diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2015-02-09 23:24:03 (GMT) |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2015-02-09 23:24:03 (GMT) |
commit | 8308756f45a12e2ff4f7749c2694fc83cdef0be9 (patch) | |
tree | 618c8f955d22927ebe73d94305d19b37c6ab65c9 /include | |
parent | 23e8fe2e16587494268510c1bc9f6952f50f0311 (diff) | |
parent | afffc6c1805d98e08e778cddb644a666e78cfcfd (diff) | |
download | linux-8308756f45a12e2ff4f7749c2694fc83cdef0be9.tar.xz |
Merge branch 'locking-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull core locking updates from Ingo Molnar:
"The main changes are:
- mutex, completions and rtmutex micro-optimizations
- lock debugging fix
- various cleanups in the MCS and the futex code"
* 'locking-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
locking/rtmutex: Optimize setting task running after being blocked
locking/rwsem: Use task->state helpers
sched/completion: Add lock-free checking of the blocking case
sched/completion: Remove unnecessary ->wait.lock serialization when reading completion state
locking/mutex: Explicitly mark task as running after wakeup
futex: Fix argument handling in futex_lock_pi() calls
doc: Fix misnamed FUTEX_CMP_REQUEUE_PI op constants
locking/Documentation: Update code path
softirq/preempt: Add missing current->preempt_disable_ip update
locking/osq: No need for load/acquire when acquire-polling
locking/mcs: Better differentiate between MCS variants
locking/mutex: Introduce ww_mutex_set_context_slowpath()
locking/mutex: Move MCS related comments to proper location
locking/mutex: Checking the stamp is WW only
Diffstat (limited to 'include')
-rw-r--r-- | include/linux/osq_lock.h | 12 |
1 files changed, 10 insertions, 2 deletions
diff --git a/include/linux/osq_lock.h b/include/linux/osq_lock.h index 90230d5..3a6490e 100644 --- a/include/linux/osq_lock.h +++ b/include/linux/osq_lock.h @@ -5,8 +5,11 @@ * An MCS like lock especially tailored for optimistic spinning for sleeping * lock implementations (mutex, rwsem, etc). */ - -#define OSQ_UNLOCKED_VAL (0) +struct optimistic_spin_node { + struct optimistic_spin_node *next, *prev; + int locked; /* 1 if lock acquired */ + int cpu; /* encoded CPU # + 1 value */ +}; struct optimistic_spin_queue { /* @@ -16,6 +19,8 @@ struct optimistic_spin_queue { atomic_t tail; }; +#define OSQ_UNLOCKED_VAL (0) + /* Init macro and function. */ #define OSQ_LOCK_UNLOCKED { ATOMIC_INIT(OSQ_UNLOCKED_VAL) } @@ -24,4 +29,7 @@ static inline void osq_lock_init(struct optimistic_spin_queue *lock) atomic_set(&lock->tail, OSQ_UNLOCKED_VAL); } +extern bool osq_lock(struct optimistic_spin_queue *lock); +extern void osq_unlock(struct optimistic_spin_queue *lock); + #endif |