[GHC] #14290: Strictness bug with existential contexts on data constructors

[GHC]

#14290: Strictness bug with existential contexts on data constructors
-------------------------------------+-------------------------------------
        Reporter:  simonmar          |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  highest           |            Milestone:  8.2.2
       Component:  Compiler          |              Version:  8.2.1
      Resolution:                    |             Keywords:
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:                    |
-------------------------------------+-------------------------------------
Description changed by simonmar:

Old description:

> The following:
>
> {{{
> {-# LANGUAGE ExistentialQuantification #-}
> module Main (main) where
>
> main :: IO ()
> main = r `seq` return ()
>
> r :: Rec
> r = Rec{ a = error "xxx", b = 3, c = True }
>
> class C t
> instance C Bool
>
> data Rec = forall t. C t => Rec
>   { a :: ()
>   , b :: !Int
>   , c :: t
>   }
> }}}
>
> Fails with `error "xxx"`, when it should succeed. The problem is that the
> strictness signature for the data con wrapper doesn't take into account
> the dictionary fields.
>
> I have a patch which I'll upload to Phabricator shortly...

New description:

 The following:

 {{{
 {-# LANGUAGE ExistentialQuantification #-}
 module Main (main) where

 main :: IO ()
 main = r `seq` return ()

 r :: Rec
 r = Rec{ a = error "xxx", b = 3, c = True }

 class C t
 instance C Bool

 data Rec = forall t. C t => Rec
   { a :: ()
   , b :: !Int
   , c :: t
   }
 }}}

 Should succesed, but fails with `error "xxx"` (but only when compiled with
 `-O`). The problem is that the strictness signature for the data con
 wrapper doesn't take into account the dictionary fields.

 I have a patch which I'll upload to Phabricator shortly...

--

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14290#comment:1>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler