the Fn key of the other keyboard produces nothing.
And xev shows nothing with either keyboard.
Vincent Lefevre <vincent@vinc17.net>:
Before the keycode, there is the scan code, and it seems that it is
evdev that maps scan codes to keycodes (the Fn key has its own scan
code). See the following file: /lib/udev/hwdb.d/60-keyboard.hwdb
There are several utilities for low-level tests:
evtest, input-events, input-kbd, lsinput.
There is some information about the Fn key there:
https://askubuntu.com/q/818413/691424
How the Fn key is handled/configured exactly depends on the hardware.
With the keyboard I'm using, there are driver options to be put in a
file under the /etc/modprobe.d directory to change the way the Fn key
behaves (and other things).
Steinar Bang <sb@dod.no>:
How the Fn key is handled/configured exactly depends on the hardware.
With the keyboard I'm using, there are driver options to be put in a
file under the /etc/modprobe.d directory to change the way the Fn key
behaves (and other things).
Hm... ok. Could a kernel module change be the cause of what affects my keyboard? (ie. my recurrent question: is it possible to fix Fn without entering the BIOS?)
| Sysop: | Jacob Catayoc |
|---|---|
| Location: | Pasay City, Metro Manila, Philippines |
| Users: | 4 |
| Nodes: | 4 (0 / 4) |
| Uptime: | 497100:53:47 |
| Calls: | 182 |
| Files: | 744 |
| D/L today: |
39 files (5,932K bytes) |
| Messages: | 73,594 |