AMD Open64 Compiler On Bulldozer FX AMD FX-8150 Bulldozer benchmarking with GCC and LLVM/Clang and the latest AMD Open64 4.2.5.2-1 compiler from August 2012. Benchmarking by Michael Larabel for a future article on Phoronix.com.
HTML result view exported from: https://openbenchmarking.org/result/1209219-RA-AMDOPEN6425&grw&sor .
AMD Open64 Compiler On Bulldozer FX Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Compiler File-System Screen Resolution GCC 4.6.3 GCC 4.7.1 LLVM-Clang 3.0 AMD Open64 4.5.2-1 AMD FX-8150 Eight-Core @ 3.60GHz (8 Cores) ASUS Crosshair V Formula AMD ATI RD890 bridge 4096MB 60GB OCZ VERTEX2 NVIDIA GeForce 9600 GSO 512MB (399/399MHz) Realtek ALC889 DELL P2210H Intel 82583V Gigabit Connection Ubuntu 12.10 3.5.0-14-generic (x86_64) Unity 6.4.0 X Server 1.13.0 nouveau 1.0.1 3.0 Mesa 8.1-devel Gallium 0.4 GCC 4.6.3 ext4 1920x1080 GCC 4.7.1 Clang 3.0-6ubuntu3 Open64 4.5.2 OpenBenchmarking.org Compiler Details - GCC 4.6.3, GCC 4.7.1: --disable-multilib --enable-checking=release --enable-languages=c,c++,fortran Processor Details - Scaling Governor: ondemand System Details - Compiz was running on this system.
AMD Open64 Compiler On Bulldozer FX scimark2: Monte Carlo scimark2: Fast Fourier Transform scimark2: Sparse Matrix Multiply scimark2: Dense LU Matrix Factorization scimark2: Jacobi Successive Over-Relaxation mafft: Multiple Sequence Alignment compress-7zip: Compress Speed Test john-the-ripper: Blowfish GCC 4.6.3 GCC 4.7.1 LLVM-Clang 3.0 AMD Open64 4.5.2-1 375.27 77.09 986.92 1763.21 758.02 7.06 17497 6346 280.72 74.73 964.97 1746.75 743.24 6.87 18053 6327 372.72 81.97 969.15 1754.40 1228.94 7.74 17968 788 307.90 74.09 965.43 1993.97 920.09 6.91 18102 4982 OpenBenchmarking.org
SciMark Computational Test: Monte Carlo OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Monte Carlo GCC 4.6.3 LLVM-Clang 3.0 AMD Open64 4.5.2-1 GCC 4.7.1 80 160 240 320 400 SE +/- 4.58, N = 4 SE +/- 6.45, N = 8 SE +/- 1.41, N = 4 SE +/- 0.87, N = 4 375.27 372.72 307.90 280.72
SciMark Computational Test: Fast Fourier Transform OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Fast Fourier Transform LLVM-Clang 3.0 GCC 4.6.3 GCC 4.7.1 AMD Open64 4.5.2-1 20 40 60 80 100 SE +/- 0.46, N = 4 SE +/- 0.37, N = 4 SE +/- 0.31, N = 4 SE +/- 0.36, N = 4 81.97 77.09 74.73 74.09
SciMark Computational Test: Sparse Matrix Multiply OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Sparse Matrix Multiply GCC 4.6.3 LLVM-Clang 3.0 AMD Open64 4.5.2-1 GCC 4.7.1 200 400 600 800 1000 SE +/- 13.26, N = 4 SE +/- 16.59, N = 4 SE +/- 12.98, N = 4 SE +/- 4.69, N = 4 986.92 969.15 965.43 964.97
SciMark Computational Test: Dense LU Matrix Factorization OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Dense LU Matrix Factorization AMD Open64 4.5.2-1 GCC 4.6.3 LLVM-Clang 3.0 GCC 4.7.1 400 800 1200 1600 2000 SE +/- 11.59, N = 4 SE +/- 8.41, N = 4 SE +/- 2.36, N = 4 SE +/- 5.04, N = 3 1993.97 1763.21 1754.40 1746.75
SciMark Computational Test: Jacobi Successive Over-Relaxation OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation LLVM-Clang 3.0 AMD Open64 4.5.2-1 GCC 4.6.3 GCC 4.7.1 300 600 900 1200 1500 SE +/- 7.48, N = 4 SE +/- 5.43, N = 4 SE +/- 4.63, N = 4 SE +/- 0.90, N = 4 1228.94 920.09 758.02 743.24
Timed MAFFT Alignment Multiple Sequence Alignment OpenBenchmarking.org Seconds, Fewer Is Better Timed MAFFT Alignment 6.864 Multiple Sequence Alignment GCC 4.7.1 AMD Open64 4.5.2-1 GCC 4.6.3 LLVM-Clang 3.0 2 4 6 8 10 SE +/- 0.05, N = 3 SE +/- 0.13, N = 6 SE +/- 0.10, N = 6 SE +/- 0.19, N = 6 6.87 6.91 7.06 7.74 1. (CC) gcc options: -O3 -lm -lpthread
7-Zip Compression Compress Speed Test OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 9.20.1 Compress Speed Test AMD Open64 4.5.2-1 GCC 4.7.1 LLVM-Clang 3.0 GCC 4.6.3 4K 8K 12K 16K 20K SE +/- 30.33, N = 3 SE +/- 104.69, N = 3 SE +/- 30.51, N = 3 SE +/- 24.50, N = 3 18102 18053 17968 17497 1. (CXX) g++ options: -pipe -lpthread
John The Ripper Test: Blowfish OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.7.9 Test: Blowfish GCC 4.6.3 GCC 4.7.1 AMD Open64 4.5.2-1 LLVM-Clang 3.0 1400 2800 4200 5600 7000 SE +/- 4.33, N = 3 SE +/- 4.33, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 6346 6327 4982 788 1. (CC) gcc options: -fopenmp -lcrypt
Phoronix Test Suite v10.8.5