[nightly] DPH Performance Test Succeeded
DPH Buildbot
dph at ouroborus.net
Fri Jul 29 07:20:45 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) : 500.8ms (----) / 500.9ms (----) / 501.2ms (----)
kernel runtime (wall clock) : 180.0ms ( -12) / 180.8ms ( -12) / 183.0ms ( -11)
dph.quicksort.vectorised.par.N1
runtime (wall clock) : 617.4ms ( -12) / 692.6ms ( -1) / 701.1ms (----)
kernel runtime (wall clock) : 383.0ms ( -6) / 384.2ms ( -9) / 385.0ms ( -10)
dph.quickhull.vector-immutable.seq.N4
runtime (wall clock) : 400.8ms (----) / 400.9ms (----) / 401.1ms (----)
kernel runtime (wall clock) : 113.0ms ( -16) / 129.1ms ( -7) / 146.0ms ( +3)
repa.laplace.par.N1
runtime (wall clock) : 6.804s ( -27) / 6.804s ( -28) / 6.806s ( -29)
kernel runtime (wall clock) : 6.159s ( -10) / 6.175s ( -11) / 6.207s ( -11)
repa.laplace.par.N2
runtime (wall clock) : 4.002s ( -23) / 4.013s ( -24) / 4.102s ( -24)
kernel runtime (wall clock) : 3.486s ( -10) / 3.515s ( -10) / 3.576s ( -10)
repa.fft2d.par.N2
runtime (wall clock) : 4.803s ( -13) / 5.084s ( -8) / 5.205s ( -9)
kernel runtime (wall clock) : 4.220s ( -14) / 4.483s ( -11) / 4.614s ( -11)
noslow.dph-prim.seq.Double.filterin_r
kernel runtime (wall clock) : 36.74us ( -14) / 36.75us ( -14) / 36.75us ( -14)
-- LOSERS (had a swing of > +10%)
dph.dotp.vectorised.par.N1
runtime (wall clock) : 1.801s (----) / 1.892s ( +4) / 1.903s (----)
kernel runtime (wall clock) : 48.00ms (----) / 93.90ms ( +95) / 99.00ms (+102)
dph.dotp.vectorised.par.N2
runtime (wall clock) : 1.801s (----) / 1.802s (----) / 1.802s (----)
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) / 30.80ms ( +13) / 33.00ms ( +6)
dph.dotp.vectorised.seq.N4
runtime (wall clock) : 2.001s ( +11) / 2.002s ( +11) / 2.002s ( +11)
kernel runtime (wall clock) : 102.0ms ( +92) / 102.0ms ( +92) / 102.0ms ( +85)
dph.dotp.vector.seq.N4
runtime (wall clock) : 1.801s (----) / 1.892s ( +5) / 1.902s ( +6)
kernel runtime (wall clock) : 53.00ms (----) / 97.10ms ( +81) / 102.0ms ( +79)
dph.sumsq.vectorised.par.N2
runtime (wall clock) : 500.6ms (+150) / 500.7ms (+150) / 500.8ms (+149)
kernel runtime (wall clock) : 418.0ms (+127) / 418.0ms (+127) / 418.0ms (+127)
dph.sumsq.vectorised.par.N4
runtime (wall clock) : 300.5ms (+199) / 300.6ms (+199) / 300.8ms (+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) / 200.6ms (+100) / 200.8ms ( +99)
kernel runtime (wall clock) : 105.0ms (+128) / 105.0ms (+126) / 105.0ms (+110)
dph.sumsq.vectorised.seq.N4
runtime (wall clock) : 1.201s (+200) / 1.201s (+200) / 1.201s (+200)
kernel runtime (wall clock) : 1.103s (+201) / 1.103s (+200) / 1.104s (+200)
dph.sumsq.vector.seq.N4
runtime (wall clock) : 1.201s (+200) / 1.201s (+200) / 1.201s (+200)
kernel runtime (wall clock) : 1.103s (+201) / 1.104s (+200) / 1.106s (+201)
dph.evens.vectorised.par.N8
runtime (wall clock) : 400.8ms (----) / 480.9ms ( +17) / 501.1ms (----)
kernel runtime (wall clock) : 66.00ms ( -3) / 74.60ms ( +4) / 79.00ms ( -8)
dph.evens.vectorised.seq.N4
runtime (wall clock) : 500.9ms (----) / 560.9ms ( +12) / 601.1ms ( +20)
kernel runtime (wall clock) : 199.0ms ( +5) / 199.5ms ( +5) / 200.0ms ( +5)
dph.quicksort.vectorised.par.N2
runtime (wall clock) : 600.8ms ( +20) / 601.1ms ( -9) / 601.4ms ( -14)
kernel runtime (wall clock) : 325.0ms ( +7) / 343.8ms ( -7) / 356.0ms ( -9)
dph.quicksort.vectorised.par.N4
runtime (wall clock) : 601.1ms ( +20) / 651.2ms ( +3) / 701.5ms (----)
kernel runtime (wall clock) : 343.0ms ( +12) / 361.1ms ( +1) / 385.0ms ( +2)
dph.smvm.vectorised.seq.N4
runtime (wall clock) : 100.4ms (----) / 100.6ms (----) / 100.8ms (----)
kernel runtime (wall clock) : 23.00ms ( +35) / 23.00ms ( +35) / 23.00ms ( +35)
dph.quickhull.vectorised.par.N2
runtime (wall clock) : 800.9ms ( +14) / 811.3ms ( +3) / 901.1ms ( +12)
kernel runtime (wall clock) : 456.0ms ( +3) / 495.0ms ( +5) / 545.0ms ( +10)
dph.quickhull.vectorised.seq.N4
runtime (wall clock) : 900.8ms ( +12) / 901.1ms ( +6) / 901.4ms (----)
kernel runtime (wall clock) : 558.0ms ( +9) / 570.2ms ( +10) / 573.0ms ( +9)
dph.quickhull.vector-mutable.seq.N4
runtime (wall clock) : 300.7ms (----) / 380.8ms ( +23) / 401.0ms (----)
kernel runtime (wall clock) : 54.00ms ( +2) / 58.60ms ( +10) / 63.00ms ( +12)
dph.quickhull.vector-forkIO.par.N4
runtime (wall clock) : 300.7ms (----) / 300.8ms (----) / 301.0ms (----)
kernel runtime (wall clock) : 31.00ms (----) / 37.50ms ( +11) / 45.00ms ( +10)
dph.quickhull.vector-forkIO.par.N8
runtime (wall clock) : 300.7ms ( -25) / 300.9ms ( -25) / 301.2ms ( -25)
kernel runtime (wall clock) : 44.00ms ( +16) / 47.20ms ( +19) / 49.00ms ( +20)
repa.mmult.par.N1
runtime (wall clock) : 21.82s (+154) / 21.92s (+154) / 22.01s (+150)
kernel runtime (wall clock) : 21.60s (+155) / 21.66s (+155) / 21.71s (+155)
repa.mmult.par.N2
runtime (wall clock) : 11.11s (+158) / 11.11s (+155) / 11.11s (+152)
kernel runtime (wall clock) : 10.85s (+156) / 10.87s (+156) / 10.89s (+156)
repa.mmult.par.N4
runtime (wall clock) : 5.703s (+200) / 5.705s (+200) / 5.706s (+200)
kernel runtime (wall clock) : 5.422s (+202) / 5.454s (+204) / 5.469s (+205)
repa.mmult.par.N8
runtime (wall clock) : 2.936s (+144) / 3.016s (+146) / 3.131s (+138)
kernel runtime (wall clock) : 2.726s (+152) / 2.744s (+150) / 2.844s (+142)
repa.laplace.par.N4
runtime (wall clock) : 2.603s ( -13) / 3.083s ( +3) / 3.403s ( +13)
kernel runtime (wall clock) : 2.180s ( +8) / 2.668s ( +32) / 3.008s ( +49)
repa.laplace.par.N6
runtime (wall clock) : 2.903s ( +4) / 2.943s ( +5) / 3.003s ( +7)
kernel runtime (wall clock) : 2.529s ( +18) / 2.550s ( +18) / 2.567s ( +16)
repa.blur.par.N1
runtime (wall clock) : 6.403s (+433) / 6.403s (+433) / 6.404s (+433)
kernel runtime (wall clock) : 6.019s (+646) / 6.021s (+645) / 6.030s (+644)
repa.blur.par.N2
runtime (wall clock) : 4.003s (+471) / 4.403s (+519) / 4.503s (+462)
kernel runtime (wall clock) : 3.677s (+732) / 4.105s (+815) / 4.206s (+789)
repa.blur.par.N4
runtime (wall clock) : 2.502s (+400) / 2.622s (+424) / 2.702s (+440)
kernel runtime (wall clock) : 2.213s (+702) / 2.369s (+738) / 2.432s (+736)
repa.blur.par.N6
runtime (wall clock) : 2.002s (+400) / 2.002s (+400) / 2.002s (+399)
kernel runtime (wall clock) : 1.741s (+766) / 1.767s (+761) / 1.781s (+740)
repa.blur.par.N8
runtime (wall clock) : 1.703s (+325) / 1.714s (+327) / 1.804s (+350)
kernel runtime (wall clock) : 1.486s (+749) / 1.504s (+727) / 1.559s (+729)
repa.canny.par.N1
runtime (wall clock) : 2.101s (+425) / 2.101s (+424) / 2.102s (+424)
kernel runtime (wall clock) : 1.664s (+6833) / 1.666s (+6843) / 1.678s (+6892)
repa.canny.par.N2
runtime (wall clock) : 1.401s (+366) / 1.401s (+366) / 1.401s (+366)
kernel runtime (wall clock) : 1.108s (+6418) / 1.111s (+5841) / 1.113s (+5465)
repa.canny.par.N4
runtime (wall clock) : 900.9ms (+200) / 901.0ms (+200) / 901.1ms (+200)
kernel runtime (wall clock) : 598.0ms (+4500) / 611.8ms (+4466) / 628.0ms (+4087)
repa.canny.par.N6
runtime (wall clock) : 700.8ms (+250) / 701.0ms (+249) / 701.2ms (+249)
kernel runtime (wall clock) : 442.0ms (+3583) / 453.3ms (+3498) / 460.0ms (+3438)
repa.canny.par.N8
runtime (wall clock) : 600.9ms (+200) / 601.1ms (+200) / 601.3ms (+199)
kernel runtime (wall clock) : 373.0ms (+2769) / 380.4ms (+2804) / 411.0ms (+2836)
repa.fft3d.par.N1
runtime (wall clock) : 33.42s ( +12) / 33.48s ( +12) / 33.53s ( +12)
kernel runtime (wall clock) : 27.71s ( +13) / 27.74s ( +12) / 27.77s ( +12)
repa.fft3d.par.N2
runtime (wall clock) : 18.71s ( +11) / 19.65s ( +13) / 20.42s ( +16)
kernel runtime (wall clock) : 15.46s ( +12) / 16.25s ( +13) / 16.95s ( +15)
repa.fft3d.par.N4
runtime (wall clock) : 11.41s ( +42) / 11.51s ( +44) / 11.71s ( +46)
kernel runtime (wall clock) : 9.501s ( +53) / 9.628s ( +55) / 9.790s ( +58)
noslow.list.*Double.find_indices
kernel runtime (wall clock) : 436.6us ( +12) / 438.3us ( +13) / 439.3us ( +13)
noslow.vector.Primitive.*Double.filterin_r
kernel runtime (wall clock) : 408.1us ( +13) / 409.6us ( +13) / 410.6us ( +13)
noslow.vector-unsafe.Primitive.*Double.filterin_r
kernel runtime (wall clock) : 409.7us ( +13) / 411.4us ( +13) / 412.5us ( +14)
noslow.dph-prim.seq.Double.scale_r
kernel runtime (wall clock) : 69.07us ( +10) / 69.08us ( +10) / 69.10us ( +10)
noslow.dph-prim.seq.Double.axpy
kernel runtime (wall clock) : 91.36us ( +11) / 91.38us ( +11) / 91.40us ( +11)
noslow.vector.Primitive.Double.scale_r
kernel runtime (wall clock) : 69.24us ( +12) / 69.26us ( +12) / 69.27us ( +12)
noslow.vector.Primitive.Double.axpy
kernel runtime (wall clock) : 91.22us ( +11) / 91.23us ( +11) / 91.25us ( +11)
noslow.vector.Unboxed.Double.scale_r
kernel runtime (wall clock) : 69.09us ( +11) / 69.13us ( +11) / 69.17us ( +11)
noslow.vector-unsafe.Primitive.Double.scale_r
kernel runtime (wall clock) : 65.71us ( +11) / 65.72us ( +11) / 65.73us ( +11)
noslow.vector-unsafe.Unboxed.Double.scale_r
kernel runtime (wall clock) : 70.14us ( +12) / 70.15us ( +12) / 70.16us ( +12)
More information about the Cvs-ghc
mailing list