Re:Compiling Rust is NP-hard
by ygra on Thursday July 08, 2021 @06:34AM
(#27770254)
I was under the impression that cases where it would matter are typically code that wouldn't usually be written. E.g. C# and Java also can solve SAT as part of overload resolution with generic types and it's an interesting peculiarity, but not a pattern that is interesting enough to optimize. Perhaps that's the case here as well.
I could also imagine the compiler code being easy to read, understand and modify are often more valuable traits than compilation speed of highly atypical patterns in a language. But that may well depend on how atypical those are.
I could also imagine the compiler code being easy to read, understand and modify are often more valuable traits than compilation speed of highly atypical patterns in a language. But that may well depend on how atypical those are.