Towards Oberon+ Concurrency, request for comments

[rochus-keller]

Here is the updated proposal: https://github.com/oberon-lang/oberon-lang.github.io/blob/main/_posts/2023-12-25-towards-concurrency.md#elaboration-of-oberon-concurrency

If you just want to look at the proposed syntax and semantics, see https://github.com/oberon-lang/oberon-lang.github.io/blob/main/_posts/2023-12-25-towards-concurrency.md#elaboration-of-oberon-concurrency

[rochus-keller]

See also #42 and https://www.quora.com/How-can-we-emulate-mutexes-and-condition-variables-using-Go-channels-only/answer/Rochus-Keller about possible limitations of the Go approach and the simulation of traditional low-level synchronization primitives using channels.

[tarkonian]

After give-up to study the Zonnon language (the compiler implementation can´t compile the examples from the documentation and many code examples), now I am looking for the implementation of concurrence in the Modula-3 (another interesting feature is the capacity to define "unsafe" blocks that "turn-off the Garbage Collector"). Maybe something of Modula-3 can be used in modula-2 (my favorite language) or Oberon+.

[rochus-keller]

Concurrency support in Modula-3 is rather conventional. There is a Thread module with Fork/Join procedures and the usual low-level synchronization features like Mutex and Condition with corresponding Acquire/Release and Wait/Signal procedures.

Concerning "unsafe", Oberon+ has a built-in FFI (foreign function interface) language with dedicated types, which allow you for one part to reuse C libraries, and for the other part to do a bit of C programming without using C. Since the FFI is part of the language, it is automatically seen as "unsafe"; I had an explicit "unsafe" keyword in an early version, but then switched to the cpointer keyword instead. See the corresponding section in the spec for more information. See also an article about the design of the FFI feature.