Re: [GHC] #11715: Constraint vs *

7 Mar 2019

      #11715: Constraint vs *
-------------------------------------+-------------------------------------
        Reporter:  bgamari           |                Owner:  goldfire
            Type:  bug               |               Status:  new
        Priority:  high              |            Milestone:
       Component:  Compiler (Type    |              Version:  8.0.1-rc1
  checker)                           |             Keywords:  Typeable,
      Resolution:                    |  LevityPolymorphism, Roles
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:  #11621, #13742,   |  Differential Rev(s):  Phab:D3316
  #16139, #16148                     |
       Wiki Page:                    |
-------------------------------------+-------------------------------------
Description changed by simonpj:

Old description:
...
I first noticed something was a bit odd with `Constraint` when chasing
#11334. Consider this testcase (`-O0` is sufficient),
{{{#!hs
import Data.Typeable
main = do
    print $ typeRep (Proxy :: Proxy Eq)
    print $ typeOf (Proxy :: Proxy Eq)
}}}
With `ghc-8.0` this will produce,
{{{
Eq
Proxy (* -> *) Eq
}}}
Notice the second line; GHC seems to be claiming that `Eq :: * -> *`,
which is clearly nonsense.
I believe this issue may be the cause of some of the testsuite failures
that I'm seeing on #11011. Unfortunately I haven't the foggiest where
this issue might originate.
See also
 * #11621
 * #12933
 * #9547
 * #13931
New description:

 I first noticed something was a bit odd with `Constraint` when chasing
 #11334. Consider this testcase (`-O0` is sufficient),
 {{{#!hs
 import Data.Typeable

 main = do
     print $ typeRep (Proxy :: Proxy Eq)
     print $ typeOf (Proxy :: Proxy Eq)
 }}}
 With `ghc-8.0` this will produce,
 {{{
 Eq
 Proxy (* -> *) Eq
 }}}
 Notice the second line; GHC seems to be claiming that `Eq :: * -> *`,
 which is clearly nonsense.

 I believe this issue may be the cause of some of the testsuite failures
 that I'm seeing on #11011. Unfortunately I haven't the foggiest where this
 issue might originate.

 See also
  * #11621
  * #12933
  * #9547
  * #13931
  * #15918 (about `mkCastTy`)

--

-- 
Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11715#comment:84
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler