1 files changed, 61 insertions, 30 deletions

diff --git a/kernel/futex/pi.c b/kernel/futex/pi.c
index ce2889f12375..90e5197f4e56 100644
--- a/kernel/futex/pi.c
+++ b/kernel/futex/pi.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0-or-later
 
 #include <linux/slab.h>
+#include <linux/sched/rt.h>
 #include <linux/sched/task.h>
 
 #include "futex.h"
@@ -610,29 +611,16 @@ int futex_lock_pi_atomic(u32 __user *uaddr, struct futex_hash_bucket *hb,
 /*
  * Caller must hold a reference on @pi_state.
  */
-static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_pi_state *pi_state)
+static int wake_futex_pi(u32 __user *uaddr, u32 uval,
+			 struct futex_pi_state *pi_state,
+			 struct rt_mutex_waiter *top_waiter)
 {
-	struct rt_mutex_waiter *top_waiter;
 	struct task_struct *new_owner;
 	bool postunlock = false;
 	DEFINE_RT_WAKE_Q(wqh);
 	u32 curval, newval;
 	int ret = 0;
 
-	top_waiter = rt_mutex_top_waiter(&pi_state->pi_mutex);
-	if (WARN_ON_ONCE(!top_waiter)) {
-		/*
-		 * As per the comment in futex_unlock_pi() this should not happen.
-		 *
-		 * When this happens, give up our locks and try again, giving
-		 * the futex_lock_pi() instance time to complete, either by
-		 * waiting on the rtmutex or removing itself from the futex
-		 * queue.
-		 */
-		ret = -EAGAIN;
-		goto out_unlock;
-	}
-
 	new_owner = top_waiter->task;
 
 	/*
@@ -945,7 +933,7 @@ int futex_lock_pi(u32 __user *uaddr, unsigned int flags, ktime_t *time, int tryl
 	to = futex_setup_timer(time, &timeout, flags, 0);
 
 retry:
-	ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &q.key, FUTEX_WRITE);
+	ret = get_futex_key(uaddr, flags, &q.key, FUTEX_WRITE);
 	if (unlikely(ret != 0))
 		goto out;
 
@@ -1002,6 +990,12 @@ retry_private:
 		goto no_block;
 	}
 
+	/*
+	 * Must be done before we enqueue the waiter, here is unfortunately
+	 * under the hb lock, but that *should* work because it does nothing.
+	 */
+	rt_mutex_pre_schedule();
+
 	rt_mutex_init_waiter(&rt_waiter);
 
 	/*
@@ -1039,19 +1033,37 @@ retry_private:
 	ret = rt_mutex_wait_proxy_lock(&q.pi_state->pi_mutex, to, &rt_waiter);
 
 cleanup:
-	spin_lock(q.lock_ptr);
 	/*
 	 * If we failed to acquire the lock (deadlock/signal/timeout), we must
-	 * first acquire the hb->lock before removing the lock from the
-	 * rt_mutex waitqueue, such that we can keep the hb and rt_mutex wait
-	 * lists consistent.
+	 * must unwind the above, however we canont lock hb->lock because
+	 * rt_mutex already has a waiter enqueued and hb->lock can itself try
+	 * and enqueue an rt_waiter through rtlock.
+	 *
+	 * Doing the cleanup without holding hb->lock can cause inconsistent
+	 * state between hb and pi_state, but only in the direction of not
+	 * seeing a waiter that is leaving.
+	 *
+	 * See futex_unlock_pi(), it deals with this inconsistency.
+	 *
+	 * There be dragons here, since we must deal with the inconsistency on
+	 * the way out (here), it is impossible to detect/warn about the race
+	 * the other way around (missing an incoming waiter).
 	 *
-	 * In particular; it is important that futex_unlock_pi() can not
-	 * observe this inconsistency.
+	 * What could possibly go wrong...
 	 */
 	if (ret && !rt_mutex_cleanup_proxy_lock(&q.pi_state->pi_mutex, &rt_waiter))
 		ret = 0;
 
+	/*
+	 * Now that the rt_waiter has been dequeued, it is safe to use
+	 * spinlock/rtlock (which might enqueue its own rt_waiter) and fix up
+	 * the
+	 */
+	spin_lock(q.lock_ptr);
+	/*
+	 * Waiter is unqueued.
+	 */
+	rt_mutex_post_schedule();
 no_block:
 	/*
 	 * Fixup the pi_state owner and possibly acquire the lock if we
@@ -1117,7 +1129,7 @@ retry:
 	if ((uval & FUTEX_TID_MASK) != vpid)
 		return -EPERM;
 
-	ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &key, FUTEX_WRITE);
+	ret = get_futex_key(uaddr, flags, &key, FUTEX_WRITE);
 	if (ret)
 		return ret;
 
@@ -1132,6 +1144,7 @@ retry:
 	top_waiter = futex_top_waiter(hb, &key);
 	if (top_waiter) {
 		struct futex_pi_state *pi_state = top_waiter->pi_state;
+		struct rt_mutex_waiter *rt_waiter;
 
 		ret = -EINVAL;
 		if (!pi_state)
@@ -1144,22 +1157,39 @@ retry:
 		if (pi_state->owner != current)
 			goto out_unlock;
 
-		get_pi_state(pi_state);
 		/*
 		 * By taking wait_lock while still holding hb->lock, we ensure
-		 * there is no point where we hold neither; and therefore
-		 * wake_futex_p() must observe a state consistent with what we
-		 * observed.
+		 * there is no point where we hold neither; and thereby
+		 * wake_futex_pi() must observe any new waiters.
+		 *
+		 * Since the cleanup: case in futex_lock_pi() removes the
+		 * rt_waiter without holding hb->lock, it is possible for
+		 * wake_futex_pi() to not find a waiter while the above does,
+		 * in this case the waiter is on the way out and it can be
+		 * ignored.
 		 *
 		 * In particular; this forces __rt_mutex_start_proxy() to
 		 * complete such that we're guaranteed to observe the
-		 * rt_waiter. Also see the WARN in wake_futex_pi().
+		 * rt_waiter.
 		 */
 		raw_spin_lock_irq(&pi_state->pi_mutex.wait_lock);
+
+		/*
+		 * Futex vs rt_mutex waiter state -- if there are no rt_mutex
+		 * waiters even though futex thinks there are, then the waiter
+		 * is leaving and the uncontended path is safe to take.
+		 */
+		rt_waiter = rt_mutex_top_waiter(&pi_state->pi_mutex);
+		if (!rt_waiter) {
+			raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock);
+			goto do_uncontended;
+		}
+
+		get_pi_state(pi_state);
 		spin_unlock(&hb->lock);
 
 		/* drops pi_state->pi_mutex.wait_lock */
-		ret = wake_futex_pi(uaddr, uval, pi_state);
+		ret = wake_futex_pi(uaddr, uval, pi_state, rt_waiter);
 
 		put_pi_state(pi_state);
 
@@ -1187,6 +1217,7 @@ retry:
 		return ret;
 	}
 
+do_uncontended:
 	/*
 	 * We have no kernel internal state, i.e. no waiters in the
 	 * kernel. Waiters which are about to queue themselves are stuck