[GHC] #12784: Typechecker regression in GHC 8.0.2 involving DefaultSignatures

[GHC]

#12784: Typechecker regression in GHC 8.0.2 involving DefaultSignatures
-------------------------------------+-------------------------------------
        Reporter:  RyanGlScott       |                Owner:
            Type:  bug               |               Status:  new
        Priority:  highest           |            Milestone:  8.0.2
       Component:  Compiler          |              Version:  8.1
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  GHC rejects       |  Unknown/Multiple
  valid program                      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------
Changes (by RyanGlScott):

 * cc: bgamari (added)


Comment:

 rwbarton, if I understand correctly, the issue here is that we're trying
 to equate `t m` with `m`? (It sort of looks like a variation of the occurs
 check, but I don't know if it's the same thing, given the nature of the
 error message.)

 It certainly seems like this program shouldn't have typechecked before,
 and that's fine. But the other issue is that we have a program which
 compiles on 8.0.1 but fails with 8.0.2, which seems a bit iffy. (See also
 https://ghc.haskell.org/trac/ghc/ticket/12768#comment:5 for another
 example.) Should we avoid backporting this fix to 8.0.2 to give users time
 to migrate their code that exhibits this bug before the next major release
 (8.2) lands? Or should we just pull the trigger and add a blurb to the
 8.0.2 release notes explaining the situation?

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12784#comment:3>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler