Non-blocking algorithm: Difference between revisions

Revision as of 06:26, 19 June 2024

Revision as of 17:56, 12 July 2024

Line 15:

Other problems are less obvious. For example, certain interactions between locks can lead to error conditions such as [[deadlock]], [[livelock]], and [[priority inversion]]. Using locks also involves a trade-off between coarse-grained locking, which can significantly reduce opportunities for [[parallel computing|parallelism]], and fine-grained locking, which requires more careful design, increases locking overhead and is more prone to bugs.

Unlike blocking algorithms, non-blocking algorithms do not suffer from these downsides, and in addition are safe for use in [[interrupt handler]]s: even though the [[Pre-emptive multitasking|preempted]] thread cannot be resumed, progress is still possible without it. In contrast, global data structures protected by mutual exclusion cannot safely be accessed in an interrupt handler, as the preempted thread may be the one holding the lock—butlock. While this can be rectified easily by masking the interrupt requestrequests during the critical section, this requires the code in the critical section to have bounded (and preferably short) running time, or excessive interrupt latency may be observed.<ref name="monit">{{cite journal | doi = 10.1145/358818.358824 | url = http://research.microsoft.com/lampson/23-ProcessesInMesa/Abstract.html | title = Experience with Processes and Monitors in Mesa | author = Butler W. Lampson | author-link = Butler W. Lampson |author2=David D. Redell |author2-link=David D. Redell | journal = Communications of the ACM | volume = 23 | issue = 2 | pages = 105–117 |date=February 1980| citeseerx = 10.1.1.142.5765 | s2cid = 1594544 }}</ref>

A lock-free data structure can be used to improve performance.