[GHC] #15019: Fix performance regressions from #14737

[GHC]

#15019: Fix performance regressions from #14737
-------------------------------------+-------------------------------------
        Reporter:  tdammers          |                Owner:  tdammers
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:  8.6.1
       Component:  Compiler          |              Version:
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Compile-time      |  Unknown/Multiple
  performance bug                    |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:  #14737            |  Differential Rev(s):  phab:D4568
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by simonpj):

 Wow. With this change I see
 {{{
 Unexpected stat failures:
    /tmp/ghctest-c0sci53m/test   spaces/./perf/compiler/T5837.run    T5837
 [stat too good] (normal)   (-7.9%)
    /tmp/ghctest-c0sci53m/test   spaces/./perf/compiler/T5321FD.run
 T5321FD [stat too good] (normal) (-12.5%)
    /tmp/ghctest-c0sci53m/test   spaces/./perf/compiler/T12425.run   T12425
 [stat too good] (optasm)  (-9.5%)
    /tmp/ghctest-c0sci53m/test   spaces/./perf/compiler/T9020.run    T9020
 [stat too good] (optasm)   (-32.5%)
    /tmp/ghctest-c0sci53m/test   spaces/./perf/compiler/T12707.run   T12707
 [stat too good] (normal)  (-8.7%)

    /tmp/ghctest-c0sci53m/test   spaces/./perf/compiler/T12227.run   T12227
 [stat not good enough] (normal) (+660%)
 }}}
 Clearly something is going pear shaped in `T12227` but the other numbers
 look fantastic for such a small change! Do you see this too?

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/15019#comment:11>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler