#8763: forM_ [1..N] does not get fused (10 times slower than go function) -------------------------------------+------------------------------------- Reporter: nh2 | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: 8.6.1 Component: Compiler | Version: 7.6.3 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: Runtime | Unknown/Multiple performance bug | Test Case: Blocked By: | Blocking: Related Tickets: #7206 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by sgraf): I just ran some quick `quickCheck $ withMaxSuccess 100000 $ \i j k -> take 1000 [i,j..k] == take 1000 (efdtInt (unI# i) (unI# j) (unI# k))` tests and both versions pass. Given that simonpj's is much more to the point, let's run with that one! Although the duplicate `n` has potential to cause pain... -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/8763#comment:55 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler