[GHC] #11715: Constraint vs *

[GHC]

#11715: Constraint vs *
-------------------------------------+-------------------------------------
        Reporter:  bgamari           |                Owner:  goldfire
            Type:  bug               |               Status:  new
        Priority:  high              |            Milestone:  8.2.1
       Component:  Compiler (Type    |              Version:  8.0.1-rc1
  checker)                           |             Keywords:  Typeable,
      Resolution:                    |  LevityPolymorphism
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 bgamari):

 While speaking to danharaj about #10359 it occurred to me that the
 semantic issues with the tuple transform that simonpj proposes in
 ticket:10359#comment:2 would disappear if dictionaries were instead of
 kind `UnliftedPtrRep`. We could then introduce a core-to-core pass which
 would look for reconstructions of unlifted data constructors (e.g. `(case
 d of (a,b) -> a, case d of (a,b) -> b)`) and reduce them safely (e.g. to
 `d`).

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11715#comment:59>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler