[GHC] #8745: GeneralizedNewtypeDeriving is still not Safe

[GHC]

#8745: GeneralizedNewtypeDeriving is still not Safe
-------------------------------------+------------------------------------
        Reporter:  goldfire          |            Owner:
            Type:  bug               |           Status:  new
        Priority:  normal            |        Milestone:
       Component:  Compiler          |          Version:  7.8.1-rc1
      Resolution:                    |         Keywords:  Safe
Operating System:  Unknown/Multiple  |     Architecture:  Unknown/Multiple
 Type of failure:  None/Unknown      |       Difficulty:  Unknown
       Test Case:                    |       Blocked By:
        Blocking:                    |  Related Tickets:
-------------------------------------+------------------------------------
Changes (by thoughtpolice):

 * cc: nomeata (added)


Comment:

 The GND change seems reasonable to me, and I believe we discussed it prior
 to this (but I can't find the mailing list post just yet) - we just need
 to remove `-XGeneralizedNewtypeDeriving` from `unsafeFlags` in `DynFlags`.

 I'm not so sure about Coerce though, at least not yet - we're still
 considering it somewhat experimental I think, so perhaps exposing it to
 Safe Haskell isn't the best idea just yet.

 Joachim, what are your thoughts on this? Do you think we should consider
 `Coercible` as safe? If so, I'm willing to defer to you here (and I'll
 even write the patch - but not sure where to export it from - perhaps a
 `Trustworthy` module called `GHC.Coerce`?)

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8745#comment:1>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler