[GHC] #14880: GHC panic: updateRole

[GHC]

#14880: GHC panic: updateRole
-------------------------------------+-------------------------------------
        Reporter:  RyanGlScott       |                Owner:  goldfire
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler (Type    |              Version:  8.2.2
  checker)                           |
      Resolution:                    |             Keywords:  TypeInType
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Compile-time      |  Unknown/Multiple
  crash or panic                     |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:  #15076            |  Differential Rev(s):  Phab:D4769
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by RyanGlScott):

 Question: is it more efficient to use `closeOverKinds (unionVarSet set1
 set2)` than `unionVarSet (closeOverKinds set1) (closeOverKinds set2)`? If
 so, I think the way `nonDetCmpTypes` is currently implemented is
 inefficient. We currently have:

 {{{#!hs
 nonDetCmpTypes :: [Type] -> [Type] -> Ordering
 nonDetCmpTypes ts1 ts2 = nonDetCmpTypesX rn_env ts1 ts2
   where
     rn_env = mkRnEnv2 (mkInScopeSet (tyCoVarsOfTypes ts1 `unionVarSet`
 tyCoVarsOfTypes ts2))
 }}}

 In particular, note this part:

 {{{#!hs
 tyCoVarsOfTypes ts1 `unionVarSet` tyCoVarsOfTypes ts2
 }}}

 `tyCoVarsOfTypes` calls `closeOverKinds`, so we're calling
 `closeOverKinds` once more than is necessary. Perhaps this should be
 written like this?

 {{{#!hs
 tyCoVarsOfTypes (ts1 ++ ts2)
 }}}

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14880#comment:46>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler