[GHC] #14201: Implement ideas from "Compiling Pattern Matching to Good Decision Trees"

[GHC]

#14201: Implement ideas from "Compiling Pattern Matching to Good Decision Trees"
-------------------------------------+-------------------------------------
        Reporter:  AndreasK          |                Owner:  AndreasK
            Type:  feature request   |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.2.1
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by bgamari):

 The assessment in comment:4 sounds pretty reasonable to me. I agree that
 there's little reason to lose sleep over the `f
 (raisesLaterCatchedException) (causesMatchFailure)` case; such code
 produce warnings under any production-worthy flag set and users will
 almost certainly understand that this sort of code is fragile.

 The `f (raisesException1) (raisesException2)` case is a tad more worrying
 since users may be mislead by the (change in the) resulting exception.
 However, as you point out, the Report does say that both exceptions are
 admissible and I don't think we should be shy about taking advantage of
 this freedom if there really is performance on the table.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14201#comment:5>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler