| | |
# A mutex has two pieces of state -- a 'locked' bit and a queue.
# When the mutex is not locked, the queue is empty.
# Otherwise, the queue contains 0 or more (function, argument) pairs
# representing functions (or methods) waiting to acquire the lock.
# When the mutex is unlocked while the queue is not empty,
# the first queue entry is removed and its function(argument) pair called,
# implying it now has the lock.
#
# Of course, no multi-threading is implied -- hence the funny interface
# for lock, where a function is called once the lock is aquired. |