5900X sysbench onednn AMD Ryzen 9 5900X 12-Core testing with a ASUS ROG CROSSHAIR VIII HERO (3202 BIOS) and Sapphire AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 6GB on Ubuntu 20.10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2103138-SYST-5900XSY95&export=txt&rdt&grs .
5900X sysbench onednn Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL Vulkan Compiler File-System Screen Resolution 1 2 3 AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads) ASUS ROG CROSSHAIR VIII HERO (3202 BIOS) AMD Starship/Matisse 16GB 1000GB Sabrent Rocket 4.0 Plus Sapphire AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 6GB (1780/875MHz) AMD Navi 10 HDMI Audio ASUS VP28U Realtek RTL8125 2.5GbE + Intel I211 Ubuntu 20.10 5.12.0-051200rc2-generic (x86_64) 20210306 GNOME Shell 3.38.1 X Server 1.20.9 4.6 Mesa 20.2.1 (LLVM 11.0.0) 1.2.131 GCC 10.2.0 ext4 3840x2160 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details - Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009 Security Details - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected
5900X sysbench onednn onednn: IP Shapes 3D - f32 - CPU onednn: Convolution Batch Shapes Auto - u8s8f32 - CPU onednn: Convolution Batch Shapes Auto - f32 - CPU onednn: Recurrent Neural Network Inference - f32 - CPU onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPU onednn: Recurrent Neural Network Training - u8s8f32 - CPU onednn: Deconvolution Batch shapes_1d - u8s8f32 - CPU onednn: Recurrent Neural Network Training - f32 - CPU onednn: Recurrent Neural Network Inference - u8s8f32 - CPU onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPU onednn: IP Shapes 1D - f32 - CPU onednn: Deconvolution Batch shapes_3d - f32 - CPU onednn: Deconvolution Batch shapes_3d - u8s8f32 - CPU sysbench: RAM / Memory onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPU sysbench: CPU onednn: IP Shapes 1D - u8s8f32 - CPU onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU onednn: Deconvolution Batch shapes_1d - f32 - CPU onednn: IP Shapes 3D - u8s8f32 - CPU 1 2 3 8.62717 17.3541 16.1940 1672.05 2897.91 2904.05 1.31629 2890.59 1707.04 1697.37 3.75267 4.27928 2.01613 13950.10 0.760181 68475.65 1.06966 1.74083 5.53845 0.524499 7.53391 16.4657 15.7201 1689.61 2873.62 2901.37 1.30940 2892.53 1698.31 1701.82 3.77107 4.26624 2.01320 13926.31 0.758823 68421.77 1.06981 1.74073 5.32047 0.862216 7.53364 16.4945 15.6990 1679.32 2875.28 2924.95 1.31421 2905.72 1701.38 1693.14 3.77032 4.28406 2.00925 13917.98 0.760394 68422.23 1.07025 1.74045 5.28668 0.517520 OpenBenchmarking.org
oneDNN Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU 1 2 3 2 4 6 8 10 SE +/- 0.01236, N = 3 SE +/- 0.00401, N = 3 SE +/- 0.01646, N = 3 8.62717 7.53391 7.53364 MIN: 8.47 MIN: 7.33 MIN: 7.32 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU 1 2 3 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.08, N = 3 SE +/- 0.09, N = 3 17.35 16.47 16.49 MIN: 16.94 MIN: 15.93 MIN: 15.81 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU 1 2 3 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 16.19 15.72 15.70 MIN: 15.68 MIN: 15.38 MIN: 15.31 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU 1 2 3 400 800 1200 1600 2000 SE +/- 3.61, N = 3 SE +/- 6.12, N = 3 SE +/- 9.45, N = 3 1672.05 1689.61 1679.32 MIN: 1639.18 MIN: 1668.94 MIN: 1646.62 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU 1 2 3 600 1200 1800 2400 3000 SE +/- 4.58, N = 3 SE +/- 15.32, N = 3 SE +/- 9.27, N = 3 2897.91 2873.62 2875.28 MIN: 2882.62 MIN: 2840.96 MIN: 2845.11 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU 1 2 3 600 1200 1800 2400 3000 SE +/- 14.19, N = 3 SE +/- 9.96, N = 3 SE +/- 17.20, N = 3 2904.05 2901.37 2924.95 MIN: 2876.45 MIN: 2874.88 MIN: 2881.73 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU 1 2 3 0.2962 0.5924 0.8886 1.1848 1.481 SE +/- 0.00569, N = 3 SE +/- 0.00533, N = 3 SE +/- 0.00800, N = 3 1.31629 1.30940 1.31421 MIN: 1.22 MIN: 1.22 MIN: 1.22 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU 1 2 3 600 1200 1800 2400 3000 SE +/- 14.19, N = 3 SE +/- 10.81, N = 3 SE +/- 12.12, N = 3 2890.59 2892.53 2905.72 MIN: 2861.19 MIN: 2864.07 MIN: 2873.26 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU 1 2 3 400 800 1200 1600 2000 SE +/- 2.15, N = 3 SE +/- 3.80, N = 3 SE +/- 10.80, N = 3 1707.04 1698.31 1701.38 MIN: 1692.22 MIN: 1678.43 MIN: 1662.08 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU 1 2 3 400 800 1200 1600 2000 SE +/- 20.36, N = 3 SE +/- 1.80, N = 3 SE +/- 4.28, N = 3 1697.37 1701.82 1693.14 MIN: 1650.11 MIN: 1684.14 MIN: 1661.05 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU 1 2 3 0.8485 1.697 2.5455 3.394 4.2425 SE +/- 0.00813, N = 3 SE +/- 0.00424, N = 3 SE +/- 0.02135, N = 3 3.75267 3.77107 3.77032 MIN: 3.45 MIN: 3.45 MIN: 3.46 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU 1 2 3 0.9639 1.9278 2.8917 3.8556 4.8195 SE +/- 0.00766, N = 3 SE +/- 0.00980, N = 3 SE +/- 0.00550, N = 3 4.27928 4.26624 4.28406 MIN: 4.15 MIN: 4.15 MIN: 4.17 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU 1 2 3 0.4536 0.9072 1.3608 1.8144 2.268 SE +/- 0.00253, N = 3 SE +/- 0.00271, N = 3 SE +/- 0.00223, N = 3 2.01613 2.01320 2.00925 MIN: 1.94 MIN: 1.91 MIN: 1.78 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Sysbench Test: RAM / Memory OpenBenchmarking.org MiB/sec, More Is Better Sysbench 1.0.20 Test: RAM / Memory 1 2 3 3K 6K 9K 12K 15K SE +/- 13.75, N = 3 SE +/- 13.11, N = 3 SE +/- 16.45, N = 3 13950.10 13926.31 13917.98 1. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU 1 2 3 0.1711 0.3422 0.5133 0.6844 0.8555 SE +/- 0.001551, N = 3 SE +/- 0.001642, N = 3 SE +/- 0.004859, N = 3 0.760181 0.758823 0.760394 MIN: 0.71 MIN: 0.72 MIN: 0.72 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Sysbench Test: CPU OpenBenchmarking.org Events Per Second, More Is Better Sysbench 1.0.20 Test: CPU 1 2 3 15K 30K 45K 60K 75K SE +/- 66.39, N = 3 SE +/- 41.87, N = 3 SE +/- 46.63, N = 3 68475.65 68421.77 68422.23 1. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm
oneDNN Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU 1 2 3 0.2408 0.4816 0.7224 0.9632 1.204 SE +/- 0.00321, N = 3 SE +/- 0.00243, N = 3 SE +/- 0.00399, N = 3 1.06966 1.06981 1.07025 MIN: 1 MIN: 1 MIN: 0.99 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU 1 2 3 0.3917 0.7834 1.1751 1.5668 1.9585 SE +/- 0.00034, N = 3 SE +/- 0.00092, N = 3 SE +/- 0.00070, N = 3 1.74083 1.74073 1.74045 MIN: 1.63 MIN: 1.64 MIN: 1.63 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU 1 2 3 1.2462 2.4924 3.7386 4.9848 6.231 SE +/- 0.21782, N = 12 SE +/- 0.19863, N = 15 SE +/- 0.19023, N = 15 5.53845 5.32047 5.28668 MIN: 3.21 MIN: 3.23 MIN: 3.24 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
oneDNN Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.1.2 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU 1 2 3 0.194 0.388 0.582 0.776 0.97 SE +/- 0.002916, N = 3 SE +/- 0.348175, N = 15 SE +/- 0.003242, N = 3 0.524499 0.862216 0.517520 MIN: 0.47 MIN: 0.46 MIN: 0.47 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
Phoronix Test Suite v10.8.5