[patch 15/20] posix-timers: Add proper comments in do_timer_create()

From: Thomas Gleixner
Date: Tue Apr 25 2023 - 14:50:17 EST


The comment about timer lifetime at the end of the function is misplaced
and uncomprehensible.

Make it understandable and put it at the right place. Add a new comment
about the visibility of the new timer ID to user space.

Signed-off-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
---
kernel/time/posix-timers.c | 20 +++++++++++---------
1 file changed, 11 insertions(+), 9 deletions(-)

--- a/kernel/time/posix-timers.c
+++ b/kernel/time/posix-timers.c
@@ -528,12 +528,17 @@ static int do_timer_create(clockid_t whi
new_timer->sigq->info.si_tid = new_timer->it_id;
new_timer->sigq->info.si_code = SI_TIMER;

- if (copy_to_user(created_timer_id,
- &new_timer_id, sizeof (new_timer_id))) {
+ if (copy_to_user(created_timer_id, &new_timer_id, sizeof (new_timer_id))) {
error = -EFAULT;
goto out;
}
-
+ /*
+ * After succesful copy out, the timer ID is visible to user space
+ * now but not yet valid because new_timer::signal is still NULL.
+ *
+ * Complete the initialization with the clock specific create
+ * callback.
+ */
error = kc->timer_create(new_timer);
if (error)
goto out;
@@ -543,14 +548,11 @@ static int do_timer_create(clockid_t whi
WRITE_ONCE(new_timer->it_signal, current->signal);
list_add(&new_timer->list, &current->signal->posix_timers);
spin_unlock_irq(&current->sighand->siglock);
-
- return 0;
/*
- * In the case of the timer belonging to another task, after
- * the task is unlocked, the timer is owned by the other task
- * and may cease to exist at any time. Don't use or modify
- * new_timer after the unlock call.
+ * After unlocking sighand::siglock @new_timer is subject to
+ * concurrent removal and cannot be touched anymore
*/
+ return 0;
out:
posix_timer_unhash_and_free(new_timer);
return error;