Re: [PATCH v3 1/2] platform/x86: Add Uniwill laptop driver

[Armin Wolf <W_Armin@gmx.de>]

Am 03.09.25 um 19:08 schrieb Werner Sembach:

> Hi,
>
> started to look into the driver regarding TUXEDO NB02 devices support, 
> starting with the FN-Keys:
>
> Am 31.08.25 um 21:27 schrieb Armin Wolf:
>> +static const struct key_entry uniwill_keymap[] = {
>> +    /* Reported via keyboard controller */
>> +    { KE_IGNORE,    UNIWILL_OSD_CAPSLOCK,                   { 
>> KEY_CAPSLOCK }},
>> +    { KE_IGNORE,    UNIWILL_OSD_NUMLOCK,                    { 
>> KEY_NUMLOCK }},
>> +
>> +    /* Reported when the user locks/unlocks the super key */
>> +    { KE_IGNORE,    UNIWILL_OSD_SUPER_KEY_LOCK_ENABLE,      { 
>> KEY_UNKNOWN }},
>> +    { KE_IGNORE,    UNIWILL_OSD_SUPER_KEY_LOCK_DISABLE,     { 
>> KEY_UNKNOWN }},
>
> Can you also add
>
> { KE_IGNORE,    UNIWILL_OSD_SUPER_KEY_LOCK_CHANGED,     { KEY_UNKNOWN }},
>
> ?
>
> UNIWILL_OSD_SUPER_KEY_LOCK_ENABLE and 
> UNIWILL_OSD_SUPER_KEY_LOCK_DISABLE are always sent in pair 
> with UNIWILL_OSD_SUPER_KEY_LOCK_CHANGED (at least on my test device) 
> and without this line an unknown key event is generated (as that is 
> not explicitly marked as KE_IGNORE without the line).

OK.

>
>> +
>> +    /* Reported in manual mode when toggling the airplane mode 
>> status */
>> +    { KE_KEY,       UNIWILL_OSD_RFKILL,                     { 
>> KEY_RFKILL }},
>> +
>> +    /* Reported when user wants to cycle the platform profile */
>> +    { KE_IGNORE,    UNIWILL_OSD_PERFORMANCE_MODE_TOGGLE,    { 
>> KEY_UNKNOWN }},
> This is a physical button on the gaming devices from Uniwill, my 
> suggestion would be to bind it to F14 (because another ODM has a very 
> similar key that already sends F14 by default) and then let userspace 
> handle it (KDE for example has energy profiles that could be bound to 
> it).
>> +
>> +    /* Reported when the user wants to adjust the brightness of the 
>> keyboard */
>> +    { KE_KEY,       UNIWILL_OSD_KBDILLUMDOWN,               { 
>> KEY_KBDILLUMDOWN }},
>> +    { KE_KEY,       UNIWILL_OSD_KBDILLUMUP,                 { 
>> KEY_KBDILLUMUP }},
>> +
>> +    /* Reported when the user wants to toggle the microphone mute 
>> status */
>> +    { KE_KEY,       UNIWILL_OSD_MIC_MUTE,                   { 
>> KEY_MICMUTE }},
>> +
>> +    /* Reported when the user locks/unlocks the Fn key */
>> +    { KE_IGNORE,    UNIWILL_OSD_FN_LOCK,                    { 
>> KEY_FN_ESC }},
>> +
>> +    /* Reported when the user wants to toggle the brightness of the 
>> keyboard */
>> +    { KE_KEY,       UNIWILL_OSD_KBDILLUMTOGGLE,             { 
>> KEY_KBDILLUMTOGGLE }},
>> +
>> +    /* FIXME: find out the exact meaning of those events */
>> +    { KE_IGNORE,    UNIWILL_OSD_BAT_CHARGE_FULL_24_H,       { 
>> KEY_UNKNOWN }},
>> +    { KE_IGNORE,    UNIWILL_OSD_BAT_ERM_UPDATE,             { 
>> KEY_UNKNOWN }},
>> +
>> +    /* Reported when the user wants to toggle the benchmark mode 
>> status */
>> +    { KE_IGNORE,    UNIWILL_OSD_BENCHMARK_MODE_TOGGLE,      { 
>> KEY_UNKNOWN }},
>> +
>> +    { KE_END }
>> +};
>
> Any reason for still having KEY_* defines even on the ignored events? 
> Looking at other drivers KE_IGNORE events usually don't have it.
>
> Best regards,
>
> Werner

I decided to ignore UNIWILL_OSD_FN_LOCK because i do not know if the Fn + Esc key presses are filtered by the EC or also received by the OS.

Thanks,
Armin Wolf