- 13.1 Concurrentgate
- 13.2 A Brief History of Data Sharing
- 13.3 Look, Ma, No (Default) Sharing
- 13.4 Starting a Thread
- 13.5 Exchanging Messages between Threads
- 13.6 Pattern Matching with receive
- 13.7 File Copyingwith a Twist
- 13.8 Thread Termination
- 13.9 Out-of-Band Communication
- 13.10 Mailbox Crowding
- 13.11 The shared Type Qualifier
- 13.12 Operations with shared Data and Their Effects
- 13.13 Lock-Based Synchronization with synchronized classes
- 13.14 Field Typing in synchronized classes
- 13.15 Deadlocks and the synchronized Statement
- 13.16 Lock-Free Coding with shared classes
- 13.17 Summary
This chapter is from the book
This chapter is from the book
This chapter is from the book
13.17 Summary
The implementation of setlsb, dirty and leaking some grease at the seams, is a fitting finale for a chapter that has started with the simple beauty of message passing and has gradually descended into the underworld of sharing.
D has an ample offering of threading amenities. For most applications on modern machines, the preferred mechanism is defining protocols built around message passing. Immutable sharing should be of great help there. You'd be well advised to use message passing for defining robust, scalable concurrent applications.
If you need to do synchronization based on mutual exclusion, you can do so with the help of synchronized classes. Be warned that support for lock-based programming is limited compared to other languages, and for good reasons.
If you need simple sharing of data, you may want to use shared values. D guarantees that operations on shared values are performed in the order specified in your code and do not cause visibility paradoxes and low-level races.
Finally, if activities such as bungee jumping, crocodile taming, or walking on coals seem sheer boredom to you, you'll be glad that lock-free programming exists, and that you can do it in D by using shared structs and classes.