[GHC] #7258: Compiling DynFlags is jolly slow

[GHC]

#7258: Compiling DynFlags is jolly slow
-------------------------------------+-------------------------------------
        Reporter:  simonpj           |                Owner:  simonpj
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  7.6.1
      Resolution:                    |             Keywords:  deriving-perf
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 tdammers):

 Comparing the latest 3 graphs, we can see that `tryToInline` more badly
 than the other cost centres, at least in the misbehaving examples, while
 the other top cost centres seem to scale somewhat proportionally.
 `simpl_binds` is the largest one in most cases, but `tryToInline` is "more
 nonlinear" for the `show` and `read` examples, surpassing it between 150
 and 200 fields in the `show` example, and (educated guess) around 500
 fields in the `read` example.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7258#comment:83>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler