[nightly] DPH Performance Test Succeeded
DPH Buildbot
dph at ouroborus.net
Wed Jul 6 17:52:32 CEST 2011
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 7.0.1
GCC i686-apple-darwin10-gcc-4.2.1 (GCC) 4.2.1 (Apple Inc. build 5666) (dot 3)
Total tests = 708
-- WINNERS (had a swing of < -10%)
repa.laplace.par.N1
runtime (wall clock) : 5.803s ( -38) / 5.804s ( -38) / 5.805s ( -40)
kernel runtime (wall clock) : 5.202s ( -24) / 5.211s ( -25) / 5.223s ( -25)
repa.laplace.par.N2
runtime (wall clock) : 3.503s ( -33) / 3.573s ( -32) / 3.604s ( -33)
kernel runtime (wall clock) : 3.035s ( -22) / 3.110s ( -20) / 3.145s ( -21)
repa.laplace.par.N8
runtime (wall clock) : 2.303s ( -12) / 2.303s ( -13) / 2.303s ( -15)
kernel runtime (wall clock) : 1.876s ( -11) / 1.882s ( -11) / 1.892s ( -13)
repa.canny.par.N1
runtime (wall clock) : 400.6ms (----) / 400.7ms (----) / 401.0ms (----)
kernel runtime (wall clock) : 21.00ms ( -12) / 21.00ms ( -12) / 21.00ms ( -12)
repa.canny.par.N2
runtime (wall clock) : 300.5ms (----) / 300.6ms (----) / 300.8ms (----)
kernel runtime (wall clock) : 17.00ms (----) / 17.20ms ( -8) / 18.00ms ( -10)
repa.fft2d.par.N2
runtime (wall clock) : 5.004s ( -9) / 5.094s ( -8) / 5.205s ( -9)
kernel runtime (wall clock) : 4.462s ( -9) / 4.526s ( -10) / 4.620s ( -10)
noslow.dph-prim.seq.Double.filterin_r
kernel runtime (wall clock) : 36.78us ( -14) / 36.79us ( -14) / 36.80us ( -14)
-- LOSERS (had a swing of > +10%)
dph.dotp.vectorised.par.N4
runtime (wall clock) : 1.801s ( -31) / 1.802s ( -31) / 1.802s ( -33)
kernel runtime (wall clock) : 22.00ms ( +5) / 33.20ms ( +22) / 35.00ms ( +13)
dph.evens.vectorised.par.N8
runtime (wall clock) : 400.9ms (----) / 490.9ms ( +19) / 501.0ms (----)
kernel runtime (wall clock) : 65.00ms ( -4) / 74.60ms ( +4) / 87.00ms ( +1)
dph.quicksort.vectorised.par.N2
runtime (wall clock) : 600.9ms ( +20) / 601.1ms ( -9) / 601.4ms ( -14)
kernel runtime (wall clock) : 350.0ms ( +16) / 366.5ms ( -1) / 377.0ms ( -4)
dph.quicksort.vectorised.par.N4
runtime (wall clock) : 601.1ms ( +20) / 661.3ms ( +5) / 701.4ms (----)
kernel runtime (wall clock) : 305.0ms (----) / 361.5ms ( +2) / 384.0ms ( +1)
dph.smvm.vectorised.par.N1
runtime (wall clock) : 100.5ms (----) / 100.6ms (----) / 100.8ms ( -4)
kernel runtime (wall clock) : 32.00ms ( +7) / 32.20ms ( +7) / 33.00ms ( +10)
dph.smvm.vectorised.par.N4
runtime (wall clock) : 100.5ms (----) / 100.6ms (----) / 100.9ms (----)
kernel runtime (wall clock) : 10.00ms ( +11) / 12.30ms ( -19) / 18.00ms (----)
dph.smvm.vectorised.par.N8
runtime (wall clock) : 100.5ms (----) / 100.6ms (----) / 100.8ms (----)
kernel runtime (wall clock) : 15.00ms ( +15) / 15.20ms ( +4) / 16.00ms (----)
dph.quickhull.vectorised.par.N2
runtime (wall clock) : 801.1ms ( +14) / 801.3ms ( +1) / 801.5ms (----)
kernel runtime (wall clock) : 436.0ms ( -1) / 482.4ms ( +2) / 535.0ms ( +8)
dph.quickhull.vector-mutable.seq.N4
runtime (wall clock) : 400.7ms ( +33) / 400.8ms ( +29) / 400.9ms (----)
kernel runtime (wall clock) : 52.00ms ( -2) / 53.20ms (----) / 56.00ms (----)
dph.quickhull.vector-forkIO.par.N4
runtime (wall clock) : 300.6ms (----) / 330.7ms ( +10) / 400.8ms ( +33)
kernel runtime (wall clock) : 32.00ms ( +3) / 36.10ms ( +6) / 43.00ms ( +5)
dph.quickhull.vector-forkIO.par.N8
runtime (wall clock) : 300.7ms ( -25) / 300.9ms ( -25) / 301.1ms ( -25)
kernel runtime (wall clock) : 38.00ms (----) / 45.40ms ( +14) / 49.00ms ( +20)
repa.mmult.par.N2
runtime (wall clock) : 4.803s ( +12) / 4.805s ( +10) / 4.806s ( +9)
kernel runtime (wall clock) : 4.480s ( +6) / 4.490s ( +6) / 4.494s ( +6)
repa.mmult.par.N4
runtime (wall clock) : 2.502s ( +32) / 2.543s ( +34) / 2.603s ( +37)
kernel runtime (wall clock) : 2.247s ( +25) / 2.259s ( +26) / 2.267s ( +26)
repa.mmult.par.N8
runtime (wall clock) : 1.409s ( +17) / 1.426s ( +16) / 1.506s ( +14)
kernel runtime (wall clock) : 1.131s ( +4) / 1.144s ( +4) / 1.167s (----)
repa.laplace.par.N4
runtime (wall clock) : 2.703s ( -10) / 3.153s ( +5) / 3.404s ( +13)
kernel runtime (wall clock) : 2.279s ( +13) / 2.725s ( +35) / 2.941s ( +46)
repa.blur.par.N6
runtime (wall clock) : 400.6ms (----) / 410.7ms ( +2) / 500.9ms ( +25)
kernel runtime (wall clock) : 199.0ms (----) / 206.6ms (----) / 220.0ms ( +4)
repa.canny.par.N6
runtime (wall clock) : 200.5ms (----) / 260.6ms ( +30) / 300.7ms ( +50)
kernel runtime (wall clock) : 12.00ms (----) / 13.00ms ( +3) / 14.00ms ( +8)
repa.fft3d.par.N1
runtime (wall clock) : 33.21s ( +11) / 33.38s ( +12) / 33.42s ( +11)
kernel runtime (wall clock) : 27.59s ( +12) / 27.64s ( +12) / 27.69s ( +12)
repa.fft3d.par.N2
runtime (wall clock) : 18.71s ( +11) / 19.68s ( +13) / 20.32s ( +15)
kernel runtime (wall clock) : 15.49s ( +12) / 16.38s ( +14) / 16.93s ( +15)
repa.fft3d.par.N4
runtime (wall clock) : 11.41s ( +43) / 11.47s ( +43) / 11.61s ( +45)
kernel runtime (wall clock) : 9.524s ( +54) / 9.606s ( +55) / 9.781s ( +58)
noslow.list.*Double.find_indices
kernel runtime (wall clock) : 436.6us ( +12) / 438.2us ( +13) / 439.3us ( +13)
noslow.vector.Primitive.*Double.filterin_r
kernel runtime (wall clock) : 408.7us ( +13) / 410.4us ( +13) / 411.4us ( +13)
noslow.vector-unsafe.Primitive.*Double.filterin_r
kernel runtime (wall clock) : 410.6us ( +13) / 412.1us ( +13) / 413.1us ( +14)
noslow.dph-prim.seq.Double.scale_r
kernel runtime (wall clock) : 69.16us ( +10) / 69.18us ( +10) / 69.20us ( +10)
noslow.dph-prim.seq.Double.axpy
kernel runtime (wall clock) : 91.46us ( +11) / 91.49us ( +11) / 91.51us ( +11)
noslow.vector.Primitive.Double.scale_r
kernel runtime (wall clock) : 69.36us ( +13) / 69.38us ( +13) / 69.39us ( +13)
noslow.vector.Primitive.Double.axpy
kernel runtime (wall clock) : 91.23us ( +11) / 91.25us ( +11) / 91.26us ( +11)
noslow.vector.Unboxed.Double.scale_r
kernel runtime (wall clock) : 69.28us ( +11) / 69.36us ( +11) / 69.44us ( +11)
noslow.vector-unsafe.Primitive.Double.scale_r
kernel runtime (wall clock) : 65.75us ( +11) / 65.76us ( +11) / 65.78us ( +11)
noslow.vector-unsafe.Unboxed.Double.scale_r
kernel runtime (wall clock) : 70.17us ( +12) / 70.19us ( +12) / 70.21us ( +12)
More information about the Cvs-ghc
mailing list