Consider the currently accepted output for `T18357a`: ``` T18357a.hs:9:10: error: [GHC-25897] • Couldn't match kind ‘r’ with ‘LiftedRep’ Expected a type, but ‘Int’ has kind ‘*’ ``` Very confusing: "expected a type, but `Int` has kind `Type`". The actual error message we are supposed to get is: ``` Expected kind ‘TYPE r’, but ‘Int’ has kind ‘*’ ``` and the issue is that the `TYPE r` got pretty-printed as `Type` instead. A similar example is in the currently accepted output of `T16074`: ``` Couldn't match type ‘a’ with ‘b’ Expected: TYPE a :~: TYPE b Actual: TYPE a :~: TYPE a ‘a’ is a rigid type variable bound by the type signature for: foo :: * :~: * at T16074.hs:9:1-24 ‘b’ is a rigid type variable bound by the type signature for: foo :: * :~: * ``` when it should be: ``` Couldn't match type ‘a’ with ‘b’ Expected: TYPE a :~: TYPE b Actual: TYPE a :~: TYPE a ‘a’ is a rigid type variable bound by the type signature for: foo :: forall (a :: RuntimeRep) (b :: RuntimeRep). TYPE a :~: TYPE b at T16074.hs:9:1-24 ‘b’ is a rigid type variable bound by the type signature for: foo :: forall (a :: RuntimeRep) (b :: RuntimeRep). TYPE a :~: TYPE b ``` This is essentially the same as #26335, but for `RuntimeRep` instead of linearity. I am fixing in !14357.