zz AMD Ryzen 7 5700G testing with a ASUS TUF GAMING B550M-PLUS (WI-FI) (2423 BIOS) and ASUS AMD Cezanne 512MB on Ubuntu 21.10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2204252-NE-ZZ773288056&grs&sro .
zz Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL Vulkan Compiler File-System Screen Resolution A B C AMD Ryzen 7 5700G @ 3.80GHz (8 Cores / 16 Threads) ASUS TUF GAMING B550M-PLUS (WI-FI) (2423 BIOS) AMD Renoir/Cezanne 16GB 1000GB Samsung SSD 980 PRO 1TB ASUS AMD Cezanne 512MB (2000/1800MHz) AMD Renoir Radeon HD Audio MX279 Realtek RTL8125 2.5GbE + Intel Wi-Fi 6 AX200 Ubuntu 21.10 5.16.0-051600rc8daily20220108-generic (x86_64) GNOME Shell 40.5 X Server 1.20.11 + Wayland 4.6 Mesa 22.0.0-devel (git-9cb9101 2022-01-08 impish-oibaf-ppa) (LLVM 13.0.0 DRM 3.44) 1.2.199 GCC 11.2.0 ext4 1920x1080 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-gcn/usr --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-build-config=bootstrap-lto-lean --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: 0xa50000c Python Details - Python 3.9.7 Security Details - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + 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
zz onnx: bertsquad-12 - CPU - Standard onednn: IP Shapes 3D - u8s8f32 - CPU onnx: GPT-2 - CPU - Standard onednn: Recurrent Neural Network Training - u8s8f32 - CPU onednn: Deconvolution Batch shapes_3d - u8s8f32 - CPU svt-av1: Preset 12 - Bosphorus 4K onednn: Deconvolution Batch shapes_1d - f32 - CPU avifenc: 10, Lossless onednn: Recurrent Neural Network Inference - u8s8f32 - CPU onednn: IP Shapes 1D - f32 - CPU onnx: ArcFace ResNet-100 - CPU - Standard svt-av1: Preset 12 - Bosphorus 1080p onnx: super-resolution-10 - CPU - Standard onednn: Deconvolution Batch shapes_3d - f32 - CPU svt-av1: Preset 8 - Bosphorus 1080p onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU svt-av1: Preset 4 - Bosphorus 1080p onednn: Recurrent Neural Network Inference - f32 - CPU onednn: IP Shapes 3D - f32 - CPU avifenc: 6, Lossless avifenc: 6 svt-av1: Preset 10 - Bosphorus 1080p onednn: IP Shapes 1D - u8s8f32 - CPU svt-av1: Preset 8 - Bosphorus 4K onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPU onednn: Matrix Multiply Batch Shapes Transformer - f32 - CPU avifenc: 0 onednn: Convolution Batch Shapes Auto - f32 - CPU onednn: Convolution Batch Shapes Auto - u8s8f32 - CPU svt-av1: Preset 10 - Bosphorus 4K onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPU onnx: yolov4 - CPU - Standard svt-av1: Preset 4 - Bosphorus 4K avifenc: 2 onednn: Recurrent Neural Network Training - f32 - CPU onednn: Deconvolution Batch shapes_1d - u8s8f32 - CPU onnx: fcn-resnet101-11 - CPU - Standard onednn: IP Shapes 1D - bf16bf16bf16 - CPU A B C 496 2.84523 6063 3580.24 2.91606 89.668 8.14632 5.733 2210.71 3.98732 1040 355.484 3273 7.03558 92.454 1.70968 6.035 2213.99 11.7131 12.537 10.148 189.409 1.39092 26.393 2213.88 4.48476 158.835 22.3508 23.4701 65.142 3584.64 299 1.981 74.461 3583.1 1.95484 47 495 2.60928 6062 3577.41 3.01126 86.968 8.36734 5.59 2214.51 4.05032 1036 350.775 3309 7.02718 93.619 1.68982 6.044 2219.82 11.7333 12.5 10.165 188.79 1.3903 26.321 2225.46 4.48147 158.875 22.3987 23.5615 64.923 3577.26 298 1.984 74.241 3580.17 1.95319 47 788 2.6052 6545 3850.32 2.86797 89.686 8.25826 5.635 2265.57 4.07036 1056 349.096 3263 7.12334 92.819 1.70049 6.091 2232.66 11.7967 12.589 10.219 190.064 1.39934 26.25 2214.16 4.50371 158.147 22.4506 23.4575 65.193 3591.14 299 1.978 74.275 3585.22 1.95551 47 OpenBenchmarking.org
ONNX Runtime Model: bertsquad-12 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.11 Model: bertsquad-12 - Device: CPU - Executor: Standard A B C 200 400 600 800 1000 496 495 788 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt
oneDNN Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU A B C 0.6402 1.2804 1.9206 2.5608 3.201 2.84523 2.60928 2.60520 MIN: 2.65 MIN: 2.53 MIN: 2.52 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
ONNX Runtime Model: GPT-2 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.11 Model: GPT-2 - Device: CPU - Executor: Standard A B C 1400 2800 4200 5600 7000 6063 6062 6545 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt
oneDNN Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU A B C 800 1600 2400 3200 4000 3580.24 3577.41 3850.32 MIN: 3565.15 MIN: 3567.8 MIN: 3827.91 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU A B C 0.6775 1.355 2.0325 2.71 3.3875 2.91606 3.01126 2.86797 MIN: 2.8 MIN: 2.84 MIN: 2.79 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
SVT-AV1 Encoder Mode: Preset 12 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.0 Encoder Mode: Preset 12 - Input: Bosphorus 4K A B C 20 40 60 80 100 89.67 86.97 89.69 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU A B C 2 4 6 8 10 8.14632 8.36734 8.25826 MIN: 5.99 MIN: 6.04 MIN: 6.03 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
libavif avifenc Encoder Speed: 10, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.10 Encoder Speed: 10, Lossless A B C 1.2899 2.5798 3.8697 5.1596 6.4495 5.733 5.590 5.635 1. (CXX) g++ options: -O3 -fPIC -lm
oneDNN Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU A B C 500 1000 1500 2000 2500 2210.71 2214.51 2265.57 MIN: 2195.33 MIN: 2204.95 MIN: 2245.96 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
oneDNN Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU A B C 0.9158 1.8316 2.7474 3.6632 4.579 3.98732 4.05032 4.07036 MIN: 3.82 MIN: 3.86 MIN: 3.91 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
ONNX Runtime Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.11 Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard A B C 200 400 600 800 1000 1040 1036 1056 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt
SVT-AV1 Encoder Mode: Preset 12 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.0 Encoder Mode: Preset 12 - Input: Bosphorus 1080p A B C 80 160 240 320 400 355.48 350.78 349.10 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
ONNX Runtime Model: super-resolution-10 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.11 Model: super-resolution-10 - Device: CPU - Executor: Standard A B C 700 1400 2100 2800 3500 3273 3309 3263 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt
oneDNN Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU A B C 2 4 6 8 10 7.03558 7.02718 7.12334 MIN: 6.84 MIN: 6.83 MIN: 6.89 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.0 Encoder Mode: Preset 8 - Input: Bosphorus 1080p A B C 20 40 60 80 100 92.45 93.62 92.82 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU A B C 0.3847 0.7694 1.1541 1.5388 1.9235 1.70968 1.68982 1.70049 MIN: 1.65 MIN: 1.63 MIN: 1.64 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
SVT-AV1 Encoder Mode: Preset 4 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.0 Encoder Mode: Preset 4 - Input: Bosphorus 1080p A B C 2 4 6 8 10 6.035 6.044 6.091 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
oneDNN Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU A B C 500 1000 1500 2000 2500 2213.99 2219.82 2232.66 MIN: 2203.86 MIN: 2206.92 MIN: 2221.51 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
oneDNN Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU A B C 3 6 9 12 15 11.71 11.73 11.80 MIN: 11.67 MIN: 11.68 MIN: 11.74 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
libavif avifenc Encoder Speed: 6, Lossless OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.10 Encoder Speed: 6, Lossless A B C 3 6 9 12 15 12.54 12.50 12.59 1. (CXX) g++ options: -O3 -fPIC -lm
libavif avifenc Encoder Speed: 6 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.10 Encoder Speed: 6 A B C 3 6 9 12 15 10.15 10.17 10.22 1. (CXX) g++ options: -O3 -fPIC -lm
SVT-AV1 Encoder Mode: Preset 10 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.0 Encoder Mode: Preset 10 - Input: Bosphorus 1080p A B C 40 80 120 160 200 189.41 188.79 190.06 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
oneDNN Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU A B C 0.3149 0.6298 0.9447 1.2596 1.5745 1.39092 1.39030 1.39934 MIN: 1.33 MIN: 1.33 MIN: 1.33 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
SVT-AV1 Encoder Mode: Preset 8 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.0 Encoder Mode: Preset 8 - Input: Bosphorus 4K A B C 6 12 18 24 30 26.39 26.32 26.25 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
oneDNN Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU A B C 500 1000 1500 2000 2500 2213.88 2225.46 2214.16 MIN: 2205.37 MIN: 2205.78 MIN: 2198.56 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
oneDNN Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU A B C 1.0133 2.0266 3.0399 4.0532 5.0665 4.48476 4.48147 4.50371 MIN: 4.36 MIN: 4.37 MIN: 4.38 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
libavif avifenc Encoder Speed: 0 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.10 Encoder Speed: 0 A B C 40 80 120 160 200 158.84 158.88 158.15 1. (CXX) g++ options: -O3 -fPIC -lm
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU A B C 5 10 15 20 25 22.35 22.40 22.45 MIN: 22 MIN: 22.05 MIN: 22.1 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
oneDNN Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU A B C 6 12 18 24 30 23.47 23.56 23.46 MIN: 23.09 MIN: 23.18 MIN: 22.97 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
SVT-AV1 Encoder Mode: Preset 10 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.0 Encoder Mode: Preset 10 - Input: Bosphorus 4K A B C 15 30 45 60 75 65.14 64.92 65.19 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
oneDNN Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU A B C 800 1600 2400 3200 4000 3584.64 3577.26 3591.14 MIN: 3574.59 MIN: 3566.21 MIN: 3578.53 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
ONNX Runtime Model: yolov4 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.11 Model: yolov4 - Device: CPU - Executor: Standard A B C 70 140 210 280 350 299 298 299 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt
SVT-AV1 Encoder Mode: Preset 4 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 1.0 Encoder Mode: Preset 4 - Input: Bosphorus 4K A B C 0.4464 0.8928 1.3392 1.7856 2.232 1.981 1.984 1.978 1. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
libavif avifenc Encoder Speed: 2 OpenBenchmarking.org Seconds, Fewer Is Better libavif avifenc 0.10 Encoder Speed: 2 A B C 20 40 60 80 100 74.46 74.24 74.28 1. (CXX) g++ options: -O3 -fPIC -lm
oneDNN Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU A B C 800 1600 2400 3200 4000 3583.10 3580.17 3585.22 MIN: 3568.77 MIN: 3567.88 MIN: 3573.75 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
oneDNN Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU A B C 0.44 0.88 1.32 1.76 2.2 1.95484 1.95319 1.95551 MIN: 1.91 MIN: 1.9 MIN: 1.91 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
ONNX Runtime Model: fcn-resnet101-11 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.11 Model: fcn-resnet101-11 - Device: CPU - Executor: Standard A B C 11 22 33 44 55 47 47 47 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt
Phoronix Test Suite v10.8.5