As I wrote that I came to the same conclusion, but decided to leave the rest in case. On Thu, Nov 26, 2015 at 12:25 AM, GHC wrote:

#11028: Refactor ConDecl -------------------------------------+------------------------------------- Reporter: simonpj | Owner: alanz Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.10.2 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+-------------------------------------

Comment (by simonpj):

...

Alternatively, everything except the `rnLHsType` in `rnConResult` can move to the type checking.

That sounds exactly right. `Note [Sorting out the result type]` in `RdrHsSyn` seems entirely redundant with the new `ConDeclGADT` form.

-- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11028#comment:25 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler