#11795: Performance issues with replicateM_ -------------------------------------+------------------------------------- Reporter: snoyberg | Owner: snoyberg Type: bug | Status: new Priority: normal | Milestone: Component: libraries/base | Version: 7.10.3 Resolution: | Keywords: Operating System: MacOS X | Architecture: x86_64 Type of failure: Runtime | (amd64) performance bug | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by snoyberg): Thanks for the quick response Simon. I'm definitely not deeply familiar with the state hack, but from what I can see the issue I'm reporting here may be orthogonal. Specifically, if I pass in `-fno-state-hack`, I still see the large-allocation behavior. I believe the issue here is twofold: * The specialize pragma is getting the way of inlining firing * The behavior of generating and consuming a list is less efficient than the simpler code I've provided here I'll be happy to add comments to any other issues as you see fit, but I don't want to add confusion to an already complicated issue if this is in fact separate from the state hack. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11795#comment:3 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler