[GHC] #14944: Compile speed regression

[GHC]

#14944: Compile speed regression
-------------------------------------+-------------------------------------
        Reporter:  br1               |                Owner:  dfeuer
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.4.1
      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 dfeuer):

 In case this helps any, the simplifier runs iteration 1,2,1,2,3,1,2,3, and
 then the next iteration (1) is the one that blows up the term count in
 8.4.1. It appears that the blow-up is non-linear. When I increase the
 number of lines in the `Main.hs` source, the ratio of terms in that
 iteration to terms in the previous one increases. Similarly, the ratio of
 time and allocation in that simplifier run compared to the previous one
 increases. I've tried looking at a very cut-down version with `-ddump-
 simpl-iterations`, but there are still an awful lot of terms to look at.
 I'll keep at it.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14944#comment:10>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler