Re: [GHC] #11292: Generics unboxed types: TEST=GEq1 WAY=optasm is failing

28 Dec 2015

      #11292: Generics unboxed types: TEST=GEq1 WAY=optasm is failing
-------------------------------------+-------------------------------------
        Reporter:  thomie            |                Owner:
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  7.11
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by rwbarton):

 It looks like GHC is happy to inline everything in your simplified example
 leaving
 {{{
 Main.main2 =
   case GHC.Prim.tagToEnum#
          @ GHC.Types.Bool (GHC.Prim.eqAddr# "a"# "a"#)
   of _ [Occ=Dead] {
     GHC.Types.False -> GHC.Show.shows26;
     GHC.Types.True -> GHC.Show.shows24
   }
 }}}
 The value `"a"#` has been duplicated, so there are now two copies of the
 string constant in the program, with different addresses.

 For your test you can presumably work around this with NOINLINE
 annotations on `u0` and `uf0`.

 This duplication seems like generally undesirable behavior though; leaving
 aside the question of whether or not it is "semantically correct", it
 makes the resulting object file unnecessarily large (imagine the string
 was much longer than a single character). Probably worth opening a
 separate ticket.

--
Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/11292#comment:2
GHC http://www.haskell.org/ghc/
The Glasgow Haskell Compiler