[GHC] #13014: Seemingly unnecessary marking of a SpecConstr specialization as a loopbreaker

[GHC]

#13014: Seemingly unnecessary marking of a SpecConstr specialization as a
loopbreaker
-------------------------------------+-------------------------------------
        Reporter:  nfrisby           |                Owner:
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.0.1
      Resolution:                    |             Keywords:  SpecConstr
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Runtime           |  Unknown/Multiple
  performance bug                    |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by nfrisby):

 Replying to [comment:5 simonpj]:
 > [snip]
 >
 > ...
 >
 > I wish I knew how to make this better, but I don't.  Yet anyway.

 One quick idea: introduce `GHC.Types.DEEP_SPEC` alongside `SPEC`. The key
 difference is that the occurrence analyzer would try very hard to not mark
 specializations involving `DEEP_SPEC` as loopbreakers. The docs for this
 would likely emphasize some of the potential downsides of this unbounded
 inlining.

 Unless you think that's promising, let's close this ticket as "Not a Bug".
 I don't recall wanting aggressive inlining of a recursive function that
 wasn't driven by a type-refining GADT, so #13016 ought to handle my needs.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13014#comment:8>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler