ARM GCC 4.7.2 Cortex-A15 Compiler Optimizations Benchmarking compiler optimization levels on GCC 4.7.2 from a Google Nexus 10 with Samsung Exynos 5 Dual SoC. It's an ARM Cortex-A15 dual-core 1.7GHz processor. Benchmarking by Michael Larabel for a future article on Phoronix.
HTML result view exported from: https://openbenchmarking.org/result/1302284-FO-ARMGCC47227&sor&grs .
ARM GCC 4.7.2 Cortex-A15 Compiler Optimizations Processor Motherboard Memory Disk OS Kernel Compiler File-System Screen Resolution -Os -O0 -O1 -O2 -O3 -Ofast ARMv7 rev 4 @ 1.70GHz (2 Cores) Manta 2048MB 16GB MAG2GA Ubuntu 12.10 3.4.5-gf68518f (armv7l) GCC 4.7 ext4 2560x1600 OpenBenchmarking.org Compiler Details - --build=arm-linux-gnueabihf --disable-libitm --disable-sjlj-exceptions --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-languages=c,c++,go,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multilib --enable-nls --enable-objc-gc --enable-plugin --enable-shared --enable-threads=posix --host=arm-linux-gnueabihf --target=arm-linux-gnueabihf --with-arch=armv7-a --with-float=hard --with-fpu=vfpv3-d16 --with-mode=thumb -v Processor Details - Scaling Governor: interactive
ARM GCC 4.7.2 Cortex-A15 Compiler Optimizations smallpt: Global Illumination Renderer; 100 Samples graphics-magick: Local Adaptive Thresholding encode-opus: WAV, Opus Encode, Opus Decode nero2d: Total Time graphics-magick: Sharpen c-ray: Total Time graphics-magick: Resizing build-php: Time To Compile himeno: Poisson Pressure Solver graphics-magick: HWB Color Space graphics-magick: Blur -Os -O0 -O1 -O2 -O3 -Ofast 784 19 55.78 3505.78 20 588.38 36 204.10 100.58 48 29 1388 6 163.43 7634.24 7 909.85 16 117.17 60.43 27 16 361 19 51.84 2300.70 19 653.17 36 191.56 108.57 47 30 329 22 47.80 2142.95 23 523.58 40 252.76 140.04 47 31 1297 22 46.64 2149.05 23 349.83 40 285.41 141.55 47 30 279 22 45.73 2145.44 23 332.87 39 286.52 129.49 55 31 OpenBenchmarking.org
Smallpt Global Illumination Renderer; 100 Samples OpenBenchmarking.org Seconds, Fewer Is Better Smallpt 1.0 Global Illumination Renderer; 100 Samples -Ofast -O2 -O1 -Os -O3 -O0 300 600 900 1200 1500 SE +/- 1.20, N = 3 SE +/- 2.73, N = 3 SE +/- 1.53, N = 3 SE +/- 3.48, N = 3 SE +/- 4.18, N = 3 SE +/- 25.31, N = 3 279 329 361 784 1297 1388 -Ofast -O2 -O1 -Os -O0 1. (CXX) g++ options: -fopenmp
GraphicsMagick Operation: Local Adaptive Thresholding OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.16 Operation: Local Adaptive Thresholding -Ofast -O3 -O2 -O1 -Os -O0 5 10 15 20 25 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 22 22 22 19 19 6 -Ofast -O2 -O2 -O1 -Os -O0 1. (CC) gcc options: -std=gnu99 -fopenmp -pthread -lXext -lSM -lICE -lX11 -lbz2 -lz -lm -lgomp -lpthread
Opus Codec Encoding WAV, Opus Encode, Opus Decode OpenBenchmarking.org Seconds, Fewer Is Better Opus Codec Encoding 1.0.1 WAV, Opus Encode, Opus Decode -Ofast -O3 -O2 -O1 -Os -O0 40 80 120 160 200 SE +/- 0.03, N = 5 SE +/- 0.02, N = 5 SE +/- 0.05, N = 5 SE +/- 0.17, N = 5 SE +/- 0.03, N = 5 SE +/- 0.17, N = 5 45.73 46.64 47.80 51.84 55.78 163.43 -Ofast -O3 -ffast-math -O2 -O1 -Os -O0 1. (CC) gcc options: -std=gnu99 -fPIE -pie -logg -lm -lopus
Open FMM Nero2D Total Time OpenBenchmarking.org Seconds, Fewer Is Better Open FMM Nero2D 2.0.2 Total Time -O2 -Ofast -O3 -O1 -Os -O0 1600 3200 4800 6400 8000 2142.95 2145.44 2149.05 2300.70 3505.78 7634.24 -O2 -Ofast -O2 -O1 -Os -O0 1. (CXX) g++ options: -lfftw3 -llapack -lblas -lgfortran -lm
GraphicsMagick Operation: Sharpen OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.16 Operation: Sharpen -Ofast -O3 -O2 -Os -O1 -O0 6 12 18 24 30 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 6 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 23 23 23 20 19 7 -Ofast -O2 -O2 -Os -O1 -O0 1. (CC) gcc options: -std=gnu99 -fopenmp -pthread -lXext -lSM -lICE -lX11 -lbz2 -lz -lm -lgomp -lpthread
C-Ray Total Time OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time -Ofast -O3 -O2 -Os -O1 -O0 200 400 600 800 1000 SE +/- 3.01, N = 3 SE +/- 4.87, N = 5 SE +/- 8.88, N = 4 SE +/- 3.84, N = 3 SE +/- 15.01, N = 6 SE +/- 1.42, N = 3 332.87 349.83 523.58 588.38 653.17 909.85 -Ofast -O2 -Os -O1 -O0 1. (CC) gcc options: -lm -lpthread -O3
GraphicsMagick Operation: Resizing OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.16 Operation: Resizing -O3 -O2 -Ofast -O1 -Os -O0 9 18 27 36 45 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 40 40 39 36 36 16 -O2 -O2 -Ofast -O1 -Os -O0 1. (CC) gcc options: -std=gnu99 -fopenmp -pthread -lXext -lSM -lICE -lX11 -lbz2 -lz -lm -lgomp -lpthread
Timed PHP Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed PHP Compilation 5.2.9 Time To Compile -O0 -O1 -Os -O2 -O3 -Ofast 60 120 180 240 300 SE +/- 0.25, N = 3 SE +/- 0.85, N = 3 SE +/- 0.90, N = 3 SE +/- 0.77, N = 3 SE +/- 1.07, N = 3 SE +/- 0.83, N = 3 117.17 191.56 204.10 252.76 285.41 286.52 -O0 -O1 -Os -O2 -O2 -Ofast 1. (CC) gcc options: -pedantic -ldl -lz -lm
Himeno Benchmark Poisson Pressure Solver OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver -O3 -O2 -Ofast -O1 -Os -O0 30 60 90 120 150 SE +/- 1.96, N = 3 SE +/- 5.17, N = 6 SE +/- 1.15, N = 3 SE +/- 1.57, N = 3 SE +/- 1.80, N = 6 SE +/- 0.29, N = 3 141.55 140.04 129.49 108.57 100.58 60.43 -O2 -Ofast -O1 -Os -O0 1. (CC) gcc options: -O3
GraphicsMagick Operation: HWB Color Space OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.16 Operation: HWB Color Space -Ofast -Os -O3 -O2 -O1 -O0 12 24 36 48 60 SE +/- 0.88, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 55 48 47 47 47 27 -Ofast -Os -O2 -O2 -O1 -O0 1. (CC) gcc options: -std=gnu99 -fopenmp -pthread -lXext -lSM -lICE -lX11 -lbz2 -lz -lm -lgomp -lpthread
GraphicsMagick Operation: Blur OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.16 Operation: Blur -Ofast -O2 -O3 -O1 -Os -O0 7 14 21 28 35 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.25, N = 4 31 31 30 30 29 16 -Ofast -O2 -O2 -O1 -Os -O0 1. (CC) gcc options: -std=gnu99 -fopenmp -pthread -lXext -lSM -lICE -lX11 -lbz2 -lz -lm -lgomp -lpthread
Phoronix Test Suite v10.8.5