[GHC] #14610: newtype wrapping of a monadic stack kills performance

[GHC]

#14610: newtype wrapping of a monadic stack kills performance
-------------------------------------+-------------------------------------
        Reporter:  mrkkrp            |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:  8.6.1
       Component:  Compiler          |              Version:  8.2.2
      Resolution:                    |             Keywords:  JoinPoints
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Runtime           |  Unknown/Multiple
  performance bug                    |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:  #14620            |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by nomeata):

 Ah, of course you are right: If we drop `isValidJoinPointType` then we get
 more join points, and still no crashes. But it would indeed prevent the
 transformation in comment:14. That leaves two options:

  * Un-join-point `go` if we want to do this transformation (no regression
 about now)
  * Actually do this transformation I propose in comment:18 in GHC, which
 means we get a join point as before and can do that transformation.

 All not carefully thought out, of course, and too much distraction here
 right now, so I am mostly conveying my gut feeling, hoping to not make a
 fool out of myself here.

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14610#comment:19>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler