#8275: Loopification breaks profiling ----------------------------------------+---------------------------------- Reporter: jstolarek | Owner: jstolarek Type: bug | Status: new Priority: highest | Milestone: Component: Profiling | Version: 7.7 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: Type of failure: Building GHC failed | Unknown/Multiple Test Case: | Difficulty: Unknown Blocking: 8298 | Blocked By: | Related Tickets: ----------------------------------------+---------------------------------- Comment (by jstolarek): Another thing I observed is that my example is very sensitive. If you modify it by removing any call to `f` (for example by changing scrutinee to `case n of...` or second branch to `5 -> 5`) or replace second pattern match in case alternatives with `_` (i.e second branch becomes `_ -> f (n - 1)`) the segfault disappears. I haven't yet examined the differences in Cmm between these versions, but if the problem is caused by code generated for `f` this should allow us to track the cause. I don't know yet how to debug this issue if the problem is hidden in the libraries :-/ -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/8275#comment:7 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler