Panic when wrongly using a type family as return types for GADTs

The following code causes the following panic:

{-# LANGUAGE TypeFamilies, GADTs #-}

type family F a

data G1 a where
  G1C :: F Int

data G2 a where
  G2C :: F Int

ghc: panic! (the 'impossible' happened)
  (GHC version 7.6.0.20120810 for x86_64-apple-darwin):
	compiler/typecheck/TcTyClsDecls.lhs:1081:5-62: Irrefutable pattern failed for pattern Data.Maybe.Just subst

Admittedly, the code causing the panic is silly, but it happened to me as I was refactoring. Having two separate GADTs with the wrong return types seems necessary to cause the panic.

Trac metadata

Trac field	Value
Version	7.6.1-rc1
Type	Bug
TypeOfFailure	OtherFailure
Priority	normal
Resolution	Unresolved
Component	Compiler
Test case
Differential revisions
BlockedBy
Related
Blocking
CC
Operating system
Architecture

changed weight to 5

added Tbug Trac import labels

mentioned in commit d0ddde58

mentioned in commit c5b92198

Thanks! Fixed. Pls merge to 7.6

Trac metadata

Trac field	Value
Test case	→ typecheck/should_fail/T7175

added 1 deleted label

closed

changed milestone to %7.6.1

Merged as 0d45533c.

Trac metadata

Trac field	Value
Resolution	Unresolved → ResolvedFixed

removed 1 deleted label

mentioned in issue #10836 (closed)

mentioned in commit a870738a

Hooray! Hooray! Hooray!

This fix is quite straightforward, of course, but somehow this dark corner has lurked a long time. Thanks. I no longer don armor and surround myself in a cloud of fear and loathing every time I need to think about editing rejigConRes.

We try to please :-)

added compiler crash label

added Pnormal label