CVE-2025-38352

{"id": "CVE-2025-38352", "cveTags": [], "metrics": {}, "published": "2025-07-22T08:15:23.577", "references": [{"url": "https://git.kernel.org/stable/c/2c72fe18cc5f9f1750f5bc148cf1c94c29e106ff", "source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}, {"url": "https://git.kernel.org/stable/c/2f3daa04a9328220de46f0d5c919a6c0073a9f0b", "source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}, {"url": "https://git.kernel.org/stable/c/460188bc042a3f40f72d34b9f7fc6ee66b0b757b", "source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}, {"url": "https://git.kernel.org/stable/c/764a7a5dfda23f69919441f2eac2a83e7db6e5bb", "source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}, {"url": "https://git.kernel.org/stable/c/78a4b8e3795b31dae58762bc091bb0f4f74a2200", "source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}, {"url": "https://git.kernel.org/stable/c/c076635b3a42771ace7d276de8dc3bc76ee2ba1b", "source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}, {"url": "https://git.kernel.org/stable/c/c29d5318708e67ac13c1b6fc1007d179fb65b4d7", "source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}, {"url": "https://git.kernel.org/stable/c/f90fff1e152dedf52b932240ebbd670d83330eca", "source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}], "vulnStatus": "Awaiting Analysis", "descriptions": [{"lang": "en", "value": "In the Linux kernel, the following vulnerability has been resolved:\n\nposix-cpu-timers: fix race between handle_posix_cpu_timers() and posix_cpu_timer_del()\n\nIf an exiting non-autoreaping task has already passed exit_notify() and\ncalls handle_posix_cpu_timers() from IRQ, it can be reaped by its parent\nor debugger right after unlock_task_sighand().\n\nIf a concurrent posix_cpu_timer_del() runs at that moment, it won't be\nable to detect timer->it.cpu.firing != 0: cpu_timer_task_rcu() and/or\nlock_task_sighand() will fail.\n\nAdd the tsk->exit_state check into run_posix_cpu_timers() to fix this.\n\nThis fix is not needed if CONFIG_POSIX_CPU_TIMERS_TASK_WORK=y, because\nexit_task_work() is called before exit_notify(). But the check still\nmakes sense, task_work_add(&tsk->posix_cputimers_work.work) will fail\nanyway in this case."}, {"lang": "es", "value": "En el kernel de Linux, se ha resuelto la siguiente vulnerabilidad: posix-cpu-timers: corregir la ejecuci\u00f3n entre handle_posix_cpu_timers() y posix_cpu_timer_del(). Si una tarea no autorecuperadora que est\u00e1 saliendo ya ha pasado exit_notify() y llama a handle_posix_cpu_timers() desde IRQ, puede ser recuperada por su padre o depurador justo despu\u00e9s de unlock_task_sighand(). Si en ese momento se ejecuta un posix_cpu_timer_del() concurrente, no podr\u00e1 detectar timer->it.cpu.firing != 0: cpu_timer_task_rcu() y/o lock_task_sighand() fallar\u00e1n. Agregue la verificaci\u00f3n tsk->exit_state en run_posix_cpu_timers() para corregir esto. Esta correcci\u00f3n no es necesaria si CONFIG_POSIX_CPU_TIMERS_TASK_WORK=y, porque exit_task_work() se llama antes de exit_notify(). Pero la verificaci\u00f3n sigue teniendo sentido, task_work_add(&tsk->posix_cputimers_work.work) fallar\u00e1 de todos modos en este caso."}], "lastModified": "2025-07-22T13:05:40.573", "sourceIdentifier": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}