rtmutex: use a trylock for waiter lock in trylock

Mike Galbraith captered the following:
| >hardkernel#11 [ffff88017b243e90] _raw_spin_lock at ffffffff815d2596
| >hardkernel#12 [ffff88017b243e90] rt_mutex_trylock at ffffffff815d15be
| >hardkernel#13 [ffff88017b243eb0] get_next_timer_interrupt at ffffffff81063b42
| >hardkernel#14 [ffff88017b243f00] tick_nohz_stop_sched_tick at ffffffff810bd1fd
| >hardkernel#15 [ffff88017b243f70] tick_nohz_irq_exit at ffffffff810bd7d2
| >hardkernel#16 [ffff88017b243f90] irq_exit at ffffffff8105b02d
| >hardkernel#17 [ffff88017b243fb0] reschedule_interrupt at ffffffff815db3dd
| >--- <IRQ stack> ---
| >hardkernel#18 [ffff88017a2a9bc8] reschedule_interrupt at ffffffff815db3dd
| >    [exception RIP: task_blocks_on_rt_mutex+51]
| >hardkernel#19 [ffff88017a2a9ce0] rt_spin_lock_slowlock at ffffffff815d183c
| >hardkernel#20 [ffff88017a2a9da0] lock_timer_base.isra.35 at ffffffff81061cbf
| >hardkernel#21 [ffff88017a2a9dd0] schedule_timeout at ffffffff815cf1ce
| >hardkernel#22 [ffff88017a2a9e50] rcu_gp_kthread at ffffffff810f9bbb
| >hardkernel#23 [ffff88017a2a9ed0] kthread at ffffffff810796d5
| >hardkernel#24 [ffff88017a2a9f50] ret_from_fork at ffffffff815da04c

lock_timer_base() does a try_lock() which deadlocks on the waiter lock
not the lock itself.
This patch takes the waiter_lock with trylock so it should work from interrupt
context as well. If the fastpath doesn't work and the waiter_lock itself is
taken then it seems that the lock itself taken.
This patch also adds a "rt_spin_try_unlock" to keep lockdep happy. If we
managed to take the wait_lock in the first place we should also be able
to take it in the unlock path.

Cc: [email protected]
Reported-by: Mike Galbraith <[email protected]>
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>

include/linux/spinlock_rt.h

@@ -22,6 +22,7 @@ extern void __lockfunc rt_spin_lock(spinlock_t *lock);
 extern unsigned long __lockfunc rt_spin_lock_trace_flags(spinlock_t *lock);
 extern void __lockfunc rt_spin_lock_nested(spinlock_t *lock, int subclass);
 extern void __lockfunc rt_spin_unlock(spinlock_t *lock);
+extern void __lockfunc rt_spin_unlock_after_trylock_in_irq(spinlock_t *lock);
 extern void __lockfunc rt_spin_unlock_wait(spinlock_t *lock);
 extern int __lockfunc rt_spin_trylock_irqsave(spinlock_t *lock, unsigned long *flags);
 extern int __lockfunc rt_spin_trylock_bh(spinlock_t *lock);

kernel/rtmutex.c

@@ -801,10 +801,8 @@ static void  noinline __sched rt_spin_lock_slowlock(struct rt_mutex *lock)
 /*
  * Slow path to release a rt_mutex spin_lock style
  */
-static void  noinline __sched rt_spin_lock_slowunlock(struct rt_mutex *lock)
+static void __sched __rt_spin_lock_slowunlock(struct rt_mutex *lock)
 {
-	raw_spin_lock(&lock->wait_lock);
-
 	debug_rt_mutex_unlock(lock);
 
 	rt_mutex_deadlock_account_unlock(current);
@@ -823,6 +821,23 @@ static void  noinline __sched rt_spin_lock_slowunlock(struct rt_mutex *lock)
 	rt_mutex_adjust_prio(current);
 }
 
+static void  noinline __sched rt_spin_lock_slowunlock(struct rt_mutex *lock)
+{
+	raw_spin_lock(&lock->wait_lock);
+	__rt_spin_lock_slowunlock(lock);
+}
+
+static void  noinline __sched rt_spin_lock_slowunlock_hirq(struct rt_mutex *lock)
+{
+	int ret;
+
+	do {
+		ret = raw_spin_trylock(&lock->wait_lock);
+	} while (!ret);
+
+	__rt_spin_lock_slowunlock(lock);
+}
+
 void __lockfunc rt_spin_lock(spinlock_t *lock)
 {
 	rt_spin_lock_fastlock(&lock->lock, rt_spin_lock_slowlock);
@@ -853,6 +868,13 @@ void __lockfunc rt_spin_unlock(spinlock_t *lock)
 }
 EXPORT_SYMBOL(rt_spin_unlock);
 
+void __lockfunc rt_spin_unlock_after_trylock_in_irq(spinlock_t *lock)
+{
+	/* NOTE: we always pass in '1' for nested, for simplicity */
+	spin_release(&lock->dep_map, 1, _RET_IP_);
+	rt_spin_lock_fastunlock(&lock->lock, rt_spin_lock_slowunlock_hirq);
+}
+
 void __lockfunc __rt_spin_unlock(struct rt_mutex *lock)
 {
 	rt_spin_lock_fastunlock(lock, rt_spin_lock_slowunlock);
@@ -1056,7 +1078,8 @@ rt_mutex_slowtrylock(struct rt_mutex *lock)
 {
 	int ret = 0;
 
-	raw_spin_lock(&lock->wait_lock);
+	if (!raw_spin_trylock(&lock->wait_lock))
+		return ret;
 	init_lists(lock);
 
 	if (likely(rt_mutex_owner(lock) != current)) {

kernel/timer.c

@@ -1398,7 +1398,7 @@ unsigned long get_next_timer_interrupt(unsigned long now)
 		expires = base->next_timer;
 	}
 #ifdef CONFIG_PREEMPT_RT_FULL
-	rt_spin_unlock(&base->lock);
+	rt_spin_unlock_after_trylock_in_irq(&base->lock);
 #else
 	spin_unlock(&base->lock);
 #endif