[GHC] #8224: Excessive system time -- new IO manager problem?

[GHC]

#8224: Excessive system time -- new IO manager problem?
-------------------------------------+-------------------------------------
        Reporter:  rrnewton          |                Owner:
            Type:  bug               |               Status:  new
        Priority:  high              |            Milestone:  8.2.1
       Component:  Runtime System    |              Version:  7.7
      Resolution:                    |             Keywords:  IO Manager,
                                     |  System Time
Operating System:  Linux             |         Architecture:  x86_64
 Type of failure:  Runtime           |  (amd64)
  performance bug                    |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:  #9221             |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by bgamari):

 I tried reproducing this with cf5eec3eaa638719fd9768c20271f8aa2b2eac1f on
 a 2-socket, 24-core machine I have access to. I was unable to see anything
 like the system time numbers like reported in comment:17,
 {{{#!bash
 $ cat >T8224.hs <<EOF
 main = return ()
 EOF
 $ inplace/bin/ghc-stage2 -threaded  -rtsopts T8224.hs
 $ time ./T8224 +RTS -N32

 real    0m0.016s
 user    0m0.008s
 sys     0m0.012s
 }}}
 and
 {{{#!bash
 $ cat >T8224a.hs <<EOF
 import Control.Monad
 import Control.Concurrent

 main = do
   n <- getNumCapabilities
   forM_ [1..n] $ \_ -> forkIO $ return ()
   threadDelay (4 * 1000 * 1000)
 EOF
 $ inplace/bin/ghc-stage2 -threaded  -rtsopts T8224a.hs
 $ time ./T8224a +RTS -N32

 real    0m4.018s
 user    0m0.160s
 sys     0m0.028s
 }}}

 All-in-all this doesn't seem terribly unreasonable to me.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8224#comment:29>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler