Concurrency TS Editor's Report
summary
Proposal adds continuations to futures via std::future then with implicit unwrapping, plus two multi-future ops: when_all/when_any (when_all returns future<vector<future<T>>> or future<tuple<...>>; when_any returns index + container). Language-level await lets asynchronous loops replace heavy continuation plumbing. New synchronization types include latch/barrier (countdown latch example for n tasks) and an atomic_shared_ptr enabling atomic compare_exchange of shared_ptrs (may not be lock free; implementations exist at Microsoft and Boost).
AI-written summary · about
chapters (5)
- 0:00Introduction to Concurrency TS
- 1:25Improvements to std::future
- 17:21Latches and barriers
- 24:32Atomic shared pointers
- 28:26Q&A
related talks
01h 05m“The Continuing Future of C++ Concurrency"
01h 28mThere Is A Better Future
01h 04mConcurrency in C++20 and Beyond
01h 24mConcurrency in C++20 and Beyond
01h 32mConcurrency, Parallelism and Coroutines
01h 05m“Concurrency, Parallelism and Coroutines”
01h 00mConcurrency Improvements in C++20: A Deep Dive
01h 07mThe C++20 Synchronization Library