Hans Wennborg: C++ switch statements under the hood in LLVM

Поделиться
HTML-код
  • Опубликовано: 2 фев 2025

Комментарии •

  • @TavishMcEwen
    @TavishMcEwen 2 месяца назад +1

    Been wanting an in-depth explanation of how switch statements are optimized, perfect!

  • @uhertlein
    @uhertlein 2 месяца назад +4

    Super interesting talk, thanks a lot!
    (It's always good to know some internals, even if it's not immediately obvious what for at this point.)

  • @arisweedler4703
    @arisweedler4703 2 месяца назад +1

    Very nice, very clear. An in depth explanation

  • @anthonyoleinik6472
    @anthonyoleinik6472 2 месяца назад +1

    Great talk great presenter, thanks!

  • @markusknoch2266
    @markusknoch2266 Месяц назад +2

    llvm turning a switch to a lookup or linear transformation is somehow so funny to me. Me colleagues would grill me for writing a lookup or linear transformation and llvm just turns it into one

    • @karoI508
      @karoI508 Месяц назад

      @@markusknoch2266 Why would they grill you? luts are one of the best optimisations you can make! ask any graphics programmer :D

    • @MrEverydaynormalguy
      @MrEverydaynormalguy Месяц назад

      I like to watch those talks about low level programming languages as I find them fascinating but professionaly I write in Python/Kotlin/Dart idk finding a linear transform or lut in those languages would be odd I guess. But still I sometimes think in those idioms, but I guess change your language change your style makes sense. However writing python after writing a type safe language makes your python just so much better. So sometimes it makes sense to mix styles