• BB_C@programming.dev
    link
    fedilink
    arrow-up
    1
    arrow-down
    1
    ·
    1 day ago

    everyone else who’s been discussing this problem for years that it isn’t a problem.

    I haven’t seen anyone else concede the performance talking point (under different banners), but proceeding to make this out to be a big issue. This includes this thread’s OP which literally starts on-topic arguments with:

    How poisoning harms every single piece of code

    As discussed in too much detail above, (un)locking mutexes is cheap as long as they’re not contended, and when contended, they are at least cheap instruction wise. This is where poisoning mutexes get you.

    This is what makes your positioning weird.

    My replies here, starting with pointing out the lack of benchmarks of real-world use-cases, come from actually bothering to read OP, IN FULL.