[nightly] DPH Performance Test Succeeded

[DPH Buildbot]

Full logs at http://log.ouroborus.net/limitingfactor/dph

Environment
  Platform
    host:      limitingfactor.cse.unsw.EDU.AU
    arch:      i386
    processor: i386
    system:    Darwin 10.7.0
  Versions
    GHC The Glorious Glasgow Haskell Compilation System, version 7.0.3
    GCC i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)

 Total tests = 702
 
 -- WINNERS (had a swing of < -10%)
 dph.evens.vectorised.par.N8
       runtime        (wall clock)    : 400.8ms (----) / 441.1ms (  +7) / 501.0ms (----)
       kernel runtime (wall clock)    : 67.00ms (  -1) / 69.30ms (  -3) / 72.00ms ( -16)
 
 dph.quickhull.vector-forkIO.par.N4
       runtime        (wall clock)    : 400.7ms (----) / 400.8ms (----) / 401.1ms (----)
       kernel runtime (wall clock)    : 32.00ms (----) / 41.30ms (  +2) / 47.00ms ( -11)
 
 repa.blur.par.N1
       runtime        (wall clock)    : 2.301s  ( -21) / 2.302s  ( -21) / 2.302s  ( -26)
       kernel runtime (wall clock)    : 1.928s  ( -26) / 1.932s  ( -26) / 1.936s  ( -26)
 
 repa.blur.par.N2
       runtime        (wall clock)    : 1.301s  ( -19) / 1.321s  ( -17) / 1.502s  (  -6)
       kernel runtime (wall clock)    : 1.034s  ( -25) / 1.050s  ( -24) / 1.172s  ( -15)
 
 repa.fft2d.par.N4
       runtime        (wall clock)    : 3.207s  ( -13) / 3.294s  ( -11) / 3.303s  ( -11)
       kernel runtime (wall clock)    : 2.874s  ( -13) / 2.895s  ( -12) / 2.925s  ( -11)
 
 repa.fft2d.par.N6
       runtime        (wall clock)    : 2.703s  ( -31) / 2.703s  ( -32) / 2.703s  ( -32)
       kernel runtime (wall clock)    : 2.364s  ( -34) / 2.386s  ( -34) / 2.408s  ( -34)
 
 repa.fft2d.par.N8
       runtime        (wall clock)    : 2.604s  ( -38) / 2.635s  ( -38) / 2.805s  ( -35)
       kernel runtime (wall clock)    : 2.283s  ( -42) / 2.327s  ( -41) / 2.464s  ( -38)
 
 noslow.vector.Primitive.*Double.filterin_r
       kernel runtime (wall clock)    : 362.3us ( -12) / 362.4us ( -13) / 362.6us ( -13)
 
 -- LOSERS  (had a swing of > +10%)
 dph.dotp.vectorised.seq.N4
       runtime        (wall clock)    : 2.702s  (----) / 2.702s  (----) / 2.703s  (----)
       kernel runtime (wall clock)    : 53.00ms (----) / 54.00ms (  +2) / 62.00ms ( +15)
 
 dph.quicksort.vectorised.par.N2
       runtime        (wall clock)    : 601.0ms ( +20) / 611.1ms (  -8) / 701.2ms (----)
       kernel runtime (wall clock)    : 352.0ms ( +16) / 369.5ms (----) / 375.0ms (  -4)
 
 dph.quicksort.vectorised.par.N8
       runtime        (wall clock)    : 801.3ms (----) / 821.5ms (  -5) / 1.002s  ( +11)
       kernel runtime (wall clock)    : 472.0ms (  -4) / 484.6ms (  -5) / 504.0ms ( -13)
 
 repa.mmult.par.N2
       runtime        (wall clock)    : 4.803s  ( +12) / 4.805s  ( +10) / 4.809s  (  +9)
       kernel runtime (wall clock)    : 4.500s  (  +6) / 4.517s  (  +6) / 4.530s  (  +6)
 
 repa.mmult.par.N4
       runtime        (wall clock)    : 2.502s  ( +32) / 3.243s  ( +71) / 4.204s  (+121)
       kernel runtime (wall clock)    : 2.255s  ( +26) / 2.952s  ( +65) / 3.880s  (+116)
 
 repa.mmult.par.N8
       runtime        (wall clock)    : 2.220s  ( +85) / 2.287s  ( +87) / 2.320s  ( +76)
       kernel runtime (wall clock)    : 1.971s  ( +82) / 1.974s  ( +80) / 1.981s  ( +68)
 
 repa.laplace.par.N1
       runtime        (wall clock)    : 12.81s  ( +38) / 12.81s  ( +36) / 12.81s  ( +33)
       kernel runtime (wall clock)    : 12.44s  ( +81) / 12.45s  ( +80) / 12.46s  ( +79)
 
 repa.laplace.par.N2
       runtime        (wall clock)    : 7.605s  ( +46) / 8.356s  ( +58) / 8.607s  ( +59)
       kernel runtime (wall clock)    : 7.286s  ( +88) / 8.056s  (+106) / 8.275s  (+109)
 
 repa.laplace.par.N4
       runtime        (wall clock)    : 4.904s  ( +63) / 4.985s  ( +66) / 5.005s  ( +67)
       kernel runtime (wall clock)    : 4.630s  (+129) / 4.749s  (+135) / 4.808s  (+138)
 
 repa.laplace.par.N6
       runtime        (wall clock)    : 4.004s  ( +43) / 4.034s  ( +44) / 4.104s  ( +46)
       kernel runtime (wall clock)    : 3.794s  ( +77) / 3.828s  ( +77) / 3.902s  ( +77)
 
 repa.laplace.par.N8
       runtime        (wall clock)    : 3.806s  ( +46) / 3.807s  ( +44) / 3.808s  ( +41)
       kernel runtime (wall clock)    : 3.597s  ( +71) / 3.619s  ( +70) / 3.640s  ( +66)
 
 repa.blur.par.N4
       runtime        (wall clock)    : 900.8ms (----) / 911.0ms (  +1) / 1.001s  ( +11)
       kernel runtime (wall clock)    : 659.0ms (  -7) / 678.1ms (  -5) / 697.0ms (  -4)
 
 repa.fft3d.par.N1
       runtime        (wall clock)    : 32.82s  ( +10) / 32.89s  ( +10) / 33.01s  ( +10)
       kernel runtime (wall clock)    : 27.53s  ( +12) / 27.60s  ( +12) / 27.64s  ( +12)
 
 repa.fft3d.par.N2
       runtime        (wall clock)    : 18.41s  (  +9) / 19.07s  ( +10) / 19.91s  ( +13)
       kernel runtime (wall clock)    : 15.38s  ( +11) / 16.03s  ( +12) / 16.91s  ( +15)
 
 repa.fft3d.par.N4
       runtime        (wall clock)    : 11.11s  ( +39) / 11.27s  ( +41) / 11.41s  ( +43)
       kernel runtime (wall clock)    : 9.387s  ( +51) / 9.534s  ( +54) / 9.644s  ( +55)
 
 noslow.list.*Double.find_indices
       kernel runtime (wall clock)    : 442.0us ( +14) / 443.9us ( +14) / 445.1us ( +15)
 
 noslow.dph-prim.seq.Double.axpy
       kernel runtime (wall clock)    : 91.12us ( +11) / 91.14us ( +11) / 91.17us ( +11)
 
 noslow.vector.Primitive.Double.axpy
       kernel runtime (wall clock)    : 91.42us ( +11) / 91.45us ( +11) / 91.48us ( +11)
 
 noslow.vector.Unboxed.Double.axpy
       kernel runtime (wall clock)    : 92.50us ( +12) / 92.52us ( +12) / 92.54us ( +12)
 
 noslow.vector-unsafe.Primitive.Double.axpy
       kernel runtime (wall clock)    : 94.51us ( +15) / 94.52us ( +15) / 94.54us ( +15)
 
 noslow.vector-unsafe.Unboxed.Double.axpy
       kernel runtime (wall clock)    : 91.75us ( +12) / 91.77us ( +12) / 91.79us ( +12)