locking: Remove ACCESS_ONCE() usage

With the new standardized functions, we can replace all
ACCESS_ONCE() calls across relevant locking - this includes
lockref and seqlock while at it.

ACCESS_ONCE() does not work reliably on non-scalar types.
For example gcc 4.6 and 4.7 might remove the volatile tag
for such accesses during the SRA (scalar replacement of
aggregates) step:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58145

Update the new calls regardless of if it is a scalar type,
this is cleaner than having three alternatives.

Signed-off-by: Davidlohr Bueso <[e-mail]>
Cc: Peter Zijlstra <[e-mail]>
Cc: Linus Torvalds <[e-mail]>
Cc: Andrew Morton <[e-mail]>
Cc: Thomas Gleixner <[e-mail]>
Cc: Paul E. McKenney <[e-mail]>
Link: http://lkml.kernel.org/r/[e-mail]
Signed-off-by: Ingo Molnar <[e-mail]>

include/linux/seqlock.h

···
 	unsigned ret;
 
 repeat:
-	ret = ACCESS_ONCE(s->sequence);
+	ret = READ_ONCE(s->sequence);
 	if (unlikely(ret & 1)) {
 		cpu_relax();
 		goto repeat;
···
  */
 static inline unsigned raw_read_seqcount(const seqcount_t *s)
 {
-	unsigned ret = ACCESS_ONCE(s->sequence);
+	unsigned ret = READ_ONCE(s->sequence);
 	smp_rmb();
 	return ret;
 }
···
  */
 static inline unsigned raw_seqcount_begin(const seqcount_t *s)
 {
-	unsigned ret = ACCESS_ONCE(s->sequence);
+	unsigned ret = READ_ONCE(s->sequence);
 	smp_rmb();
 	return ret & ~1;
 }

kernel/locking/mcs_spinlock.h

···
 		 */
 		return;
 	}
-	ACCESS_ONCE(prev->next) = node;
+	WRITE_ONCE(prev->next, node);
 
 	/* Wait until the lock holder passes the lock down. */
 	arch_mcs_spin_lock_contended(&node->locked);
···
 static inline
 void mcs_spin_unlock(struct mcs_spinlock **lock, struct mcs_spinlock *node)
 {
-	struct mcs_spinlock *next = ACCESS_ONCE(node->next);
+	struct mcs_spinlock *next = READ_ONCE(node->next);
 
 	if (likely(!next)) {
 		/*
···
 		if (likely(cmpxchg(lock, node, NULL) == node))
 			return;
 		/* Wait until the next pointer is set */
-		while (!(next = ACCESS_ONCE(node->next)))
+		while (!(next = READ_ONCE(node->next)))
 			cpu_relax_lowlatency();
 	}
 

kernel/locking/mutex.c

···
 		return 0;
 
 	rcu_read_lock();
-	owner = ACCESS_ONCE(lock->owner);
+	owner = READ_ONCE(lock->owner);
 	if (owner)
 		retval = owner->on_cpu;
 	rcu_read_unlock();
···
 			 * As such, when deadlock detection needs to be
 			 * performed the optimistic spinning cannot be done.
 			 */
-			if (ACCESS_ONCE(ww->ctx))
+			if (READ_ONCE(ww->ctx))
 				break;
 		}
 
···
 		 * If there's an owner, wait for it to either
 		 * release the lock or go to sleep.
 		 */
-		owner = ACCESS_ONCE(lock->owner);
+		owner = READ_ONCE(lock->owner);
 		if (owner && !mutex_spin_on_owner(lock, owner))
 			break;
 
···
 __ww_mutex_lock_check_stamp(struct mutex *lock, struct ww_acquire_ctx *ctx)
 {
 	struct ww_mutex *ww = container_of(lock, struct ww_mutex, base);
-	struct ww_acquire_ctx *hold_ctx = ACCESS_ONCE(ww->ctx);
+	struct ww_acquire_ctx *hold_ctx = READ_ONCE(ww->ctx);
 
 	if (!hold_ctx)
 		return 0;

kernel/locking/osq_lock.c

···
 
 	prev = decode_cpu(old);
 	node->prev = prev;
-	ACCESS_ONCE(prev->next) = node;
+	WRITE_ONCE(prev->next, node);
 
 	/*
 	 * Normally @prev is untouchable after the above store; because at that
···
 	 * cmpxchg in an attempt to undo our queueing.
 	 */
 
-	while (!ACCESS_ONCE(node->locked)) {
+	while (!READ_ONCE(node->locked)) {
 		/*
 		 * If we need to reschedule bail... so we can block.
 		 */
···
 		 * Or we race against a concurrent unqueue()'s step-B, in which
 		 * case its step-C will write us a new @node->prev pointer.
 		 */
-		prev = ACCESS_ONCE(node->prev);
+		prev = READ_ONCE(node->prev);
 	}
 
 	/*
···
 	 * it will wait in Step-A.
 	 */
 
-	ACCESS_ONCE(next->prev) = prev;
-	ACCESS_ONCE(prev->next) = next;
+	WRITE_ONCE(next->prev, prev);
+	WRITE_ONCE(prev->next, next);
 
 	return false;
 }
···
 	node = this_cpu_ptr(&osq_node);
 	next = xchg(&node->next, NULL);
 	if (next) {
-		ACCESS_ONCE(next->locked) = 1;
+		WRITE_ONCE(next->locked, 1);
 		return;
 	}
 
 	next = osq_wait_next(lock, node, NULL);
 	if (next)
-		ACCESS_ONCE(next->locked) = 1;
+		WRITE_ONCE(next->locked, 1);
 }

kernel/locking/rwsem-xadd.c

···
  */
 static inline bool rwsem_try_write_lock_unqueued(struct rw_semaphore *sem)
 {
-	long old, count = ACCESS_ONCE(sem->count);
+	long old, count = READ_ONCE(sem->count);
 
 	while (true) {
 		if (!(count == 0 || count == RWSEM_WAITING_BIAS))
···
 		return false;
 
 	rcu_read_lock();
-	owner = ACCESS_ONCE(sem->owner);
+	owner = READ_ONCE(sem->owner);
 	if (!owner) {
-		long count = ACCESS_ONCE(sem->count);
+		long count = READ_ONCE(sem->count);
 		/*
 		 * If sem->owner is not set, yet we have just recently entered the
 		 * slowpath with the lock being active, then there is a possibility
···
 		goto done;
 
 	while (true) {
-		owner = ACCESS_ONCE(sem->owner);
+		owner = READ_ONCE(sem->owner);
 		if (owner && !rwsem_spin_on_owner(sem, owner))
 			break;
 
···
 
 	/* we're now waiting on the lock, but no longer actively locking */
 	if (waiting) {
-		count = ACCESS_ONCE(sem->count);
+		count = READ_ONCE(sem->count);
 
 		/*
 		 * If there were already threads queued before us and there are

lib/lockref.c

···
 #define CMPXCHG_LOOP(CODE, SUCCESS) do {					\
 	struct lockref old;							\
 	BUILD_BUG_ON(sizeof(old) != 8);						\
-	old.lock_count = ACCESS_ONCE(lockref->lock_count);			\
+	old.lock_count = READ_ONCE(lockref->lock_count);			\
 	while (likely(arch_spin_value_unlocked(old.lock.rlock.raw_lock))) {  	\
 		struct lockref new = old, prev = old;				\
 		CODE								\