[GHC] #13092: family instance consistency checks are too pessimistic

[GHC]

#13092: family instance consistency checks are too pessimistic
-------------------------------------+-------------------------------------
        Reporter:  rwbarton          |                Owner:  rwbarton
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:  8.2.1
       Component:  Compiler          |              Version:  8.1
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Compile-time      |  Unknown/Multiple
  performance bug                    |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by rwbarton):

 Replying to [comment:8 simonpj]:
 > > When Main is compiled the interface file for B is not read at all!
 >
 > Ah, that is the bit I was missing! (It's not read, because nobody
 mentions `X`.)  If B.hi was read we would (I hope) get an overlap error
 when solving `(A (a,y))`.

 Earlier I had a simpler version (without the proxy/CPS stuff in `x` and
 `y`, and using visible type application in `z`) that ''was'' reading
 `B.hi`, because the type of `x` was `A (X, b) -> ()` which mentions `X`,
 which is defined in `B`. But it didn't read it until it got to `z`, at
 which point it had already type checked `y`, it seems. So it never noticed
 the conflict.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13092#comment:9>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler