[GHC] #14279: Type families interfere with specialisation rewrite rules

[GHC]

#14279: Type families interfere with specialisation rewrite rules
-------------------------------------+-------------------------------------
        Reporter:  IvanTimokhin      |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.2.1
      Resolution:                    |             Keywords:  TypeFamilies
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  Other             |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by goldfire):

 I argue that (related to my musings in #14119), patterns -- such as that
 LHS of rules -- should not mention type families at all. In other words,
 we should change any `F ty` in rule LHS to a fresh type variable `a` and
 then require `F ty ~ a`. Which, as it turns out, is precisely what's done
 in my [http://cs.brynmawr.edu/~rae/papers/2017/partiality/partiality.pdf
 Constrained Type Families] paper.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14279#comment:3>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler