mcs_spinlock.h 3.53 KB
Newer Older
1
/* SPDX-License-Identifier: GPL-2.0 */
2 3 4 5 6 7 8 9 10 11 12 13 14 15
/*
 * MCS lock defines
 *
 * This file contains the main data structure and API definitions of MCS lock.
 *
 * The MCS lock (proposed by Mellor-Crummey and Scott) is a simple spin-lock
 * with the desirable properties of being fair, and with each cpu trying
 * to acquire the lock spinning on a local variable.
 * It avoids expensive cache bouncings that common test-and-set spin-lock
 * implementations incur.
 */
#ifndef __LINUX_MCS_SPINLOCK_H
#define __LINUX_MCS_SPINLOCK_H

16 17
#include <asm/mcs_spinlock.h>

18 19 20
struct mcs_spinlock {
	struct mcs_spinlock *next;
	int locked; /* 1 if lock acquired */
21
	int count;  /* nesting count, see qspinlock.c */
22 23
};

24 25 26 27 28 29 30 31
#ifndef arch_mcs_spin_lock_contended
/*
 * Using smp_load_acquire() provides a memory barrier that ensures
 * subsequent operations happen after the lock is acquired.
 */
#define arch_mcs_spin_lock_contended(l)					\
do {									\
	while (!(smp_load_acquire(l)))					\
32
		cpu_relax();						\
33 34 35 36 37 38 39 40 41 42 43 44 45
} while (0)
#endif

#ifndef arch_mcs_spin_unlock_contended
/*
 * smp_store_release() provides a memory barrier to ensure all
 * operations in the critical section has been completed before
 * unlocking.
 */
#define arch_mcs_spin_unlock_contended(l)				\
	smp_store_release((l), 1)
#endif

46 47 48 49 50 51 52 53
/*
 * Note: the smp_load_acquire/smp_store_release pair is not
 * sufficient to form a full memory barrier across
 * cpus for many architectures (except x86) for mcs_unlock and mcs_lock.
 * For applications that need a full barrier across multiple cpus
 * with mcs_unlock and mcs_lock pair, smp_mb__after_unlock_lock() should be
 * used after mcs_lock.
 */
54 55 56 57 58 59 60 61

/*
 * In order to acquire the lock, the caller should declare a local node and
 * pass a reference of the node to this function in addition to the lock.
 * If the lock has already been acquired, then this will proceed to spin
 * on this node->locked until the previous lock holder sets the node->locked
 * in mcs_spin_unlock().
 */
62 63 64 65 66 67 68 69 70
static inline
void mcs_spin_lock(struct mcs_spinlock **lock, struct mcs_spinlock *node)
{
	struct mcs_spinlock *prev;

	/* Init node */
	node->locked = 0;
	node->next   = NULL;

71 72 73 74 75 76 77
	/*
	 * We rely on the full barrier with global transitivity implied by the
	 * below xchg() to order the initialization stores above against any
	 * observation of @node. And to provide the ACQUIRE ordering associated
	 * with a LOCK primitive.
	 */
	prev = xchg(lock, node);
78
	if (likely(prev == NULL)) {
79 80 81 82 83 84 85 86
		/*
		 * Lock acquired, don't need to set node->locked to 1. Threads
		 * only spin on its own node->locked value for lock acquisition.
		 * However, since this thread can immediately acquire the lock
		 * and does not proceed to spin on its own node->locked, this
		 * value won't be used. If a debug mode is needed to
		 * audit lock status, then set node->locked value here.
		 */
87 88
		return;
	}
89
	WRITE_ONCE(prev->next, node);
90 91 92

	/* Wait until the lock holder passes the lock down. */
	arch_mcs_spin_lock_contended(&node->locked);
93 94
}

95 96 97 98
/*
 * Releases the lock. The caller should pass in the corresponding node that
 * was used to acquire the lock.
 */
99 100 101
static inline
void mcs_spin_unlock(struct mcs_spinlock **lock, struct mcs_spinlock *node)
{
102
	struct mcs_spinlock *next = READ_ONCE(node->next);
103 104 105 106 107

	if (likely(!next)) {
		/*
		 * Release the lock by setting it to NULL
		 */
108
		if (likely(cmpxchg_release(lock, node, NULL) == node))
109 110
			return;
		/* Wait until the next pointer is set */
111
		while (!(next = READ_ONCE(node->next)))
112
			cpu_relax();
113
	}
114 115 116

	/* Pass lock to next waiter. */
	arch_mcs_spin_unlock_contended(&next->locked);
117 118 119
}

#endif /* __LINUX_MCS_SPINLOCK_H */