New blog post series on how I made an interpreter written in #Haskell 700× faster
http://www.joachim-breitner.de/blog/758-Winter_is_coming_even_more_quickly …
-
-
The forth entry is a small one: Export lists help improving performance: http://www.joachim-breitner.de/blog/762-Faster_Winter_3__Export_lists …
Prikaži ovu nit -
In the fifth entry, I talk about how ReaderT gets in the way, and why eta-expansion is so important for performance: http://www.joachim-breitner.de/blog/763-Faster_Winter_5__Eta-Expanding_ReaderT …
Prikaži ovu nit -
Time for the sixth entry, a small one showing that sometimes, simpler code is better code: http://www.joachim-breitner.de/blog/764-Faster_Winter_6__Simpler_Code …
Prikaži ovu nit -
After a week, I can publish the seventh and (for now) last entry, where we introduce a “control stack” (or should I say, “zipper-like data structure”)? http://www.joachim-breitner.de/blog/765-Faster_Winter_7__The_Zipper … I am also sharing the code used for the benchmarks, stats and the graph: http://www.joachim-breitner.de/blog/766-Faster_Winter__Statistics_the_making-of …
Prikaži ovu nit
Kraj razgovora
Novi razgovor -
-
-
Actually, I like `OrdList` https://hackage.haskell.org/package/ghc-8.6.5/docs/OrdList.html#t:OrdList … (which is a terrible name, it should have been called `ListBuilder`) much better, because it's more intuitive. Every usage of difference lists basically inlines `OrdList`s `toList` implementation.
-
Agreed, horrible name. But if you really only consume it once via toList, why is it better than a difference list? (I guess it boils down to not having to import a package vs. explicit, more declarative types)
- Još 1 odgovor
Novi razgovor -
-
-
I'm trying to apply this one right now!
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.