[PATCH 5.10.y, 5.4.y, 4.19.y] rtmutex: Drop rt_mutex::wait_lock before scheduling

[Thomas Gleixner <tglx@linutronix.de>]

From: Roland Xu <mu001999@outlook.com>

commit d33d26036a0274b472299d7dcdaa5fb34329f91b upstream.

rt_mutex_handle_deadlock() is called with rt_mutex::wait_lock held.  In the
good case it returns with the lock held and in the deadlock case it emits a
warning and goes into an endless scheduling loop with the lock held, which
triggers the 'scheduling in atomic' warning.

Unlock rt_mutex::wait_lock in the dead lock case before issuing the warning
and dropping into the schedule for ever loop.

[ tglx: Moved unlock before the WARN(), removed the pointless comment,
  	massaged changelog, added Fixes tag ]

Fixes: 3d5c9340d194 ("rtmutex: Handle deadlock detection smarter")
Signed-off-by: Roland Xu <mu001999@outlook.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/all/ME0P300MB063599BEF0743B8FA339C2CECC802@ME0P300MB0635.AUSP300.PROD.OUTLOOK.COM
---
Backport to 5.10.y, 5.4.y, 4.19.y
---
 kernel/locking/rtmutex.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

--- a/kernel/locking/rtmutex.c
+++ b/kernel/locking/rtmutex.c
@@ -1205,6 +1205,7 @@ static int __sched
 }
 
 static void rt_mutex_handle_deadlock(int res, int detect_deadlock,
+				     struct rt_mutex *lock,
 				     struct rt_mutex_waiter *w)
 {
 	/*
@@ -1214,6 +1215,7 @@ static void rt_mutex_handle_deadlock(int
 	if (res != -EDEADLOCK || detect_deadlock)
 		return;
 
+	raw_spin_unlock_irq(&lock->wait_lock);
 	/*
 	 * Yell lowdly and stop the task right here.
 	 */
@@ -1269,7 +1271,7 @@ rt_mutex_slowlock(struct rt_mutex *lock,
 	if (unlikely(ret)) {
 		__set_current_state(TASK_RUNNING);
 		remove_waiter(lock, &waiter);
-		rt_mutex_handle_deadlock(ret, chwalk, &waiter);
+		rt_mutex_handle_deadlock(ret, chwalk, lock, &waiter);
 	}
 
 	/*