Clang 6.0 AMD EPYC Tuning Comparison vm-other Xen 4.9.0 Hypervisor testing on Ubuntu 17.10 via the Phoronix Test Suite. -march=znver1: Processor: AMD EPYC 7601 32-Core @ 2.20GHz (64 Cores), Motherboard: TYAN B8026T70AE24HR, Chipset: AMD Device 1450, Memory: 126976MB, Disk: 280GB INTEL SSDPE21D280GA, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom Limited NetXtreme BCM5720 Gigabit PCIe OS: Ubuntu 17.10, Kernel: 4.13.0-21-generic (x86_64), Desktop: GNOME Shell 3.26.1, Display Driver: modesetting 1.19.5, OpenCL: OpenCL 1.2 pocl 1.0 LLVM 5.0.0, Compiler: Clang 6.0.0 (SVN 321623) + LLVM 6.0.0svn, File-System: ext4, Screen Resolution: 1920x1080, System Layer: vm-other Xen 4.9.0 Hypervisor -march=haswell: Processor: AMD EPYC 7601 32-Core @ 2.20GHz (64 Cores), Motherboard: TYAN B8026T70AE24HR, Chipset: AMD Device 1450, Memory: 126976MB, Disk: 280GB INTEL SSDPE21D280GA, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom Limited NetXtreme BCM5720 Gigabit PCIe OS: Ubuntu 17.10, Kernel: 4.13.0-21-generic (x86_64), Desktop: GNOME Shell 3.26.1, Display Driver: modesetting 1.19.5, OpenCL: OpenCL 1.2 pocl 1.0 LLVM 5.0.0, Compiler: Clang 6.0.0 (SVN 321623) + LLVM 6.0.0svn, File-System: ext4, Screen Resolution: 1920x1080, System Layer: vm-other Xen 4.9.0 Hypervisor -march=x86-64: Processor: AMD EPYC 7601 32-Core @ 2.20GHz (64 Cores), Motherboard: TYAN B8026T70AE24HR, Chipset: AMD Device 1450, Memory: 126976MB, Disk: 280GB INTEL SSDPE21D280GA, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom Limited NetXtreme BCM5720 Gigabit PCIe OS: Ubuntu 17.10, Kernel: 4.13.0-21-generic (x86_64), Desktop: GNOME Shell 3.26.1, Display Driver: modesetting 1.19.5, OpenCL: OpenCL 1.2 pocl 1.0 LLVM 5.0.0, Compiler: Clang 6.0.0 (SVN 321623) + LLVM 6.0.0svn, File-System: ext4, Screen Resolution: 1920x1080, System Layer: vm-other Xen 4.9.0 Hypervisor SQLite 3.8.10.2 Test Target: Default Test Directory Seconds < Lower Is Better -march=znver1 .. 7.48 |======================================================== -march=haswell . 7.28 |====================================================== -march=x86-64 .. 7.53 |======================================================== PolyBench-C 3.2 Test: 3 Matrix Multiplications Seconds < Lower Is Better -march=znver1 .. 62.75 |======================================================= -march=haswell . 62.33 |====================================================== -march=x86-64 .. 62.98 |======================================================= FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 4096 Mflops > Higher Is Better -march=znver1 .. 5031.60 |===================================================== -march=haswell . 4839.53 |=================================================== -march=x86-64 .. 4660.83 |================================================= FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 Mflops > Higher Is Better -march=znver1 .. 12481 |================================================== -march=haswell . 12393 |================================================== -march=x86-64 .. 13649 |======================================================= Timed HMMer Search 2.3.2 Pfam Database Search Seconds < Lower Is Better -march=znver1 .. 11.09 |============================================ -march=haswell . 13.83 |======================================================= -march=x86-64 .. 12.85 |=================================================== SciMark 2.0 Computational Test: Composite Mflops > Higher Is Better -march=znver1 .. 1699.32 |==================================================== -march=haswell . 1739.34 |===================================================== -march=x86-64 .. 1479.53 |============================================= SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better -march=znver1 .. 552.19 |====================================================== -march=haswell . 555.76 |====================================================== -march=x86-64 .. 531.38 |==================================================== SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better -march=znver1 .. 226.68 |====================================================== -march=haswell . 226.73 |====================================================== -march=x86-64 .. 179.29 |=========================================== SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better -march=znver1 .. 2258.64 |===================================================== -march=haswell . 2207.31 |==================================================== -march=x86-64 .. 2190.10 |=================================================== SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better -march=znver1 .. 4034.89 |================================================== -march=haswell . 4285.18 |===================================================== -march=x86-64 .. 3190.43 |======================================= SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better -march=znver1 .. 1424.21 |===================================================== -march=haswell . 1421.72 |===================================================== -march=x86-64 .. 1110.65 |========================================= TSCP 1.81 AI Chess Performance Nodes Per Second > Higher Is Better -march=znver1 .. 918269 |====================================================== -march=haswell . 917963 |====================================================== -march=x86-64 .. 917658 |====================================================== GraphicsMagick 1.3.19 Operation: Blur Iterations Per Minute > Higher Is Better -march=znver1 .. 104 |========================================================= -march=haswell . 102 |======================================================== -march=x86-64 .. 101 |======================================================= GraphicsMagick 1.3.19 Operation: Sharpen Iterations Per Minute > Higher Is Better -march=znver1 .. 136 |========================================================= -march=haswell . 135 |========================================================= -march=x86-64 .. 131 |======================================================= GraphicsMagick 1.3.19 Operation: HWB Color Space Iterations Per Minute > Higher Is Better -march=znver1 .. 155 |========================================================= -march=haswell . 155 |========================================================= -march=x86-64 .. 150 |======================================================= GraphicsMagick 1.3.19 Operation: Local Adaptive Thresholding Iterations Per Minute > Higher Is Better -march=znver1 .. 98 |========================================================== -march=haswell . 94 |======================================================== -march=x86-64 .. 97 |========================================================= Himeno Benchmark 3.0 Poisson Pressure Solver MFLOPS > Higher Is Better -march=znver1 .. 1052.47 |===================================================== -march=haswell . 1031.68 |==================================================== -march=x86-64 .. 1032.71 |==================================================== ebizzy 0.3 Records/s > Higher Is Better -march=znver1 .. 1145405 |===================================================== -march=haswell . 1120570 |==================================================== -march=x86-64 .. 1076648 |================================================== C-Ray 1.1 Total Time Seconds < Lower Is Better -march=znver1 .. 4.48 |======================================================= -march=haswell . 4.49 |======================================================== -march=x86-64 .. 4.53 |======================================================== Bullet Physics Engine 2.81 Test: Raytests Seconds < Lower Is Better -march=znver1 .. 3.18 |======================================================= -march=haswell . 3.08 |====================================================== -march=x86-64 .. 3.22 |======================================================== Bullet Physics Engine 2.81 Test: 3000 Fall Seconds < Lower Is Better -march=znver1 .. 5.34 |======================================================= -march=haswell . 5.38 |======================================================= -march=x86-64 .. 5.48 |======================================================== Bullet Physics Engine 2.81 Test: 1000 Stack Seconds < Lower Is Better -march=znver1 .. 6.08 |====================================================== -march=haswell . 6.05 |====================================================== -march=x86-64 .. 6.30 |======================================================== Bullet Physics Engine 2.81 Test: 1000 Convex Seconds < Lower Is Better -march=znver1 .. 5.31 |======================================================= -march=haswell . 5.01 |==================================================== -march=x86-64 .. 5.43 |======================================================== Bullet Physics Engine 2.81 Test: 136 Ragdolls Seconds < Lower Is Better -march=znver1 .. 3.23 |======================================================= -march=haswell . 3.23 |======================================================= -march=x86-64 .. 3.28 |======================================================== Bullet Physics Engine 2.81 Test: Prim Trimesh Seconds < Lower Is Better -march=znver1 .. 1.09 |======================================================= -march=haswell . 1.09 |======================================================= -march=x86-64 .. 1.10 |======================================================== Bullet Physics Engine 2.81 Test: Convex Trimesh Seconds < Lower Is Better -march=znver1 .. 1.32 |======================================================== -march=haswell . 1.26 |===================================================== -march=x86-64 .. 1.33 |======================================================== FLAC Audio Encoding 1.3.1 WAV To FLAC Seconds < Lower Is Better -march=znver1 .. 6.63 |=============================================== -march=haswell . 6.69 |=============================================== -march=x86-64 .. 7.94 |======================================================== LAME MP3 Encoding 3.99.5 WAV To MP3 Seconds < Lower Is Better -march=znver1 .. 12.81 |======================================================= -march=haswell . 12.83 |======================================================= -march=x86-64 .. 11.33 |================================================= Apache Benchmark 2.4.7 Static Web Page Serving Requests Per Second > Higher Is Better -march=znver1 .. 9663.93 |===================================================== -march=haswell . 9410.04 |==================================================== -march=x86-64 .. 9531.43 |====================================================