C++ MythBusters Strike 2
summary
Makes-or-break specifics: modules aren’t yet a universal win—real-world compiler support is uneven so the “restructure and huge build speedups” claim is largely busted; coroutines show plausible benefit but current libs cover only slices of use cases. Use concrete tooling: run clang -ftime-trace or Clang Build Analyzer, VCPerf/Windows Performance Analyzer, and include-what-you-use to find hot headers. Debug builds historically call std::move/std::forward but MSVC/Clang now treat them as intrinsics. Chrome/Android and Windows experiments—e.g. DirectWrite rewritten in Rust and GDI work—show Rust reduces memory bugs in practice.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:08C++ community perspectives
- 4:01First series recap
- 8:15C++ tooling challenges
- 11:28Slow compilation myth
- 23:47Debugging performance issues
- 30:01C++ safety limits
- 39:03Rust in Windows/Android
- 50:13C++ successor languages
- 53:26Ranges vs iterators
- 57:46CMake debugging
- 1:00:57Balancing C++ features
- 1:03:25Q&A
related talks
57mC++ MythBusters Strike 2
26mC++ Modules Myth Busting
01h 21mC++ MythBusters
01h 44mImproving C++ Compilation Times: Tools & Techniques
58mSo, You Want to Use C++ Modules … Cross-Platform?
01h 30mTuning C++: Benchmarks, and CPUs, and Compilers! Oh My!
50mPlenary: C++ MythBusters
01h 11mDuck-Tape Chronicles - Rust C++ Interop