Clang znver1 Scheduler Model Tests AMD Ryzen 7 1700 Eight-Core testing with the new Zen znver1 scheduler model. Tests by Michael Larabel for a future article. Clang 4.0: -O3 -march=x86-64: Processor: AMD Ryzen 7 1700 Eight-Core @ 3.00GHz (16 Cores), Motherboard: MSI B350 TOMAHAWK (MS-7A34) v1.0, Chipset: AMD Device 1450, Memory: 16384MB, Disk: 120GB Samsung SSD 840, Graphics: HIS AMD Radeon HD 7750/8740 / R7 250E 1024MB, Audio: AMD Cape Verde/Pitcairn, Network: Realtek RTL8111/8168/8411 OS: Ubuntu 17.04, Kernel: 4.12.0-999-generic (x86_64) 20170622, Desktop: Unity 7.5.0, Display Driver: modesetting 1.19.3, Compiler: Clang 4.0.0-1ubuntu1, File-System: ext4, Screen Resolution: 1024x768 Clang 4.0: -O3 -march=znver1: Processor: AMD Ryzen 7 1700 Eight-Core @ 3.00GHz (16 Cores), Motherboard: MSI B350 TOMAHAWK (MS-7A34) v1.0, Chipset: AMD Device 1450, Memory: 16384MB, Disk: 120GB Samsung SSD 840, Graphics: HIS AMD Radeon HD 7750/8740 / R7 250E 1024MB, Audio: AMD Cape Verde/Pitcairn, Network: Realtek RTL8111/8168/8411 OS: Ubuntu 17.04, Kernel: 4.12.0-999-generic (x86_64) 20170622, Desktop: Unity 7.5.0, Display Driver: modesetting 1.19.3, Compiler: Clang 4.0.0-1ubuntu1, File-System: ext4, Screen Resolution: 1024x768 Clang SVN: -O3 -march=x86-64: Processor: AMD Ryzen 7 1700 Eight-Core @ 3.00GHz (16 Cores), Motherboard: MSI B350 TOMAHAWK (MS-7A34) v1.0, Chipset: AMD Device 1450, Memory: 16384MB, Disk: 120GB Samsung SSD 840, Graphics: HIS AMD Radeon HD 7750/8740 / R7 250E 1024MB, Audio: AMD Cape Verde/Pitcairn, Network: Realtek RTL8111/8168/8411 OS: Ubuntu 17.04, Kernel: 4.12.0-999-generic (x86_64) 20170622, Desktop: Unity 7.5.0, Display Driver: modesetting 1.19.3, Compiler: Clang 6.0.0 (SVN 308494), File-System: ext4, Screen Resolution: 1024x768 Clang SVN: -O3 -march=znver1: Processor: AMD Ryzen 7 1700 Eight-Core @ 3.00GHz (16 Cores), Motherboard: MSI B350 TOMAHAWK (MS-7A34) v1.0, Chipset: AMD Device 1450, Memory: 16384MB, Disk: 120GB Samsung SSD 840, Graphics: HIS AMD Radeon HD 7750/8740 / R7 250E 1024MB, Audio: AMD Cape Verde/Pitcairn, Network: Realtek RTL8111/8168/8411 OS: Ubuntu 17.04, Kernel: 4.12.0-999-generic (x86_64) 20170622, Desktop: Unity 7.5.0, Display Driver: modesetting 1.19.3, Compiler: Clang 6.0.0 (SVN 308494), File-System: ext4, Screen Resolution: 1024x768 FFTW 3.3.4 Build: Float + SSE - Size: 2D FFT Size 4096 Mflops > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 12975 |======================================= Clang 4.0: -O3 -march=znver1 . 13072 |======================================= Clang SVN: -O3 -march=x86-64 . 13690 |========================================= Clang SVN: -O3 -march=znver1 . 13398 |======================================== SciMark 2.0 Computational Test: Composite Mflops > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 2124.54 |====================================== Clang 4.0: -O3 -march=znver1 . 2174.86 |======================================= Clang SVN: -O3 -march=x86-64 . 2129.48 |====================================== Clang SVN: -O3 -march=znver1 . 2152.36 |======================================= SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 623.66 |====================================== Clang 4.0: -O3 -march=znver1 . 651.73 |======================================== Clang SVN: -O3 -march=x86-64 . 642.55 |======================================= Clang SVN: -O3 -march=znver1 . 657.96 |======================================== SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 134.39 |======================================== Clang 4.0: -O3 -march=znver1 . 135.84 |======================================== Clang SVN: -O3 -march=x86-64 . 133.36 |======================================= Clang SVN: -O3 -march=znver1 . 134.24 |======================================== SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 2654.48 |======================================= Clang 4.0: -O3 -march=znver1 . 2616.81 |====================================== Clang SVN: -O3 -march=x86-64 . 2623.75 |======================================= Clang SVN: -O3 -march=znver1 . 2597.54 |====================================== SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 5534.54 |===================================== Clang 4.0: -O3 -march=znver1 . 5792.78 |======================================= Clang SVN: -O3 -march=x86-64 . 5572.72 |====================================== Clang SVN: -O3 -march=znver1 . 5692.83 |====================================== SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 1675.63 |======================================= Clang 4.0: -O3 -march=znver1 . 1677.13 |======================================= Clang SVN: -O3 -march=x86-64 . 1675.03 |======================================= Clang SVN: -O3 -march=znver1 . 1679.21 |======================================= TSCP 1.81 AI Chess Performance Nodes Per Second > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 1054730 |===================================== Clang 4.0: -O3 -march=znver1 . 1019964 |==================================== Clang SVN: -O3 -march=x86-64 . 1103331 |======================================= Clang SVN: -O3 -march=znver1 . 1096303 |======================================= x264 2017-05-19 H.264 Video Encoding Frames Per Second > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 252.21 |======================================== Clang 4.0: -O3 -march=znver1 . 252.73 |======================================== Clang SVN: -O3 -march=x86-64 . 252.33 |======================================== Clang SVN: -O3 -march=znver1 . 252.52 |======================================== GraphicsMagick 1.3.19 Operation: Blur Iterations Per Minute > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 105 |================================ Clang 4.0: -O3 -march=znver1 . 101 |=============================== Clang SVN: -O3 -march=x86-64 . 136 |========================================== Clang SVN: -O3 -march=znver1 . 140 |=========================================== GraphicsMagick 1.3.19 Operation: Sharpen Iterations Per Minute > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 57 |================== Clang 4.0: -O3 -march=znver1 . 60 |=================== Clang SVN: -O3 -march=x86-64 . 137 |========================================== Clang SVN: -O3 -march=znver1 . 139 |=========================================== GraphicsMagick 1.3.19 Operation: Resizing Iterations Per Minute > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 132 |========================================== Clang 4.0: -O3 -march=znver1 . 135 |=========================================== GraphicsMagick 1.3.19 Operation: HWB Color Space Iterations Per Minute > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 163 |======================================= Clang 4.0: -O3 -march=znver1 . 148 |==================================== Clang SVN: -O3 -march=x86-64 . 173 |========================================== Clang SVN: -O3 -march=znver1 . 178 |=========================================== GraphicsMagick 1.3.19 Operation: Local Adaptive Thresholding Iterations Per Minute > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 133 |========================================== Clang 4.0: -O3 -march=znver1 . 134 |========================================== Clang SVN: -O3 -march=x86-64 . 133 |========================================== Clang SVN: -O3 -march=znver1 . 136 |=========================================== Himeno Benchmark 3.0 Poisson Pressure Solver MFLOPS > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 1114.03 |======================================= Clang 4.0: -O3 -march=znver1 . 1115.36 |======================================= Clang SVN: -O3 -march=x86-64 . 1110.92 |======================================= Clang SVN: -O3 -march=znver1 . 1022.65 |==================================== ebizzy 0.3 Records/s > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 408278 |======================================== Clang 4.0: -O3 -march=znver1 . 406802 |======================================== Clang SVN: -O3 -march=x86-64 . 408509 |======================================== Clang SVN: -O3 -march=znver1 . 382960 |===================================== C-Ray 1.1 Total Time Seconds < Lower Is Better Clang 4.0: -O3 -march=x86-64 . 14.72 |========================================= Clang 4.0: -O3 -march=znver1 . 14.62 |========================================= Clang SVN: -O3 -march=x86-64 . 13.68 |====================================== Clang SVN: -O3 -march=znver1 . 13.62 |====================================== Stockfish 2014-11-26 Total Time ms < Lower Is Better Clang 4.0: -O3 -march=x86-64 . 3704 |========================================== Clang 4.0: -O3 -march=znver1 . 3671 |========================================== Clang SVN: -O3 -march=x86-64 . 3710 |========================================== Clang SVN: -O3 -march=znver1 . 3670 |========================================== Bullet Physics Engine 2.81 Test: Raytests Seconds < Lower Is Better Clang 4.0: -O3 -march=x86-64 . 2.85 |========================================== Clang 4.0: -O3 -march=znver1 . 2.79 |========================================= Clang SVN: -O3 -march=x86-64 . 2.86 |========================================== Clang SVN: -O3 -march=znver1 . 2.78 |========================================= Bullet Physics Engine 2.81 Test: 3000 Fall Seconds < Lower Is Better Clang 4.0: -O3 -march=x86-64 . 4.62 |========================================== Clang 4.0: -O3 -march=znver1 . 4.53 |========================================= Clang SVN: -O3 -march=x86-64 . 4.62 |========================================== Clang SVN: -O3 -march=znver1 . 4.53 |========================================= Bullet Physics Engine 2.81 Test: 1000 Stack Seconds < Lower Is Better Clang 4.0: -O3 -march=x86-64 . 5.33 |========================================== Clang 4.0: -O3 -march=znver1 . 5.14 |========================================= Clang SVN: -O3 -march=x86-64 . 5.27 |========================================== Clang SVN: -O3 -march=znver1 . 5.10 |======================================== Bullet Physics Engine 2.81 Test: 1000 Convex Seconds < Lower Is Better Clang 4.0: -O3 -march=x86-64 . 4.89 |========================================== Clang 4.0: -O3 -march=znver1 . 4.69 |======================================== Clang SVN: -O3 -march=x86-64 . 4.84 |========================================== Clang SVN: -O3 -march=znver1 . 4.64 |======================================== Bullet Physics Engine 2.81 Test: 136 Ragdolls Seconds < Lower Is Better Clang 4.0: -O3 -march=x86-64 . 2.97 |========================================== Clang 4.0: -O3 -march=znver1 . 2.93 |========================================= Clang SVN: -O3 -march=x86-64 . 2.96 |========================================== Clang SVN: -O3 -march=znver1 . 2.93 |========================================= Bullet Physics Engine 2.81 Test: Prim Trimesh Seconds < Lower Is Better Clang 4.0: -O3 -march=x86-64 . 0.99 |========================================== Clang 4.0: -O3 -march=znver1 . 0.99 |========================================== Clang SVN: -O3 -march=x86-64 . 1.00 |========================================== Clang SVN: -O3 -march=znver1 . 0.99 |========================================== Bullet Physics Engine 2.81 Test: Convex Trimesh Seconds < Lower Is Better Clang 4.0: -O3 -march=x86-64 . 1.19 |========================================== Clang 4.0: -O3 -march=znver1 . 1.17 |========================================= Clang SVN: -O3 -march=x86-64 . 1.20 |========================================== Clang SVN: -O3 -march=znver1 . 1.16 |========================================= FLAC Audio Encoding 1.3.1 WAV To FLAC Seconds < Lower Is Better Clang 4.0: -O3 -march=x86-64 . 6.77 |========================================== Clang 4.0: -O3 -march=znver1 . 5.72 |=================================== Clang SVN: -O3 -march=x86-64 . 6.77 |========================================== Clang SVN: -O3 -march=znver1 . 5.69 |=================================== LAME MP3 Encoding 3.99.3 WAV To MP3 Seconds < Lower Is Better Clang 4.0: -O3 -march=x86-64 . 9.62 |==================================== Clang 4.0: -O3 -march=znver1 . 10.74 |======================================== Clang SVN: -O3 -march=x86-64 . 9.75 |===================================== Clang SVN: -O3 -march=znver1 . 10.92 |========================================= WavPack Audio Encoding 5.1 WAV To WavPack Seconds < Lower Is Better Clang 4.0: -O3 -march=x86-64 . 6.50 |========================================== Clang 4.0: -O3 -march=znver1 . 6.44 |========================================= Clang SVN: -O3 -march=x86-64 . 6.52 |========================================== Clang SVN: -O3 -march=znver1 . 6.45 |========================================== Tachyon 0.98.9 Total Time Seconds < Lower Is Better Clang 4.0: -O3 -march=x86-64 . 5.23 |========================================== Clang 4.0: -O3 -march=znver1 . 5.25 |========================================== Clang SVN: -O3 -march=x86-64 . 5.29 |========================================== Clang SVN: -O3 -march=znver1 . 5.29 |========================================== libjpeg-turbo tjbench 1.5.1 Test: Decompression Throughput Megapixels/sec > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 163.71 |======================================= Clang 4.0: -O3 -march=znver1 . 166.39 |======================================== Clang SVN: -O3 -march=x86-64 . 161.78 |======================================= Clang SVN: -O3 -march=znver1 . 165.30 |======================================== PostgreSQL pgbench 9.6.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write TPS > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 1899.66 |====================================== Clang 4.0: -O3 -march=znver1 . 1938.55 |======================================= Clang SVN: -O3 -march=x86-64 . 1865.09 |====================================== Clang SVN: -O3 -march=znver1 . 1813.34 |==================================== PostgreSQL pgbench 9.6.3 Scaling: Buffer Test - Test: Single Thread - Mode: Read Write TPS > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 226.16 |======================================== Clang 4.0: -O3 -march=znver1 . 226.21 |======================================== Clang SVN: -O3 -march=x86-64 . 226.04 |======================================== Clang SVN: -O3 -march=znver1 . 224.89 |======================================== PostgreSQL pgbench 9.6.3 Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Write TPS > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 2021.12 |====================================== Clang 4.0: -O3 -march=znver1 . 2061.89 |======================================= Clang SVN: -O3 -march=x86-64 . 1938.93 |===================================== Clang SVN: -O3 -march=znver1 . 1875.27 |=================================== Redis 3.0.1 Test: LPOP Requests Per Second > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 1890860.67 |=================================== Clang 4.0: -O3 -march=znver1 . 1908429.46 |=================================== Clang SVN: -O3 -march=x86-64 . 1968693.38 |==================================== Clang SVN: -O3 -march=znver1 . 1960971.50 |==================================== Redis 3.0.1 Test: GET Requests Per Second > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 1900142.80 |=================================== Clang 4.0: -O3 -march=znver1 . 1896426.08 |=================================== Clang SVN: -O3 -march=x86-64 . 1953368.38 |==================================== Clang SVN: -O3 -march=znver1 . 1973143.29 |==================================== Redis 3.0.1 Test: SET Requests Per Second > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 1415121.33 |==================================== Clang 4.0: -O3 -march=znver1 . 1400655.09 |==================================== Clang SVN: -O3 -march=x86-64 . 1385134.04 |=================================== Clang SVN: -O3 -march=znver1 . 1413713.29 |==================================== Hierarchical INTegration 1.0 Test: FLOAT QUIPs > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 301996595.09 |================================== Clang 4.0: -O3 -march=znver1 . 301686748.92 |================================== Clang SVN: -O3 -march=x86-64 . 301795000.85 |================================== Clang SVN: -O3 -march=znver1 . 300351844.05 |================================== Apache Benchmark 2.4.7 Static Web Page Serving Requests Per Second > Higher Is Better Clang 4.0: -O3 -march=x86-64 . 27221.41 |===================================== Clang 4.0: -O3 -march=znver1 . 26808.65 |===================================== Clang SVN: -O3 -march=x86-64 . 26994.80 |===================================== Clang SVN: -O3 -march=znver1 . 27700.93 |======================================