[GHC] #9630: compile-time performance regression (probably due to Generics)

[GHC]

#9630: compile-time performance regression (probably due to Generics)
-------------------------------------+-------------------------------------
        Reporter:  hvr               |                   Owner:  simonpj
            Type:  bug               |                  Status:  new
        Priority:  high              |               Milestone:  7.12.1
       Component:  Compiler          |                 Version:  7.9
      Resolution:                    |                Keywords:
Operating System:  Unknown/Multiple  |            Architecture:
 Type of failure:  Compile-time      |  Unknown/Multiple
  performance bug                    |               Test Case:
      Blocked By:                    |                Blocking:
 Related Tickets:  #9583             |  Differential Revisions:
-------------------------------------+-------------------------------------

Comment (by simonpj):

 I would love for someone to investigate this:
  * What exactly is happening?
  * Is it due to over-aggressive inlining in the `binary`?  comment:7
 suggests that might be so
  * Are there underlying issues in GHC itself?  comment:10 suggests so.
  * If so, is the issue only that GHC is specialising too much? Or is it
 doing so uselessly or redundantly?  And is it doing so by itself, or in
 response to user pragmas?
  * Even if GHC is behaving precisely as advertised, what should we do to
 stop this kind of code blow-up happening, or tell the user that it is
 happening, and give some clue about what to do?

 We are a bit stuck here until someone has the time and motivation to get
 some data.

 Thanks

 Simon

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9630#comment:11>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler