ARM64 gcc codegen comparison

gcc 5.4/6.3/7.0 benchmarks running on a Cortex-A53

HTML result view exported from: https://openbenchmarking.org/result/1701128-TA-GCCCOMPAR79&grt&sor.

ARM64 gcc codegen comparisonProcessorMotherboardMemoryDiskOSKernelCompilerFile-SystemScreen Resolutiongcc5 A57 vectorizegcc5 thunderx vectorizegcc5 A72 LTOgcc6 A53gcc6 A53 mtune/vectorizegcc6 A57 vectorizegcc7 A53 vectorizegcc7 thunderx vectorizegcc7 A53 vectorize LTOAArch64 rev 4 @ 1.55GHz (4 Cores)Amlogic2048MB32GB 00000 + 16GB NCardUbuntu 16.043.14.29 (aarch64)GCC 5.4.0 20160609 + Clang 3.8.0-2ubuntu4 + LLVM 3.8.0ext41920x3240AArch64 rev 4 @ 1.50GHz (4 Cores)AArch64 rev 4 @ 1.55GHz (4 Cores)GCC 6.3.0 + Clang 3.8.0-2ubuntu4 + LLVM 3.8.0AArch64 rev 4 @ 1.50GHz (4 Cores)GCC 7.0.0 20170110 + Clang 3.8.0-2ubuntu4 + LLVM 3.8.0AArch64 rev 4 @ 1.55GHz (4 Cores)OpenBenchmarking.orgCompiler Details- gcc5 A57 vectorize: --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 - gcc5 thunderx vectorize: --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 - gcc5 A72 LTO: --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 - gcc6 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-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 - gcc6 A53 mtune/vectorize: --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-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 - gcc6 A57 vectorize: --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-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 - gcc7 A53 vectorize: --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-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 - gcc7 thunderx vectorize: --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-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 - gcc7 A53 vectorize LTO: --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-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 Disk Details- gcc5 A57 vectorize: DEADLINE / commit=30,errors=remount-ro,noatime,nodiratime,rw- gcc5 thunderx vectorize: DEADLINE / commit=30,errors=remount-ro,noatime,nodiratime,rw- gcc5 A72 LTO: DEADLINE / commit=30,errors=remount-ro,noatime,nodiratime,rw- gcc6 A53: DEADLINE / commit=30,errors=remount-ro,noatime,nodiratime,rw- gcc6 A53 mtune/vectorize: DEADLINE / commit=30,errors=remount-ro,noatime,nodiratime,rw- gcc6 A57 vectorize: DEADLINE / commit=30,errors=remount-ro,noatime,nodiratime,rw- gcc7 A53 vectorize: CFQ / commit=30,errors=remount-ro,noatime,nodiratime,rw- gcc7 thunderx vectorize: CFQ / commit=30,errors=remount-ro,noatime,nodiratime,rw- gcc7 A53 vectorize LTO: CFQ / commit=30,errors=remount-ro,noatime,nodiratime,rwProcessor Details- Scaling Governor: meson_cpufreq performance

ARM64 gcc codegen comparisonc-ray: Total Timefftw: Stock - 2D FFT Size 2048fhourstones: Complex Connect-4 Solvinggmpbench: Total Timeopenssl: RSA 4096-bit Performancepostmark: Disk Transaction Performanceprimesieve: 1e12 Prime Number Generationramspeed: Copy - Integerramspeed: Copy - Floating Pointredis: GETsmallpt: Global Illumination Renderer; 100 Samplessudokut: Total Timetachyon: Total Timemafft: Multiple Sequence Alignmentttsiod-renderer: Phong Rendering With Soft-Shadow Mappinggcc5 A57 vectorizegcc5 thunderx vectorizegcc5 A72 LTOgcc6 A53gcc6 A53 mtune/vectorizegcc6 A57 vectorizegcc7 A53 vectorizegcc7 thunderx vectorizegcc7 A53 vectorize LTO150.16189.293052.83554.3121.301361574.614614.264613.36311665.33173102.0579.4734.0122.48152.96193.883045.83554.4421.201351591.804472.854497.17303529.91171103.6381.4934.7821.86223.05185.553048.70549.9621.201363604.682916.642917.54305506.49172102.6582.0334.6222.57200.00172.733129.50554.9421.301378610.664847.384844.14315587.06169101.9971.6536.1022.01199.00175.093125.77553.0221.231378592.004812.664809.44317672.24169101.6171.8234.2321.99144.49164.613123.23555.0521.231356571.714621.044624.88324752.05168101.9976.9434.9421.71187.97196.903212.10552.8421.501363543.164581.324580.39310344.73167101.9569.2735.4223.16149.82190.633210.20554.8321.501351566.212821.432817.45318926.02167102.7571.4134.4623.01184.81180.533213.77554.3721.501378540.954829.914825.13311785.02168101.7567.6433.1623.77OpenBenchmarking.org

C-Ray

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Timegcc6 A57 vectorizegcc7 thunderx vectorizegcc5 A57 vectorizegcc5 thunderx vectorizegcc7 A53 vectorize LTOgcc7 A53 vectorizegcc6 A53 mtune/vectorizegcc6 A53gcc5 A72 LTO50100150200250SE +/- 0.08, N = 3SE +/- 1.37, N = 3SE +/- 0.19, N = 3SE +/- 2.40, N = 6SE +/- 0.17, N = 3SE +/- 0.69, N = 3SE +/- 1.80, N = 3SE +/- 1.39, N = 3SE +/- 0.09, N = 3144.49149.82150.16152.96184.81187.97199.00200.00223.05-Ofast -mcpu=cortex-a57 -ftree-vectorize-Ofast -mcpu=thunderx -ftree-vectorize-Ofast -mcpu=cortex-a57 -ftree-vectorize-Ofast -mcpu=thunderx -ftree-vectorize-mcpu=cortex-a53 -ftree-vectorize -flto -ffat-lto-objects-Ofast -mcpu=cortex-a53 -ftree-vectorize-mtune=cortex-a53 -ftree-vectorize-mcpu=cortex-a53-Ofast -mcpu=cortex-a72 -flto -ffat-lto-objects -fuse-linker-plugin1. (CC) gcc options: -lm -lpthread -O3 -fomit-frame-pointer -fipa-pta -march=armv8-a+crc

FFTW

Build: Stock - Size: 2D FFT Size 2048

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.4Build: Stock - Size: 2D FFT Size 2048gcc7 A53 vectorizegcc5 thunderx vectorizegcc7 thunderx vectorizegcc5 A57 vectorizegcc5 A72 LTOgcc7 A53 vectorize LTOgcc6 A53 mtune/vectorizegcc6 A53gcc6 A57 vectorize4080120160200SE +/- 0.99, N = 5SE +/- 0.04, N = 5SE +/- 1.10, N = 5SE +/- 0.26, N = 5SE +/- 0.18, N = 5SE +/- 0.49, N = 5SE +/- 0.28, N = 5SE +/- 0.29, N = 5SE +/- 0.17, N = 5196.90193.88190.63189.29185.55180.53175.09172.73164.61-Ofast -mcpu=cortex-a53 -ftree-vectorize-Ofast -mcpu=thunderx -ftree-vectorize-Ofast -mcpu=thunderx -ftree-vectorize-Ofast -mcpu=cortex-a57 -ftree-vectorize-Ofast -mcpu=cortex-a72 -flto -ffat-lto-objects -fuse-linker-plugin-O3 -mcpu=cortex-a53 -ftree-vectorize -flto -ffat-lto-objects-O3 -mtune=cortex-a53 -ftree-vectorize-O3 -mcpu=cortex-a53-Ofast -mcpu=cortex-a57 -ftree-vectorize1. (CC) gcc options: -fomit-frame-pointer -fipa-pta -march=armv8-a+crc -lm

Fhourstones

Complex Connect-4 Solving

OpenBenchmarking.orgKpos / sec, More Is BetterFhourstones 3.1Complex Connect-4 Solvinggcc7 A53 vectorize LTOgcc7 A53 vectorizegcc7 thunderx vectorizegcc6 A53gcc6 A53 mtune/vectorizegcc6 A57 vectorizegcc5 A57 vectorizegcc5 A72 LTOgcc5 thunderx vectorize7001400210028003500SE +/- 0.22, N = 3SE +/- 0.35, N = 3SE +/- 0.76, N = 3SE +/- 1.20, N = 3SE +/- 3.94, N = 3SE +/- 2.96, N = 3SE +/- 1.51, N = 3SE +/- 1.26, N = 3SE +/- 0.86, N = 33213.773212.103210.203129.503125.773123.233052.833048.703045.831. (CC) gcc options: -O3

GMPbench

Total Time

OpenBenchmarking.orgGMPbench Score, More Is BetterGMPbench 0.2Total Timegcc6 A57 vectorizegcc6 A53gcc7 thunderx vectorizegcc5 thunderx vectorizegcc7 A53 vectorize LTOgcc5 A57 vectorizegcc6 A53 mtune/vectorizegcc7 A53 vectorizegcc5 A72 LTO120240360480600555.05554.94554.83554.44554.37554.31553.02552.84549.96-Ofast -mcpu=cortex-a57 -ftree-vectorize-O3 -mcpu=cortex-a53-Ofast -mcpu=thunderx -ftree-vectorize-Ofast -mcpu=thunderx -ftree-vectorize-O3 -mcpu=cortex-a53 -ftree-vectorize -flto -ffat-lto-objects-Ofast -mcpu=cortex-a57 -ftree-vectorize-O3 -mtune=cortex-a53 -ftree-vectorize-Ofast -mcpu=cortex-a53 -ftree-vectorize-Ofast -mcpu=cortex-a72 -flto -ffat-lto-objects -fuse-linker-plugin1. (CC) gcc options: -fomit-frame-pointer -fipa-pta -march=armv8-a+crc -lm

OpenSSL

RSA 4096-bit Performance

OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.0.1gRSA 4096-bit Performancegcc7 A53 vectorize LTOgcc7 thunderx vectorizegcc7 A53 vectorizegcc6 A53gcc5 A57 vectorizegcc6 A57 vectorizegcc6 A53 mtune/vectorizegcc5 A72 LTOgcc5 thunderx vectorize510152025SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 321.5021.5021.5021.3021.3021.2321.2321.2021.201. (CC) gcc options: -O3 -fomit-frame-pointer -lssl -lcrypto -ldl

PostMark

Disk Transaction Performance

OpenBenchmarking.orgTPS, More Is BetterPostMark 1.51Disk Transaction Performancegcc7 A53 vectorize LTOgcc6 A53 mtune/vectorizegcc6 A53gcc7 A53 vectorizegcc5 A72 LTOgcc5 A57 vectorizegcc6 A57 vectorizegcc7 thunderx vectorizegcc5 thunderx vectorize30060090012001500SE +/- 2.67, N = 3SE +/- 2.67, N = 3SE +/- 2.67, N = 3SE +/- 2.67, N = 3SE +/- 2.67, N = 3SE +/- 6.49, N = 3SE +/- 2.33, N = 3SE +/- 0.00, N = 3SE +/- 4.04, N = 31378137813781363136313611356135113511. (CC) gcc options: -O3

Primesieve

1e12 Prime Number Generation

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 5.4.21e12 Prime Number Generationgcc7 A53 vectorize LTOgcc7 A53 vectorizegcc7 thunderx vectorizegcc6 A57 vectorizegcc5 A57 vectorizegcc5 thunderx vectorizegcc6 A53 mtune/vectorizegcc5 A72 LTOgcc6 A53130260390520650SE +/- 8.42, N = 3SE +/- 3.01, N = 3SE +/- 2.99, N = 3SE +/- 4.25, N = 3SE +/- 8.11, N = 6SE +/- 10.05, N = 6SE +/- 17.98, N = 6SE +/- 9.47, N = 4SE +/- 15.34, N = 6540.95543.16566.21571.71574.61591.80592.00604.68610.66-O3 -mcpu=cortex-a53 -ftree-vectorize -flto -ffat-lto-objects-Ofast -mcpu=cortex-a53 -ftree-vectorize-Ofast -mcpu=thunderx -ftree-vectorize-Ofast -mcpu=cortex-a57 -ftree-vectorize-Ofast -mcpu=cortex-a57 -ftree-vectorize-Ofast -mcpu=thunderx -ftree-vectorize-O3 -mtune=cortex-a53 -ftree-vectorize-Ofast -mcpu=cortex-a72 -flto -ffat-lto-objects -fuse-linker-plugin-O3 -mcpu=cortex-a531. (CXX) g++ options: -fomit-frame-pointer -fipa-pta -march=armv8-a+crc -fopenmp

RAMspeed SMP

Type: Copy - Benchmark: Integer

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Copy - Benchmark: Integergcc6 A53gcc7 A53 vectorize LTOgcc6 A53 mtune/vectorizegcc6 A57 vectorizegcc5 A57 vectorizegcc7 A53 vectorizegcc5 thunderx vectorizegcc5 A72 LTOgcc7 thunderx vectorize100020003000400050004847.384829.914812.664621.044614.264581.324472.852916.642821.43

RAMspeed SMP

Type: Copy - Benchmark: Floating Point

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Copy - Benchmark: Floating Pointgcc6 A53gcc7 A53 vectorize LTOgcc6 A53 mtune/vectorizegcc6 A57 vectorizegcc5 A57 vectorizegcc7 A53 vectorizegcc5 thunderx vectorizegcc5 A72 LTOgcc7 thunderx vectorize100020003000400050004844.144825.134809.444624.884613.364580.394497.172917.542817.45

Redis

Test: GET

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 3.0.1Test: GETgcc6 A57 vectorizegcc7 thunderx vectorizegcc6 A53 mtune/vectorizegcc6 A53gcc7 A53 vectorize LTOgcc5 A57 vectorizegcc7 A53 vectorizegcc5 A72 LTOgcc5 thunderx vectorize70K140K210K280K350KSE +/- 1024.45, N = 3SE +/- 2784.59, N = 3SE +/- 1273.24, N = 3SE +/- 3145.13, N = 3SE +/- 2239.53, N = 3SE +/- 1214.02, N = 3SE +/- 4662.92, N = 6SE +/- 3249.19, N = 3SE +/- 5780.31, N = 3324752.05318926.02317672.24315587.06311785.02311665.33310344.73305506.49303529.91-Ofast -mcpu=cortex-a57 -ftree-vectorize-Ofast -mcpu=thunderx -ftree-vectorize-O3 -mtune=cortex-a53 -ftree-vectorize-O3 -mcpu=cortex-a53-O3 -mcpu=cortex-a53 -ftree-vectorize -flto -ffat-lto-objects-Ofast -mcpu=cortex-a57 -ftree-vectorize-Ofast -mcpu=cortex-a53 -ftree-vectorize-Ofast -mcpu=cortex-a72 -flto -ffat-lto-objects -fuse-linker-plugin-Ofast -mcpu=thunderx -ftree-vectorize1. (CC) gcc options: -ggdb -rdynamic -lm -pthread -ldl -O2 -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 Samplesgcc7 A53 vectorizegcc7 thunderx vectorizegcc6 A57 vectorizegcc7 A53 vectorize LTOgcc6 A53gcc6 A53 mtune/vectorizegcc5 thunderx vectorizegcc5 A72 LTOgcc5 A57 vectorize4080120160200SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3167167168168169169171172173-Ofast -mcpu=cortex-a53 -ftree-vectorize-Ofast -mcpu=thunderx -ftree-vectorize-Ofast -mcpu=cortex-a57 -ftree-vectorize-O3 -mcpu=cortex-a53 -ftree-vectorize -flto -ffat-lto-objects-O3 -mcpu=cortex-a53-O3 -mtune=cortex-a53 -ftree-vectorize-Ofast -mcpu=thunderx -ftree-vectorize-Ofast -mcpu=cortex-a72 -flto -ffat-lto-objects -fuse-linker-plugin-Ofast -mcpu=cortex-a57 -ftree-vectorize1. (CXX) g++ options: -fopenmp -fomit-frame-pointer -fipa-pta -march=armv8-a+crc

Sudokut

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterSudokut 0.4Total Timegcc6 A53 mtune/vectorizegcc7 A53 vectorize LTOgcc7 A53 vectorizegcc6 A53gcc6 A57 vectorizegcc5 A57 vectorizegcc5 A72 LTOgcc7 thunderx vectorizegcc5 thunderx vectorize20406080100SE +/- 0.05, N = 3SE +/- 0.21, N = 3SE +/- 0.20, N = 3SE +/- 0.23, N = 3SE +/- 0.17, N = 3SE +/- 0.11, N = 3SE +/- 0.73, N = 3SE +/- 0.76, N = 3SE +/- 0.78, N = 3101.61101.75101.95101.99101.99102.05102.65102.75103.63

Tachyon

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterTachyon 0.98.9Total Timegcc7 A53 vectorize LTOgcc7 A53 vectorizegcc7 thunderx vectorizegcc6 A53gcc6 A53 mtune/vectorizegcc6 A57 vectorizegcc5 A57 vectorizegcc5 thunderx vectorizegcc5 A72 LTO20406080100SE +/- 0.11, N = 3SE +/- 0.08, N = 3SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 0.24, N = 3SE +/- 0.20, N = 3SE +/- 0.09, N = 3SE +/- 0.53, N = 3SE +/- 0.09, N = 367.6469.2771.4171.6571.8276.9479.4781.4982.03

Timed MAFFT Alignment

Multiple Sequence Alignment

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MAFFT Alignment 6.864Multiple Sequence Alignmentgcc7 A53 vectorize LTOgcc5 A57 vectorizegcc6 A53 mtune/vectorizegcc7 thunderx vectorizegcc5 A72 LTOgcc5 thunderx vectorizegcc6 A57 vectorizegcc7 A53 vectorizegcc6 A53816243240SE +/- 0.70, N = 6SE +/- 0.54, N = 6SE +/- 0.48, N = 6SE +/- 0.73, N = 6SE +/- 0.01, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.80, N = 6SE +/- 0.61, N = 633.1634.0134.2334.4634.6234.7834.9435.4236.101. (CC) gcc options: -O3 -lm -lpthread

TTSIOD 3D Renderer

Phong Rendering With Soft-Shadow Mapping

OpenBenchmarking.orgFPS, More Is BetterTTSIOD 3D Renderer 2.3aPhong Rendering With Soft-Shadow Mappinggcc7 A53 vectorize LTOgcc7 A53 vectorizegcc7 thunderx vectorizegcc5 A72 LTOgcc5 A57 vectorizegcc6 A53gcc6 A53 mtune/vectorizegcc5 thunderx vectorizegcc6 A57 vectorize612182430SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.13, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 323.7723.1623.0122.5722.4822.0121.9921.8621.71-O3 -mcpu=cortex-a53 -ftree-vectorize -ffat-lto-objects-Ofast -mcpu=cortex-a53 -ftree-vectorize-Ofast -mcpu=thunderx -ftree-vectorize-Ofast -mcpu=cortex-a72 -ffat-lto-objects -fuse-linker-plugin-Ofast -mcpu=cortex-a57 -ftree-vectorize-O3 -mcpu=cortex-a53-O3 -mtune=cortex-a53 -ftree-vectorize-Ofast -mcpu=thunderx -ftree-vectorize-Ofast -mcpu=cortex-a57 -ftree-vectorize1. (CXX) g++ options: -fomit-frame-pointer -fipa-pta -march=armv8-a+crc -flto -ffast-math -mtune=native -lSDL -lstdc++

Phoronix Test Suite v10.8.4