The Old New Thing

Visual Studio 2005 gives you acquire and release semantics for free on volatile memory access

April 19, 2011 Apr 19, 2011 04/19/11

Raymond Chen

If you are using Visual Studio 2005 or later, then you don't need the weird function because Visual Studio 2005 and later automatically impose acquire semantics on reads from volatile locations. It also imposes release semantics on writes to volatile locations. In other words, you can replace the old function with the following: This ...

Don't forget to include the message queue in your lock hierarchy

April 18, 2011 Apr 18, 2011 04/18/11

Raymond Chen

In addition to the loader lock, the message queue is another resource that people often forget to incorporate into their lock hierarchy. If your code runs on a UI thread, then it implicitly owns the message queue whenever it is running, because messages cannot be dispatched to a thread until it calls a message-retrieval function such as or...

Lock-free algorithms: The try/commit/(hand off) model

April 15, 2011 Apr 15, 2011 04/15/11

Raymond Chen

The last lock-free pattern for this week isn't actually lock-free, but it does run without blocking. The pattern for what I'll call try/commit/(hand off) is more complicated than the other patterns, so I'll start off by describing it in words rather than in code, because the code tends to make things more complicated. First, you take the ...

Lock-free algorithms: The opportunistic cache

April 14, 2011 Apr 14, 2011 04/14/11

Raymond Chen

Suppose profiling reveals that a specific calculation is responsible for a significant portion of your CPU time, and instrumentation says that most of the time, it's just being asked to calculate the same thing over and over. A simple one-level cache would do the trick here. Of course, this isn't thread-safe, because if one thread is pre-...

Lock-free algorithms: Update if you can I'm feeling down

April 13, 2011 Apr 13, 2011 04/13/11

Raymond Chen

A customer was looking for advice on this synchronization problem: We have a small amount of data that we need to share among multiple processes. One way to protect the data is to use a spin lock. However, that has potential for deadlock if the process which holds the spinlock doesn't get a chance to release it. For example, it might be ...

Lock-free algorithms: The try/commit/(try again) pattern

April 12, 2011 Apr 12, 2011 04/12/11

Raymond Chen

The singleton constructor pattern and the example we saw some time ago are really special cases of the more general pattern which I'll call try/commit/(try again). I don't know if this pattern has a real name, but that's what I'm calling it for today. The general form of this pattern goes like this: We calculate the desired new value ...

Patterns for using the InitOnce functions

April 8, 2011 Apr 8, 2011 04/8/11

Raymond Chen

Letting the smart people do the work for you.