[GHC] #14816: Missed Called Arity opportunity?

[GHC]

#14816: Missed Called Arity opportunity?
-------------------------------------+-------------------------------------
        Reporter:  dfeuer            |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:  8.6.1
       Component:  Compiler          |              Version:  8.2.2
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Runtime           |  Unknown/Multiple
  performance bug                    |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by sgraf):

 OK, this is due to the call to
 [https://github.com/ghc/ghc/blob/81a5e05d376c075a38e55bc124ea6226c1f3bef7/compiler/stranal/DmdAnal.hs#L659
 reuseEnv here] (that function effectively duplicates every demand, so
 `1*C1(U(U)) -> C(U(U))`) with the explanation in `Aggregated demand for
 cardinality`, the gist of which is that we have to treat free variables of
 let-bound thunks differently in usage analysis than we would like to for
 strictness analysis.

 I think this should only be relevant for thunks, e.g. we should first
 `splitFVs is_thunk rhs_fv` and only then `reuseEnv` the `lazy_fv`. I'll
 give that a shot.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14816#comment:11>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler