Hi Arnd,
On Mon, 3 Feb 2025 at 13:46, Arnd Bergmann <arnd@xxxxxxxx> wrote:
>
> On Mon, Feb 3, 2025, at 09:00, Sumit Garg wrote:
> > OP-TEE supplicant is a user-space daemon and it's possible for it
> > being hung or crashed or killed in the middle of processing an OP-TEE
> > RPC call. It becomes more complicated when there is incorrect shutdown
> > ordering of the supplicant process vs the OP-TEE client application which
> > can eventually lead to system hang-up waiting for the closure of the
> > client application.
> >
> > Allow the client process waiting in kernel for supplicant response to
> > be killed rather than indefinitetly waiting in an unkillable state.
>
> It would be good to mention that the existing version ends up in
> a busy-loop here because of the broken wait_for_completion_interruptible()
> loop.
>
> A normal uninterruptible wait should not have resulted in the hung-task
> watchdog getting triggered, but the endless loop would.
Sure, I will add that.
>
> > + if (wait_for_completion_killable(&req->c)) {
> > + if (!mutex_lock_killable(&supp->mutex)) {
> > if (req->in_queue) {
>
> Using mutex_trylock() here is probably clearer than
> mutex_lock_killable(), since a task that is already in the
> process of getting killed won't ever wait for the mutex.
> mutex_lock_killable() does try to get the lock first though
> if nobody else holds it already, which is the same as trylock.
Let's follow-up on this in the other thread.
-Sumit
>
> Arnd
