lockref: implement lockless reference count updates using cmpxchg()

Instead of taking the spinlock, the lockless versions atomically check
that the lock is not taken, and do the reference count update using a
cmpxchg() loop.  This is semantically identical to doing the reference
count update protected by the lock, but avoids the "wait for lock"
contention that you get when accesses to the reference count are
contended.

Note that a "lockref" is absolutely _not_ equivalent to an atomic_t.
Even when the lockref reference counts are updated atomically with
cmpxchg, the fact that they also verify the state of the spinlock means
that the lockless updates can never happen while somebody else holds the
spinlock.

So while "lockref_put_or_lock()" looks a lot like just another name for
"atomic_dec_and_lock()", and both optimize to lockless updates, they are
fundamentally different: the decrement done by atomic_dec_and_lock() is
truly independent of any lock (as long as it doesn't decrement to zero),
so a locked region can still see the count change.

The lockref structure, in contrast, really is a *locked* reference
count.  If you hold the spinlock, the reference count will be stable and
you can modify the reference count without using atomics, because even
the lockless updates will see and respect the state of the lock.

In order to enable the cmpxchg lockless code, the architecture needs to
do three things:

 (1) Make sure that the "arch_spinlock_t" and an "unsigned int" can fit
     in an aligned u64, and have a "cmpxchg()" implementation that works
     on such a u64 data type.

 (2) define a helper function to test for a spinlock being unlocked
     ("arch_spin_value_unlocked()")

 (3) select the "ARCH_USE_CMPXCHG_LOCKREF" config variable in its
     Kconfig file.

This enables it for x86-64 (but not 32-bit, we'd need to make sure
cmpxchg() turns into the proper cmpxchg8b in order to enable it for
32-bit mode).

Signed-off-by: Linus Torvalds <[email protected]>

Author: torvalds

arch/x86/Kconfig

@@ -16,6 +16,7 @@ config X86_64
 	def_bool y
 	depends on 64BIT
 	select X86_DEV_DMA_OPS
+	select ARCH_USE_CMPXCHG_LOCKREF
 
 ### Arch settings
 config X86

arch/x86/include/asm/spinlock.h

@@ -34,6 +34,11 @@
 # define UNLOCK_LOCK_PREFIX
 #endif
 
+static __always_inline int arch_spin_value_unlocked(arch_spinlock_t lock)
+{
+	return lock.tickets.head == lock.tickets.tail;
+}
+
 /*
  * Ticket locks are conceptually two parts, one indicating the current head of
  * the queue, and the other indicating the current tail. The lock is acquired

include/linux/lockref.h

@@ -17,8 +17,15 @@
 #include <linux/spinlock.h>
 
 struct lockref {
-	spinlock_t lock;
-	unsigned int count;
+	union {
+#ifdef CONFIG_CMPXCHG_LOCKREF
+		aligned_u64 lock_count;
+#endif
+		struct {
+			spinlock_t lock;
+			unsigned int count;
+		};
+	};
 };
 
 extern void lockref_get(struct lockref *);

lib/Kconfig

@@ -48,6 +48,16 @@ config STMP_DEVICE
 config PERCPU_RWSEM
 	boolean
 
+config ARCH_USE_CMPXCHG_LOCKREF
+	bool
+
+config CMPXCHG_LOCKREF
+	def_bool y if ARCH_USE_CMPXCHG_LOCKREF
+	depends on SMP
+	depends on !GENERIC_LOCKBREAK
+	depends on !DEBUG_SPINLOCK
+	depends on !DEBUG_LOCK_ALLOC
+
 config CRC_CCITT
 	tristate "CRC-CCITT functions"
 	help

lib/lockref.c

@@ -1,6 +1,33 @@
 #include <linux/export.h>
 #include <linux/lockref.h>
 
+#ifdef CONFIG_CMPXCHG_LOCKREF
+
+/*
+ * Note that the "cmpxchg()" reloads the "old" value for the
+ * failure case.
+ */
+#define CMPXCHG_LOOP(CODE, SUCCESS) do {					\
+	struct lockref old;							\
+	BUILD_BUG_ON(sizeof(old) != 8);						\
+	old.lock_count = ACCESS_ONCE(lockref->lock_count);			\
+	while (likely(arch_spin_value_unlocked(old.lock.rlock.raw_lock))) {  	\
+		struct lockref new = old, prev = old;				\
+		CODE								\
+		old.lock_count = cmpxchg(&lockref->lock_count,			\
+					 old.lock_count, new.lock_count);	\
+		if (likely(old.lock_count == prev.lock_count)) {		\
+			SUCCESS;						\
+		}								\
+	}									\
+} while (0)
+
+#else
+
+#define CMPXCHG_LOOP(CODE, SUCCESS) do { } while (0)
+
+#endif
+
 /**
  * lockref_get - Increments reference count unconditionally
  * @lockcnt: pointer to lockref structure
@@ -10,6 +37,12 @@
  */
 void lockref_get(struct lockref *lockref)
 {
+	CMPXCHG_LOOP(
+		new.count++;
+	,
+		return;
+	);
+
 	spin_lock(&lockref->lock);
 	lockref->count++;
 	spin_unlock(&lockref->lock);
@@ -23,9 +56,18 @@ EXPORT_SYMBOL(lockref_get);
  */
 int lockref_get_not_zero(struct lockref *lockref)
 {
-	int retval = 0;
+	int retval;
+
+	CMPXCHG_LOOP(
+		new.count++;
+		if (!old.count)
+			return 0;
+	,
+		return 1;
+	);
 
 	spin_lock(&lockref->lock);
+	retval = 0;
 	if (lockref->count) {
 		lockref->count++;
 		retval = 1;
@@ -43,6 +85,14 @@ EXPORT_SYMBOL(lockref_get_not_zero);
  */
 int lockref_get_or_lock(struct lockref *lockref)
 {
+	CMPXCHG_LOOP(
+		new.count++;
+		if (!old.count)
+			break;
+	,
+		return 1;
+	);
+
 	spin_lock(&lockref->lock);
 	if (!lockref->count)
 		return 0;
@@ -59,6 +109,14 @@ EXPORT_SYMBOL(lockref_get_or_lock);
  */
 int lockref_put_or_lock(struct lockref *lockref)
 {
+	CMPXCHG_LOOP(
+		new.count--;
+		if (old.count <= 1)
+			break;
+	,
+		return 1;
+	);
+
 	spin_lock(&lockref->lock);
 	if (lockref->count <= 1)
 		return 0;