CVE-2026-46202

In the Linux kernel, the following vulnerability has been resolved:

HID: appletb-kbd: run inactivity autodim from workqueues

The autodim code in hid-appletb-kbd takes backlightdevice->opslock via backlightdevicesetbrightness() -> mutexlock() from two different atomic contexts:

• appletbinactivitytimer() is a struct timer_list callback, so it runs in softirq context. Every expiry triggers

  BUG: sleeping function called from invalid context at kernel/locking/mutex.c:591 Call Trace: <IRQ> __might_resched __mutexlock backlightdevicesetbrightness appletbinactivitytimer calltimerfn runtimersoftirq

• resetinactivitytimer() is called from appletbkbdhidevent() and appletbkbdinpevent(). On real USB hardware these run in softirq/IRQ context (URB completion and input-event dispatch). When the Touch Bar has already been dimmed or turned off, the reset path calls backlightdeviceset_brightness() directly to restore brightness, producing the same warning.

Both call sites hit the same mutex_lock()-from-atomic bug. Fix them together by moving the blocking work onto the system workqueue:

• Convert the inactivity timer from struct timerlist to struct delayedwork; the callback (appletbinactivitywork) now runs in process context where mutex_lock() is legal.
• Add a dedicated struct workstruct restorebrightnesswork and have resetinactivitytimer() schedule it instead of calling backlightdevicesetbrightness() directly.

Cancel both works synchronously during driver tear-down alongside the existing backlight reference drop.

The semantics are unchanged (same delays, same state transitions on dim, turn-off and user activity); only the execution context of the sleeping call changes. The timer field and callback are renamed to match their new type; resetinactivitytimer() keeps its name because it is invoked from input event paths that read naturally as "reset the inactivity timer".