ARM Cortex A15 GCC Compiler Benchmarks Benchmarks of GCC 4.6.3 vs. GCC 4.7.2 from a Samsung Exynos 5 Dual SoC with ARM Cortex A15 dual-core ARMv7 1.7GHz processor. Cortex A15 ARM compiler benchmarking for a future article on Phoronix.com.
HTML result view exported from: https://openbenchmarking.org/result/1211249-RA-CORTEX15G24&grr&sor .
ARM Cortex A15 GCC Compiler Benchmarks Processor Motherboard Memory Disk OS Kernel Desktop Display Server Compiler File-System Screen Resolution GCC 4.6.3 GCC 4.7.2 Exynos ARMv7 rev 4 @ 1.70GHz (2 Cores) SAMSUNG EXYNOS5 (Flattened Device Tree) 2048MB 16GB SEM16G Ubuntu 12.04 3.4.0 (armv7l) Unity 2D 5.16.0 X Server 1.11.3 GCC 4.6 ext4 1366x768 GCC 4.7 OpenBenchmarking.org Compiler Details - GCC 4.6.3: --build=arm-linux-gnueabihf --disable-sjlj-exceptions --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-languages=c,c++,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 - GCC 4.7.2: --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: ondemand System Details - GCC 4.6.3: Python 2.7.3.
ARM Cortex A15 GCC Compiler Benchmarks nginx: Static Web Page Serving openssl: RSA 4096-bit Performance ffmpeg: H.264 HD To NTSC DV encode-ogg: WAV To Ogg encode-mp3: WAV To MP3 encode-flac: WAV To FLAC crafty: Elapsed Time smallpt: Global Illumination Renderer; 100 Samples primesieve: 1e12 Prime Number Generation compress-pbzip2: 256MB File Compression c-ray: Total Time build-php: Time To Compile compress-7zip: Compress Speed Test x264: H.264 Video Encoding vpxenc: vpxenc tscp: AI Chess Performance scimark2: Jacobi Successive Over-Relaxation scimark2: Dense LU Matrix Factorization scimark2: Sparse Matrix Multiply scimark2: Fast Fourier Transform scimark2: Monte Carlo cachebench: Read / Modify / Write cachebench: Write cachebench: Read gmpbench: Total Time ffte: N=64, 1D Complex FFT Routine dolfyn: Computational Fluid Dynamics clomp: Static OMP Speedup GCC 4.6.3 GCC 4.7.2 3535.77 10 162.54 31.50 52.30 19.81 0.06 1208 957.91 88.41 447.38 255.79 1951 10.62 1.21 146968 361.97 435.03 333.34 37.65 167.90 6979.64 3521.37 2570.83 451.29 1257.00 158.91 1.70 3578.55 9.90 157.40 31.75 54.48 19.18 0.06 1181 810.27 84.30 348.61 284.15 1703 10.83 1.29 146096 344.10 450.12 338.23 44.62 160.62 7092.44 3722.67 2582.33 452.59 1476.59 171.65 1.81 OpenBenchmarking.org
NGINX Benchmark Static Web Page Serving OpenBenchmarking.org Requests Per Second, More Is Better NGINX Benchmark 1.0.11 Static Web Page Serving GCC 4.7.2 GCC 4.6.3 800 1600 2400 3200 4000 SE +/- 70.90, N = 3 SE +/- 9.45, N = 3 3578.55 3535.77 1. (CC) gcc options: -lpthread -lcrypt -lcrypto -lz
OpenSSL RSA 4096-bit Performance OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.0.1c RSA 4096-bit Performance GCC 4.6.3 GCC 4.7.2 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 10.00 9.90 1. (CC) gcc options: -march=armv7-a -O3 -lssl -lcrypto -ldl
FFmpeg H.264 HD To NTSC DV OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 1.0 H.264 HD To NTSC DV GCC 4.7.2 GCC 4.6.3 40 80 120 160 200 SE +/- 0.77, N = 3 SE +/- 0.55, N = 3 157.40 162.54 1. (CC) gcc options: -lavdevice -lavfilter -lavformat -lavcodec -lswresample -lswscale -lavutil -ldl -lm -pthread -lbz2 -lrt
Ogg Encoding WAV To Ogg OpenBenchmarking.org Seconds, Fewer Is Better Ogg Encoding 1.3.0 WAV To Ogg GCC 4.6.3 GCC 4.7.2 7 14 21 28 35 SE +/- 0.13, N = 5 SE +/- 0.16, N = 5 31.50 31.75 1. (CC) gcc options: -O2 -ffast-math -fsigned-char -lm -logg
LAME MP3 Encoding WAV To MP3 OpenBenchmarking.org Seconds, Fewer Is Better LAME MP3 Encoding 3.99.3 WAV To MP3 GCC 4.6.3 GCC 4.7.2 12 24 36 48 60 SE +/- 0.05, N = 5 SE +/- 0.05, N = 5 52.30 54.48
FLAC Audio Encoding WAV To FLAC OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.2.1 WAV To FLAC GCC 4.7.2 GCC 4.6.3 5 10 15 20 25 SE +/- 0.06, N = 5 SE +/- 0.04, N = 5 19.18 19.81 1. (CXX) g++ options: -O2 -logg -lm
Crafty Elapsed Time OpenBenchmarking.org Seconds, Fewer Is Better Crafty 23.4 Elapsed Time GCC 4.6.3 GCC 4.7.2 0.0135 0.027 0.0405 0.054 0.0675 SE +/- 0.01, N = 6 SE +/- 0.00, N = 6 0.06 0.06 1. (CC) gcc options: -lstdc++ -lm
Smallpt Global Illumination Renderer; 100 Samples OpenBenchmarking.org Seconds, Fewer Is Better Smallpt 1.0 Global Illumination Renderer; 100 Samples GCC 4.7.2 GCC 4.6.3 300 600 900 1200 1500 SE +/- 2.08, N = 3 SE +/- 1.45, N = 3 1181 1208 1. (CXX) g++ options: -fopenmp
Primesieve 1e12 Prime Number Generation OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 4.0 1e12 Prime Number Generation GCC 4.7.2 GCC 4.6.3 200 400 600 800 1000 SE +/- 9.65, N = 3 SE +/- 6.82, N = 3 810.27 957.91 1. (CXX) g++ options: -O2 -fopenmp
Parallel BZIP2 Compression 256MB File Compression OpenBenchmarking.org Seconds, Fewer Is Better Parallel BZIP2 Compression 1.1.6 256MB File Compression GCC 4.7.2 GCC 4.6.3 20 40 60 80 100 SE +/- 0.42, N = 3 SE +/- 0.38, N = 3 84.30 88.41 1. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread
C-Ray Total Time OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time GCC 4.7.2 GCC 4.6.3 100 200 300 400 500 SE +/- 16.52, N = 6 SE +/- 7.03, N = 6 348.61 447.38 1. (CC) gcc options: -lm -lpthread -O3
Timed PHP Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed PHP Compilation 5.2.9 Time To Compile GCC 4.6.3 GCC 4.7.2 60 120 180 240 300 SE +/- 1.85, N = 3 SE +/- 0.26, N = 3 255.79 284.15 1. (CC) gcc options: -O2 -pedantic -ldl -lz -lm
7-Zip Compression Compress Speed Test OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 9.20.1 Compress Speed Test GCC 4.6.3 GCC 4.7.2 400 800 1200 1600 2000 SE +/- 6.69, N = 3 SE +/- 13.53, N = 3 1951 1703 1. (CXX) g++ options: -pipe -lpthread
x264 H.264 Video Encoding OpenBenchmarking.org Frames Per Second, More Is Better x264 2012-10-03 H.264 Video Encoding GCC 4.7.2 GCC 4.6.3 3 6 9 12 15 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 10.83 10.62
VP8 libvpx Encoding vpxenc OpenBenchmarking.org Frames Per Second, More Is Better VP8 libvpx Encoding 1.1.0 vpxenc GCC 4.7.2 GCC 4.6.3 0.2903 0.5806 0.8709 1.1612 1.4515 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 1.29 1.21 1. (CC) gcc options: -lvpx -lm -lpthread
TSCP AI Chess Performance OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance GCC 4.6.3 GCC 4.7.2 30K 60K 90K 120K 150K SE +/- 150.74, N = 5 SE +/- 233.82, N = 5 146968 146096
SciMark Computational Test: Jacobi Successive Over-Relaxation OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation GCC 4.6.3 GCC 4.7.2 80 160 240 320 400 SE +/- 0.82, N = 4 SE +/- 1.15, N = 4 361.97 344.10
SciMark Computational Test: Dense LU Matrix Factorization OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Dense LU Matrix Factorization GCC 4.7.2 GCC 4.6.3 100 200 300 400 500 SE +/- 2.60, N = 4 SE +/- 0.92, N = 4 450.12 435.03
SciMark Computational Test: Sparse Matrix Multiply OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Sparse Matrix Multiply GCC 4.7.2 GCC 4.6.3 70 140 210 280 350 SE +/- 0.53, N = 4 SE +/- 2.68, N = 4 338.23 333.34
SciMark Computational Test: Fast Fourier Transform OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Fast Fourier Transform GCC 4.7.2 GCC 4.6.3 10 20 30 40 50 SE +/- 0.29, N = 4 SE +/- 0.82, N = 4 44.62 37.65
SciMark Computational Test: Monte Carlo OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Monte Carlo GCC 4.6.3 GCC 4.7.2 40 80 120 160 200 SE +/- 0.13, N = 4 SE +/- 0.12, N = 4 167.90 160.62
CacheBench Test: Read / Modify / Write OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Read / Modify / Write GCC 4.7.2 GCC 4.6.3 1500 3000 4500 6000 7500 SE +/- 39.96, N = 3 SE +/- 42.99, N = 3 7092.44 6979.64 1. (CC) gcc options: -lrt
CacheBench Test: Write OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Write GCC 4.7.2 GCC 4.6.3 800 1600 2400 3200 4000 SE +/- 17.55, N = 3 SE +/- 6.16, N = 3 3722.67 3521.37 1. (CC) gcc options: -lrt
CacheBench Test: Read OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Read GCC 4.7.2 GCC 4.6.3 600 1200 1800 2400 3000 SE +/- 0.34, N = 3 SE +/- 6.33, N = 3 2582.33 2570.83 1. (CC) gcc options: -lrt
GMPbench Total Time OpenBenchmarking.org GMPbench Score, More Is Better GMPbench 0.2 Total Time GCC 4.7.2 GCC 4.6.3 100 200 300 400 500 452.59 451.29 1. (CC) gcc options: -O3 -fomit-frame-pointer
FFTE Test: N=64, 1D Complex FFT Routine OpenBenchmarking.org MFLOPS, More Is Better FFTE 5.0 Test: N=64, 1D Complex FFT Routine GCC 4.7.2 GCC 4.6.3 300 600 900 1200 1500 SE +/- 2.17, N = 3 SE +/- 4.27, N = 3 1476.59 1257.00 1. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp
Dolfyn Computational Fluid Dynamics OpenBenchmarking.org Seconds, Fewer Is Better Dolfyn 0.527 Computational Fluid Dynamics GCC 4.6.3 GCC 4.7.2 40 80 120 160 200 SE +/- 3.19, N = 6 SE +/- 1.08, N = 3 158.91 171.65
CLOMP Static OMP Speedup OpenBenchmarking.org Speedup, More Is Better CLOMP 3.3 Static OMP Speedup GCC 4.7.2 GCC 4.6.3 0.4073 0.8146 1.2219 1.6292 2.0365 SE +/- 0.01, N = 5 SE +/- 0.06, N = 10 1.81 1.70 1. (CC) gcc options: --openmp -O3 -lm
Phoronix Test Suite v10.8.5