CVE-2026-98859 in Linux
Summary
by MITRE • 10/18/2026
In the Linux kernel, the following vulnerability has been resolved:
HID: multitouch: stop the release timer from being rearmed on remove
mt_remove() quiesces the sticky-finger timer before stopping the hardware:
timer_delete_sync(&td->release_timer);
sysfs_remove_group(&hdev->dev.kobj, &mt_attribute_group); hid_hw_stop(hdev);
timer_delete_sync() waits for a running callback and dequeues the timer, but it does not stop the timer from being armed again. The transport is still delivering reports at that point, and the report path rearms it:
if (app->quirks & MT_QUIRK_STICKY_FINGERS) {
if (td->mt_io_flags & MT_IO_SLOTS_MASK) mod_timer(&td->release_timer, jiffies + msecs_to_jiffies(100));
A report that arrives after timer_delete_sync() has returned therefore leaves the timer queued. td is allocated with devm_kzalloc() against hdev->dev, so it is freed when the driver is unbound, after mt_remove() returns. When the timer fires afterwards, mt_expired_timeout() dereferences the freed td:
struct mt_device *td = timer_container_of(td, t, release_timer); struct hid_device *hdev = td->hdev;
if (test_and_set_bit_lock(MT_IO_FLAGS_RUNNING, &td->mt_io_flags))
Simply moving the teardown after hid_hw_stop() does not fix this on its own, because mt_expired_timeout() calls mt_release_contacts(), which walks hdev->inputs; the timer still has to be quiesced before hid_hw_stop() tears the input devices down.
Use timer_shutdown_sync() instead, which additionally makes any later mod_timer() a no-op, so neither ordering constraint has to be traded off against the other. This is the final-teardown pattern the function was introduced for, and hid-wiimote already uses it for the same reason.
Once again VulDB remains the best source for vulnerability data.
Analysis
by VulDB Data Team • 10/18/2026
The vulnerability described involves a use-after-free condition within the Linux kernel's HID multitouch driver subsystem, specifically affecting devices that implement sticky-finger quirks. This flaw arises from an improper sequence of operations during device removal and driver unbinding. When a HID-multitouch device is disconnected or its driver is unloaded, the mt_remove function attempts to clean up resources by first deleting the release timer using timer_delete_sync and subsequently stopping the hardware interface via hid_hw_stop. While timer_delete_sync effectively waits for any currently executing callback to finish and dequeues the pending timer instance, it does not prevent future re-arming of that same timer structure. Consequently, if a HID report arrives after timer_delete_sync returns but before the underlying data structures are fully invalidated or freed, the input processing path may invoke mod_timer to reschedule the release timer based on sticky-finger logic.
The core technical flaw lies in the lifecycle management of the mt_device structure, referenced as td. This structure is allocated using devm_kzalloc bound to the device object hdev->dev. As a result, the memory for td remains valid only until the driver unbinds and the device release callback executes. Because hid_hw_stop is called after timer_delete_sync but before the final cleanup of input devices and potentially other dependent structures, there exists a race window where incoming reports can trigger mod_timer on an already logically removed timer. When this re-armed timer eventually fires, it invokes mt_expired_timeout, which attempts to dereference td via timer_container_of. Since td has likely been freed by that point due to the driver unbinding process completing or being in progress, this results in a use-after-free scenario where kernel memory is accessed after its deallocation.
This vulnerability poses significant operational risks including potential system crashes, denial of service through kernel oops or panic, and potentially arbitrary code execution if an attacker can control the contents of the freed memory region to manipulate function pointers or critical data structures during the dereference operation. The issue stems from a failure to properly synchronize timer lifecycle with device teardown sequences, violating fundamental principles of safe resource management in concurrent environments. It aligns with CWE-416: Use After Free and relates to ATT&CK technique T1059 if exploited for initial access or persistence mechanisms involving kernel-level exploitation chains.
The resolution involves replacing the use of timer_delete_sync with timer_shutdown_sync within the mt_remove function. Unlike its predecessor, timer_shutdown_sync not only waits for running callbacks and dequeues pending timers but also marks the timer as shut down such that any subsequent calls to mod_timer become no-ops. This ensures that even if reports arrive after shutdown is initiated, they cannot re-arm the timer or cause further access to freed memory. Additionally, this change respects the necessary ordering constraints where mt_expired_timeout must still be quiesced before hid_hw_stop tears down input devices, thereby maintaining system stability without introducing new race conditions.
Mitigation strategies for organizations relying on affected Linux distributions include applying vendor-provided kernel updates that incorporate this patch as soon as they become available. For environments unable to immediately update kernels, disabling the specific HID-multitouch driver module or physically disconnecting vulnerable peripheral devices can reduce exposure until patches are deployed. Security teams should monitor for signs of instability in systems utilizing touchscreens or similar multi-touch input devices, particularly during device removal events such as unplugging USB-connected tablets or wireless dongles. Regular auditing of kernel logs for use-after-free warnings and ensuring strict adherence to driver unbinding protocols will help minimize the risk associated with this class of vulnerabilities.