* Re: [PATCH v1] platform/x86: redmi-wmi: add display-switch key binding
2026-09-28 22:14 [PATCH v1] platform/x86: redmi-wmi: add display-switch key binding Ilya Gladyshev
@ 2026-09-29 23:42 ` Антон Карасев
0 siblings, 0 replies; 2+ messages in thread
From: Антон Карасев @ 2026-09-29 23:42 UTC (permalink / raw)
To: Ilya Gladyshev
Cc: linux-kernel, platform-driver-x86, ilpo.jarvinen, hansg,
qby140326, W_Armin, vlku.milos.fun
Tested-by: Anton Karasev <uselessfire@gmail.com>
Tested on the reporting machine (Xiaomi Redmi Book Pro 16 2024, DMI
XIAOMI/TM2309, BIOS RMAMT6B0P0B0B), kernel 7.2.7, by building redmi-wmi
from mainline with this patch applied and loading it in place of the
in-tree module.
Before: pressing the display-switch key gave "Unknown WMI event with
payload 257" and no input event at all.
After: the key reports KEY_SWITCHVIDEOMODE. The keymap read back from the
device contains both forms now:
scancode 0x101 -> 227
scancode 0x10101 -> 227
and the desktop reacts to the key (KDE opens its display-switch menu).
The other hotkeys are unaffected: 0x201 -> KEY_SELECTIVE_SCREENSHOT,
0x1b01 -> KEY_CONFIG, 0x11801/0x11901 -> KEY_ASSISTANT all still work,
and nothing new showed up in dmesg.
Two notes that did not fit the fix, in case they are useful:
1. The merge series "Merge redmi-wmi into bitland-mifs-wmi" (v6,
20260929134503.17249-1-qby140326@gmail.com) drops redmi-wmi, and its
combined keymap only carries the long form:
{ KE_KEY, BI_HOTKEY_CODE(WMI_EVENT_RESERVED_1, 1, 0), {
KEY_SWITCHVIDEOMODE } },
With WMI_EVENT_RESERVED_1 = 1 and WMI_EVENT_TYPE_HOTKEY = 1 that expands
to 0x00010101; the short 0x00000101 has no entry there. If that series
lands, this fix is lost again on this model. I am replying to that
thread separately.
2. Two payloads are still unhandled here: 0x00000901 and 0x00010901. They
are not key presses -- the EC echoes back the Caps Lock LED state that
the host itself has just set (third byte 1 = on, 0 = off, the same
scheme as the Fn Lock events). Verified by switching between windows
with per-window keyboard layouts, which changes the LED without anyone
touching the key: the events still arrive, 200-500 ms after the LED
change. On a system where Caps Lock switches the layout this is a
steady stream of dev_dbg lines; KE_IGNORE would silence it.
Details and traces: https://bugzilla.kernel.org/show_bug.cgi?id=222062
Thanks for the quick fix!
Anton
вт, 29 сент. 2026 г. в 01:14, Ilya Gladyshev <ilya.gladyshev@linux.dev>:
>
> As reported on the mailing list [1], on the Xiaomi Redmi Book Pro 16
> 2024, the display-switch key emits 0x101 as a WMI payload. Add a
> mapping for it to the keymap.
>
> [1]: https://lore.kernel.org/all/CAFsq4KBXCB=aK8ENHx__6+RcC-ybT7eKjcgsa3TgcJ54x-PLEg@mail.gmail.com/
>
> Reported-by: Антон Карасев <uselessfire@gmail.com>
> Closes: https://bugzilla.kernel.org/show_bug.cgi?id=222062
> Signed-off-by: Ilya Gladyshev <ilya.gladyshev@linux.dev>
> ---
>
> This patch conflicts with the unification of redmi-wmi and
> bitland-mifs-wmi [2]. However, that patch appears to be stuck for now,
> so updating a single mapping in the old driver seems acceptable.
>
> Cc its author, Mingyou Chen.
>
> Also, I can only test that it doesn't break on my hardware, so any
> additional testing is highly welcome.
>
> [2]: https://lore.kernel.org/all/20260816100813.300450-1-qby140326@gmail.com/
> ---
> drivers/platform/x86/redmi-wmi.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/platform/x86/redmi-wmi.c b/drivers/platform/x86/redmi-wmi.c
> index cc82ef502002..155777bf90f3 100644
> --- a/drivers/platform/x86/redmi-wmi.c
> +++ b/drivers/platform/x86/redmi-wmi.c
> @@ -22,6 +22,7 @@ static const struct key_entry redmi_wmi_keymap[] = {
> {KE_KEY, 0x00000301, {KEY_ALL_APPLICATIONS}},
> {KE_KEY, 0x00001b01, {KEY_CONFIG}},
> {KE_KEY, 0x00011b01, {KEY_CONFIG}},
> + {KE_KEY, 0x00000101, {KEY_SWITCHVIDEOMODE}},
> {KE_KEY, 0x00010101, {KEY_SWITCHVIDEOMODE}},
> {KE_KEY, 0x00001a01, {KEY_REFRESH_RATE_TOGGLE}},
>
> --
> 2.55.0
>
^ permalink raw reply [flat|nested] 2+ messages in thread