[GHC] #13589: Possible inconsistency in CSE's treatment of NOINLINE

[GHC]

#13589: Possible inconsistency in CSE's treatment of NOINLINE
-------------------------------------+-------------------------------------
        Reporter:  bgamari           |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.0.1
      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 simonpj):

 I think the rule does address the issue. If CSE fires we'll get
 {{{
 bar = rhs
 foo = bar
 }}}
 Before CSE, if `foo` is inlined, then the context into which it inlines
 will see `<rhs>`.  After CSE, if we inline `foo` we'll see `bar` not
 `<rhs>` so we want `bar` to be able to inline too.

 >  Afterall, NOINLINE bar is always active,

 No: a NOINLINE pragma cause a `NeverActive` activation.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13589#comment:1>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler