[GHC] #15299: GHCi :print produces variables that cause a panic

21 Jun 2018

      #15299: GHCi :print produces variables that cause a panic
--------------------------------------+-------------------------------
           Reporter:  Omf             |             Owner:  (none)
               Type:  bug             |            Status:  new
           Priority:  normal          |         Milestone:  8.6.1
          Component:  GHCi            |           Version:  8.2.2
           Keywords:                  |  Operating System:  Linux
       Architecture:  x86_64 (amd64)  |   Type of failure:  GHCi crash
          Test Case:                  |        Blocked By:
           Blocking:                  |   Related Tickets:
Differential Rev(s):                  |         Wiki Page:
--------------------------------------+-------------------------------
 Using ghci installed by Stack, I noted some strange behaviour from the
 :sprint and :print commands:

 {{{#!hs
 λ> xs = [1,2,3]
 λ> :sprint xs
 xs = _
 λ> head xs
 1
 λ> :sprint xs
 xs = _
 λ> :print xs
 xs = (_t1::Num a => [a])
 λ> _t1

 <interactive>:6:1: error:
     • No instance for (Num a) arising from a use of ‘_t1’
     • In the expression: _t1
       In an equation for ‘it’: it = _t1
 λ> :t _t1

 <interactive>:1:1: error:
     No instance for (Num a) arising from a use of ‘_t1’
 λ> :info _t1
 _t1 :: Num a => [a]     -- Defined in ‘interactive:Ghci3’
 λ> _t1 :: [Int]

 <interactive>:9:1: error:ghc: panic! (the 'impossible' happened)
   (GHC version 8.2.2 for x86_64-unknown-linux):
         No skolem info:
   a_a1rR
   Call stack:
       CallStack (from HasCallStack):
         prettyCurrentCallStack, called at
 compiler/utils/Outputable.hs:1133:58 in ghc:Outputable
         callStackDoc, called at compiler/utils/Outputable.hs:1137:37 in
 ghc:Outputable
         pprPanic, called at compiler/typecheck/TcErrors.hs:2653:5 in
 ghc:TcErrors

 Please report this as a GHC bug:  http://www.haskell.org/ghc/reportabug
 }}}

 I don't really understand the error, or why even the :type command fails,
 but it appears to be related to the way numeric types are polymorphic
 until a specific type is forced. Note that the both the :sprint and :print
 commands don't seem to recognise that we've evaluated any of xs, just
 printing an underscore no matter what.

 The panic doesn't happen if I force a type when defining xs:

 {{{#!hs
 λ> xs = [1,2,3] :: [Int]
 λ> :sp xs
 xs = _
 λ> head xs
 1
 λ> :sp xs
 xs = 1 : _
 λ> :print xs
 xs = 1 : (_t2::[Int])
 λ> _t2
 [2,3]
 λ> :t _t2
 _t2 :: [Int]
 λ> :in _t2
 _t2 :: [Int]    -- Defined in ‘interactive:Ghci6’
 λ> _t2 :: [Integer]
 }}}

-- 
Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/15299
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler