Editing Double-checked locking (section)

{{Short description|Software design pattern}}
In [[software engineering]], '''double-checked locking''' (also known as "double-checked locking optimization"<ref>Schmidt, D et al. Pattern-Oriented Software Architecture Vol 2, 2000 pp353-363</ref>) is a [[software design pattern]] used to reduce the overhead of acquiring a [[Lock (computer science)|lock]] by testing the locking criterion (the "lock hint") before acquiring the lock.  Locking occurs only if the locking criterion check indicates that locking is required.

The original form of the pattern, appearing in ''Pattern Languages of Program Design 3'',<ref>{{cite book |title=Pattern languages of program design. 3 |date=1998 |publisher=Addison-Wesley |location=Reading, Mass |isbn=978-0201310115 |edition=Nachdr. |url=https://www.dre.vanderbilt.edu/~schmidt/PDF/DC-Locking.pdf}}</ref> has [[data race]]s, depending on the [[Memory model (programming)|memory model]] in use, and it is hard to get right. Some consider it to be an [[anti-pattern]].<ref>{{cite book |last1=Gregoire |first1=Marc |title=Professional C++ |date=24 February 2021 |publisher=John Wiley & Sons |isbn=978-1-119-69545-5 |url=https://books.google.com/books?id=MyEgEAAAQBAJ&pg=PA946 |language=en}}</ref> There are valid forms of the pattern, including the use of the {{java|volatile}} keyword in Java and explicit memory barriers in C++.<ref name="bdec">David Bacon et al. [http://www.cs.umd.edu/~pugh/java/memoryModel/DoubleCheckedLocking.html The "Double-Checked Locking is Broken" Declaration].</ref>

The pattern is typically used to reduce locking overhead when implementing "[[lazy initialization]]" in a multi-threaded environment, especially as part of the [[Singleton pattern]]. Lazy initialization avoids initializing a value until the first time it is accessed.