Merge branch 'next' (accumulated 3.16 merge window patches) into master

Now that 3.15 is released, this merges the 'next' branch into 'master',
bringing us to the normal situation where my 'master' branch is the
merge window.

* accumulated work in next: (6809 commits)
  ufs: sb mutex merge + mutex_destroy
  powerpc: update comments for generic idle conversion
  cris: update comments for generic idle conversion
  idle: remove cpu_idle() forward declarations
  nbd: zero from and len fields in NBD_CMD_DISCONNECT.
  mm: convert some level-less printks to pr_*
  MAINTAINERS: adi-buildroot-devel is moderated
  MAINTAINERS: add linux-api for review of API/ABI changes
  mm/kmemleak-test.c: use pr_fmt for logging
  fs/dlm/debug_fs.c: replace seq_printf by seq_puts
  fs/dlm/lockspace.c: convert simple_str to kstr
  fs/dlm/config.c: convert simple_str to kstr
  mm: mark remap_file_pages() syscall as deprecated
  mm: memcontrol: remove unnecessary memcg argument from soft limit functions
  mm: memcontrol: clean up memcg zoneinfo lookup
  mm/memblock.c: call kmemleak directly from memblock_(alloc|free)
  mm/mempool.c: update the kmemleak stack trace for mempool allocations
  lib/radix-tree.c: update the kmemleak stack trace for radix tree allocations
  mm: introduce kmemleak_update_trace()
  mm/kmemleak.c: use %u to print ->checksum
  ...

1 files changed, 87 insertions, 70 deletions

diff --git a/arch/s390/lib/spinlock.c b/arch/s390/lib/spinlock.c
index f709983f41f..5b0e445bc3f 100644
--- a/arch/s390/lib/spinlock.c
+++ b/arch/s390/lib/spinlock.c
@@ -26,83 +26,81 @@ __setup("spin_retry=", spin_retry_setup);
 
 void arch_spin_lock_wait(arch_spinlock_t *lp)
 {
-	int count = spin_retry;
-	unsigned int cpu = ~smp_processor_id();
+	unsigned int cpu = SPINLOCK_LOCKVAL;
 	unsigned int owner;
+	int count;
 
 	while (1) {
-		owner = lp->owner_cpu;
-		if (!owner || smp_vcpu_scheduled(~owner)) {
-			for (count = spin_retry; count > 0; count--) {
-				if (arch_spin_is_locked(lp))
-					continue;
-				if (_raw_compare_and_swap(&lp->owner_cpu, 0,
-							  cpu) == 0)
-					return;
-			}
-			if (MACHINE_IS_LPAR)
-				continue;
+		owner = ACCESS_ONCE(lp->lock);
+		/* Try to get the lock if it is free. */
+		if (!owner) {
+			if (_raw_compare_and_swap(&lp->lock, 0, cpu))
+				return;
+			continue;
 		}
-		owner = lp->owner_cpu;
-		if (owner)
+		/* Check if the lock owner is running. */
+		if (!smp_vcpu_scheduled(~owner)) {
+			smp_yield_cpu(~owner);
+			continue;
+		}
+		/* Loop for a while on the lock value. */
+		count = spin_retry;
+		do {
+			owner = ACCESS_ONCE(lp->lock);
+		} while (owner && count-- > 0);
+		if (!owner)
+			continue;
+		/*
+		 * For multiple layers of hypervisors, e.g. z/VM + LPAR
+		 * yield the CPU if the lock is still unavailable.
+		 */
+		if (!MACHINE_IS_LPAR)
 			smp_yield_cpu(~owner);
-		if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0)
-			return;
 	}
 }
 EXPORT_SYMBOL(arch_spin_lock_wait);
 
 void arch_spin_lock_wait_flags(arch_spinlock_t *lp, unsigned long flags)
 {
-	int count = spin_retry;
-	unsigned int cpu = ~smp_processor_id();
+	unsigned int cpu = SPINLOCK_LOCKVAL;
 	unsigned int owner;
+	int count;
 
 	local_irq_restore(flags);
 	while (1) {
-		owner = lp->owner_cpu;
-		if (!owner || smp_vcpu_scheduled(~owner)) {
-			for (count = spin_retry; count > 0; count--) {
-				if (arch_spin_is_locked(lp))
-					continue;
-				local_irq_disable();
-				if (_raw_compare_and_swap(&lp->owner_cpu, 0,
-							  cpu) == 0)
-					return;
-				local_irq_restore(flags);
-			}
-			if (MACHINE_IS_LPAR)
-				continue;
+		owner = ACCESS_ONCE(lp->lock);
+		/* Try to get the lock if it is free. */
+		if (!owner) {
+			local_irq_disable();
+			if (_raw_compare_and_swap(&lp->lock, 0, cpu))
+				return;
+			local_irq_restore(flags);
 		}
-		owner = lp->owner_cpu;
-		if (owner)
+		/* Check if the lock owner is running. */
+		if (!smp_vcpu_scheduled(~owner)) {
 			smp_yield_cpu(~owner);
-		local_irq_disable();
-		if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0)
-			return;
-		local_irq_restore(flags);
-	}
-}
-EXPORT_SYMBOL(arch_spin_lock_wait_flags);
-
-int arch_spin_trylock_retry(arch_spinlock_t *lp)
-{
-	unsigned int cpu = ~smp_processor_id();
-	int count;
-
-	for (count = spin_retry; count > 0; count--) {
-		if (arch_spin_is_locked(lp))
 			continue;
-		if (_raw_compare_and_swap(&lp->owner_cpu, 0, cpu) == 0)
-			return 1;
+		}
+		/* Loop for a while on the lock value. */
+		count = spin_retry;
+		do {
+			owner = ACCESS_ONCE(lp->lock);
+		} while (owner && count-- > 0);
+		if (!owner)
+			continue;
+		/*
+		 * For multiple layers of hypervisors, e.g. z/VM + LPAR
+		 * yield the CPU if the lock is still unavailable.
+		 */
+		if (!MACHINE_IS_LPAR)
+			smp_yield_cpu(~owner);
 	}
-	return 0;
 }
-EXPORT_SYMBOL(arch_spin_trylock_retry);
+EXPORT_SYMBOL(arch_spin_lock_wait_flags);
 
-void arch_spin_relax(arch_spinlock_t *lock)
+void arch_spin_relax(arch_spinlock_t *lp)
 {
-	unsigned int cpu = lock->owner_cpu;
+	unsigned int cpu = lp->lock;
 	if (cpu != 0) {
 		if (MACHINE_IS_VM || MACHINE_IS_KVM ||
 		    !smp_vcpu_scheduled(~cpu))
@@ -111,6 +109,17 @@ void arch_spin_relax(arch_spinlock_t *lock)
 }
 EXPORT_SYMBOL(arch_spin_relax);
 
+int arch_spin_trylock_retry(arch_spinlock_t *lp)
+{
+	int count;
+
+	for (count = spin_retry; count > 0; count--)
+		if (arch_spin_trylock_once(lp))
+			return 1;
+	return 0;
+}
+EXPORT_SYMBOL(arch_spin_trylock_retry);
+
 void _raw_read_lock_wait(arch_rwlock_t *rw)
 {
 	unsigned int old;
@@ -121,10 +130,10 @@ void _raw_read_lock_wait(arch_rwlock_t *rw)
 			smp_yield();
 			count = spin_retry;
 		}
-		if (!arch_read_can_lock(rw))
+		old = ACCESS_ONCE(rw->lock);
+		if ((int) old < 0)
 			continue;
-		old = rw->lock & 0x7fffffffU;
-		if (_raw_compare_and_swap(&rw->lock, old, old + 1) == old)
+		if (_raw_compare_and_swap(&rw->lock, old, old + 1))
 			return;
 	}
 }
@@ -141,12 +150,13 @@ void _raw_read_lock_wait_flags(arch_rwlock_t *rw, unsigned long flags)
 			smp_yield();
 			count = spin_retry;
 		}
-		if (!arch_read_can_lock(rw))
+		old = ACCESS_ONCE(rw->lock);
+		if ((int) old < 0)
 			continue;
-		old = rw->lock & 0x7fffffffU;
 		local_irq_disable();
-		if (_raw_compare_and_swap(&rw->lock, old, old + 1) == old)
+		if (_raw_compare_and_swap(&rw->lock, old, old + 1))
 			return;
+		local_irq_restore(flags);
 	}
 }
 EXPORT_SYMBOL(_raw_read_lock_wait_flags);
@@ -157,10 +167,10 @@ int _raw_read_trylock_retry(arch_rwlock_t *rw)
 	int count = spin_retry;
 
 	while (count-- > 0) {
-		if (!arch_read_can_lock(rw))
+		old = ACCESS_ONCE(rw->lock);
+		if ((int) old < 0)
 			continue;
-		old = rw->lock & 0x7fffffffU;
-		if (_raw_compare_and_swap(&rw->lock, old, old + 1) == old)
+		if (_raw_compare_and_swap(&rw->lock, old, old + 1))
 			return 1;
 	}
 	return 0;
@@ -169,6 +179,7 @@ EXPORT_SYMBOL(_raw_read_trylock_retry);
 
 void _raw_write_lock_wait(arch_rwlock_t *rw)
 {
+	unsigned int old;
 	int count = spin_retry;
 
 	while (1) {
@@ -176,9 +187,10 @@ void _raw_write_lock_wait(arch_rwlock_t *rw)
 			smp_yield();
 			count = spin_retry;
 		}
-		if (!arch_write_can_lock(rw))
+		old = ACCESS_ONCE(rw->lock);
+		if (old)
 			continue;
-		if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000) == 0)
+		if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000))
 			return;
 	}
 }
@@ -186,6 +198,7 @@ EXPORT_SYMBOL(_raw_write_lock_wait);
 
 void _raw_write_lock_wait_flags(arch_rwlock_t *rw, unsigned long flags)
 {
+	unsigned int old;
 	int count = spin_retry;
 
 	local_irq_restore(flags);
@@ -194,23 +207,27 @@ void _raw_write_lock_wait_flags(arch_rwlock_t *rw, unsigned long flags)
 			smp_yield();
 			count = spin_retry;
 		}
-		if (!arch_write_can_lock(rw))
+		old = ACCESS_ONCE(rw->lock);
+		if (old)
 			continue;
 		local_irq_disable();
-		if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000) == 0)
+		if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000))
 			return;
+		local_irq_restore(flags);
 	}
 }
 EXPORT_SYMBOL(_raw_write_lock_wait_flags);
 
 int _raw_write_trylock_retry(arch_rwlock_t *rw)
 {
+	unsigned int old;
 	int count = spin_retry;
 
 	while (count-- > 0) {
-		if (!arch_write_can_lock(rw))
+		old = ACCESS_ONCE(rw->lock);
+		if (old)
 			continue;
-		if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000) == 0)
+		if (_raw_compare_and_swap(&rw->lock, 0, 0x80000000))
 			return 1;
 	}
 	return 0;