[GHC] #9221: (super!) linear slowdown of parallel builds on 40 core machine

[GHC]

#9221: (super!) linear slowdown of parallel builds on 40 core machine
-------------------------------------+-------------------------------------
        Reporter:  carter            |                   Owner:
            Type:  bug               |                  Status:  new
        Priority:  high              |               Milestone:  7.12.1
       Component:  Compiler          |                 Version:  7.8.2
      Resolution:                    |                Keywords:
Operating System:  Unknown/Multiple  |            Architecture:
 Type of failure:  Compile-time      |  Unknown/Multiple
  performance bug                    |               Test Case:
      Blocked By:                    |                Blocking:
 Related Tickets:  #910, #8224       |  Differential Revisions:
-------------------------------------+-------------------------------------

Comment (by simonmar):

 I talked with @bgamari about this a bit today.  What I think you're seeing
 in the above data is entries of *indirections*, not blackholes.
 Indirections and blackholes look the same, except that in a true black
 hole, the indirectee points to a TSO rather than a value.  Looking for the
 cases where the indirectee is a TSO will tell you how many times we get
 blocked on a black hole.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9221#comment:39>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler