[GHC] #9233: Compiler performance regression

GHC ghc-devs at haskell.org
Tue Jun 24 10:23:09 UTC 2014


#9233: Compiler performance regression
---------------------------------+------------------------------------
        Reporter:  augustss      |            Owner:
            Type:  bug           |           Status:  new
        Priority:  low           |        Milestone:
       Component:  Compiler      |          Version:  7.8.2
      Resolution:                |         Keywords:
Operating System:  Windows       |     Architecture:  Unknown/Multiple
 Type of failure:  None/Unknown  |       Difficulty:  Unknown
       Test Case:                |       Blocked By:
        Blocking:                |  Related Tickets:
---------------------------------+------------------------------------

Comment (by simonpj):

 I guess that Austin might be able to put up a Windows version of GHC 7.8 (
 plus I suppose 7.6) that has profiling enabled.  You would not want to run
 that regularly, but it would mean you could compile both modules with a
 compiler that dumps its profile.

 Are you using -O2?  The `SpecConstr` thing can make a pretty big
 difference.  You could try switching it off globally and see what happens.
 Using `-dshow-passes` shows you the size of each intermediate can also be
 illuminating, because it shows size blow-ups clearly.  Probably worth a
 try before the profiled-compiler stuff.

 Simon

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9233#comment:3>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler


More information about the ghc-tickets mailing list