[GHC] #14967: Optimizer Casting Caf with nominal type

[GHC]

#14967: Optimizer Casting Caf with nominal type
-------------------------------------+-------------------------------------
        Reporter:  etn               |                Owner:  (none)
            Type:  bug               |               Status:  closed
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.4.1
      Resolution:  invalid           |             Keywords:  coerce,
                                     |  nominal
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Incorrect result  |  Unknown/Multiple
  at runtime                         |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------
Changes (by simonpj):

 * status:  new => closed
 * resolution:   => invalid


Comment:

 Looks ok to me. We get
 {{{
 T14967.$fGenNom_$cgen :: forall a. Nom a
 T14967.$fGenNom_$cgen
   = \ (@ a_a2zl) ->
       trace @ (Nom a_a2zl) T14967.$fGenNom3 (T14967.$fGenNom1 @ a_a2zl)
 }}}
 and we invoke this polymorphic function at three different types.  But
 types are erased before code generation, so this turns into
 {{{
 T14967.$fGenNom_$cgen = trace T14967.$fGenNom3 T14967.$fGenNom1
 }}}
 and the three uses of this thunk are of course shared.

 There no casting involved.  Only polymorphic functions.

 Does that make sense?  I'll close as not-a-bug but re-open if you
 disagree.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14967#comment:2>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler