GCC 6.1 Compiler Optimization Benchmarks GCC 6.1.0 compiler benchmarks with different optimization flags. Intel Xeon E5-2687W v3 GCC compiler benchmarks on Debian. Tests by Michael Larabel of Phoronix for a future article.
HTML result view exported from: https://openbenchmarking.org/result/1605151-GA-1605083HA39&sor&grs .
GCC 6.1 Compiler Optimization Benchmarks Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Display Server Display Driver OpenGL Compiler File-System Screen Resolution Desktop -O0 -Os -Og -O1 -O2 -O3 -O3 -march=native -O3 -march=native -flto -Ofast -march=native s10 Intel Xeon E5-2687W v3 @ 3.50GHz (20 Cores) MSI X99S SLI PLUS (MS-7885) v1.0 Intel Xeon E7 v3/Xeon 16384MB PNY CS1211 120GB + 80GB INTEL SSDSCKGW08 AMD FirePro V7900 2048MB Realtek ALC892 ASUS PB278 Intel Connection Debian testing 4.5.0-1-amd64 (x86_64) X Server 1.18.3 modesetting 1.18.3 3.3 Mesa 11.1.3 Gallium 0.4 GCC 6.1.0 ext4 2560x1440 Intel Xeon E31245 @ 3.70GHz (8 Cores) ASUS P8B WS 3001GB Hitachi HDS72303 + 128GB SAMSUNG MZNTE128 Intel Sandybridge Server (1350MHz) Realtek Generic SyncMaster Gentoo 2.2 4.5.0-gentoo (x86_64) KDE Frameworks 5 intel 2.99.917 3.3 Mesa 11.2.2 GCC 5.3.0 + Clang 3.8.0 + LLVM 3.8.0 1920x1080 OpenBenchmarking.org Compiler Details - -O0: --disable-multilib --enable-checking=release - -Os: --disable-multilib --enable-checking=release - -Og: --disable-multilib --enable-checking=release - -O1: --disable-multilib --enable-checking=release - -O2: --disable-multilib --enable-checking=release - -O3: --disable-multilib --enable-checking=release - -O3 -march=native: --disable-multilib --enable-checking=release - -O3 -march=native -flto: --disable-multilib --enable-checking=release - -Ofast -march=native: --disable-multilib --enable-checking=release - s10: --bindir=/usr/x86_64-pc-linux-gnu/gcc-bin/5.3.0 --build=x86_64-pc-linux-gnu --datadir=/usr/share/gcc-data/x86_64-pc-linux-gnu/5.3.0 --disable-altivec --disable-fixed-point --disable-libcilkrts --disable-libmpx --disable-libmudflap --disable-libssp --disable-werror --enable-__cxa_atexit --enable-checking=release --enable-clocale=gnu --enable-languages=c,c++,java,objc,fortran --enable-libgomp --enable-libsanitizer --enable-libstdcxx-time --enable-libvtv --enable-lto --enable-multilib --enable-nls --enable-obsolete --enable-secureplt --enable-shared --enable-targets=all --enable-threads=posix --enable-vtable-verify --host=x86_64-pc-linux-gnu --includedir=/usr/lib/gcc/x86_64-pc-linux-gnu/5.3.0/include --mandir=/usr/share/gcc-data/x86_64-pc-linux-gnu/5.3.0/man --with-multilib-list=m32,m64 --with-python-dir=/share/gcc-data/x86_64-pc-linux-gnu/5.3.0/python --without-isl Processor Details - Scaling Governor: intel_pstate powersave
GCC 6.1 Compiler Optimization Benchmarks encode-flac: WAV To FLAC pgbench: Buffer Test - Normal Load - Read Write pgbench: Buffer Test - Heavy Contention - Read Write himeno: Poisson Pressure Solver graphics-magick: Local Adaptive Thresholding pgbench: Buffer Test - Single Thread - Read Write encode-mp3: WAV To MP3 hint: FLOAT redis: GET redis: LPOP redis: SADD redis: LPUSH redis: SET graphics-magick: Sharpen scimark2: Fast Fourier Transform build-php: Time To Compile build-imagemagick: Time To Compile graphics-magick: Resizing graphics-magick: HWB Color Space graphics-magick: Blur scimark2: Sparse Matrix Multiply scimark2: Composite scimark2: Dense LU Matrix Factorization scimark2: Monte Carlo scimark2: Jacobi Successive Over-Relaxation hmmer: Pfam Database Search -O0 -Os -Og -O1 -O2 -O3 -O3 -march=native -O3 -march=native -flto -Ofast -march=native s10 46.74 4468.97 4840.76 424.57 17 303.93 36.02 103731655.85 548655.63 547091.88 491891.31 476295.97 479934.92 71 440.57 5.58 9.34 97 110 82 2565.94 1407.02 2454.12 545.51 1028.95 13.82 10.62 4275.85 4497.15 1181.18 68 346.32 16.28 303914359.33 645755.96 642758.08 607478.25 598808.52 596757.10 124 456.35 11.61 32.43 168 188 110 2589.56 1426.79 2482.13 552.04 1053.88 10.68 8.11 4364.38 4538.18 1102.64 54 353.15 17.15 326497871.93 652904.39 637156.51 582272.77 602047.96 592312.44 100 456.61 8.18 13.35 149 168 113 2580.32 1427.70 2521.75 546.26 1033.56 8.23 7.68 4257.86 4494.38 1060.93 76 351.47 15.14 242450705.97 655681.23 655900.13 605722.93 589230.57 597265.48 135 447.31 9.76 27.24 168 187 137 2609.45 1437.73 2534.08 551.44 1046.36 10.19 6.68 4322.67 4494.08 1916.56 82 363.87 14.26 317711776.83 628643.96 649030.83 598759.46 599526.56 586099.29 134 461.55 16.08 38.55 174 186 131 2571.06 1426.80 2534.23 537.90 1029.27 11.63 6.83 4495.93 4720.08 1895.45 83 351.89 12.52 312279718.27 669846.73 646935.48 605861.48 593409.02 587251.75 136 458.56 17.59 55.45 171 185 130 2622.50 1442.30 2531.29 547.85 1051.27 13.08 7.01 4281.30 4539.62 2113.04 85 349.97 12.45 310268777.87 631189.52 655097.69 615258.45 584299.37 584905.04 143 443.71 18.10 55.40 180 190 138 2440.96 1388.10 2468.30 547.58 1039.94 13.04 2150.96 312975471.93 465.50 82.86 121.45 2511.39 1445.10 2586.62 614.59 1047.43 7.03 2019.61 86 11.34 309403432.89 631191.87 656696.79 616016.48 598935.85 588019.67 145 468.61 17.99 55.89 182 204 144 2517.37 1421.80 2519.72 553.28 1050.05 8.32 8.97 683.11 791.05 1474.95 78 91.81 14.45 287050317.81 1570800.87 1559720.21 1216564.21 1169238.33 1149452.71 103 231.80 34.49 74.78 151 159 125 1544.66 1023.06 1835.64 464.45 1038.75 15.29 OpenBenchmarking.org
FLAC Audio Encoding WAV To FLAC OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.3.1 WAV To FLAC -O2 -O3 -O3 -march=native -Ofast -march=native -O1 -Og s10 -Os -O0 11 22 33 44 55 SE +/- 0.07, N = 5 SE +/- 0.04, N = 5 SE +/- 0.10, N = 5 SE +/- 0.04, N = 5 SE +/- 0.04, N = 5 SE +/- 0.07, N = 5 SE +/- 0.02, N = 5 SE +/- 0.05, N = 5 SE +/- 0.12, N = 5 6.68 6.83 7.01 7.03 7.68 8.11 8.97 10.62 46.74 -O2 -O3 -O3 -march=native -Ofast -march=native -O1 -Og -O2 -logg -Os -O0 1. (CXX) g++ options: -fvisibility=hidden -lm
PostgreSQL pgbench Scaling: Buffer Test - Test: Normal Load - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 9.4.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write -O3 -O0 -Og -O2 -O3 -march=native -Os -O1 s10 1000 2000 3000 4000 5000 SE +/- 64.78, N = 6 SE +/- 44.94, N = 3 SE +/- 64.39, N = 5 SE +/- 62.15, N = 5 SE +/- 18.80, N = 3 SE +/- 32.55, N = 3 SE +/- 80.41, N = 6 SE +/- 10.47, N = 6 4495.93 4468.97 4364.38 4322.67 4281.30 4275.85 4257.86 683.11 -O3 -O0 -Og -O2 -O3 -march=native -Os -O1 -O2 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -pthread -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
PostgreSQL pgbench Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 9.4.3 Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Write -O0 -O3 -O3 -march=native -Og -Os -O1 -O2 s10 1000 2000 3000 4000 5000 SE +/- 21.91, N = 3 SE +/- 40.24, N = 3 SE +/- 66.74, N = 3 SE +/- 54.29, N = 3 SE +/- 27.80, N = 3 SE +/- 67.44, N = 4 SE +/- 19.05, N = 3 SE +/- 15.48, N = 6 4840.76 4720.08 4539.62 4538.18 4497.15 4494.38 4494.08 791.05 -O0 -O3 -O3 -march=native -Og -Os -O1 -O2 -O2 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -pthread -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
Himeno Benchmark Poisson Pressure Solver OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver -O3 -march=native -flto -O3 -march=native -Ofast -march=native -O2 -O3 s10 -Os -Og -O1 -O0 500 1000 1500 2000 2500 SE +/- 7.85, N = 3 SE +/- 9.13, N = 3 SE +/- 3.96, N = 3 SE +/- 4.78, N = 3 SE +/- 6.74, N = 3 SE +/- 0.96, N = 3 SE +/- 3.40, N = 3 SE +/- 1.19, N = 3 SE +/- 1.09, N = 3 SE +/- 0.64, N = 3 2150.96 2113.04 2019.61 1916.56 1895.45 1474.95 1181.18 1102.64 1060.93 424.57 -march=native -flto -mavx2 -march=native -mavx2 -Ofast -march=native -mavx2 -O2 -mavx2 -mavx2 -Os -mavx2 -Og -mavx2 -O1 -mavx2 -O0 -mavx2 1. (CC) gcc options: -O3
GraphicsMagick Operation: Local Adaptive Thresholding OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.19 Operation: Local Adaptive Thresholding -Ofast -march=native -O3 -march=native -O3 -O2 s10 -O1 -Os -Og -O0 20 40 60 80 100 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, 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 86 85 83 82 78 76 68 54 17 -Ofast -march=native -ldl -O3 -march=native -ldl -O3 -ldl -O2 -ldl -O2 -ljbig -lwebp -llcms2 -ltiff -lfreetype -ljasper -ljpeg -lwmflite -llzma -lbz2 -lxml2 -lgomp -O1 -ldl -Os -ldl -Og -ldl -O0 -ldl 1. (CC) gcc options: -fopenmp -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread
PostgreSQL pgbench Scaling: Buffer Test - Test: Single Thread - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 9.4.3 Scaling: Buffer Test - Test: Single Thread - Mode: Read Write -O2 -Og -O3 -O1 -O3 -march=native -Os -O0 s10 80 160 240 320 400 SE +/- 2.06, N = 3 SE +/- 5.68, N = 6 SE +/- 2.78, N = 3 SE +/- 5.58, N = 3 SE +/- 1.70, N = 3 SE +/- 4.13, N = 3 SE +/- 4.48, N = 6 SE +/- 0.53, N = 3 363.87 353.15 351.89 351.47 349.97 346.32 303.93 91.81 -O2 -Og -O3 -O1 -O3 -march=native -Os -O0 -O2 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -pthread -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
LAME MP3 Encoding WAV To MP3 OpenBenchmarking.org Seconds, Fewer Is Better LAME MP3 Encoding 3.99.3 WAV To MP3 -Ofast -march=native -O3 -march=native -O3 -O2 s10 -O1 -Os -Og -O0 8 16 24 32 40 SE +/- 0.07, N = 5 SE +/- 0.10, N = 5 SE +/- 0.09, N = 5 SE +/- 0.10, N = 5 SE +/- 0.03, N = 5 SE +/- 0.08, N = 5 SE +/- 0.06, N = 5 SE +/- 0.06, N = 5 SE +/- 0.09, N = 5 11.34 12.45 12.52 14.26 14.45 15.14 16.28 17.15 36.02 -Ofast -march=native -O3 -march=native -O3 -O2 -O3 -ffast-math -funroll-loops -lncurses -O1 -Os -Og -O0 1. (CC) gcc options: -pipe -lm
Hierarchical INTegration Test: FLOAT OpenBenchmarking.org QUIPs, More Is Better Hierarchical INTegration 1.0 Test: FLOAT -Og -O2 -O3 -march=native -flto -O3 -O3 -march=native -Ofast -march=native -Os s10 -O1 -O0 70M 140M 210M 280M 350M SE +/- 179064.26, N = 3 SE +/- 1047951.03, N = 3 SE +/- 100984.48, N = 3 SE +/- 978215.41, N = 3 SE +/- 619811.97, N = 3 SE +/- 547786.00, N = 3 SE +/- 119724.21, N = 3 SE +/- 89816.12, N = 3 SE +/- 222306.33, N = 3 SE +/- 246206.62, N = 3 326497871.93 317711776.83 312975471.93 312279718.27 310268777.87 309403432.89 303914359.33 287050317.81 242450705.97 103731655.85 -Og -O2 -O3 -march=native -flto -O3 -O3 -march=native -Ofast -march=native -Os -O3 -march=native -O1 -O0 1. (CC) gcc options: -lm
Redis Test: GET OpenBenchmarking.org Requests Per Second, More Is Better Redis 3.0.1 Test: GET s10 -O3 -O1 -Og -Os -Ofast -march=native -O3 -march=native -O2 -O0 300K 600K 900K 1200K 1500K SE +/- 27357.36, N = 3 SE +/- 4253.76, N = 3 SE +/- 5335.15, N = 3 SE +/- 2604.35, N = 3 SE +/- 3565.12, N = 3 SE +/- 6785.13, N = 3 SE +/- 10870.06, N = 4 SE +/- 5831.74, N = 3 SE +/- 1566.53, N = 3 1570800.87 669846.73 655681.23 652904.39 645755.96 631191.87 631189.52 628643.96 548655.63 -std=gnu99 -pipe -g3 -O3 -funroll-loops 1. (CC) gcc options: -ggdb -rdynamic -lm -pthread -ldl
Redis Test: LPOP OpenBenchmarking.org Requests Per Second, More Is Better Redis 3.0.1 Test: LPOP s10 -Ofast -march=native -O1 -O3 -march=native -O2 -O3 -Os -Og -O0 300K 600K 900K 1200K 1500K SE +/- 19069.23, N = 3 SE +/- 5651.00, N = 3 SE +/- 2503.49, N = 3 SE +/- 10960.53, N = 3 SE +/- 5733.33, N = 3 SE +/- 11650.40, N = 6 SE +/- 12533.78, N = 3 SE +/- 8860.98, N = 6 SE +/- 3542.72, N = 3 1559720.21 656696.79 655900.13 655097.69 649030.83 646935.48 642758.08 637156.51 547091.88 -std=gnu99 -pipe -g3 -O3 -funroll-loops 1. (CC) gcc options: -ggdb -rdynamic -lm -pthread -ldl
Redis Test: SADD OpenBenchmarking.org Requests Per Second, More Is Better Redis 3.0.1 Test: SADD s10 -Ofast -march=native -O3 -march=native -Os -O3 -O1 -O2 -Og -O0 300K 600K 900K 1200K 1500K SE +/- 3417.96, N = 3 SE +/- 126.46, N = 3 SE +/- 126.17, N = 3 SE +/- 4522.27, N = 3 SE +/- 3715.65, N = 3 SE +/- 2982.88, N = 3 SE +/- 9681.71, N = 3 SE +/- 11078.60, N = 6 SE +/- 1350.43, N = 3 1216564.21 616016.48 615258.45 607478.25 605861.48 605722.93 598759.46 582272.77 491891.31 -std=gnu99 -pipe -g3 -O3 -funroll-loops 1. (CC) gcc options: -ggdb -rdynamic -lm -pthread -ldl
Redis Test: LPUSH OpenBenchmarking.org Requests Per Second, More Is Better Redis 3.0.1 Test: LPUSH s10 -Og -O2 -Ofast -march=native -Os -O3 -O1 -O3 -march=native -O0 300K 600K 900K 1200K 1500K SE +/- 17952.95, N = 3 SE +/- 553.25, N = 3 SE +/- 1359.59, N = 3 SE +/- 2036.63, N = 3 SE +/- 1356.34, N = 3 SE +/- 4012.35, N = 3 SE +/- 8590.61, N = 6 SE +/- 11218.12, N = 6 SE +/- 2663.98, N = 3 1169238.33 602047.96 599526.56 598935.85 598808.52 593409.02 589230.57 584299.37 476295.97 -std=gnu99 -pipe -g3 -O3 -funroll-loops 1. (CC) gcc options: -ggdb -rdynamic -lm -pthread -ldl
Redis Test: SET OpenBenchmarking.org Requests Per Second, More Is Better Redis 3.0.1 Test: SET s10 -O1 -Os -Og -Ofast -march=native -O3 -O2 -O3 -march=native -O0 200K 400K 600K 800K 1000K SE +/- 3977.33, N = 3 SE +/- 1956.89, N = 3 SE +/- 5397.69, N = 3 SE +/- 1909.49, N = 3 SE +/- 2097.48, N = 3 SE +/- 8111.26, N = 5 SE +/- 6845.46, N = 3 SE +/- 8041.99, N = 3 SE +/- 7680.16, N = 3 1149452.71 597265.48 596757.10 592312.44 588019.67 587251.75 586099.29 584905.04 479934.92 -std=gnu99 -pipe -g3 -O3 -funroll-loops 1. (CC) gcc options: -ggdb -rdynamic -lm -pthread -ldl
GraphicsMagick Operation: Sharpen OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.19 Operation: Sharpen -Ofast -march=native -O3 -march=native -O3 -O1 -O2 -Os s10 -Og -O0 30 60 90 120 150 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 145 143 136 135 134 124 103 100 71 -Ofast -march=native -ldl -O3 -march=native -ldl -O3 -ldl -O1 -ldl -O2 -ldl -Os -ldl -O2 -ljbig -lwebp -llcms2 -ltiff -lfreetype -ljasper -ljpeg -lwmflite -llzma -lbz2 -lxml2 -lgomp -Og -ldl -O0 -ldl 1. (CC) gcc options: -fopenmp -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread
SciMark Computational Test: Fast Fourier Transform OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Fast Fourier Transform -Ofast -march=native -O3 -march=native -flto -O2 -O3 -Og -Os -O1 -O3 -march=native -O0 s10 100 200 300 400 500 SE +/- 0.63, N = 4 SE +/- 3.19, N = 4 SE +/- 2.20, N = 4 SE +/- 2.05, N = 4 SE +/- 3.34, N = 4 SE +/- 1.29, N = 4 SE +/- 3.48, N = 4 SE +/- 1.70, N = 4 SE +/- 5.60, N = 4 SE +/- 6.76, N = 4 468.61 465.50 461.55 458.56 456.61 456.35 447.31 443.71 440.57 231.80 -Ofast -march=native -O3 -march=native -flto -O2 -O3 -Og -Os -O1 -O3 -march=native -O0 1. (CXX) g++ options:
Timed PHP Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed PHP Compilation 5.2.9 Time To Compile -O0 -Og -O1 -Os -O2 -O3 -Ofast -march=native -O3 -march=native s10 -O3 -march=native -flto 20 40 60 80 100 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.19, N = 3 SE +/- 0.05, N = 3 SE +/- 0.10, N = 3 5.58 8.18 9.76 11.61 16.08 17.59 17.99 18.10 34.49 82.86 -O1 -Os -O2 -O3 -Ofast -march=native -O3 -march=native -O2 -O3 -march=native -flto 1. (CC) gcc options: -pedantic -ldl -lz -lm
Timed ImageMagick Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed ImageMagick Compilation 6.9.0 Time To Compile -O0 -Og -O1 -Os -O2 -O3 -march=native -O3 -Ofast -march=native s10 -O3 -march=native -flto 30 60 90 120 150 SE +/- 0.10, N = 3 SE +/- 0.13, N = 3 SE +/- 0.21, N = 3 SE +/- 0.08, N = 3 SE +/- 0.06, N = 3 SE +/- 0.23, N = 3 SE +/- 0.14, N = 3 SE +/- 0.09, N = 3 SE +/- 0.13, N = 3 SE +/- 0.25, N = 3 9.34 13.35 27.24 32.43 38.55 55.40 55.45 55.89 74.78 121.45
GraphicsMagick Operation: Resizing OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.19 Operation: Resizing -Ofast -march=native -O3 -march=native -O2 -O3 -O1 -Os s10 -Og -O0 40 80 120 160 200 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.67, N = 3 SE +/- 0.00, N = 3 SE +/- 1.00, N = 3 SE +/- 0.58, N = 3 SE +/- 0.88, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 182 180 174 171 168 168 151 149 97 -Ofast -march=native -ldl -O3 -march=native -ldl -O2 -ldl -O3 -ldl -O1 -ldl -Os -ldl -O2 -ljbig -lwebp -llcms2 -ltiff -lfreetype -ljasper -ljpeg -lwmflite -llzma -lbz2 -lxml2 -lgomp -Og -ldl -O0 -ldl 1. (CC) gcc options: -fopenmp -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread
GraphicsMagick Operation: HWB Color Space OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.19 Operation: HWB Color Space -Ofast -march=native -O3 -march=native -Os -O1 -O2 -O3 -Og s10 -O0 40 80 120 160 200 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 204 190 188 187 186 185 168 159 110 -Ofast -march=native -ldl -O3 -march=native -ldl -Os -ldl -O1 -ldl -O2 -ldl -O3 -ldl -Og -ldl -O2 -ljbig -lwebp -llcms2 -ltiff -lfreetype -ljasper -ljpeg -lwmflite -llzma -lbz2 -lxml2 -lgomp -O0 -ldl 1. (CC) gcc options: -fopenmp -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread
GraphicsMagick Operation: Blur OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.19 Operation: Blur -Ofast -march=native -O3 -march=native -O1 -O2 -O3 s10 -Og -Os -O0 30 60 90 120 150 SE +/- 0.33, N = 3 SE +/- 1.20, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 SE +/- 0.33, N = 3 SE +/- 1.53, N = 3 SE +/- 0.58, N = 3 SE +/- 0.00, N = 3 144 138 137 131 130 125 113 110 82 -Ofast -march=native -ldl -O3 -march=native -ldl -O1 -ldl -O2 -ldl -O3 -ldl -O2 -ljbig -lwebp -llcms2 -ltiff -lfreetype -ljasper -ljpeg -lwmflite -llzma -lbz2 -lxml2 -lgomp -Og -ldl -Os -ldl -O0 -ldl 1. (CC) gcc options: -fopenmp -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread
SciMark Computational Test: Sparse Matrix Multiply OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Sparse Matrix Multiply -O3 -O1 -Os -Og -O2 -O0 -Ofast -march=native -O3 -march=native -flto -O3 -march=native s10 600 1200 1800 2400 3000 SE +/- 9.06, N = 4 SE +/- 3.88, N = 4 SE +/- 21.77, N = 4 SE +/- 13.12, N = 4 SE +/- 32.30, N = 4 SE +/- 9.07, N = 4 SE +/- 19.13, N = 4 SE +/- 24.76, N = 4 SE +/- 12.32, N = 4 SE +/- 16.80, N = 4 2622.50 2609.45 2589.56 2580.32 2571.06 2565.94 2517.37 2511.39 2440.96 1544.66 -O3 -O1 -Os -Og -O2 -O0 -Ofast -march=native -O3 -march=native -flto -O3 -march=native 1. (CXX) g++ options:
SciMark Computational Test: Composite OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Composite -O3 -march=native -flto -O3 -O1 -Og -O2 -Os -Ofast -march=native -O0 -O3 -march=native s10 300 600 900 1200 1500 SE +/- 7.57, N = 4 SE +/- 2.58, N = 4 SE +/- 1.73, N = 4 SE +/- 3.98, N = 4 SE +/- 7.42, N = 4 SE +/- 2.83, N = 4 SE +/- 5.38, N = 4 SE +/- 8.59, N = 4 SE +/- 4.89, N = 4 SE +/- 15.87, N = 4 1445.10 1442.30 1437.73 1427.70 1426.80 1426.79 1421.80 1407.02 1388.10 1023.06 -O3 -march=native -flto -O3 -O1 -Og -O2 -Os -Ofast -march=native -O0 -O3 -march=native 1. (CXX) g++ options:
SciMark Computational Test: Dense LU Matrix Factorization OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Dense LU Matrix Factorization -O3 -march=native -flto -O2 -O1 -O3 -Og -Ofast -march=native -Os -O3 -march=native -O0 s10 600 1200 1800 2400 3000 SE +/- 28.87, N = 4 SE +/- 3.32, N = 4 SE +/- 7.15, N = 4 SE +/- 7.24, N = 4 SE +/- 4.63, N = 4 SE +/- 11.60, N = 4 SE +/- 18.22, N = 4 SE +/- 10.80, N = 4 SE +/- 26.42, N = 4 SE +/- 52.11, N = 4 2586.62 2534.23 2534.08 2531.29 2521.75 2519.72 2482.13 2468.30 2454.12 1835.64 -O3 -march=native -flto -O2 -O1 -O3 -Og -Ofast -march=native -Os -O3 -march=native -O0 1. (CXX) g++ options:
SciMark Computational Test: Monte Carlo OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Monte Carlo -O3 -march=native -flto -Ofast -march=native -Os -O1 -O3 -O3 -march=native -Og -O0 -O2 s10 130 260 390 520 650 SE +/- 6.00, N = 4 SE +/- 1.67, N = 4 SE +/- 1.68, N = 4 SE +/- 1.22, N = 4 SE +/- 3.64, N = 4 SE +/- 5.84, N = 4 SE +/- 5.79, N = 4 SE +/- 6.16, N = 4 SE +/- 9.59, N = 4 SE +/- 3.57, N = 4 614.59 553.28 552.04 551.44 547.85 547.58 546.26 545.51 537.90 464.45 -O3 -march=native -flto -Ofast -march=native -Os -O1 -O3 -O3 -march=native -Og -O0 -O2 1. (CXX) g++ options:
SciMark Computational Test: Jacobi Successive Over-Relaxation OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation -Os -O3 -Ofast -march=native -O3 -march=native -flto -O1 -O3 -march=native s10 -Og -O2 -O0 200 400 600 800 1000 SE +/- 1.16, N = 4 SE +/- 2.16, N = 4 SE +/- 0.76, N = 4 SE +/- 6.06, N = 4 SE +/- 3.32, N = 4 SE +/- 6.48, N = 4 SE +/- 11.16, N = 4 SE +/- 8.62, N = 4 SE +/- 12.72, N = 4 SE +/- 9.29, N = 4 1053.88 1051.27 1050.05 1047.43 1046.36 1039.94 1038.75 1033.56 1029.27 1028.95 -Os -O3 -Ofast -march=native -O3 -march=native -flto -O1 -O3 -march=native -Og -O2 -O0 1. (CXX) g++ options:
Timed HMMer Search Pfam Database Search OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 2.3.2 Pfam Database Search -Og -Ofast -march=native -O1 -Os -O2 -O3 -march=native -O3 -O0 s10 4 8 12 16 20 SE +/- 0.05, N = 3 SE +/- 0.27, N = 6 SE +/- 0.16, N = 6 SE +/- 0.19, N = 3 SE +/- 0.51, N = 6 SE +/- 0.68, N = 6 SE +/- 0.35, N = 6 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 8.23 8.32 10.19 10.68 11.63 13.04 13.08 13.82 15.29 -Og -Ofast -march=native -O1 -Os -O2 -O3 -march=native -O3 -O0 -O2 1. (CC) gcc options: -pthread -lhmmer -lsquid -lm
Phoronix Test Suite v10.8.4