Re: [PATCH 1/2] Input: Add a helper for reporting a screen lock key sequence

[Armin Wolf <W_Armin@gmx.de>]

Am 16.04.25 um 22:43 schrieb Mario Limonciello:

> On 4/16/2025 3:39 PM, Armin Wolf wrote:
>> Am 07.04.25 um 17:27 schrieb Mario Limonciello:
>>
>>> From: Mario Limonciello <mario.limonciello@amd.com>
>>>
>>> In the PC industry KEY_SCREENLOCK isn't used as frequently as it used
>>> to be. Modern versions of Windows [1], GNOME and KDE support "META" 
>>> + "L"
>>> to lock the screen. Modern hardware [2] also sends this sequence of
>>> events for keys with a silkscreen for screen lock.
>>>
>>> Introduced a helper input_report_lock_sequence() for drivers to utilize
>>> if they want to send this sequence.
>>>
>>> Link: 
>>> https://support.microsoft.com/en-us/windows/keyboard-shortcuts- 
>>> in-windows-dcc61a57-8ff0-cffe-9796-cb9706c75eec [1]
>>> Link: https://www.logitech.com/en-us/shop/p/k860-split- 
>>> ergonomic.920-009166 [2]
>>> Suggested-by: Armin Wolf <W_Armin@gmx.de>
>>> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
>>> ---
>>>   drivers/input/input.c | 20 ++++++++++++++++++++
>>>   include/linux/input.h |  2 ++
>>>   2 files changed, 22 insertions(+)
>>>
>>> diff --git a/drivers/input/input.c b/drivers/input/input.c
>>> index ec4346f20efdd..dfeace85c4710 100644
>>> --- a/drivers/input/input.c
>>> +++ b/drivers/input/input.c
>>> @@ -508,6 +508,26 @@ void input_copy_abs(struct input_dev *dst, 
>>> unsigned int dst_axis,
>>>   }
>>>   EXPORT_SYMBOL(input_copy_abs);
>>> +/**
>>> + * input_report_lock_sequence - Report key combination to lock the 
>>> screen
>>> + * @dev: input device
>>> + *
>>> + * Key combination used in the PC industry since Windows 7 for 
>>> locking display
>>> + * is META + L. This is also used in GNOME and KDE by default.
>>> + * See https://support.microsoft.com/en-us/windows/keyboard- 
>>> shortcuts-in-windows-dcc61a57-8ff0-cffe-9796-cb9706c75eec
>>> + */
>>
>> Hi,
>>
>> maybe it would make more sense to have the input subsystem 
>> transparently translate between KEY_SCREENLOCK and this
>> special screen lock sequence. This way there would be no room for 
>> regressions as people could enable/disable this
>> behavior via Kconfig.
>
> Wouldn't all drivers that emit KEY_SCREENLOCK still need to be 
> modified to have support to emit KEY_LEFTMETA and KEY_L?
>
> Wouldn't that mean conditional code in every single driver based on 
> the Kconfig?
>
> Is that worth it?
>
This translation code could live inside the input code itself. This way all drivers would benefit from it.

Thanks,
Armin Wolf

>>
>> Thanks,
>> Armin Wolf
>>
>>> +void input_report_lock_sequence(struct input_dev *dev)
>>> +{
>>> +    input_report_key(dev, KEY_LEFTMETA, 1);
>>> +    input_report_key(dev, KEY_L, 1);
>>> +    input_sync(dev);
>>> +    input_report_key(dev, KEY_L, 0);
>>> +    input_sync(dev);
>>> +    input_report_key(dev, KEY_LEFTMETA, 0);
>>> +    input_sync(dev);
>>> +}
>>> +EXPORT_SYMBOL(input_report_lock_sequence);
>>> +
>>>   /**
>>>    * input_grab_device - grabs device for exclusive use
>>>    * @handle: input handle that wants to own the device
>>> diff --git a/include/linux/input.h b/include/linux/input.h
>>> index 7d7cb0593a63e..16f7bef12f1c1 100644
>>> --- a/include/linux/input.h
>>> +++ b/include/linux/input.h
>>> @@ -492,6 +492,8 @@ void input_set_abs_params(struct input_dev *dev, 
>>> unsigned int axis,
>>>   void input_copy_abs(struct input_dev *dst, unsigned int dst_axis,
>>>               const struct input_dev *src, unsigned int src_axis);
>>> +void input_report_lock_sequence(struct input_dev *dev);
>>> +
>>>   #define INPUT_GENERATE_ABS_ACCESSORS(_suffix, _item)            \
>>>   static inline int input_abs_get_##_suffix(struct input_dev *dev,    \
>>>                         unsigned int axis)        \
>