Kwiecień 2009

As one example, the code in Figure 4 shows a version of this code that doesn't suffer from the same problem. Instead, it allocates a bunch of extra Random instances between those that we care about, thus ensuring that the ones we do care about are far enough apart so as to not be subject to false sharing (at least on this machine). For our tests, this updated version produced significantly better results, running up to six times faster than the code from Figure 3 on our dual-core test machine.

It's far easier and much less trouble to find and use a bug-ridden, poorly implemented snippet of code written by a 13 year old blogger on the other side of the world, than it is to find and use the equivalent piece of code, written by your team leader on the other side of a cubicle partition.

On the world’s largest VLCC trading route, from the Middle East to Japan, rates have fallen to under $8,000 per day, just over 10% of the $70,000 per day seen at the beginning of 2009.   “There’s no reason why the market should be where it is,” said Frontline acting chief executive Jens Martin Jensen.   “It doesn’t make any sense. If [a crude oil trader] pays $50 a barrel and you have 2m barrels on board [a tanker] worth $100m, who cares if you are paying $10,000 or $20,000 per day? It doesn’t make any difference. It’s weakness in certain owners minds.

The most practical rule to avoid deadlock is to make sure that the locks are always acquired in the same order. In our example, it means that either the score or character lock must be acquired first—it doesn't matter which as long as we are consistent. This implies the need for a lock hierarchy—meaning that locks are not only protecting their individual items but are also keeping an order to the items. The score lock protects not only the values of the score, but the character lock as well.

While lock leveling works, it does not come without challenges. Dynamic composition of software components can lead to unexpected runtime failures. If a low-level component holds a lock and makes a virtual method call against a user-supplied object, and that user object then attempts to acquire a lock at a higher level, a lock hierarchy-violation exception will be generated. A deadlock won't occur, but a run-time failure will. This is one reason that making virtual method calls while holding a lock is generally considered a bad practice. Although this is still better than running the risk of a deadlock, this is a primary reason databases do not employ this technique: they must enable entirely dynamic composition of user transactions.

The easiest way to deal with this is to always lock the mutexes in the same order. This is especially easy if the order can be hard-coded, and some uses naturally lend themselves towards this choice. For example, if the mutexes protect objects with different roles, it is relatively easy to always lock the mutex protecting one set of data before locking the other one. In such a situation, Lock hierarchies can be used to enforce the ordering — with a lock hierarchy, a thread cannot acquire a lock on a mutex with a higher hierarchy level than any mutexes currently locked by that thread.