Commit e0fd4d42 authored Nov 14, 2025 by Eslam Khafagy Committed by Thomas Gleixner Nov 14, 2025

posix-timers: Plug potential memory leak in do_timer_create()



When posix timer creation is set to allocate a given timer ID and the
access to the user space value faults, the function terminates without
freeing the already allocated posix timer structure.

Move the allocation after the user space access to cure that.

[ tglx: Massaged change log ]

Fixes: ec2d0c04 ("posix-timers: Provide a mechanism to allocate a given timer ID")
Reported-by:  <syzbot+9c47ad18f978d4394986@syzkaller.appspotmail.com>
Suggested-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: Eslam Khafagy <eslam.medhat1993@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Link: https://patch.msgid.link/20251114122739.994326-1-eslam.medhat1993@gmail.com
Closes: https://lore.kernel.org/all/69155df4.a70a0220.3124cb.0017.GAE@google.com/T/

parent e9a6fb0b

kernel/time/posix-timers.c

+6 −6

Original line number	Diff line number	Diff line
		@@ -475,12 +475,6 @@ static int do_timer_create(clockid_t which_clock, struct sigevent *event,
		if (!kc->timer_create)
		return -EOPNOTSUPP;

		new_timer = alloc_posix_timer();
		if (unlikely(!new_timer))
		return -EAGAIN;

		spin_lock_init(&new_timer->it_lock);

		/* Special case for CRIU to restore timers with a given timer ID. */
		if (unlikely(current->signal->timer_create_restore_ids)) {
		if (copy_from_user(&req_id, created_timer_id, sizeof(req_id)))
		@@ -490,6 +484,12 @@ static int do_timer_create(clockid_t which_clock, struct sigevent *event,
		return -EINVAL;
		}

		new_timer = alloc_posix_timer();
		if (unlikely(!new_timer))
		return -EAGAIN;

		spin_lock_init(&new_timer->it_lock);

		/*
		* Add the timer to the hash table. The timer is not yet valid
		* after insertion, but has a unique ID allocated.