1 files changed, 101 insertions, 82 deletions

diff --git a/arch/arm/include/asm/spinlock.h b/arch/arm/include/asm/spinlock.h
index 65fa3c88095..ac4bfae2670 100644
--- a/arch/arm/include/asm/spinlock.h
+++ b/arch/arm/include/asm/spinlock.h
@@ -5,21 +5,13 @@
 #error SMP not supported on pre-ARMv6 CPUs
 #endif
 
-#include <asm/processor.h>
+#include <linux/prefetch.h>
 
 /*
  * sev and wfe are ARMv6K extensions.  Uniprocessor ARMv6 may not have the K
  * extensions, so when running on UP, we have to patch these instructions away.
  */
-#define ALT_SMP(smp, up)					\
-	"9998:	" smp "\n"					\
-	"	.pushsection \".alt.smp.init\", \"a\"\n"	\
-	"	.long	9998b\n"				\
-	"	" up "\n"					\
-	"	.popsection\n"
-
 #ifdef CONFIG_THUMB2_KERNEL
-#define SEV		ALT_SMP("sev.w", "nop.w")
 /*
  * For Thumb-2, special care is needed to ensure that the conditional WFE
  * instruction really does assemble to exactly 4 bytes (as required by
@@ -31,46 +23,33 @@
  * the assembler won't change IT instructions which are explicitly present
  * in the input.
  */
-#define WFE(cond)	ALT_SMP(		\
+#define WFE(cond)	__ALT_SMP_ASM(		\
 	"it " cond "\n\t"			\
 	"wfe" cond ".n",			\
 						\
 	"nop.w"					\
 )
 #else
-#define SEV		ALT_SMP("sev", "nop")
-#define WFE(cond)	ALT_SMP("wfe" cond, "nop")
+#define WFE(cond)	__ALT_SMP_ASM("wfe" cond, "nop")
 #endif
 
+#define SEV		__ALT_SMP_ASM(WASM(sev), WASM(nop))
+
 static inline void dsb_sev(void)
 {
-#if __LINUX_ARM_ARCH__ >= 7
-	__asm__ __volatile__ (
-		"dsb\n"
-		SEV
-	);
-#else
-	__asm__ __volatile__ (
-		"mcr p15, 0, %0, c7, c10, 4\n"
-		SEV
-		: : "r" (0)
-	);
-#endif
+
+	dsb(ishst);
+	__asm__(SEV);
 }
 
 /*
- * ARMv6 Spin-locking.
- *
- * We exclusively read the old value.  If it is zero, we may have
- * won the lock, so we try exclusively storing it.  A memory barrier
- * is required after we get a lock, and before we release it, because
- * V6 CPUs are assumed to have weakly ordered memory.
+ * ARMv6 ticket-based spin-locking.
  *
- * Unlocked value: 0
- * Locked value: 1
+ * A memory barrier is required after we get a lock, and before we
+ * release it, because V6 CPUs are assumed to have weakly ordered
+ * memory.
  */
 
-#define arch_spin_is_locked(x)		((x)->lock != 0)
 #define arch_spin_unlock_wait(lock) \
 	do { while (arch_spin_is_locked(lock)) cpu_relax(); } while (0)
 
@@ -79,34 +58,47 @@ static inline void dsb_sev(void)
 static inline void arch_spin_lock(arch_spinlock_t *lock)
 {
 	unsigned long tmp;
+	u32 newval;
+	arch_spinlock_t lockval;
 
+	prefetchw(&lock->slock);
 	__asm__ __volatile__(
-"1:	ldrex	%0, [%1]\n"
-"	teq	%0, #0\n"
-	WFE("ne")
-"	strexeq	%0, %2, [%1]\n"
-"	teqeq	%0, #0\n"
+"1:	ldrex	%0, [%3]\n"
+"	add	%1, %0, %4\n"
+"	strex	%2, %1, [%3]\n"
+"	teq	%2, #0\n"
 "	bne	1b"
-	: "=&r" (tmp)
-	: "r" (&lock->lock), "r" (1)
+	: "=&r" (lockval), "=&r" (newval), "=&r" (tmp)
+	: "r" (&lock->slock), "I" (1 << TICKET_SHIFT)
 	: "cc");
 
+	while (lockval.tickets.next != lockval.tickets.owner) {
+		wfe();
+		lockval.tickets.owner = ACCESS_ONCE(lock->tickets.owner);
+	}
+
 	smp_mb();
 }
 
 static inline int arch_spin_trylock(arch_spinlock_t *lock)
 {
-	unsigned long tmp;
-
-	__asm__ __volatile__(
-"	ldrex	%0, [%1]\n"
-"	teq	%0, #0\n"
-"	strexeq	%0, %2, [%1]"
-	: "=&r" (tmp)
-	: "r" (&lock->lock), "r" (1)
-	: "cc");
-
-	if (tmp == 0) {
+	unsigned long contended, res;
+	u32 slock;
+
+	prefetchw(&lock->slock);
+	do {
+		__asm__ __volatile__(
+		"	ldrex	%0, [%3]\n"
+		"	mov	%2, #0\n"
+		"	subs	%1, %0, %0, ror #16\n"
+		"	addeq	%0, %0, %4\n"
+		"	strexeq	%2, %0, [%3]"
+		: "=&r" (slock), "=&r" (contended), "=&r" (res)
+		: "r" (&lock->slock), "I" (1 << TICKET_SHIFT)
+		: "cc");
+	} while (res);
+
+	if (!contended) {
 		smp_mb();
 		return 1;
 	} else {
@@ -117,15 +109,26 @@ static inline int arch_spin_trylock(arch_spinlock_t *lock)
 static inline void arch_spin_unlock(arch_spinlock_t *lock)
 {
 	smp_mb();
+	lock->tickets.owner++;
+	dsb_sev();
+}
 
-	__asm__ __volatile__(
-"	str	%1, [%0]\n"
-	:
-	: "r" (&lock->lock), "r" (0)
-	: "cc");
+static inline int arch_spin_value_unlocked(arch_spinlock_t lock)
+{
+	return lock.tickets.owner == lock.tickets.next;
+}
 
-	dsb_sev();
+static inline int arch_spin_is_locked(arch_spinlock_t *lock)
+{
+	return !arch_spin_value_unlocked(ACCESS_ONCE(*lock));
+}
+
+static inline int arch_spin_is_contended(arch_spinlock_t *lock)
+{
+	struct __raw_tickets tickets = ACCESS_ONCE(lock->tickets);
+	return (tickets.next - tickets.owner) > 1;
 }
+#define arch_spin_is_contended	arch_spin_is_contended
 
 /*
  * RWLOCKS
@@ -139,6 +142,7 @@ static inline void arch_write_lock(arch_rwlock_t *rw)
 {
 	unsigned long tmp;
 
+	prefetchw(&rw->lock);
 	__asm__ __volatile__(
 "1:	ldrex	%0, [%1]\n"
 "	teq	%0, #0\n"
@@ -155,17 +159,21 @@ static inline void arch_write_lock(arch_rwlock_t *rw)
 
 static inline int arch_write_trylock(arch_rwlock_t *rw)
 {
-	unsigned long tmp;
-
-	__asm__ __volatile__(
-"1:	ldrex	%0, [%1]\n"
-"	teq	%0, #0\n"
-"	strexeq	%0, %2, [%1]"
-	: "=&r" (tmp)
-	: "r" (&rw->lock), "r" (0x80000000)
-	: "cc");
-
-	if (tmp == 0) {
+	unsigned long contended, res;
+
+	prefetchw(&rw->lock);
+	do {
+		__asm__ __volatile__(
+		"	ldrex	%0, [%2]\n"
+		"	mov	%1, #0\n"
+		"	teq	%0, #0\n"
+		"	strexeq	%1, %3, [%2]"
+		: "=&r" (contended), "=&r" (res)
+		: "r" (&rw->lock), "r" (0x80000000)
+		: "cc");
+	} while (res);
+
+	if (!contended) {
 		smp_mb();
 		return 1;
 	} else {
@@ -187,7 +195,7 @@ static inline void arch_write_unlock(arch_rwlock_t *rw)
 }
 
 /* write_can_lock - would write_trylock() succeed? */
-#define arch_write_can_lock(x)		((x)->lock == 0)
+#define arch_write_can_lock(x)		(ACCESS_ONCE((x)->lock) == 0)
 
 /*
  * Read locks are a bit more hairy:
@@ -205,6 +213,7 @@ static inline void arch_read_lock(arch_rwlock_t *rw)
 {
 	unsigned long tmp, tmp2;
 
+	prefetchw(&rw->lock);
 	__asm__ __volatile__(
 "1:	ldrex	%0, [%2]\n"
 "	adds	%0, %0, #1\n"
@@ -225,6 +234,7 @@ static inline void arch_read_unlock(arch_rwlock_t *rw)
 
 	smp_mb();
 
+	prefetchw(&rw->lock);
 	__asm__ __volatile__(
 "1:	ldrex	%0, [%2]\n"
 "	sub	%0, %0, #1\n"
@@ -241,22 +251,31 @@ static inline void arch_read_unlock(arch_rwlock_t *rw)
 
 static inline int arch_read_trylock(arch_rwlock_t *rw)
 {
-	unsigned long tmp, tmp2 = 1;
-
-	__asm__ __volatile__(
-"1:	ldrex	%0, [%2]\n"
-"	adds	%0, %0, #1\n"
-"	strexpl	%1, %0, [%2]\n"
-	: "=&r" (tmp), "+r" (tmp2)
-	: "r" (&rw->lock)
-	: "cc");
-
-	smp_mb();
-	return tmp2 == 0;
+	unsigned long contended, res;
+
+	prefetchw(&rw->lock);
+	do {
+		__asm__ __volatile__(
+		"	ldrex	%0, [%2]\n"
+		"	mov	%1, #0\n"
+		"	adds	%0, %0, #1\n"
+		"	strexpl	%1, %0, [%2]"
+		: "=&r" (contended), "=&r" (res)
+		: "r" (&rw->lock)
+		: "cc");
+	} while (res);
+
+	/* If the lock is negative, then it is already held for write. */
+	if (contended < 0x80000000) {
+		smp_mb();
+		return 1;
+	} else {
+		return 0;
+	}
 }
 
 /* read_can_lock - would read_trylock() succeed? */
-#define arch_read_can_lock(x)		((x)->lock < 0x80000000)
+#define arch_read_can_lock(x)		(ACCESS_ONCE((x)->lock) < 0x80000000)
 
 #define arch_read_lock_flags(lock, flags) arch_read_lock(lock)
 #define arch_write_lock_flags(lock, flags) arch_write_lock(lock)