Cortex A53 running different codegen options

AARCH64 codegen comparison

HTML result view exported from: https://openbenchmarking.org/result/1609182-LO-1609128HA85&sor&grs.

Cortex A53 running different codegen optionsProcessorMemoryDiskOSKernelCompilerFile-SystemScreen ResolutionA53 codegenA73 codegenmtune A53mtune A53 GCC7Unknown @ 1.50GHz (4 Cores)2048MB16GB NCard + 32GB 00000Ubuntu 16.043.14.65-61 (aarch64)GCC 7.0.0 20160904 + LLVM 3.8.0ext41280x1440GCC 5.3.1 20160413 + LLVM 3.8.0GCC 7.0.0 20160915 + LLVM 3.8.0OpenBenchmarking.orgCompiler Details- A53 codegen: --build=aarch64-linux-gnu --disable-bootstrap --disable-browser-plugin --disable-libquadmath --disable-werror --enable-checking=release --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,c++,fortran --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --target=aarch64-linux-gnu --with-arch-directory=aarch64 --with-default-libstdcxx-abi=new - A73 codegen: --build=aarch64-linux-gnu --disable-bootstrap --disable-browser-plugin --disable-libquadmath --disable-werror --enable-checking=release --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,c++,fortran --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --target=aarch64-linux-gnu --with-arch-directory=aarch64 --with-default-libstdcxx-abi=new - mtune A53: --build=aarch64-linux-gnu --disable-browser-plugin --disable-libquadmath --disable-werror --enable-checking=release --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-gtk-cairo --enable-java-awt=gtk --enable-java-home --enable-languages=c,ada,c++,java,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --target=aarch64-linux-gnu --with-arch-directory=aarch64 --with-default-libstdcxx-abi=new -v - mtune A53 GCC7: --build=aarch64-linux-gnu --disable-bootstrap --disable-browser-plugin --disable-libquadmath --disable-werror --enable-checking=release --enable-clocale=gnu --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,c++,fortran --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --target=aarch64-linux-gnu --with-arch-directory=aarch64 --with-default-libstdcxx-abi=new Processor Details- Scaling Governor: meson_cpufreq performance

Cortex A53 running different codegen optionsbuild-apache: Time To Compilec-ray: Total Timestockfish: Total Timefftw: Stock - 2D FFT Size 2048encode-flac: WAV To FLACvpxenc: vpxencredis: GETn-queens: Elapsed Timesmallpt: Global Illumination Renderer; 100 Samplesredis: SETfhourstones: Complex Connect-4 Solvingffmpeg: H.264 HD To NTSC DVpgbench: Buffer Test - Single Thread - Read Writepgbench: Buffer Test - Normal Load - Read Writeprimesieve: 1e12 Prime Number GenerationA53 codegenA73 codegenmtune A53mtune A53 GCC7245.49192.7020504196.93153.486.81264703.30156.09169194408.303136.27191.79144.95598.25575.31248.20184.8721094187.98162.417.06270145.88157.93173198057.423140.63192.25204.69584.80579.67390.57240.3422423179.22154.837.15252844.60158.04176191749.733045.83193.89205.87599.02556.08376.56187.3020372190.17149.157.36261242.76149.22169192951.043144.40191.80216.23657.04567.25OpenBenchmarking.org

Timed Apache Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Apache Compilation 2.4.7Time To CompileA53 codegenA73 codegenmtune A53 GCC7mtune A5380160240320400SE +/- 0.03, N = 3SE +/- 0.29, N = 3SE +/- 0.26, N = 3SE +/- 0.45, N = 3245.49248.20376.56390.57

C-Ray

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total TimeA73 codegenmtune A53 GCC7A53 codegenmtune A5350100150200250SE +/- 1.45, N = 3SE +/- 0.97, N = 3SE +/- 0.20, N = 3SE +/- 1.93, N = 3184.87187.30192.70240.34-mcpu=cortex-a73-mtune=cortex-a53 -flto -ffat-lto-objects-mcpu=cortex-a53-mtune=cortex-a53 -flto -ffat-lto-objects1. (CC) gcc options: -lm -lpthread -O3 -fomit-frame-pointer -fipa-pta -march=armv8-a+crc

Stockfish

Total Time

OpenBenchmarking.orgms, Fewer Is BetterStockfish 2014-11-26Total Timemtune A53 GCC7A53 codegenA73 codegenmtune A535K10K15K20K25KSE +/- 84.05, N = 3SE +/- 78.03, N = 3SE +/- 120.77, N = 3SE +/- 54.96, N = 320372205042109422423-mtune=cortex-a53 -ffat-lto-objects-mcpu=cortex-a53-mcpu=cortex-a73-mtune=cortex-a53 -ffat-lto-objects1. (CXX) g++ options: -lpthread -O3 -fomit-frame-pointer -fipa-pta -march=armv8-a+crc -flto -fno-exceptions -fno-rtti -ansi -pedantic

FFTW

Build: Stock - Size: 2D FFT Size 2048

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.4Build: Stock - Size: 2D FFT Size 2048A53 codegenmtune A53 GCC7A73 codegenmtune A534080120160200SE +/- 0.13, N = 5SE +/- 0.16, N = 5SE +/- 0.05, N = 5SE +/- 0.09, N = 5196.93190.17187.98179.22-mcpu=cortex-a53-mtune=cortex-a53 -flto -ffat-lto-objects-mcpu=cortex-a73-mtune=cortex-a53 -flto -ffat-lto-objects1. (CC) gcc options: -O3 -fomit-frame-pointer -fipa-pta -march=armv8-a+crc -lm

FLAC Audio Encoding

WAV To FLAC

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.1WAV To FLACmtune A53 GCC7A53 codegenmtune A53A73 codegen4080120160200SE +/- 0.06, N = 5SE +/- 0.10, N = 5SE +/- 0.07, N = 5SE +/- 0.12, N = 5149.15153.48154.83162.41-mtune=cortex-a53 -flto -ffat-lto-objects-mcpu=cortex-a53-mtune=cortex-a53 -flto -ffat-lto-objects-mcpu=cortex-a731. (CXX) g++ options: -O3 -fomit-frame-pointer -fipa-pta -march=armv8-a+crc -fvisibility=hidden -lm

VP8 libvpx Encoding

vpxenc

OpenBenchmarking.orgFrames Per Second, More Is BetterVP8 libvpx Encoding 1.3.0vpxencmtune A53 GCC7mtune A53A73 codegenA53 codegen246810SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.09, N = 37.367.157.066.81-mtune=cortex-a53 -flto -ffat-lto-objects-mtune=cortex-a53 -flto -ffat-lto-objects-mcpu=cortex-a73-mcpu=cortex-a531. (CXX) g++ options: -lvpx -lgtest -lpthread -lm -O3 -fomit-frame-pointer -fipa-pta -march=armv8-a+crc

Redis

Test: GET

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 3.0.1Test: GETA73 codegenA53 codegenmtune A53 GCC7mtune A5360K120K180K240K300KSE +/- 3076.94, N = 3SE +/- 1308.37, N = 3SE +/- 1125.65, N = 3SE +/- 110.73, N = 3270145.88264703.30261242.76252844.60-mcpu=cortex-a73-mcpu=cortex-a53-mtune=cortex-a53 -flto -ffat-lto-objects-mtune=cortex-a53 -flto -ffat-lto-objects1. (CC) gcc options: -ggdb -rdynamic -lm -pthread -ldl -O2 -O3 -fomit-frame-pointer -fipa-pta -march=armv8-a+crc

N-Queens

Elapsed Time

OpenBenchmarking.orgSeconds, Fewer Is BetterN-Queens 1.0Elapsed Timemtune A53 GCC7A53 codegenA73 codegenmtune A53306090120150SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3149.22156.09157.93158.04-mcpu=cortex-a53-mcpu=cortex-a731. (CC) gcc options: -static -fopenmp -O3 -fomit-frame-pointer -fipa-pta -march=armv8-a+crc

Smallpt

Global Illumination Renderer; 100 Samples

OpenBenchmarking.orgSeconds, Fewer Is BetterSmallpt 1.0Global Illumination Renderer; 100 SamplesA53 codegenmtune A53 GCC7A73 codegenmtune A534080120160200SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3169169173176-mcpu=cortex-a53-mtune=cortex-a53 -flto -ffat-lto-objects-mcpu=cortex-a73-mtune=cortex-a53 -flto -ffat-lto-objects1. (CXX) g++ options: -fopenmp -O3 -fomit-frame-pointer -fipa-pta -march=armv8-a+crc

Redis

Test: SET

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 3.0.1Test: SETA73 codegenA53 codegenmtune A53 GCC7mtune A5340K80K120K160K200KSE +/- 1065.37, N = 3SE +/- 826.93, N = 3SE +/- 1104.96, N = 3SE +/- 845.07, N = 3198057.42194408.30192951.04191749.73-mcpu=cortex-a73-mcpu=cortex-a53-mtune=cortex-a53 -flto -ffat-lto-objects-mtune=cortex-a53 -flto -ffat-lto-objects1. (CC) gcc options: -ggdb -rdynamic -lm -pthread -ldl -O2 -O3 -fomit-frame-pointer -fipa-pta -march=armv8-a+crc

Fhourstones

Complex Connect-4 Solving

OpenBenchmarking.orgKpos / sec, More Is BetterFhourstones 3.1Complex Connect-4 Solvingmtune A53 GCC7A73 codegenA53 codegenmtune A537001400210028003500SE +/- 3.60, N = 3SE +/- 1.36, N = 3SE +/- 2.43, N = 3SE +/- 1.05, N = 33144.403140.633136.273045.831. (CC) gcc options: -O3

FFmpeg

H.264 HD To NTSC DV

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 2.6.2H.264 HD To NTSC DVA53 codegenmtune A53 GCC7A73 codegenmtune A534080120160200SE +/- 0.90, N = 3SE +/- 1.30, N = 3SE +/- 0.96, N = 3SE +/- 0.53, N = 3191.79191.80192.25193.89-mcpu=cortex-a53-lSDL -mtune=cortex-a53 -flto -ffat-lto-objects-lSDL -mcpu=cortex-a73-lSDL -mtune=cortex-a53 -flto -ffat-lto-objects1. (CC) gcc options: -lavdevice -lavfilter -lavformat -lavcodec -lswresample -lswscale -lavutil -lxcb -lxcb-shm -lX11 -lm -pthread -O3 -fomit-frame-pointer -fipa-pta -march=armv8-a+crc -std=c99 -fno-math-errno -fno-signed-zeros -fno-tree-vectorize -MMD -MF -MT

PostgreSQL pgbench

Scaling: Buffer Test - Test: Single Thread - Mode: Read Write

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 9.4.3Scaling: Buffer Test - Test: Single Thread - Mode: Read Writemtune A53 GCC7mtune A53A73 codegenA53 codegen50100150200250SE +/- 5.47, N = 6SE +/- 7.11, N = 6SE +/- 5.82, N = 6SE +/- 6.47, N = 6216.23205.87204.69144.95-mtune=cortex-a53 -flto -ffat-lto-objects-mtune=cortex-a53 -flto -ffat-lto-objects-mcpu=cortex-a72-mcpu=cortex-a531. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -fomit-frame-pointer -fipa-pta -march=armv8-a+crc -pthread -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

PostgreSQL pgbench

Scaling: Buffer Test - Test: Normal Load - Mode: Read Write

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 9.4.3Scaling: Buffer Test - Test: Normal Load - Mode: Read Writemtune A53 GCC7mtune A53A53 codegenA73 codegen140280420560700SE +/- 73.52, N = 6SE +/- 78.60, N = 6SE +/- 74.34, N = 6SE +/- 75.55, N = 6657.04599.02598.25584.80-mtune=cortex-a53 -flto -ffat-lto-objects-mtune=cortex-a53 -flto -ffat-lto-objects-mcpu=cortex-a53-mcpu=cortex-a721. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -fomit-frame-pointer -fipa-pta -march=armv8-a+crc -pthread -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

Primesieve

1e12 Prime Number Generation

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 5.4.21e12 Prime Number Generationmtune A53mtune A53 GCC7A53 codegenA73 codegen130260390520650SE +/- 9.30, N = 3SE +/- 0.68, N = 3SE +/- 9.24, N = 4SE +/- 16.21, N = 6556.08567.25575.31579.67-mtune=cortex-a53 -flto -ffat-lto-objects-mtune=cortex-a53 -flto -ffat-lto-objects-mcpu=cortex-a53-mcpu=cortex-a731. (CXX) g++ options: -O3 -fomit-frame-pointer -fipa-pta -march=armv8-a+crc -fopenmp

Phoronix Test Suite v10.8.4