[GHC] #12506: Compile time regression in GHC 8.

[GHC]

#12506: Compile time regression in GHC 8.
-------------------------------------+-------------------------------------
        Reporter:  deech             |                Owner:  bgamari
            Type:  bug               |               Status:  new
        Priority:  high              |            Milestone:
       Component:  Compiler          |              Version:  8.0.1
      Resolution:                    |             Keywords:
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 bgamari):

 I clarify that I suspect the large term counts here are really the result
 of the library itself making heavy use of type classes, essentially making
 this yet another manfestation of #8095. The high desugaring times observed
 in compiling tree are due to the fact that we run `simpleOptPgm` during
 desugaring.

 I didn't have a chance to look at compilation time regressed from 7.8 to
 7.10 to 8.0. When I get a chance the best next step will likely to focus
 on the 7.10 to 8.0 regression (since it is both newer (and therefore
 easier to debug) as well as larger). First verify that the desugared Core
 is equivalent, then look at compiler profiles to see if the compiler
 implementation has regressed.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/12506#comment:10>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler