[GHC] #14831: QuantifiedConstraints: Odd superclass constraint

[GHC]

#14831: QuantifiedConstraints: Odd superclass constraint
-------------------------------------+-------------------------------------
        Reporter:  Iceland_jack      |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.5
      Resolution:                    |             Keywords:
                                     |  QuantifiedConstraints wipT2893
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by simonpj):

 > but my hunch is that you're experiencing the same bug as in...

 Indeed: see comment:4 above, which gives exactly the same diagnosis as I
 gave in https://ghc.haskell.org/trac/ghc/ticket/5927#comment:33.

 In both comments I point out that a local instance declaration that claims
 to produce evidence for `Semigroup xx` '''for any xx''' is very suspicious
 (comment:4 above).

 Now you have two ways of proving `SemiGroup (Free ...)`: from the top-
 level instance or from the local instance (QC).  And currently the local
 one "wins".

 For now I'm saying "not a bug" and "you probably didn't really want that
 program anyway".  But I could obviously be wrong about the latter.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14831#comment:7>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler