{"id":"CVE-2026-78118","aliases":["EUVD-2026-107485","GHSA-gm6w-pv43-m3vf"],"status":"published","title":"Race condition on kernel wait queues: unlocked z_waitq_head() read in the timer expiration handler can yield a stale or sentinel thread pointer on SMP","description":"The kernel helper z_waitq_head() (kernel/include/wait_q.h) reads the head of a thread wait queue, but on SMP systems it was called from several places without holding the scheduler spinlock _sched_spinlock that serializes wait-queue mutation. The consequential call site is z_timer_expiration_handler() in kernel/timer.c, which held only the timer subsystem lock timer_lock while reading timer-\u003ewait_q and then unpending, setting the return value of, and readying the thread it obtained. halt_thread() in kernel/sched.c removes an aborted thread from its wait queue while holding only _sched_spinlock, so the two paths are not mutually exclusive.\n\nWhen the last waiter is removed on another CPU inside that window, sys_dlist_peek_head() can return the list sentinel — a pointer into the k_timer object itself — typed as a struct k_thread * (with CONFIG_WAITQ_SCALABLE, rb_get_min() can return an arbitrary node of a tree being restructured), or it can return a pointer to a thread that has already been unpended and, for dynamically allocated threads, freed. The handler then calls unpend_thread_no_timeout(), which performs a sys_dlist_remove() through pointers read from that object, and arch_thread_return_value_set(), which writes unconditionally through the pointer.\n\nThe result is out-of-bounds and use-after-free writes in supervisor context, plus scheduler state corruption; a milder outcome is a thread that has re-pended elsewhere being woken with a success return value from an unrelated blocking call. All of the primitives needed to drive the interleaving — k_timer_start(), k_timer_status_sync(), k_thread_create() and k_thread_abort() — are system calls, so with CONFIG_USERSPACE enabled an unprivileged user-mode thread can attempt the race against kernel-owned data.\n\nExploitation requires a multi-CPU build (CONFIG_MP_MAX_NUM_CPUS greater than one; the affected region is a no-op on uniprocessor targets), local code execution, and winning a narrow timing window, so the practical outcome is most often a kernel crash rather than controlled corruption. The fix splits the helper into z_waitq_head_locked() for callers that already hold the scheduler spinlock and a z_waitq_head() that acquires it, and brings the timer expiration handler's head read, unpend and ready under a single LOCK_SCHED_SPINLOCK region.","severity":{"version":"3.1","score":7,"vector":"CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H","level":"high","assessor":"cna","from":"cve"},"epss":{"day":"2026-11-12","score":0.00104,"percentile":0.01039,"model":"v2026.06.15"},"facts":{"in_kev":false,"known_exploited":false,"public_exploit":false,"fix_available":false},"cwes":["CWE-362"],"published":"2026-10-21T18:19:32Z","modified":"2026-11-04T10:47:23Z","products":[{"vendor":"zephyrproject","product":"zephyr"},{"vendor":"zephyrproject-rtos","product":"zephyr"}],"references":[{"url":"https://github.com/zephyrproject-rtos/zephyr/commit/a9b21fb354e8468dd15db307c3e599dc68b7ef33","tags":["patch","fix"]},{"url":"https://github.com/zephyrproject-rtos/zephyr/security/advisories/GHSA-gm6w-pv43-m3vf","tags":["advisory"]},{"url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/78xxx/CVE-2026-78118.json","tags":["advisory"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-78118","tags":["advisory"]},{"url":"https://github.com/zephyrproject-rtos/zephyr","tags":["package"]}],"sources":{"cve":[{"id":"CVE-2026-78118","title":"Race condition on kernel wait queues: unlocked z_waitq_head() read in the timer expiration handler can yield a stale or sentinel thread pointer on SMP","description":"The kernel helper z_waitq_head() (kernel/include/wait_q.h) reads the head of a thread wait queue, but on SMP systems it was called from several places without holding the scheduler spinlock _sched_spinlock that serializes wait-queue mutation. The consequential call site is z_timer_expiration_handler() in kernel/timer.c, which held only the timer subsystem lock timer_lock while reading timer-\u003ewait_q and then unpending, setting the return value of, and readying the thread it obtained. halt_thread() in kernel/sched.c removes an aborted thread from its wait queue while holding only _sched_spinlock, so the two paths are not mutually exclusive.\n\nWhen the last waiter is removed on another CPU inside that window, sys_dlist_peek_head() can return the list sentinel — a pointer into the k_timer object itself — typed as a struct k_thread * (with CONFIG_WAITQ_SCALABLE, rb_get_min() can return an arbitrary node of a tree being restructured), or it can return a pointer to a thread that has already been unpended and, for dynamically allocated threads, freed. The handler then calls unpend_thread_no_timeout(), which performs a sys_dlist_remove() through pointers read from that object, and arch_thread_return_value_set(), which writes unconditionally through the pointer.\n\nThe result is out-of-bounds and use-after-free writes in supervisor context, plus scheduler state corruption; a milder outcome is a thread that has re-pended elsewhere being woken with a success return value from an unrelated blocking call. All of the primitives needed to drive the interleaving — k_timer_start(), k_timer_status_sync(), k_thread_create() and k_thread_abort() — are system calls, so with CONFIG_USERSPACE enabled an unprivileged user-mode thread can attempt the race against kernel-owned data.\n\nExploitation requires a multi-CPU build (CONFIG_MP_MAX_NUM_CPUS greater than one; the affected region is a no-op on uniprocessor targets), local code execution, and winning a narrow timing window, so the practical outcome is most often a kernel crash rather than controlled corruption. The fix splits the helper into z_waitq_head_locked() for callers that already hold the scheduler spinlock and a z_waitq_head() that acquires it, and brings the timer expiration handler's head read, unpend and ready under a single LOCK_SCHED_SPINLOCK region.","status":"published","published":"2026-10-21T18:19:32.2Z","modified":"2026-10-23T21:28:10.701Z","cvss":[{"version":"3.1","score":7,"vector":"CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H","level":"high","assessor":"cna"}],"cwes":["CWE-362"],"updated_at":"2026-11-02T13:34:59.78118Z"}],"euvd":[{"id":"EUVD-2026-107485","description":"The kernel helper z_waitq_head() (kernel/include/wait_q.h) reads the head of a thread wait queue, but on SMP systems it was called from several places without holding the scheduler spinlock _sched_spinlock that serializes wait-queue mutation. The consequential call site is z_timer_expiration_handler() in kernel/timer.c, which held only the timer subsystem lock timer_lock while reading timer-\u003ewait_q and then unpending, setting the return value of, and readying the thread it obtained. halt_thread() in kernel/sched.c removes an aborted thread from its wait queue while holding only _sched_spinlock, so the two paths are not mutually exclusive.\n\nWhen the last waiter is removed on another CPU inside that window, sys_dlist_peek_head() can return the list sentinel — a pointer into the k_timer object itself — typed as a struct k_thread * (with CONFIG_WAITQ_SCALABLE, rb_get_min() can return an arbitrary node of a tree being restructured), or it can return a pointer to a thread that has already been unpended and, for dynamically allocated threads, freed. The handler then calls unpend_thread_no_timeout(), which performs a sys_dlist_remove() through pointers read from that object, and arch_thread_return_value_set(), which writes unconditionally through the pointer.\n\nThe result is out-of-bounds and use-after-free writes in supervisor context, plus scheduler state corruption; a milder outcome is a thread that has re-pended elsewhere being woken with a success return value from an unrelated blocking call. All of the primitives needed to drive the interleaving — k_timer_start(), k_timer_status_sync(), k_thread_create() and k_thread_abort() — are system calls, so with CONFIG_USERSPACE enabled an unprivileged user-mode thread can attempt the race against kernel-owned data.\n\nExploitation requires a multi-CPU build (CONFIG_MP_MAX_NUM_CPUS greater than one; the affected region is a no-op on uniprocessor targets), local code execution, and winning a narrow timing window, so the practical outcome is most often a kernel crash rather than controlled corruption. The fix splits the helper into z_waitq_head_locked() for callers that already hold the scheduler spinlock and a z_waitq_head() that acquires it, and brings the timer expiration handler's head read, unpend and ready under a single LOCK_SCHED_SPINLOCK region.","status":"published","published":"2026-10-21T18:19:32Z","modified":"2026-10-23T21:28:10Z","cvss":[{"version":"3.1","score":7,"vector":"CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H","level":"high"}],"updated_at":"2026-11-02T16:59:31.52463Z"}],"nvd":[{"id":"CVE-2026-78118","description":"The kernel helper z_waitq_head() (kernel/include/wait_q.h) reads the head of a thread wait queue, but on SMP systems it was called from several places without holding the scheduler spinlock _sched_spinlock that serializes wait-queue mutation. The consequential call site is z_timer_expiration_handler() in kernel/timer.c, which held only the timer subsystem lock timer_lock while reading timer-\u003ewait_q and then unpending, setting the return value of, and readying the thread it obtained. halt_thread() in kernel/sched.c removes an aborted thread from its wait queue while holding only _sched_spinlock, so the two paths are not mutually exclusive.\n\nWhen the last waiter is removed on another CPU inside that window, sys_dlist_peek_head() can return the list sentinel — a pointer into the k_timer object itself — typed as a struct k_thread * (with CONFIG_WAITQ_SCALABLE, rb_get_min() can return an arbitrary node of a tree being restructured), or it can return a pointer to a thread that has already been unpended and, for dynamically allocated threads, freed. The handler then calls unpend_thread_no_timeout(), which performs a sys_dlist_remove() through pointers read from that object, and arch_thread_return_value_set(), which writes unconditionally through the pointer.\n\nThe result is out-of-bounds and use-after-free writes in supervisor context, plus scheduler state corruption; a milder outcome is a thread that has re-pended elsewhere being woken with a success return value from an unrelated blocking call. All of the primitives needed to drive the interleaving — k_timer_start(), k_timer_status_sync(), k_thread_create() and k_thread_abort() — are system calls, so with CONFIG_USERSPACE enabled an unprivileged user-mode thread can attempt the race against kernel-owned data.\n\nExploitation requires a multi-CPU build (CONFIG_MP_MAX_NUM_CPUS greater than one; the affected region is a no-op on uniprocessor targets), local code execution, and winning a narrow timing window, so the practical outcome is most often a kernel crash rather than controlled corruption. The fix splits the helper into z_waitq_head_locked() for callers that already hold the scheduler spinlock and a z_waitq_head() that acquires it, and brings the timer expiration handler's head read, unpend and ready under a single LOCK_SCHED_SPINLOCK region.","status":"published","published":"2026-10-21T19:17:03.787Z","modified":"2026-10-23T22:17:00.45Z","cvss":[{"version":"3.1","score":7,"vector":"CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H","level":"high","assessor":"secondary"}],"cwes":["CWE-362"],"updated_at":"2026-11-02T14:13:02.371082Z"}],"osv":[{"id":"CVE-2026-78118","title":"Race condition on kernel wait queues: unlocked z_waitq_head() read in the timer expiration handler can yield a stale or sentinel thread pointer on SMP","description":"The kernel helper z_waitq_head() (kernel/include/wait_q.h) reads the head of a thread wait queue, but on SMP systems it was called from several places without holding the scheduler spinlock _sched_spinlock that serializes wait-queue mutation. The consequential call site is z_timer_expiration_handler() in kernel/timer.c, which held only the timer subsystem lock timer_lock while reading timer-\u003ewait_q and then unpending, setting the return value of, and readying the thread it obtained. halt_thread() in kernel/sched.c removes an aborted thread from its wait queue while holding only _sched_spinlock, so the two paths are not mutually exclusive.\n\nWhen the last waiter is removed on another CPU inside that window, sys_dlist_peek_head() can return the list sentinel — a pointer into the k_timer object itself — typed as a struct k_thread * (with CONFIG_WAITQ_SCALABLE, rb_get_min() can return an arbitrary node of a tree being restructured), or it can return a pointer to a thread that has already been unpended and, for dynamically allocated threads, freed. The handler then calls unpend_thread_no_timeout(), which performs a sys_dlist_remove() through pointers read from that object, and arch_thread_return_value_set(), which writes unconditionally through the pointer.\n\nThe result is out-of-bounds and use-after-free writes in supervisor context, plus scheduler state corruption; a milder outcome is a thread that has re-pended elsewhere being woken with a success return value from an unrelated blocking call. All of the primitives needed to drive the interleaving — k_timer_start(), k_timer_status_sync(), k_thread_create() and k_thread_abort() — are system calls, so with CONFIG_USERSPACE enabled an unprivileged user-mode thread can attempt the race against kernel-owned data.\n\nExploitation requires a multi-CPU build (CONFIG_MP_MAX_NUM_CPUS greater than one; the affected region is a no-op on uniprocessor targets), local code execution, and winning a narrow timing window, so the practical outcome is most often a kernel crash rather than controlled corruption. The fix splits the helper into z_waitq_head_locked() for callers that already hold the scheduler spinlock and a z_waitq_head() that acquires it, and brings the timer expiration handler's head read, unpend and ready under a single LOCK_SCHED_SPINLOCK region.","status":"published","published":"2026-10-21T18:19:32.2Z","modified":"2026-11-04T13:05:59.50759876Z","cvss":[{"version":"3.1","score":7,"vector":"CVSS:3.1/AV:L/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H","level":"high"}],"cwes":["CWE-362"],"updated_at":"2026-11-04T13:40:00.290922Z"}]}}