toshiba_acpi: fn key started working inverse in satellite L50-B

toshiba_acpi: fn key started working inverse in satellite L50-B

[kemal]

in my TOSHIBA SATELLITE L50-B, the fn key started working
the inverse of how it should after upgrading to kernel 6.9.7
fn+function key used to emit an actual function key press, now instead
it emits a hotkey press and the key itself emits an actual function key
press. there are some recent commits to toshiba_acpi, but i cant see how
it could affect my hardware.
what could be the reason i lost the old behavior?

Re: toshiba_acpi: fn key started working inverse in satellite L50-B

[Armin Wolf]

Am 01.07.24 um 17:53 schrieb kemal:

> in my TOSHIBA SATELLITE L50-B, the fn key started working
> the inverse of how it should after upgrading to kernel 6.9.7
> fn+function key used to emit an actual function key press, now instead
> it emits a hotkey press and the key itself emits an actual function key
> press. there are some recent commits to toshiba_acpi, but i cant see how
> it could affect my hardware.
> what could be the reason i lost the old behavior?

Hi,

it seems that commit 23f1d8b47d12 ("platform/x86: toshiba_acpi: Add quirk for buttons on Z830")
botched up the DMI quirk handling.

The global quick flag is tested in toshiba_acpi_enable_hotkeys(), which is called by
toshiba_acpi_setup_keyboard() before the global quirk flag is properly initialized.

Can you compile kernel modules on your device? If yes, then i can send you a prototype version
of this driver which should fix this issue.

Thanks,
Armin Wolf

Re: toshiba_acpi: fn key started working inverse in satellite L50-B

[Armin Wolf]

[-- Attachment #1: Type: text/plain, Size: 1406 bytes --]

Am 01.07.24 um 19:14 schrieb Armin Wolf:

> Am 01.07.24 um 17:53 schrieb kemal:
>
>> in my TOSHIBA SATELLITE L50-B, the fn key started working
>> the inverse of how it should after upgrading to kernel 6.9.7
>> fn+function key used to emit an actual function key press, now instead
>> it emits a hotkey press and the key itself emits an actual function key
>> press. there are some recent commits to toshiba_acpi, but i cant see how
>> it could affect my hardware.
>> what could be the reason i lost the old behavior?
>
> Hi,
>
> it seems that commit 23f1d8b47d12 ("platform/x86: toshiba_acpi: Add
> quirk for buttons on Z830")
> botched up the DMI quirk handling.
>
> The global quick flag is tested in toshiba_acpi_enable_hotkeys(),
> which is called by
> toshiba_acpi_setup_keyboard() before the global quirk flag is properly
> initialized.
>
> Can you compile kernel modules on your device? If yes, then i can send
> you a prototype version
> of this driver which should fix this issue.
>
> Thanks,
> Armin Wolf
>
I attached a prototype version of the driver which should fix the issue. You can build the module using
"make" and load the resulting kernel module using "sudo insmod toshiba_acpi.ko" (remember to unload the
old driver first using "sudo modprobe -r toshiba_acpi").

Can you verify that the prototype driver indeed fixes the issue?

Thanks,
Armin Wolf

[-- Attachment #2: toshiba_acpi.zip --]
[-- Type: application/zip, Size: 18982 bytes --]

Re: toshiba_acpi: fn key started working inverse in satellite L50-B

[kemal]

im very sorry, i think i failed to mail my reply to the previous 
message.
also huge thanks for the quick response and fix.

yes, the prototype driver does indeed fix the issue.
On Pzt, Tem 1 2024 at 21:09:40 +02:00:00, Armin Wolf <W_Armin@gmx.de> 
wrote:
> Am 01.07.24 um 19:14 schrieb Armin Wolf:
> 
>> Am 01.07.24 um 17:53 schrieb kemal:
>> 
>>> in my TOSHIBA SATELLITE L50-B, the fn key started working
>>> the inverse of how it should after upgrading to kernel 6.9.7
>>> fn+function key used to emit an actual function key press, now 
>>> instead
>>> it emits a hotkey press and the key itself emits an actual function 
>>> key
>>> press. there are some recent commits to toshiba_acpi, but i cant 
>>> see how
>>> it could affect my hardware.
>>> what could be the reason i lost the old behavior?
>> 
>> Hi,
>> 
>> it seems that commit 23f1d8b47d12 ("platform/x86: toshiba_acpi: Add
>> quirk for buttons on Z830")
>> botched up the DMI quirk handling.
>> 
>> The global quick flag is tested in toshiba_acpi_enable_hotkeys(),
>> which is called by
>> toshiba_acpi_setup_keyboard() before the global quirk flag is 
>> properly
>> initialized.
>> 
>> Can you compile kernel modules on your device? If yes, then i can 
>> send
>> you a prototype version
>> of this driver which should fix this issue.
>> 
>> Thanks,
>> Armin Wolf
>> 
> I attached a prototype version of the driver which should fix the 
> issue. You can build the module using
> "make" and load the resulting kernel module using "sudo insmod 
> toshiba_acpi.ko" (remember to unload the
> old driver first using "sudo modprobe -r toshiba_acpi").
> 
> Can you verify that the prototype driver indeed fixes the issue?
> 
> Thanks,
> Armin Wolf