[GHC] #14037: Fix fusion for GHC's utility functions

[GHC]

#14037: Fix fusion for GHC's utility functions
-------------------------------------+-------------------------------------
        Reporter:  goldfire          |                Owner:  rockbmb
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.3
      Resolution:                    |             Keywords:  newcomer
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Compile-time      |  Unknown/Multiple
  performance bug                    |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by rockbmb):

 Replying to [comment:5 bgamari]:
 > There is nothing that tests these functions in isolation although it's
 possible improvements would be reflected in the compiler allocations
 measured by the Nofib suite.

 I've been reading
 [https://ghc.haskell.org/trac/ghc/wiki/Building/RunningNoFib this]
 resource on NoFib, very useful. I'll keep what you said in mind.

 Since the OP also mentions `testsuite/tests/perf/compiler/T5631`, I'm also
 reading https://ghc.haskell.org/trac/ghc/wiki/Building/RunningTests so I
 can test the specific case described.

 I've already added some pragma code to `zipWith3, zipWith3M, zipwith4,
 zipwith4M`. In the coming days I'll run benchmarks.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14037#comment:6>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler