AMD EPYC compiler benchmarks for a future article on Phoronix.com.
GCC 8.2 RC1: -O2 Environment Notes: CXXFLAGS=-O2 CFLAGS=-O2Compiler Notes: --disable-multilib --enable-checking=releaseDisk Notes: NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemandSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB Protection
GCC 8.2 RC1: -O3 Environment Notes: CXXFLAGS=-O3 CFLAGS=-O3Compiler Notes: --disable-multilib --enable-checking=releaseDisk Notes: NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemandSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB Protection
GCC 8.2 RC1: -O3 native Environment Notes: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=nativeCompiler Notes: --disable-multilib --enable-checking=releaseDisk Notes: NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemandSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB Protection
GCC 8.2 RC1: -O3 native + PGO Processor: AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads), Motherboard: TYAN B8026T70AE24HR (V1.02.B10 BIOS), Chipset: AMD Family 17h, Memory: 129024MB, Disk: 280GB INTEL SSDPE21D280GA, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom Limited NetXtreme BCM5720 Gigabit PCIe
OS: Ubuntu 18.10, Kernel: 4.16.0-041600-generic (x86_64), Compiler: GCC 8.1.1 20180720, File-System: ext4, Screen Resolution: 1920x1080
GCC 9.0 SVN: -O2 Environment Notes: CXXFLAGS=-O2 CFLAGS=-O2Compiler Notes: --disable-multilib --enable-checking=releaseDisk Notes: NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemandSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB Protection
GCC 9.0 SVN: -O3 Environment Notes: CXXFLAGS=-O3 CFLAGS=-O3Compiler Notes: --disable-multilib --enable-checking=releaseDisk Notes: NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemandSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB Protection
GCC 9.0 SVN: -O3 native Environment Notes: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=nativeCompiler Notes: --disable-multilib --enable-checking=releaseDisk Notes: NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemandSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB Protection
GCC 9.0 SVN: -O3 native + PGO OS: Ubuntu 18.10, Kernel: 4.16.0-041600-generic (x86_64), Compiler: GCC 9.0.0 20180720, File-System: ext4, Screen Resolution: 1920x1080
SciMark This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Monte Carlo -O3 -O2 -O3 native -O3 native + PGO 120 240 360 480 600 SE +/- 0.01, N = 3 SE +/- 0.12, N = 3 SE +/- 0.10, N = 3 SE +/- 0.25, N = 3 SE +/- 0.53, N = 3 SE +/- 0.02, N = 3 SE +/- 9.75, N = 3 SE +/- 0.33, N = 3 561.85 561.79 556.37 213.98 561.29 556.37 552.08 213.95 1. (CC) gcc options: -lm
Result Confidence
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Monte Carlo -O3 -O2 -O3 native -O3 native + PGO 100 200 300 400 500 Min: 561.83 / Avg: 561.85 / Max: 561.87 Min: 561.56 / Avg: 561.79 / Max: 561.97 Min: 556.21 / Avg: 556.37 / Max: 556.54 Min: 213.48 / Avg: 213.98 / Max: 214.26 Min: 560.23 / Avg: 561.29 / Max: 561.86 Min: 556.33 / Avg: 556.37 / Max: 556.41 Min: 532.59 / Avg: 552.08 / Max: 561.83 Min: 213.29 / Avg: 213.95 / Max: 214.3 1. (CC) gcc options: -lm
C-Ray
Result
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time -O3 native + PGO -O3 native -O3 -O2 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 2.50 2.52 2.83 6.02 2.51 2.52 2.84 6.00 1. (CC) gcc options: -lm -lpthread -O3
Result Confidence
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time -O3 native + PGO -O3 native -O3 -O2 2 4 6 8 10 Min: 2.48 / Avg: 2.5 / Max: 2.51 Min: 2.51 / Avg: 2.52 / Max: 2.54 Min: 2.81 / Avg: 2.83 / Max: 2.85 Min: 6.01 / Avg: 6.02 / Max: 6.03 Min: 2.49 / Avg: 2.51 / Max: 2.53 Min: 2.5 / Avg: 2.52 / Max: 2.54 Min: 2.82 / Avg: 2.84 / Max: 2.86 Min: 5.98 / Avg: 6 / Max: 6.01 1. (CC) gcc options: -lm -lpthread -O3
SciMark This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation -O3 native -O3 native + PGO -O3 -O2 400 800 1200 1600 2000 SE +/- 0.72, N = 3 SE +/- 0.23, N = 3 SE +/- 0.17, N = 3 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 SE +/- 0.26, N = 3 SE +/- 0.16, N = 3 SE +/- 0.07, N = 3 1688.09 1674.99 1429.31 918.96 1687.64 1676.15 1429.81 919.02 1. (CC) gcc options: -lm
Result Confidence
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation -O3 native -O3 native + PGO -O3 -O2 300 600 900 1200 1500 Min: 1687.18 / Avg: 1688.09 / Max: 1689.51 Min: 1674.57 / Avg: 1674.99 / Max: 1675.35 Min: 1428.99 / Avg: 1429.31 / Max: 1429.59 Min: 918.93 / Avg: 918.96 / Max: 918.99 Min: 1687.56 / Avg: 1687.64 / Max: 1687.75 Min: 1675.64 / Avg: 1676.15 / Max: 1676.46 Min: 1429.49 / Avg: 1429.81 / Max: 1429.98 Min: 918.93 / Avg: 919.02 / Max: 919.16 1. (CC) gcc options: -lm
Result
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Dense LU Matrix Factorization -O3 native + PGO -O3 native -O3 -O2 1000 2000 3000 4000 5000 SE +/- 17.02, N = 3 SE +/- 20.82, N = 3 SE +/- 27.07, N = 3 SE +/- 14.02, N = 3 SE +/- 24.64, N = 3 SE +/- 25.52, N = 3 SE +/- 19.27, N = 3 SE +/- 3.21, N = 3 4865.24 4831.75 4683.55 3462.39 4845.49 4781.45 4723.98 3432.88 1. (CC) gcc options: -lm
Result Confidence
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Dense LU Matrix Factorization -O3 native + PGO -O3 native -O3 -O2 800 1600 2400 3200 4000 Min: 4834.45 / Avg: 4865.24 / Max: 4893.19 Min: 4800.85 / Avg: 4831.75 / Max: 4871.38 Min: 4637.01 / Avg: 4683.55 / Max: 4730.77 Min: 3435.97 / Avg: 3462.39 / Max: 3483.72 Min: 4798.01 / Avg: 4845.49 / Max: 4880.63 Min: 4730.45 / Avg: 4781.45 / Max: 4808.47 Min: 4692.88 / Avg: 4723.98 / Max: 4759.23 Min: 3428.82 / Avg: 3432.88 / Max: 3439.21 1. (CC) gcc options: -lm
SciMark This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Composite -O3 native -O3 native + PGO -O3 -O2 400 800 1200 1600 2000 SE +/- 5.96, N = 3 SE +/- 3.46, N = 3 SE +/- 5.53, N = 3 SE +/- 3.37, N = 3 SE +/- 4.00, N = 3 SE +/- 4.50, N = 3 SE +/- 5.95, N = 3 SE +/- 2.46, N = 3 1943.30 1877.54 1874.09 1464.06 1940.65 1877.03 1870.83 1499.96 1. (CC) gcc options: -lm
Result Confidence
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Composite -O3 native -O3 native + PGO -O3 -O2 300 600 900 1200 1500 Min: 1932.73 / Avg: 1943.3 / Max: 1953.34 Min: 1870.74 / Avg: 1877.54 / Max: 1882.04 Min: 1864.78 / Avg: 1874.09 / Max: 1883.92 Min: 1457.75 / Avg: 1464.06 / Max: 1469.27 Min: 1934.22 / Avg: 1940.65 / Max: 1947.98 Min: 1868.86 / Avg: 1877.03 / Max: 1884.37 Min: 1860.26 / Avg: 1870.83 / Max: 1880.86 Min: 1496.74 / Avg: 1499.96 / Max: 1504.78 1. (CC) gcc options: -lm
Himeno Benchmark The Himeno benchmark is a linear solver of pressure Poisson using a point-Jacobi method. Learn more via the OpenBenchmarking.org test page.
Result
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver -O2 -O3 native -O3 -O3 native + PGO 200 400 600 800 1000 SE +/- 1.94, N = 3 SE +/- 2.17, N = 3 SE +/- 0.78, N = 3 SE +/- 0.73, N = 3 SE +/- 0.47, N = 3 SE +/- 4.57, N = 3 SE +/- 5.36, N = 3 SE +/- 0.64, N = 3 1014.66 998.89 982.67 852.63 1014.38 1007.74 993.22 869.07 1. (CC) gcc options: -O3 -mavx2
Result Confidence
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver -O2 -O3 native -O3 -O3 native + PGO 200 400 600 800 1000 Min: 1012.65 / Avg: 1014.66 / Max: 1018.54 Min: 994.56 / Avg: 998.89 / Max: 1001.25 Min: 981.38 / Avg: 982.67 / Max: 984.07 Min: 851.79 / Avg: 852.63 / Max: 854.08 Min: 1013.47 / Avg: 1014.38 / Max: 1015.05 Min: 1002.23 / Avg: 1007.74 / Max: 1016.82 Min: 985.07 / Avg: 993.22 / Max: 1003.33 Min: 867.87 / Avg: 869.07 / Max: 870.08 1. (CC) gcc options: -O3 -mavx2
SciMark This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Sparse Matrix Multiply -O3 -O3 native -O3 native + PGO -O2 500 1000 1500 2000 2500 SE +/- 1.01, N = 3 SE +/- 11.03, N = 3 SE +/- 10.35, N = 3 SE +/- 2.57, N = 3 SE +/- 11.52, N = 3 SE +/- 13.99, N = 3 SE +/- 7.59, N = 3 SE +/- 14.14, N = 3 2473.98 2421.38 2413.24 2156.61 2458.50 2458.23 2400.46 2365.87 1. (CC) gcc options: -lm
Result Confidence
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Sparse Matrix Multiply -O3 -O3 native -O3 native + PGO -O2 400 800 1200 1600 2000 Min: 2472.12 / Avg: 2473.98 / Max: 2475.59 Min: 2399.32 / Avg: 2421.38 / Max: 2432.66 Min: 2396.93 / Avg: 2413.24 / Max: 2432.42 Min: 2151.82 / Avg: 2156.61 / Max: 2160.6 Min: 2439.35 / Avg: 2458.5 / Max: 2479.17 Min: 2430.56 / Avg: 2458.23 / Max: 2475.7 Min: 2390.18 / Avg: 2400.46 / Max: 2415.28 Min: 2342.97 / Avg: 2365.87 / Max: 2391.68 1. (CC) gcc options: -lm
AOBench AOBench is a lightweight ambient occlusion renderer, written in C. The test profile is using a size of 2048 x 2048. Learn more via the OpenBenchmarking.org test page.
Result
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better AOBench Size: 2048 x 2048 - Total Time -O3 native + PGO -O3 native -O3 -O2 12 24 36 48 60 SE +/- 0.16, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.20, N = 3 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 SE +/- 0.00, N = 3 48.51 50.72 53.03 55.52 49.05 50.80 52.60 55.52 1. (CC) gcc options: -lm -O3
Result Confidence
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better AOBench Size: 2048 x 2048 - Total Time -O3 native + PGO -O3 native -O3 -O2 11 22 33 44 55 Min: 48.35 / Avg: 48.51 / Max: 48.84 Min: 50.69 / Avg: 50.72 / Max: 50.75 Min: 53.01 / Avg: 53.03 / Max: 53.05 Min: 55.52 / Avg: 55.52 / Max: 55.52 Min: 48.76 / Avg: 49.05 / Max: 49.43 Min: 50.78 / Avg: 50.8 / Max: 50.81 Min: 52.53 / Avg: 52.6 / Max: 52.74 Min: 55.52 / Avg: 55.52 / Max: 55.53 1. (CC) gcc options: -lm -O3
Zstd Compression This test measures the time needed to compress a sample file (an Ubuntu file-system image) using Zstd compression. Learn more via the OpenBenchmarking.org test page.
Result
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 -O3 native + PGO -O3 native -O2 -O3 40 80 120 160 200 SE +/- 0.24, N = 3 SE +/- 1.43, N = 3 SE +/- 0.46, N = 3 SE +/- 0.51, N = 3 SE +/- 0.54, N = 3 SE +/- 0.62, N = 3 SE +/- 0.68, N = 3 147.78 156.58 168.14 168.20 155.64 166.12 166.70 1. (CC) gcc options: -pthread -lz -llzma
Result Confidence
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 -O3 native + PGO -O3 native -O2 -O3 30 60 90 120 150 Min: 147.29 / Avg: 147.78 / Max: 148.05 Min: 153.74 / Avg: 156.58 / Max: 158.33 Min: 167.6 / Avg: 168.14 / Max: 169.05 Min: 167.18 / Avg: 168.2 / Max: 168.81 Min: 154.92 / Avg: 155.64 / Max: 156.71 Min: 164.87 / Avg: 166.12 / Max: 166.81 Min: 165.35 / Avg: 166.7 / Max: 167.45 1. (CC) gcc options: -pthread -lz -llzma
PolyBench-C PolyBench-C is a C-language polyhedral benchmark suite made at the Ohio State University. Learn more via the OpenBenchmarking.org test page.
Result
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better PolyBench-C 4.2 Test: 3 Matrix Multiplications -O3 native + PGO -O3 native -O3 -O2 0.828 1.656 2.484 3.312 4.14 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 3.30 3.63 3.66 3.68 3.35 3.58 3.59 3.62 1. (CC) gcc options: -O3 -march=native
Result Confidence
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better PolyBench-C 4.2 Test: 3 Matrix Multiplications -O3 native + PGO -O3 native -O3 -O2 2 4 6 8 10 Min: 3.28 / Avg: 3.3 / Max: 3.34 Min: 3.59 / Avg: 3.63 / Max: 3.69 Min: 3.61 / Avg: 3.66 / Max: 3.75 Min: 3.63 / Avg: 3.68 / Max: 3.76 Min: 3.3 / Avg: 3.35 / Max: 3.44 Min: 3.56 / Avg: 3.58 / Max: 3.61 Min: 3.58 / Avg: 3.59 / Max: 3.62 Min: 3.56 / Avg: 3.62 / Max: 3.72 1. (CC) gcc options: -O3 -march=native
TTSIOD 3D Renderer A portable GPL 3D software renderer that supports OpenMP and Intel Threading Building Blocks with many different rendering modes. This version does not use OpenGL but is entirely CPU/software based. Learn more via the OpenBenchmarking.org test page.
Stockfish This is a test of Stockfish, an advanced C++11 chess benchmark that can scale up to 128 CPU cores. Learn more via the OpenBenchmarking.org test page.
Result
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 9 Total Time -O3 native + PGO -O3 -O3 native -O2 13M 26M 39M 52M 65M SE +/- 512120.63, N = 3 SE +/- 551889.61, N = 3 SE +/- 576746.80, N = 3 SE +/- 219692.26, N = 3 SE +/- 756265.41, N = 3 SE +/- 236832.99, N = 3 SE +/- 346634.55, N = 3 SE +/- 419233.35, N = 3 61080869 59837512 59730290 59007314 59201545 58412975 58372089 56149550 1. (CXX) g++ options: -m64 -lpthread -O3 -fno-exceptions -std=c++11 -pedantic -msse -msse3 -mpopcnt -flto
Result Confidence
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 9 Total Time -O3 native + PGO -O3 -O3 native -O2 11M 22M 33M 44M 55M Min: 60073285 / Avg: 61080868.67 / Max: 61743984 Min: 59240734 / Avg: 59837511.67 / Max: 60940039 Min: 58613110 / Avg: 59730290 / Max: 60537561 Min: 58674829 / Avg: 59007313.67 / Max: 59422322 Min: 58025227 / Avg: 59201544.67 / Max: 60613123 Min: 57970432 / Avg: 58412975.33 / Max: 58780487 Min: 57888823 / Avg: 58372089 / Max: 59044195 Min: 55521505 / Avg: 56149550 / Max: 56944654 1. (CXX) g++ options: -m64 -lpthread -O3 -fno-exceptions -std=c++11 -pedantic -msse -msse3 -mpopcnt -flto
GraphicsMagick GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.28 Operation: Sharpen -O3 native -O2 -O3 40 80 120 160 200 188 181 177 182 179 176 1. (CC) gcc options: -fopenmp -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -ldl -lpthread
ebizzy This is a test of ebizzy, a program to generate workloads resembling web server workloads. Learn more via the OpenBenchmarking.org test page.
Result
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Records/s, More Is Better ebizzy 0.3 -O3 native + PGO -O3 native -O2 -O3 200K 400K 600K 800K 1000K SE +/- 4974.04, N = 3 SE +/- 17876.02, N = 3 SE +/- 23788.74, N = 6 SE +/- 14780.83, N = 3 SE +/- 28582.82, N = 6 SE +/- 18183.13, N = 4 SE +/- 20234.15, N = 6 SE +/- 16592.93, N = 4 1166998 1124222 1111662 1109148 1147009 1113356 1112448 1106391 1. (CC) gcc options: -pthread -lpthread -O3 -march=native
Result Confidence
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Records/s, More Is Better ebizzy 0.3 -O3 native + PGO -O3 native -O2 -O3 200K 400K 600K 800K 1000K Min: 1157206 / Avg: 1166998.33 / Max: 1173413 Min: 1091160 / Avg: 1124221.67 / Max: 1152536 Min: 1044233 / Avg: 1111661.67 / Max: 1216394 Min: 1087433 / Avg: 1109148.33 / Max: 1137377 Min: 1042789 / Avg: 1147009.33 / Max: 1229755 Min: 1071327 / Avg: 1113356 / Max: 1150235 Min: 1043008 / Avg: 1112448 / Max: 1157261 Min: 1056857 / Avg: 1106391 / Max: 1127338 1. (CC) gcc options: -pthread -lpthread -O3 -march=native
Crafty This is a performance test of Crafty, an advanced open-source chess engine. Learn more via the OpenBenchmarking.org test page.
Result
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Nodes Per Second, More Is Better Crafty 25.2 Elapsed Time -O3 native + PGO -O3 native -O2 -O3 1.3M 2.6M 3.9M 5.2M 6.5M SE +/- 11021.16, N = 3 SE +/- 5146.22, N = 3 SE +/- 8016.22, N = 3 SE +/- 13110.39, N = 3 SE +/- 2607.25, N = 3 SE +/- 13470.58, N = 3 SE +/- 2577.32, N = 3 SE +/- 9192.37, N = 3 6026655 6014866 6004700 5989207 5892825 5876297 5841423 5780824 1. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
Result Confidence
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Nodes Per Second, More Is Better Crafty 25.2 Elapsed Time -O3 native + PGO -O3 native -O2 -O3 1000K 2000K 3000K 4000K 5000K Min: 6006995 / Avg: 6026655 / Max: 6045117 Min: 6005526 / Avg: 6014865.67 / Max: 6023281 Min: 5990315 / Avg: 6004700 / Max: 6018023 Min: 5963012 / Avg: 5989206.67 / Max: 6003317 Min: 5888236 / Avg: 5892825.33 / Max: 5897264 Min: 5858239 / Avg: 5876297.33 / Max: 5902641 Min: 5836946 / Avg: 5841422.67 / Max: 5845874 Min: 5763418 / Avg: 5780823.67 / Max: 5794653 1. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
PolyBench-C PolyBench-C is a C-language polyhedral benchmark suite made at the Ohio State University. Learn more via the OpenBenchmarking.org test page.
Result
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Seconds, Fewer Is Better PolyBench-C 4.2 Test: Correlation Computation -O3 native + PGO -O3 native -O3 -O2 1.2488 2.4976 3.7464 4.9952 6.244 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 5.33 5.34 5.40 5.55 5.38 5.40 5.46 5.50 1. (CC) gcc options: -O3 -march=native
Result Confidence
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Seconds, Fewer Is Better PolyBench-C 4.2 Test: Correlation Computation -O3 native + PGO -O3 native -O3 -O2 2 4 6 8 10 Min: 5.29 / Avg: 5.33 / Max: 5.4 Min: 5.34 / Avg: 5.34 / Max: 5.35 Min: 5.38 / Avg: 5.4 / Max: 5.42 Min: 5.53 / Avg: 5.55 / Max: 5.58 Min: 5.36 / Avg: 5.38 / Max: 5.42 Min: 5.4 / Avg: 5.4 / Max: 5.4 Min: 5.38 / Avg: 5.46 / Max: 5.51 Min: 5.42 / Avg: 5.5 / Max: 5.55 1. (CC) gcc options: -O3 -march=native
7-Zip Compression This is a test of 7-Zip using p7zip with its integrated benchmark feature or upstream 7-Zip for the Windows x64 build. Learn more via the OpenBenchmarking.org test page.
Result
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 16.02 Compress Speed Test -O3 native -O3 native + PGO -O2 -O3 20K 40K 60K 80K 100K SE +/- 715.28, N = 3 SE +/- 555.15, N = 3 SE +/- 771.81, N = 3 SE +/- 586.49, N = 3 SE +/- 926.79, N = 3 SE +/- 935.09, N = 3 SE +/- 465.88, N = 3 SE +/- 1357.08, N = 3 100684 99762 98480 97917 98423 97663 97385 96774 1. (CXX) g++ options: -pipe -lpthread
Result Confidence
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 16.02 Compress Speed Test -O3 native -O3 native + PGO -O2 -O3 20K 40K 60K 80K 100K Min: 99292 / Avg: 100684.33 / Max: 101665 Min: 99038 / Avg: 99762 / Max: 100853 Min: 96944 / Avg: 98479.67 / Max: 99383 Min: 96763 / Avg: 97917 / Max: 98676 Min: 96909 / Avg: 98422.67 / Max: 100106 Min: 96247 / Avg: 97663 / Max: 99429 Min: 96497 / Avg: 97385.33 / Max: 98073 Min: 94636 / Avg: 96774 / Max: 99291 1. (CXX) g++ options: -pipe -lpthread
PolyBench-C PolyBench-C is a C-language polyhedral benchmark suite made at the Ohio State University. Learn more via the OpenBenchmarking.org test page.
Result
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Seconds, Fewer Is Better PolyBench-C 4.2 Test: Covariance Computation -O3 native -O3 native + PGO -O3 -O2 1.2443 2.4886 3.7329 4.9772 6.2215 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 5.34 5.39 5.42 5.53 5.37 5.41 5.49 5.50 1. (CC) gcc options: -O3 -march=native
Result Confidence
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Seconds, Fewer Is Better PolyBench-C 4.2 Test: Covariance Computation -O3 native -O3 native + PGO -O3 -O2 2 4 6 8 10 Min: 5.31 / Avg: 5.34 / Max: 5.36 Min: 5.36 / Avg: 5.39 / Max: 5.42 Min: 5.4 / Avg: 5.42 / Max: 5.44 Min: 5.52 / Avg: 5.53 / Max: 5.55 Min: 5.33 / Avg: 5.37 / Max: 5.41 Min: 5.38 / Avg: 5.41 / Max: 5.43 Min: 5.41 / Avg: 5.49 / Max: 5.57 Min: 5.45 / Avg: 5.5 / Max: 5.54 1. (CC) gcc options: -O3 -march=native
VP9 libvpx Encoding This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9/WebM format. Learn more via the OpenBenchmarking.org test page.
Result
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.7.0 vpxenc -O2 -O3 native -O3 -O3 native + PGO 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 SE +/- 0.14, N = 3 SE +/- 0.01, N = 3 12.43 12.38 12.36 12.29 12.37 12.31 12.21 12.07 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE
Result Confidence
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.7.0 vpxenc -O2 -O3 native -O3 -O3 native + PGO 4 8 12 16 20 Min: 12.42 / Avg: 12.43 / Max: 12.44 Min: 12.35 / Avg: 12.38 / Max: 12.41 Min: 12.26 / Avg: 12.36 / Max: 12.45 Min: 12.19 / Avg: 12.29 / Max: 12.34 Min: 12.35 / Avg: 12.37 / Max: 12.39 Min: 12.23 / Avg: 12.31 / Max: 12.43 Min: 11.93 / Avg: 12.21 / Max: 12.38 Min: 12.06 / Avg: 12.07 / Max: 12.09 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE
m-queens A solver for the N-queens problem with multi-threading support via the OpenMP library. Learn more via the OpenBenchmarking.org test page.
Result
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better m-queens 1.1 Time To Solve -O3 native -O2 -O3 -O3 native + PGO 8 16 24 32 40 SE +/- 0.16, N = 3 SE +/- 0.15, N = 3 SE +/- 0.13, N = 3 SE +/- 0.16, N = 3 SE +/- 0.14, N = 3 SE +/- 0.14, N = 3 SE +/- 0.09, N = 3 SE +/- 0.19, N = 3 31.56 31.59 31.64 32.31 32.02 32.09 32.22 32.44 1. (CXX) g++ options: -fopenmp -march=native -O2
Result Confidence
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Seconds, Fewer Is Better m-queens 1.1 Time To Solve -O3 native -O2 -O3 -O3 native + PGO 7 14 21 28 35 Min: 31.34 / Avg: 31.56 / Max: 31.87 Min: 31.39 / Avg: 31.59 / Max: 31.89 Min: 31.42 / Avg: 31.64 / Max: 31.88 Min: 31.98 / Avg: 32.31 / Max: 32.5 Min: 31.85 / Avg: 32.02 / Max: 32.3 Min: 31.81 / Avg: 32.09 / Max: 32.29 Min: 32.06 / Avg: 32.22 / Max: 32.37 Min: 32.05 / Avg: 32.44 / Max: 32.66 1. (CXX) g++ options: -fopenmp -march=native -O2
Apache Benchmark This is a test of ab, which is the Apache benchmark program. This test profile measures how many requests per second a given system can sustain when carrying out 1,000,000 requests with 100 requests being carried out concurrently. Learn more via the OpenBenchmarking.org test page.
SciMark This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
Result
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Fast Fourier Transform -O3 -O2 -O3 native -O3 native + PGO 50 100 150 200 250 SE +/- 0.09, N = 3 SE +/- 0.20, N = 3 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.13, N = 3 SE +/- 0.03, N = 3 221.30 220.52 219.38 219.09 221.28 220.80 219.24 219.09 1. (CC) gcc options: -lm
Result Confidence
GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Fast Fourier Transform -O3 -O2 -O3 native -O3 native + PGO 40 80 120 160 200 Min: 221.11 / Avg: 221.3 / Max: 221.42 Min: 220.21 / Avg: 220.52 / Max: 220.89 Min: 219.2 / Avg: 219.38 / Max: 219.55 Min: 218.99 / Avg: 219.09 / Max: 219.17 Min: 221.24 / Avg: 221.28 / Max: 221.33 Min: 220.78 / Avg: 220.8 / Max: 220.82 Min: 219.06 / Avg: 219.24 / Max: 219.5 Min: 219.03 / Avg: 219.09 / Max: 219.13 1. (CC) gcc options: -lm
Tinymembench GCC 8.2 RC1 GCC 9.0 SVN OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memset -O3 native -O3 -O2 -O3 native + PGO 2K 4K 6K 8K 10K 10063.10 10037.70 9978.60 9975.20 10039.70 9989.40 9982.90 9982.80 1. (CC) gcc options: -O2 -lm
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test measures the RSA 4096-bit performance of OpenSSL. Learn more via the OpenBenchmarking.org test page.
Result
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.1.0f RSA 4096-bit Performance -O3 native + PGO -O3 native -O2 -O3 1000 2000 3000 4000 5000 SE +/- 19.76, N = 3 SE +/- 23.13, N = 3 SE +/- 22.12, N = 3 SE +/- 23.64, N = 3 SE +/- 23.93, N = 3 SE +/- 21.56, N = 3 SE +/- 22.71, N = 3 SE +/- 24.30, N = 3 4623.60 4616.47 4612.20 4602.30 4617.23 4610.63 4609.73 4609.30 1. (CC) gcc options: -O3 -pthread -m64 -lcrypto -ldl
Result Confidence
GCC 9.0 SVN GCC 8.2 RC1 OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.1.0f RSA 4096-bit Performance -O3 native + PGO -O3 native -O2 -O3 800 1600 2400 3200 4000 Min: 4587.6 / Avg: 4623.6 / Max: 4655.7 Min: 4575.1 / Avg: 4616.47 / Max: 4655.1 Min: 4573.4 / Avg: 4612.2 / Max: 4650 Min: 4559.8 / Avg: 4602.3 / Max: 4641.5 Min: 4573.6 / Avg: 4617.23 / Max: 4656.1 Min: 4572.3 / Avg: 4610.63 / Max: 4646.9 Min: 4569.5 / Avg: 4609.73 / Max: 4648.1 Min: 4570.9 / Avg: 4609.3 / Max: 4654.3 1. (CC) gcc options: -O3 -pthread -m64 -lcrypto -ldl
GCC 8.2 RC1: -O2 Environment Notes: CXXFLAGS=-O2 CFLAGS=-O2Compiler Notes: --disable-multilib --enable-checking=releaseDisk Notes: NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemandSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB Protection
Testing initiated at 21 July 2018 11:29 by user phoronix.
GCC 8.2 RC1: -O3 Environment Notes: CXXFLAGS=-O3 CFLAGS=-O3Compiler Notes: --disable-multilib --enable-checking=releaseDisk Notes: NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemandSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB Protection
Testing initiated at 20 July 2018 18:42 by user phoronix.
GCC 8.2 RC1: -O3 native Environment Notes: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=nativeCompiler Notes: --disable-multilib --enable-checking=releaseDisk Notes: NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemandSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB Protection
Testing initiated at 20 July 2018 11:08 by user phoronix.
GCC 8.2 RC1: -O3 native + PGO Processor: AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads), Motherboard: TYAN B8026T70AE24HR (V1.02.B10 BIOS), Chipset: AMD Family 17h, Memory: 129024MB, Disk: 280GB INTEL SSDPE21D280GA, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom Limited NetXtreme BCM5720 Gigabit PCIe
OS: Ubuntu 18.10, Kernel: 4.16.0-041600-generic (x86_64), Compiler: GCC 8.1.1 20180720, File-System: ext4, Screen Resolution: 1920x1080
Environment Notes: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=nativeCompiler Notes: --disable-multilib --enable-checking=releaseDisk Notes: NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemandSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB Protection
Testing initiated at 20 July 2018 17:05 by user phoronix.
GCC 9.0 SVN: -O2 Environment Notes: CXXFLAGS=-O2 CFLAGS=-O2Compiler Notes: --disable-multilib --enable-checking=releaseDisk Notes: NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemandSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB Protection
Testing initiated at 21 July 2018 09:03 by user phoronix.
GCC 9.0 SVN: -O3 Environment Notes: CXXFLAGS=-O3 CFLAGS=-O3Compiler Notes: --disable-multilib --enable-checking=releaseDisk Notes: NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemandSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB Protection
Testing initiated at 21 July 2018 06:35 by user phoronix.
GCC 9.0 SVN: -O3 native Environment Notes: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=nativeCompiler Notes: --disable-multilib --enable-checking=releaseDisk Notes: NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemandSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB Protection
Testing initiated at 20 July 2018 21:03 by user phoronix.
GCC 9.0 SVN: -O3 native + PGO Processor: AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads), Motherboard: TYAN B8026T70AE24HR (V1.02.B10 BIOS), Chipset: AMD Family 17h, Memory: 129024MB, Disk: 280GB INTEL SSDPE21D280GA, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom Limited NetXtreme BCM5720 Gigabit PCIe
OS: Ubuntu 18.10, Kernel: 4.16.0-041600-generic (x86_64), Compiler: GCC 9.0.0 20180720, File-System: ext4, Screen Resolution: 1920x1080
Environment Notes: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=nativeCompiler Notes: --disable-multilib --enable-checking=releaseDisk Notes: NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemandSecurity Notes: __user pointer sanitization + Full AMD retpoline IBPB Protection
Testing initiated at 21 July 2018 01:40 by user phoronix.