Opened 5 months ago

Closed 5 months ago

#14563 closed bug (fixed)

GHC Panic with TypeInType / levity polymorphism

Reported by: Iceland_jack Owned by:
Priority: normal Milestone:
Component: Compiler Version: 8.2.1
Keywords: LevityPolymorphism, TypeInType Cc:
Operating System: Unknown/Multiple Architecture: Unknown/Multiple
Type of failure: None/Unknown Test Case: testsuite/tests/polykinds/T14563
Blocked By: Blocking:
Related Tickets: #14555 Differential Rev(s):
Wiki Page:


This code is rightfully rejected.

-- {-# Language TypeInType #-}
{-# Language RankNTypes, KindSignatures, PolyKinds #-}

import GHC.Types (TYPE)
import Data.Kind

data Lan (g::TYPE rep -> TYPE rep') (h::TYPE rep -> TYPE rep'') a where Lan :: forall xx (g::TYPE rep -> TYPE rep') (h::TYPE rep -> Type) a. (g xx -> a) -> h xx -> Lan g h a

But uncomment first line and it panics

$ ghci2 ~/hs/132-bug.hs
GHCi, version 8.3.20171122:  :? for help
[1 of 1] Compiling Main             ( /home/baldur/hs/132-bug.hs, interpreted )
ghc-stage2: panic! (the 'impossible' happened)
  (GHC version 8.3.20171122 for x86_64-unknown-linux):
  Call stack:
      CallStack (from HasCallStack):
        callStackDoc, called at compiler/utils/Outputable.hs:1147:37 in ghc:Outputable
        pprPanic, called at compiler/typecheck/TcType.hs:1640:17 in ghc:TcType

Please report this as a GHC bug:


Maybe the same as #14555.

Change History (2)

comment:1 Changed 5 months ago by Simon Peyton Jones <simonpj@…>

In 8361b2c5/ghc:

Fix SigTvs at the kind level

This patch fixes two bugs in the treatment of SigTvs at the
kind level:

- We should always generalise them, never default them
  (Trac #14555, #14563)

- We should check if they get unified with each other
  (Trac #11203)

Both are described in TcHsType
   Note [Kind generalisation and SigTvs]

comment:2 Changed 5 months ago by simonpj

Resolution: fixed
Status: newclosed
Test Case: testsuite/tests/polykinds/T14563
Note: See TracTickets for help on using tickets.