[nightly] DPH Performance Test Succeeded
DPH Buildbot
dph at ouroborus.net
Sun Jul 31 16:36:12 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%)
dph.evens.vectorised.par.N1
runtime (wall clock) : 500.7ms (----) / 500.9ms (----) / 501.1ms (----)
kernel runtime (wall clock) : 181.0ms ( -11) / 181.7ms ( -12) / 182.0ms ( -12)
dph.evens.vectorised.par.N2
runtime (wall clock) : 400.7ms (----) / 400.9ms ( -11) / 401.1ms ( -20)
kernel runtime (wall clock) : 102.0ms ( -9) / 106.9ms ( -9) / 112.0ms ( -9)
dph.quicksort.vectorised.par.N1
runtime (wall clock) : 600.9ms ( -14) / 690.9ms ( -1) / 701.1ms (----)
kernel runtime (wall clock) : 379.0ms ( -7) / 384.0ms ( -9) / 386.0ms ( -10)
dph.quicksort.vectorised.par.N8
runtime (wall clock) : 801.3ms (----) / 801.4ms ( -1) / 801.6ms ( -11)
kernel runtime (wall clock) : 457.0ms ( -3) / 474.9ms ( -2) / 483.0ms ( -4)
dph.smvm.vectorised.par.N4
runtime (wall clock) : 100.5ms (----) / 100.6ms (----) / 100.9ms (----)
kernel runtime (wall clock) : 9.000ms (----) / 11.00ms ( -27) / 16.00ms ( -11)
dph.quickhull.vector-immutable.seq.N4
runtime (wall clock) : 400.7ms (----) / 400.8ms (----) / 401.0ms (----)
kernel runtime (wall clock) : 114.0ms ( -16) / 128.9ms ( -8) / 136.0ms ( -4)
repa.laplace.par.N1
runtime (wall clock) : 7.004s ( -25) / 7.024s ( -25) / 7.104s ( -26)
kernel runtime (wall clock) : 6.375s ( -7) / 6.414s ( -7) / 6.468s ( -7)
repa.laplace.par.N2
runtime (wall clock) : 4.003s ( -23) / 4.174s ( -21) / 4.204s ( -22)
kernel runtime (wall clock) : 3.543s ( -9) / 3.685s ( -6) / 3.717s ( -6)
repa.fft2d.par.N2
runtime (wall clock) : 4.803s ( -13) / 4.924s ( -11) / 5.105s ( -10)
kernel runtime (wall clock) : 4.189s ( -15) / 4.347s ( -13) / 4.549s ( -12)
noslow.dph-prim.seq.Double.filterin_r
kernel runtime (wall clock) : 36.76us ( -14) / 36.78us ( -14) / 36.80us ( -14)
-- LOSERS (had a swing of > +10%)
dph.dotp.vectorised.par.N1
runtime (wall clock) : 1.901s ( +6) / 1.902s ( +4) / 1.903s (----)
kernel runtime (wall clock) : 99.00ms (+106) / 99.00ms (+106) / 99.00ms (+102)
dph.dotp.vectorised.par.N2
runtime (wall clock) : 1.901s ( +6) / 1.902s ( +6) / 1.902s ( +5)
kernel runtime (wall clock) : 49.00ms (+104) / 50.20ms ( +54) / 51.00ms ( +24)
dph.dotp.vectorised.par.N4
runtime (wall clock) : 1.801s ( -31) / 1.802s ( -31) / 1.802s ( -33)
kernel runtime (wall clock) : 27.00ms ( +29) / 30.30ms ( +11) / 32.00ms ( +3)
dph.dotp.vectorised.seq.N4
runtime (wall clock) : 1.901s ( +6) / 1.902s ( +6) / 1.902s ( +6)
kernel runtime (wall clock) : 102.0ms ( +92) / 102.0ms ( +92) / 102.0ms ( +85)
dph.dotp.vector.seq.N4
runtime (wall clock) : 1.901s ( +6) / 1.972s ( +9) / 2.002s ( +11)
kernel runtime (wall clock) : 53.00ms (----) / 87.30ms ( +63) / 102.0ms ( +79)
dph.sumsq.vectorised.par.N2
runtime (wall clock) : 500.6ms (+150) / 500.6ms (+150) / 500.7ms (+149)
kernel runtime (wall clock) : 418.0ms (+127) / 418.1ms (+127) / 419.0ms (+128)
dph.sumsq.vectorised.par.N4
runtime (wall clock) : 300.5ms (+199) / 300.6ms (+199) / 300.9ms (+199)
kernel runtime (wall clock) : 209.0ms (+127) / 209.0ms (+127) / 209.0ms (+127)
dph.sumsq.vectorised.par.N8
runtime (wall clock) : 200.5ms (+100) / 200.5ms (+100) / 200.6ms ( +99)
kernel runtime (wall clock) : 105.0ms (+128) / 105.2ms (+127) / 107.0ms (+114)
dph.sumsq.vectorised.seq.N4
runtime (wall clock) : 1.201s (+200) / 1.201s (+200) / 1.201s (+200)
kernel runtime (wall clock) : 1.103s (+201) / 1.104s (+200) / 1.106s (+201)
dph.sumsq.vector.seq.N4
runtime (wall clock) : 1.201s (+200) / 1.201s (+200) / 1.201s (+200)
kernel runtime (wall clock) : 1.103s (+201) / 1.103s (+200) / 1.104s (+200)
dph.evens.vectorised.par.N8
runtime (wall clock) : 400.8ms (----) / 480.9ms ( +17) / 501.1ms (----)
kernel runtime (wall clock) : 66.00ms ( -3) / 73.50ms ( +3) / 80.00ms ( -7)
dph.evens.vectorised.seq.N4
runtime (wall clock) : 500.8ms (----) / 570.9ms ( +14) / 601.0ms ( +20)
kernel runtime (wall clock) : 199.0ms ( +5) / 199.7ms ( +5) / 202.0ms ( +6)
dph.quicksort.vectorised.par.N2
runtime (wall clock) : 601.1ms ( +20) / 611.1ms ( -8) / 701.1ms (----)
kernel runtime (wall clock) : 327.0ms ( +8) / 355.1ms ( -4) / 367.0ms ( -6)
dph.quicksort.vectorised.par.N4
runtime (wall clock) : 601.0ms ( +20) / 671.3ms ( +6) / 701.6ms (----)
kernel runtime (wall clock) : 357.0ms ( +16) / 366.9ms ( +3) / 382.0ms (----)
dph.smvm.vectorised.seq.N4
runtime (wall clock) : 100.4ms (----) / 100.5ms (----) / 100.8ms (----)
kernel runtime (wall clock) : 23.00ms ( +35) / 23.00ms ( +35) / 23.00ms ( +35)
dph.quickhull.vectorised.par.N2
runtime (wall clock) : 801.1ms ( +14) / 811.3ms ( +3) / 901.4ms ( +12)
kernel runtime (wall clock) : 475.0ms ( +8) / 504.7ms ( +7) / 547.0ms ( +10)
dph.quickhull.vectorised.seq.N4
runtime (wall clock) : 900.8ms ( +13) / 901.1ms ( +6) / 901.5ms (----)
kernel runtime (wall clock) : 558.0ms ( +9) / 569.7ms ( +10) / 573.0ms ( +9)
dph.quickhull.vector-mutable.seq.N4
runtime (wall clock) : 301.0ms (----) / 390.8ms ( +26) / 401.1ms (----)
kernel runtime (wall clock) : 54.00ms ( +2) / 57.20ms ( +7) / 61.00ms ( +9)
dph.quickhull.vector-forkIO.par.N8
runtime (wall clock) : 300.7ms ( -25) / 310.8ms ( -22) / 400.9ms (----)
kernel runtime (wall clock) : 45.00ms ( +18) / 49.60ms ( +25) / 57.00ms ( +39)
repa.mmult.par.N1
runtime (wall clock) : 21.91s (+155) / 21.93s (+154) / 22.01s (+150)
kernel runtime (wall clock) : 21.61s (+155) / 21.65s (+155) / 21.70s (+155)
repa.mmult.par.N2
runtime (wall clock) : 11.11s (+158) / 11.13s (+155) / 11.21s (+155)
kernel runtime (wall clock) : 10.86s (+156) / 10.88s (+156) / 10.91s (+156)
repa.mmult.par.N4
runtime (wall clock) : 5.703s (+200) / 5.705s (+200) / 5.706s (+200)
kernel runtime (wall clock) : 5.446s (+204) / 5.457s (+204) / 5.465s (+205)
repa.mmult.par.N8
runtime (wall clock) : 3.004s (+150) / 3.039s (+148) / 3.121s (+137)
kernel runtime (wall clock) : 2.727s (+152) / 2.752s (+151) / 2.845s (+142)
repa.laplace.par.N4
runtime (wall clock) : 2.603s ( -13) / 3.203s ( +7) / 3.704s ( +23)
kernel runtime (wall clock) : 2.225s ( +10) / 2.783s ( +38) / 3.312s ( +64)
repa.laplace.par.N6
runtime (wall clock) : 3.003s ( +7) / 3.033s ( +8) / 3.104s ( +11)
kernel runtime (wall clock) : 2.608s ( +22) / 2.642s ( +22) / 2.684s ( +22)
repa.blur.par.N1
runtime (wall clock) : 6.603s (+450) / 6.604s (+450) / 6.604s (+450)
kernel runtime (wall clock) : 6.249s (+674) / 6.254s (+673) / 6.263s (+673)
repa.blur.par.N2
runtime (wall clock) : 3.902s (+457) / 4.003s (+463) / 4.503s (+462)
kernel runtime (wall clock) : 3.595s (+713) / 3.706s (+727) / 4.233s (+795)
repa.blur.par.N4
runtime (wall clock) : 2.602s (+420) / 2.682s (+436) / 2.702s (+439)
kernel runtime (wall clock) : 2.390s (+766) / 2.435s (+761) / 2.490s (+756)
repa.blur.par.N6
runtime (wall clock) : 2.002s (+400) / 2.013s (+402) / 2.102s (+424)
kernel runtime (wall clock) : 1.777s (+784) / 1.791s (+773) / 1.825s (+761)
repa.blur.par.N8
runtime (wall clock) : 1.703s (+325) / 1.714s (+327) / 1.804s (+350)
kernel runtime (wall clock) : 1.496s (+755) / 1.514s (+733) / 1.532s (+715)
repa.canny.par.N1
runtime (wall clock) : 2.101s (+425) / 2.101s (+424) / 2.102s (+424)
kernel runtime (wall clock) : 1.680s (+6900) / 1.682s (+6908) / 1.686s (+6925)
repa.canny.par.N2
runtime (wall clock) : 1.201s (+300) / 1.201s (+300) / 1.201s (+299)
kernel runtime (wall clock) : 937.0ms (+5412) / 940.5ms (+4929) / 945.0ms (+4625)
repa.canny.par.N4
runtime (wall clock) : 900.9ms (+200) / 901.0ms (+200) / 901.2ms (+200)
kernel runtime (wall clock) : 595.0ms (+4477) / 615.7ms (+4495) / 632.0ms (+4113)
repa.canny.par.N6
runtime (wall clock) : 700.8ms (+250) / 701.0ms (+250) / 701.2ms (+249)
kernel runtime (wall clock) : 449.0ms (+3642) / 455.0ms (+3511) / 458.0ms (+3423)
repa.canny.par.N8
runtime (wall clock) : 600.9ms (+200) / 601.1ms (+200) / 601.4ms (+200)
kernel runtime (wall clock) : 371.0ms (+2754) / 376.4ms (+2773) / 392.0ms (+2700)
repa.fft3d.par.N1
runtime (wall clock) : 34.21s ( +15) / 34.36s ( +15) / 34.42s ( +15)
kernel runtime (wall clock) : 28.70s ( +17) / 28.77s ( +17) / 28.83s ( +17)
repa.fft3d.par.N2
runtime (wall clock) : 19.11s ( +13) / 20.18s ( +16) / 20.91s ( +19)
kernel runtime (wall clock) : 16.01s ( +15) / 16.93s ( +18) / 17.56s ( +20)
repa.fft3d.par.N4
runtime (wall clock) : 11.51s ( +44) / 11.68s ( +46) / 11.81s ( +48)
kernel runtime (wall clock) : 9.690s ( +56) / 9.836s ( +59) / 9.992s ( +61)
repa.fft3d.par.N6
runtime (wall clock) : 8.809s ( +10) / 8.849s ( +9) / 8.909s ( +9)
kernel runtime (wall clock) : 7.369s ( +12) / 7.453s ( +12) / 7.540s ( +12)
noslow.dph-prim.seq.Double.scale_r
kernel runtime (wall clock) : 69.01us ( +10) / 69.02us ( +10) / 69.03us ( +10)
noslow.dph-prim.seq.Double.axpy
kernel runtime (wall clock) : 91.41us ( +11) / 91.47us ( +11) / 91.53us ( +11)
noslow.vector.Primitive.Double.scale_r
kernel runtime (wall clock) : 69.22us ( +12) / 69.23us ( +12) / 69.25us ( +12)
noslow.vector.Primitive.Double.axpy
kernel runtime (wall clock) : 91.20us ( +11) / 91.21us ( +11) / 91.22us ( +11)
noslow.vector.Unboxed.Double.scale_r
kernel runtime (wall clock) : 69.19us ( +11) / 69.23us ( +11) / 69.28us ( +11)
noslow.vector.Unboxed.Double.axpy
kernel runtime (wall clock) : 91.13us ( +10) / 91.16us ( +10) / 91.19us ( +10)
noslow.vector-unsafe.Primitive.Double.scale_r
kernel runtime (wall clock) : 65.76us ( +11) / 65.78us ( +11) / 65.79us ( +11)
noslow.vector-unsafe.Unboxed.Double.scale_r
kernel runtime (wall clock) : 70.21us ( +12) / 70.23us ( +12) / 70.25us ( +12)
More information about the Cvs-ghc
mailing list