[PATCH -v5 10/14] futex: Pull rt_mutex_futex_unlock() out from under hb->lock

[Peter Zijlstra <peterz@infradead.org>]

[-- Attachment #1: peterz-futex-pi-unlock-3.patch --]
[-- Type: text/plain, Size: 10042 bytes --]

There's a number of 'interesting' problems, all caused by holding
hb->lock while doing the rt_mutex_unlock() equivalient.

Notably:

 - a PI inversion on hb->lock; and,

 - a DL crash because of pointer instability.

Because of all the previous patches that:

 - allow us to do rt_mutex_futex_unlock() without dropping wait_lock;
   which in turn allows us to rely on wait_lock atomicy.

 - changed locking rules to cover {uval,pi_state} with wait_lock.

 - simplified the waiter conundrum.

We can now quite simply pull rt_mutex_futex_unlock() out from under
hb->lock, a pi_state reference and wait_lock are sufficient.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
 kernel/futex.c |  144 ++++++++++++++++++++++++++++++++++++---------------------
 1 file changed, 91 insertions(+), 53 deletions(-)

--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -919,10 +919,12 @@ void exit_pi_state_list(struct task_stru
 		pi_state->owner = NULL;
 		raw_spin_unlock_irq(&curr->pi_lock);
 
-		rt_mutex_futex_unlock(&pi_state->pi_mutex);
-
+		get_pi_state(pi_state);
 		spin_unlock(&hb->lock);
 
+		rt_mutex_futex_unlock(&pi_state->pi_mutex);
+		put_pi_state(pi_state);
+
 		raw_spin_lock_irq(&curr->pi_lock);
 	}
 	raw_spin_unlock_irq(&curr->pi_lock);
@@ -1035,6 +1037,9 @@ static int attach_to_pi_state(u32 __user
 	 * has dropped the hb->lock in between queue_me() and unqueue_me_pi(),
 	 * which in turn means that futex_lock_pi() still has a reference on
 	 * our pi_state.
+	 *
+	 * IOW, we cannot race against the unlocked put_pi_state() in
+	 * futex_unlock_pi().
 	 */
 	WARN_ON(!atomic_read(&pi_state->refcount));
 
@@ -1378,47 +1383,33 @@ static void mark_wake_futex(struct wake_
 	smp_store_release(&q->lock_ptr, NULL);
 }
 
-static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *top_waiter,
-			 struct futex_hash_bucket *hb)
+static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_pi_state *pi_state)
 {
-	struct task_struct *new_owner;
-	struct futex_pi_state *pi_state = top_waiter->pi_state;
 	u32 uninitialized_var(curval), newval;
+	struct task_struct *new_owner;
+	bool deboost = false;
 	DEFINE_WAKE_Q(wake_q);
-	bool deboost;
 	int ret = 0;
 
-	if (!pi_state)
-		return -EINVAL;
-
-	/*
-	 * If current does not own the pi_state then the futex is
-	 * inconsistent and user space fiddled with the futex value.
-	 */
-	if (pi_state->owner != current)
-		return -EINVAL;
-
 	raw_spin_lock_irq(&pi_state->pi_mutex.wait_lock);
 	new_owner = rt_mutex_next_owner(&pi_state->pi_mutex);
-
-	/*
-	 * When we interleave with futex_lock_pi() where it does
-	 * rt_mutex_timed_futex_lock(), we might observe @this futex_q waiter,
-	 * but the rt_mutex's wait_list can be empty (either still, or again,
-	 * depending on which side we land).
-	 *
-	 * When this happens, give up our locks and try again, giving the
-	 * futex_lock_pi() instance time to complete and unqueue_me().
-	 */
 	if (!new_owner) {
-		raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock);
-		return -EAGAIN;
+		/*
+		 * Since we held neither hb->lock nor wait_lock when coming
+		 * into this function, we could have raced with futex_lock_pi()
+		 * such that it will have removed the waiter that brought us
+		 * here.
+		 *
+		 * In this case, retry the entire operation.
+		 */
+		ret = -EAGAIN;
+		goto out_unlock;
 	}
 
 	/*
-	 * We pass it to the next owner. The WAITERS bit is always
-	 * kept enabled while there is PI state around. We cleanup the
-	 * owner died bit, because we are the owner.
+	 * We pass it to the next owner. The WAITERS bit is always kept
+	 * enabled while there is PI state around. We cleanup the owner
+	 * died bit, because we are the owner.
 	 */
 	newval = FUTEX_WAITERS | task_pid_vnr(new_owner);
 
@@ -1441,10 +1432,8 @@ static int wake_futex_pi(u32 __user *uad
 			ret = -EINVAL;
 	}
 
-	if (ret) {
-		raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock);
-		return ret;
-	}
+	if (ret)
+		goto out_unlock;
 
 	raw_spin_lock(&pi_state->owner->pi_lock);
 	WARN_ON(list_empty(&pi_state->list));
@@ -1462,15 +1451,15 @@ static int wake_futex_pi(u32 __user *uad
 	 */
 	deboost = __rt_mutex_futex_unlock(&pi_state->pi_mutex, &wake_q);
 
+out_unlock:
 	raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock);
-	spin_unlock(&hb->lock);
 
 	if (deboost) {
 		wake_up_q(&wake_q);
 		rt_mutex_adjust_prio(current);
 	}
 
-	return 0;
+	return ret;
 }
 
 /*
@@ -2229,7 +2218,8 @@ static int fixup_pi_state_owner(u32 __us
 	/*
 	 * We are here either because we stole the rtmutex from the
 	 * previous highest priority waiter or we are the highest priority
-	 * waiter but failed to get the rtmutex the first time.
+	 * waiter but have failed to get the rtmutex the first time.
+	 *
 	 * We have to replace the newowner TID in the user space variable.
 	 * This must be atomic as we have to preserve the owner died bit here.
 	 *
@@ -2246,7 +2236,7 @@ static int fixup_pi_state_owner(u32 __us
 	if (get_futex_value_locked(&uval, uaddr))
 		goto handle_fault;
 
-	while (1) {
+	for (;;) {
 		newval = (uval & FUTEX_OWNER_DIED) | newtid;
 
 		if (cmpxchg_futex_value_locked(&curval, uaddr, uval, newval))
@@ -2336,6 +2326,10 @@ static int fixup_owner(u32 __user *uaddr
 		/*
 		 * Got the lock. We might not be the anticipated owner if we
 		 * did a lock-steal - fix up the PI-state in that case:
+		 *
+		 * We can safely read pi_state->owner without holding wait_lock
+		 * because we now own the rt_mutex, only the owner will attempt
+		 * to change it.
 		 */
 		if (q->pi_state->owner != current)
 			ret = fixup_pi_state_owner(uaddr, q, current);
@@ -2575,6 +2569,7 @@ static int futex_lock_pi(u32 __user *uad
 			 ktime_t *time, int trylock)
 {
 	struct hrtimer_sleeper timeout, *to = NULL;
+	struct futex_pi_state *pi_state = NULL;
 	struct futex_hash_bucket *hb;
 	struct futex_q q = futex_q_init;
 	int res, ret;
@@ -2661,12 +2656,19 @@ static int futex_lock_pi(u32 __user *uad
 	 * If fixup_owner() faulted and was unable to handle the fault, unlock
 	 * it and return the fault to userspace.
 	 */
-	if (ret && (rt_mutex_owner(&q.pi_state->pi_mutex) == current))
-		rt_mutex_futex_unlock(&q.pi_state->pi_mutex);
+	if (ret && (rt_mutex_owner(&q.pi_state->pi_mutex) == current)) {
+		pi_state = q.pi_state;
+		get_pi_state(pi_state);
+	}
 
 	/* Unqueue and drop the lock */
 	unqueue_me_pi(&q);
 
+	if (pi_state) {
+		rt_mutex_futex_unlock(&pi_state->pi_mutex);
+		put_pi_state(pi_state);
+	}
+
 	goto out_put_key;
 
 out_unlock_put_key:
@@ -2729,10 +2731,36 @@ static int futex_unlock_pi(u32 __user *u
 	 */
 	top_waiter = futex_top_waiter(hb, &key);
 	if (top_waiter) {
-		ret = wake_futex_pi(uaddr, uval, top_waiter, hb);
+		struct futex_pi_state *pi_state = top_waiter->pi_state;
+
+		ret = -EINVAL;
+		if (!pi_state)
+			goto out_unlock;
+
+		/*
+		 * If current does not own the pi_state then the futex is
+		 * inconsistent and user space fiddled with the futex value.
+		 */
+		if (pi_state->owner != current)
+			goto out_unlock;
+
 		/*
-		 * In case of success wake_futex_pi dropped the hash
-		 * bucket lock.
+		 * Grab a reference on the pi_state and drop hb->lock.
+		 *
+		 * The reference ensures pi_state lives, dropping the hb->lock
+		 * is tricky.. wake_futex_pi() will take rt_mutex::wait_lock to
+		 * close the races against futex_lock_pi(), but in case of
+		 * _any_ fail we'll abort and retry the whole deal.
+		 */
+		get_pi_state(pi_state);
+		spin_unlock(&hb->lock);
+
+		ret = wake_futex_pi(uaddr, uval, pi_state);
+
+		put_pi_state(pi_state);
+
+		/*
+		 * Success, we're done! No tricky corner cases.
 		 */
 		if (!ret)
 			goto out_putkey;
@@ -2747,7 +2775,6 @@ static int futex_unlock_pi(u32 __user *u
 		 * setting the FUTEX_WAITERS bit. Try again.
 		 */
 		if (ret == -EAGAIN) {
-			spin_unlock(&hb->lock);
 			put_futex_key(&key);
 			goto retry;
 		}
@@ -2755,7 +2782,7 @@ static int futex_unlock_pi(u32 __user *u
 		 * wake_futex_pi has detected invalid state. Tell user
 		 * space.
 		 */
-		goto out_unlock;
+		goto out_putkey;
 	}
 
 	/*
@@ -2765,8 +2792,10 @@ static int futex_unlock_pi(u32 __user *u
 	 * preserve the WAITERS bit not the OWNER_DIED one. We are the
 	 * owner.
 	 */
-	if (cmpxchg_futex_value_locked(&curval, uaddr, uval, 0))
+	if (cmpxchg_futex_value_locked(&curval, uaddr, uval, 0)) {
+		spin_unlock(&hb->lock);
 		goto pi_faulted;
+	}
 
 	/*
 	 * If uval has changed, let user space handle it.
@@ -2780,7 +2809,6 @@ static int futex_unlock_pi(u32 __user *u
 	return ret;
 
 pi_faulted:
-	spin_unlock(&hb->lock);
 	put_futex_key(&key);
 
 	ret = fault_in_user_writeable(uaddr);
@@ -2884,6 +2912,7 @@ static int futex_wait_requeue_pi(u32 __u
 				 u32 __user *uaddr2)
 {
 	struct hrtimer_sleeper timeout, *to = NULL;
+	struct futex_pi_state *pi_state = NULL;
 	struct rt_mutex_waiter rt_waiter;
 	struct futex_hash_bucket *hb;
 	union futex_key key2 = FUTEX_KEY_INIT;
@@ -2968,8 +2997,10 @@ static int futex_wait_requeue_pi(u32 __u
 		if (q.pi_state && (q.pi_state->owner != current)) {
 			spin_lock(q.lock_ptr);
 			ret = fixup_pi_state_owner(uaddr2, &q, current);
-			if (ret && rt_mutex_owner(&q.pi_state->pi_mutex) == current)
-				rt_mutex_futex_unlock(&q.pi_state->pi_mutex);
+			if (ret && rt_mutex_owner(&q.pi_state->pi_mutex) == current) {
+				pi_state = q.pi_state;
+				get_pi_state(pi_state);
+			}
 			/*
 			 * Drop the reference to the pi state which
 			 * the requeue_pi() code acquired for us.
@@ -3008,13 +3039,20 @@ static int futex_wait_requeue_pi(u32 __u
 		 * the fault, unlock the rt_mutex and return the fault to
 		 * userspace.
 		 */
-		if (ret && rt_mutex_owner(pi_mutex) == current)
-			rt_mutex_futex_unlock(pi_mutex);
+		if (ret && rt_mutex_owner(&q.pi_state->pi_mutex) == current) {
+			pi_state = q.pi_state;
+			get_pi_state(pi_state);
+		}
 
 		/* Unqueue and drop the lock. */
 		unqueue_me_pi(&q);
 	}
 
+	if (pi_state) {
+		rt_mutex_futex_unlock(&pi_state->pi_mutex);
+		put_pi_state(pi_state);
+	}
+
 	if (ret == -EINTR) {
 		/*
 		 * We've already been requeued, but cannot restart by calling