| Date | Wed, 14 Jul 2021 12:59:24 +0200 | From | Peter Zijlstra <> | Subject | Re: [patch 26/50] locking/rtmutex: Provide the spin/rwlock core lock function |
| |
On Tue, Jul 13, 2021 at 05:11:20PM +0200, Thomas Gleixner wrote: > +/** > + * rtlock_slowlock_locked - Slow path lock acquisition for RT locks > + * @lock: The underlying rt mutex > + */ > +static void __sched rtlock_slowlock_locked(struct rt_mutex *lock) > +{ > + struct rt_mutex_waiter waiter; > + > + lockdep_assert_held(&lock->wait_lock); > + > + if (try_to_take_rt_mutex(lock, current, NULL)) > + return; > + > + rt_mutex_init_rtlock_waiter(&waiter); > + > + /* Save current state and set state to TASK_RTLOCK_WAIT */ > + current_save_and_set_rtlock_wait_state(); > + > + task_blocks_on_rt_mutex(lock, &waiter, current, RT_MUTEX_MIN_CHAINWALK); > + > + for (;;) { > + /* Try to acquire the lock again. */ > + if (try_to_take_rt_mutex(lock, current, &waiter)) > + break; > + > + raw_spin_unlock_irq(&lock->wait_lock); > + > + schedule_rtlock(); > + > + raw_spin_lock_irq(&lock->wait_lock); > + set_current_state(TASK_RTLOCK_WAIT);
Since both waker and waiter are fully serialized on ->wait_lock, this can be __set_current_state(), there is no race condition AFAICT.
Might not be worth it though..
> + } > + > + /* Restore the task state */ > + current_restore_rtlock_saved_state(); > + > + /* > + * try_to_take_rt_mutex() sets the waiter bit unconditionally. We > + * might have to fix that up: > + */ > + fixup_rt_mutex_waiters(lock); > + debug_rt_mutex_free_waiter(&waiter); > +}
|