[BUG] HP Pavilion x360: Built-in Keyboard and Touchpad disabled permanently after screen removal (Stuck in Tablet Mode)

[BUG] HP Pavilion x360: Built-in Keyboard and Touchpad disabled permanently after screen removal (Stuck in Tablet Mode)

[Ngon]

Hi Dmitry and Linux-Input developers,

  I am reporting a hardware state issue on the HP Pavilion x360
Convertible 14-cd0xxx that causes the built-in keyboard and
  touchpad to become completely non-functional under Linux.


  The Situation:
  I have physically removed the original laptop screen (using the
device as a headless mini-PC). Because the lid/hall sensors and
  rotation sensors are located in the screen assembly, the Embedded
Controller (EC) or BIOS seems to default to a permanent
  "Tablet Mode" or "Lid Closed" state.


  Observations:
   1. Running libinput debug-events clearly shows:
     event20 SWITCH_TOGGLE switch tablet-mode state 1
   2. The keyboard (AT Raw Set 2) and touchpad (SynPS/2 Synaptics) are
detected by the kernel but do not emit any events
      (verified via libinput debug-events).
   3. On Windows, the keyboard still works despite driver warnings,
likely due to a proprietary HP override.
   4. On Linux, common kernel parameters like i8042.nopnp,
i8042.dumbkbd, or button.lid_init_state=open do not restore
      functionality. Blacklisting intel_vbtn and hp_wmi also does not
force the system back to "Laptop Mode."


  Question:
  Is there a way to implement a kernel quirk to force-ignore the
Tablet Mode switch for this specific model, or a way to bypass
  the EC's hard-lock on the i8042 bus when sensors are missing?

  I am available to test any patches or provide further DSDT/log details.

  Best regards,
  kemzsitink

Re: [BUG] HP Pavilion x360: Built-in Keyboard and Touchpad disabled permanently after screen removal (Stuck in Tablet Mode)

[Dmitry Torokhov]

Hi,

On Tue, Mar 03, 2026 at 03:32:42AM +0700, Ngon wrote:
> Hi Dmitry and Linux-Input developers,
> 
>   I am reporting a hardware state issue on the HP Pavilion x360
> Convertible 14-cd0xxx that causes the built-in keyboard and
>   touchpad to become completely non-functional under Linux.
> 
> 
>   The Situation:
>   I have physically removed the original laptop screen (using the
> device as a headless mini-PC). Because the lid/hall sensors and
>   rotation sensors are located in the screen assembly, the Embedded
> Controller (EC) or BIOS seems to default to a permanent
>   "Tablet Mode" or "Lid Closed" state.

So based on the above I can not see how this is classified as a "BUG".
You modified the hardware to the state it is was not designed to operate
in, so it should not be a surprise that some things are not working.

> 
> 
>   Observations:
>    1. Running libinput debug-events clearly shows:
>      event20 SWITCH_TOGGLE switch tablet-mode state 1
>    2. The keyboard (AT Raw Set 2) and touchpad (SynPS/2 Synaptics) are
> detected by the kernel but do not emit any events
>       (verified via libinput debug-events).
>    3. On Windows, the keyboard still works despite driver warnings,
> likely due to a proprietary HP override.
>    4. On Linux, common kernel parameters like i8042.nopnp,
> i8042.dumbkbd, or button.lid_init_state=open do not restore
>       functionality. Blacklisting intel_vbtn and hp_wmi also does not
> force the system back to "Laptop Mode."

With blacklisted intel_vbtn and hp_wmi do you still see an event device
reporting the tablet mode switch?

> 
> 
>   Question:
>   Is there a way to implement a kernel quirk to force-ignore the
> Tablet Mode switch for this specific model, or a way to bypass
>   the EC's hard-lock on the i8042 bus when sensors are missing?

Do you see anything from i8042 in dmesg after you do

	echo 1 > /sys/module/i8042/parameters/debug

If not then you'll need help from folks at HP to see if there is a
EC command re-enabling keyboard and mouse.

Thanks.

-- 
Dmitry