kernel/locking/rwsem.c

38  *  - Bit 0: RWSEM_READER_OWNED - rwsem may be owned by readers (just a hint)
55  * is involved. Ideally we would like to track all the readers that own
109  * 1) rwsem_mark_wake() for readers		-- set, clear
296  * The lock is owned by readers when
301  * Having some reader bits set is not enough to guarantee a readers owned
302  * lock as the readers may be in the process of backing out from the count
350 	RWSEM_WAKE_READERS,	/* Wake readers only */
362  * Magic number to batch-wakeup waiting readers, even when writers are
409  * Implies rwsem_del_waiter() for all woken readers.
451 	 * We prefer to do the first reader grant before counting readers
478 		 * indication that readers now have the lock.
487 	 * Grant up to MAX_READERS_WAKEUP read locks to all the readers in the
490 	 * number of readers before waking any processes up.
493 	 * reader phase (first waiter is a reader), all readers are eligible
518 		 * Limit # of readers that can be woken up per wakeup call.
797  * The more readers own the rwsem, the longer it will take for them to
803  * The limit is capped to a maximum of 25us (30 readers). This is just
809 	int readers = count >> RWSEM_READER_SHIFT;
812 	if (readers > 30)
813 		readers = 30;
814 	delta = (20 + readers) * NSEC_PER_USEC / 2;
834 	 *  2) readers own the lock and spinning time has exceeded limit.
1004 	 * To prevent a constant stream of readers from starving a sleeping
1006 	 * very likely owned by readers.
1020 		 * Wake up other readers in the wait queue if it is
1137 			 * when a large number of readers are to be woken up.
1222  * - just wake up any readers at the front of the queue
1513 	int count = atomic_read(&sem->rwbase.readers);