[GHC] #13016: SPECIALIZE INLINE doesn't necessarily inline specializations of a recursive function

[GHC]

#13016: SPECIALIZE INLINE doesn't necessarily inline specializations of a recursive
function
-------------------------------------+-------------------------------------
        Reporter:  nfrisby           |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.0.1
      Resolution:                    |             Keywords:  Inlining
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Runtime           |  Unknown/Multiple
  performance bug                    |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:  #13014            |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by dfeuer):

 mpickering, I don't think it's quite a duplicate. In particular, I believe
 we want `SPECIALIZE INLINE` to actually ''force'' the specialization, even
 if it makes a lot of code and even if it risks an infinite loop in the
 simplifier. The idea here seems pretty cool: it lets you get the
 guaranteed loop unrolling you'd get from the class-based definition I
 wrote above when the types are known, but falls back on recursion when
 they're not.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13016#comment:8>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler