#14626: No need to enter a scrutinised value -------------------------------------+------------------------------------- Reporter: heisenbug | Owner: heisenbug Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.2 Resolution: | Keywords: performance Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: #13861 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by heisenbug): Replying to [comment:36 simonpj]:
(That would explain why I often get crashes when mixing -O0 and -O1)
The strict-field unboxing choice should be made once and for all at the module declaring the data constructor. If client modules made a different choice there'd be chaos. If you think that is happening can you demonstrate?
I'll watch out for it. I've crashed my GHC in many very different ways in the last weeks, so it is impossible to remember. My taggedness-check on strict constructor fields will need to deal with this anyway, so we'll possibly get hard data soon. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/14626#comment:41 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler