[GHC] #13623: join points produce bad code for stream fusion

[GHC]

#13623: join points produce bad code for stream fusion
-------------------------------------+-------------------------------------
        Reporter:  choenerzs         |                Owner:  (none)
            Type:  bug               |               Status:  closed
        Priority:  highest           |            Milestone:  8.2.1
       Component:  Compiler          |              Version:  8.2.1-rc1
      Resolution:  fixed             |             Keywords:  JoinPoints
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  Runtime           |            Test Case:
  performance bug                    |  perf/should_runt/T13623
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by simonpj):

 Yes, it's hard for GHC to predict what "sufficient specialisation" is, and
 its heuristics are a bit ad hoc.  Improvements there would be welcome.

 I'm not quite sure what changed since 8.0, but it's a squishy area so I'm
 not surprised that something has.  Seeing where and why GHC 8.2 stopped
 sort of doing the full job, and what heuristic might help it continue,
 would be a useful piece of work.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13623#comment:14>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler