[GHC] #15155: How untagged pointers sneak into banged fields

[GHC]

#15155: How untagged pointers sneak into banged fields
-------------------------------------+-------------------------------------
        Reporter:  heisenbug         |                Owner:  heisenbug
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:  8.8.1
       Component:  Compiler          |              Version:  8.4.2
      Resolution:                    |             Keywords:  CodeGen
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:  14677
 Related Tickets:  #13027 #7308      |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by simonpj):

 I agree with comment:17.

 But we also agreed that our short term plan is

 * Do ''not'' require the invariant that every strict constructor field
 must be a correctly tagged pointer to a data value.

 Fewer invariants means less to think about - and this is a delicate one!
 Moreover, it's not clear whether efficiency would be better or worse with
 it.

 But yes, it'd be good to try it out.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15155#comment:18>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler