[GHC] #15696: Derived Ord instance for enumerations with more than 8 elements seems to be incorrect

[GHC]

#15696: Derived Ord instance for enumerations with more than 8 elements seems to be
incorrect
-------------------------------------+-------------------------------------
        Reporter:  mrkkrp            |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  highest           |            Milestone:  8.6.1
       Component:  Compiler          |              Version:  8.6.1
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Incorrect result  |  Unknown/Multiple
  at runtime                         |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------
Changes (by osa1):

 * cc: osa1 (added)


Comment:

 The problem is in this Core generated for this program:

 {{{
 -- RHS size: {terms: 5, types: 2, coercions: 0, joins: 0/0}
 f_r6li
 f_r6li = \ ds_d3M6 -> case ds_d3M6 of { () -> T2 }

 -- RHS size: {terms: 2, types: 0, coercions: 0, joins: 0/0}
 lvl_r6lj
 lvl_r6lj = f_r6li ()

 main2
 main2
   = case dataToTag# lvl_r6lj of a#_a2rY { __DEFAULT ->
     case dataToTag# lvl_r6lj of b#_a2rZ { __DEFAULT ->
     ...
 }}}

 We get the tag of a CAF (`lvl_r6lj`) before evaluating it, so we get tag
 of a thunk. The need for evaluating argument of `dataToTag#` is explained
 in `Note [dataToTag#]` in primops.txt.pp. It seems like we're inlining
 `getTag`, and then somehow eliminating the case expression in `getTag` (to
 evaluate its argument). If I change the `INLINE` annotation of `getTag` to
 `NOINLINE` this works as expected.

 I don't know why we're elminating the `case` in `getTag` after inlining it
 yet.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15696#comment:3>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler