#9370: unfolding info as seen when building a module depends on flags in a previously-compiled module -------------------------------------+------------------------------------- Reporter: carter | Owner: richardfung Type: bug | Status: new Priority: high | Milestone: 8.2.1 Component: Compiler | Version: 7.8.3 Resolution: | Keywords: newcomer, | Inlining Operating System: Unknown/Multiple | Architecture: Type of failure: Compile-time | Unknown/Multiple performance bug | Test Case: Blocked By: | Blocking: Related Tickets: #8635 | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by richardfung): Sorry even though you basically did the ticket for me already I still have more questions.. I'm trying to think of a good test case to add for this but it seems like the tests are limited to comparing stdout and stderr and checking if it compiled or ran successfully, at least if I understand correctly. That being the case, I was thinking I could make a test that runs with -ddump-inlinings and compare the output. Would that work? Ideally I would like to run --make twice (with the scenario that someone else mentioned where we force the module with optimizations to run first and where it runs second) and compare the inlining output to make sure they're the same. This would be nice in case the -ddump-inlinings format ever changes but I'm not sure this kind of test is supported and I also don't know how likely it is that the format would change. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/9370#comment:30 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler