#12665: Make Read instances faster, and make them fail fast -------------------------------------+------------------------------------- Reporter: dfeuer | Owner: dfeuer Type: feature request | Status: new Priority: high | Milestone: 8.2.1 Component: Core Libraries | Version: 8.0.1 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: Runtime | Unknown/Multiple performance bug | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by int-e): Replying to [comment:11 dfeuer]:
There's an unfortunate problem with the memoized base list. [...]
One can also store the bases in the stack itself, see `valInteger''` in the (updated) http://int-e.eu/~bf3/haskell/FromDigits.hs file. This causes some extra allocations, so the code becomes slower, but almost imperceptibly so. In my crude measurements (code compiled with -O2, run by using the `test*` functions in `ghci`), converting a list of 10,000,000 digits in base 10 takes 8.1s with the code currently in base (`Data.Read.Lex.valInteger`), 2.0s with `valInteger'` and 2.1s with `valInteger''`. Converting 2^18^ 18 digit integers takes 0.47s with the original code, 0.23s with `valInteger'` and `valInteger''`. Note that no actual lexing is done by this code. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/12665#comment:13 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler