Re: [patch 04/44] posix-cpu-timers: Fixup stale comment

[Frederic Weisbecker <frederic@kernel.org>]

On Mon, Aug 19, 2019 at 04:31:45PM +0200, Thomas Gleixner wrote:
> The comment above cleanup_timers() is outdated. The timers are only removed
> from the task/process list heads but not modified in any other way.
> 
> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
> ---
>  kernel/time/posix-cpu-timers.c |    7 ++++---
>  1 file changed, 4 insertions(+), 3 deletions(-)
> 
> --- a/kernel/time/posix-cpu-timers.c
> +++ b/kernel/time/posix-cpu-timers.c
> @@ -412,9 +412,10 @@ static void cleanup_timers_list(struct l
>  }
>  
>  /*
> - * Clean out CPU timers still ticking when a thread exited.  The task
> - * pointer is cleared, and the expiry time is replaced with the residual
> - * time for later timer_gettime calls to return.
> + * Clean out CPU timers which are still armed when a thread exits. The
> + * timers are only removed from the list. No other updates are done. The
> + * corresponding posix timers are still accessible, but cannot be rearmed.
> + *
>   * This must be called with the siglock held.
>   */
>  static void cleanup_timers(struct list_head *head)

Indeed and I believe we could avoid that step. We remove the sighand at the same
time so those can't be accessed anymore anyway.

exit_itimers() takes care of the last call release and could force remove from
the list (although it might be taken care of in your series, haven't checked yet):

diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c
index 0a426f4e3125..f8f4a07025fd 100644
--- a/kernel/time/posix-cpu-timers.c
+++ b/kernel/time/posix-cpu-timers.c
@@ -383,11 +383,7 @@ static int posix_cpu_timer_del(struct k_itimer *timer)
 	 */
 	sighand = lock_task_sighand(p, &flags);
 	if (unlikely(sighand == NULL)) {
-		/*
-		 * We raced with the reaping of the task.
-		 * The deletion should have cleared us off the list.
-		 */
-		WARN_ON_ONCE(!list_empty(&timer->it.cpu.entry));
+		list_del(&timer->it.cpu.entry);
 	} else {
 		if (timer->it.cpu.firing)
 			ret = TIMER_RETRY;


Reviewed-by: Frederic Weisbecker <frederic@kernel.org>