GHC Panic on malformed newtype and StrictData

The following source causes a panic:

{-# LANGUAGE StrictData #-}

module Main where

newtype Duration = Foo
data Literal = LitDuration Duration

Result:

[1 of 1] Compiling Main             ( src/Main.hs, .stack-work/dist/x86_64-linux-tinfo6/Cabal-2.2.0.1/build/tcalc/tcalc-tmp/Main.o )
ghc: panic! (the 'impossible' happened)
  (GHC version 8.4.3 for x86_64-unknown-linux):
        mkNewTyConRhs
  Foo []
  Call stack:
      CallStack (from HasCallStack):
        callStackDoc, called at compiler/utils/Outputable.hs:1150:37 in ghc:Outputable
        pprPanic, called at compiler/iface/BuildTyCl.hs:77:27 in ghc:BuildTyCl

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

If we remove the StrictData pragma, then we get a syntax error as expected:

/home/reuben/tcalc/src/Main.hs:3:20: error:
    • The constructor of a newtype must have exactly one field
        but ‘Foo’ has none
    • In the definition of data constructor ‘Foo’
      In the newtype declaration for ‘Duration’
  |
3 | newtype Duration = Foo
  |                    ^^^

I was able to reproduce this with GHC 8.4.3 and 8.2.2 (using Stack LTS snapshots). GHC 8.0.2 is not affected.

A test case buildable with Stack can be found at https://github.com/rdnetto/tcalc/tree/ghc-bug (i.e. on the ghc-bug branch).

Trac metadata

Trac field	Value
Version	8.4.3
Type	Bug
TypeOfFailure	OtherFailure
Priority	normal
Resolution	Unresolved
Component	Compiler
Test case
Differential revisions
BlockedBy
Related
Blocking
CC
Operating system
Architecture