[PATCH 2/2] ipc/sem.c: optimize sem_lock().

2013-09-14 Thread Manfred Spraul
Operations that need access to the whole array must guarantee that there are no simple operations ongoing. Right now this is achieved by spin_unlock_wait(sem->lock) on all semaphores. If complex_count is nonzero, then this spin_unlock_wait() is not necessary, because it was already performed in

[PATCH 2/2] ipc/sem.c: optimize sem_lock().

2013-09-14 Thread Manfred Spraul
Operations that need access to the whole array must guarantee that there are no simple operations ongoing. Right now this is achieved by spin_unlock_wait(sem-lock) on all semaphores. If complex_count is nonzero, then this spin_unlock_wait() is not necessary, because it was already performed in