[GHC] #8827: Inferring Safe mode with GeneralizedNewtypeDeriving is wrong

[GHC]

#8827: Inferring Safe mode with GeneralizedNewtypeDeriving is wrong
-------------------------------------+-------------------------------------
        Reporter:  goldfire          |                Owner:
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  7.9
      Resolution:                    |             Keywords:  SafeHaskell
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:  #8226, #8745      |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by goldfire):

 I don't have time to remind myself of this whole conversation now, but I
 wanted to share a point that came up in a related conversation, by
 @nomeata, about the horrible recursive find-all-constructors-out-to-the-
 leaves check: "But, wouldn’t we, at such a check, only have to look at
 things that are already imported? If anything is not imported, the check
 would fail and we could stop. (This now assumes that imported things are
 loaded, no extra loading would happen.)"

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8827#comment:50>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler