3300X oneDNN SVT Stuff

AMD Ryzen 3 3300X 4-Core testing with a MSI B350M GAMING PRO (MS-7A39) v1.0 (2.NR BIOS) and AMD FirePro V3800 512MB on Ubuntu 20.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2103158-HA-3300XONED31&grs&sro.

3300X oneDNN SVT StuffProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLCompilerFile-SystemScreen Resolution123AMD Ryzen 3 3300X 4-Core @ 3.80GHz (4 Cores / 8 Threads)MSI B350M GAMING PRO (MS-7A39) v1.0 (2.NR BIOS)AMD Starship/Matisse8GB256GB INTEL SSDPEKKW256G7AMD FirePro V3800 512MBAMD Redwood HDMI AudioDELL S2409WRealtek RTL8111/8168/8411Ubuntu 20.045.9.0-rc5-14sep-patch (x86_64) 20200914GNOME Shell 3.36.4X Server 1.20.93.3 Mesa 20.0.8 (LLVM 10.0.0)GCC 9.3.0ext41920x1080OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler 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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/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 ondemand (Boost: Enabled) - CPU Microcode: 0x8701021 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 STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

3300X oneDNN SVT Stuffonednn: IP Shapes 3D - f32 - CPUonednn: IP Shapes 3D - u8s8f32 - CPUsvt-vp9: VMAF Optimized - Bosphorus 1080psysbench: RAM / Memoryonednn: IP Shapes 1D - u8s8f32 - CPUonednn: Convolution Batch Shapes Auto - f32 - CPUsvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080ponednn: Convolution Batch Shapes Auto - u8s8f32 - CPUonednn: IP Shapes 1D - f32 - CPUonednn: Deconvolution Batch shapes_3d - f32 - CPUsvt-vp9: Visual Quality Optimized - Bosphorus 1080ponednn: Recurrent Neural Network Training - u8s8f32 - CPUonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: Recurrent Neural Network Inference - u8s8f32 - CPUonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUsvt-hevc: 1 - Bosphorus 1080ponednn: Recurrent Neural Network Training - f32 - CPUonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUsvt-hevc: 7 - Bosphorus 1080psvt-hevc: 10 - Bosphorus 1080psysbench: CPUonednn: Deconvolution Batch shapes_1d - f32 - CPU12310.40172.73839108.0016393.954.9175321.7320110.6722.65646.7805111.331587.766032.296.797253101.943113.936.098895.028276030.313109.224.035993.829.1579562.02133.198918.8211.0432010.47062.71709109.6216345.464.9084821.8783110.5722.53456.7509611.347188.036025.136.788503107.963110.226.116215.033406010.233118.204.045983.999.1729061.91133.048918.5513.009510.24412.68835109.1016159.084.8811521.7457111.2522.61506.7860011.295687.676008.786.771093113.523102.456.121415.016536015.973119.404.045981.799.1562061.92133.208921.3413.0919OpenBenchmarking.org

oneDNN

Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU1233691215SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 310.4010.4710.24MIN: 10.22MIN: 10.32MIN: 10.041. (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.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU1230.61611.23221.84832.46443.0805SE +/- 0.02085, N = 3SE +/- 0.01683, N = 3SE +/- 0.02173, N = 32.738392.717092.68835MIN: 2.39MIN: 2.39MIN: 2.341. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

SVT-VP9

Tuning: VMAF Optimized - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 1080p12320406080100SE +/- 1.12, N = 3SE +/- 0.39, N = 3SE +/- 0.22, N = 3108.00109.62109.101. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

Sysbench

Test: RAM / Memory

OpenBenchmarking.orgMiB/sec, More Is BetterSysbench 1.0.20Test: RAM / Memory1234K8K12K16K20KSE +/- 34.72, N = 3SE +/- 36.56, N = 3SE +/- 257.28, N = 316393.9516345.4616159.081. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm

oneDNN

Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU1231.10642.21283.31924.42565.532SE +/- 0.02875, N = 3SE +/- 0.02509, N = 3SE +/- 0.00783, N = 34.917534.908484.88115MIN: 4.74MIN: 4.73MIN: 4.741. (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.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU123510152025SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 321.7321.8821.75MIN: 21.32MIN: 21.34MIN: 21.341. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

SVT-VP9

Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p12320406080100SE +/- 0.17, N = 3SE +/- 0.16, N = 3SE +/- 0.05, N = 3110.67110.57111.251. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU123510152025SE +/- 0.04, N = 3SE +/- 0.10, N = 3SE +/- 0.02, N = 322.6622.5322.62MIN: 21.83MIN: 21.82MIN: 21.871. (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.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU123246810SE +/- 0.00993, N = 3SE +/- 0.00381, N = 3SE +/- 0.03383, N = 36.780516.750966.78600MIN: 6.54MIN: 6.52MIN: 6.481. (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.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU1233691215SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 311.3311.3511.30MIN: 11.07MIN: 11.05MIN: 111. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

SVT-VP9

Tuning: Visual Quality Optimized - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: Visual Quality Optimized - Input: Bosphorus 1080p12320406080100SE +/- 0.12, N = 3SE +/- 0.19, N = 3SE +/- 0.09, N = 387.7688.0387.671. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

oneDNN

Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU12313002600390052006500SE +/- 9.85, N = 3SE +/- 1.80, N = 3SE +/- 9.07, N = 36032.296025.136008.78MIN: 5963.68MIN: 5961.24MIN: 5954.041. (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.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU123246810SE +/- 0.01459, N = 3SE +/- 0.02051, N = 3SE +/- 0.01342, N = 36.797256.788506.77109MIN: 6.54MIN: 6.55MIN: 6.541. (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.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU1237001400210028003500SE +/- 11.59, N = 3SE +/- 2.08, N = 3SE +/- 9.34, N = 33101.943107.963113.52MIN: 3058.41MIN: 3073MIN: 3078.021. (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.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU1237001400210028003500SE +/- 4.61, N = 3SE +/- 2.85, N = 3SE +/- 6.91, N = 33113.933110.223102.45MIN: 3078.41MIN: 3064.84MIN: 3067.541. (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.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU123246810SE +/- 0.02165, N = 3SE +/- 0.02430, N = 3SE +/- 0.02421, N = 36.098896.116216.12141MIN: 5.9MIN: 5.92MIN: 5.921. (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: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU1231.13252.2653.39754.535.6625SE +/- 0.01117, N = 3SE +/- 0.01604, N = 3SE +/- 0.00923, N = 35.028275.033405.01653MIN: 4.89MIN: 4.9MIN: 4.881. (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.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU12313002600390052006500SE +/- 2.70, N = 3SE +/- 7.25, N = 3SE +/- 17.06, N = 36030.316010.236015.97MIN: 5964.79MIN: 5940.42MIN: 5932.881. (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.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU1237001400210028003500SE +/- 5.50, N = 3SE +/- 6.83, N = 3SE +/- 4.28, N = 33109.223118.203119.40MIN: 3068.21MIN: 3081.35MIN: 3075.081. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

SVT-HEVC

Tuning: 1 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 1080p1230.9091.8182.7273.6364.545SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 34.034.044.041. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

oneDNN

Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU12313002600390052006500SE +/- 24.14, N = 3SE +/- 10.53, N = 3SE +/- 15.20, N = 35993.825983.995981.79MIN: 5896.31MIN: 5903.92MIN: 5900.151. (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.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU1233691215SE +/- 0.04201, N = 3SE +/- 0.02925, N = 3SE +/- 0.05208, N = 39.157959.172909.15620MIN: 8.8MIN: 8.82MIN: 8.831. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

SVT-HEVC

Tuning: 7 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080p1231428425670SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.10, N = 362.0261.9161.921. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

SVT-HEVC

Tuning: 10 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080p123306090120150SE +/- 0.16, N = 3SE +/- 0.44, N = 3SE +/- 0.09, N = 3133.19133.04133.201. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

Sysbench

Test: CPU

OpenBenchmarking.orgEvents Per Second, More Is BetterSysbench 1.0.20Test: CPU1232K4K6K8K10KSE +/- 2.49, N = 3SE +/- 2.00, N = 3SE +/- 0.88, N = 38918.828918.558921.341. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU1233691215SE +/- 0.48, N = 15SE +/- 0.03, N = 3SE +/- 0.05, N = 311.0413.0113.09MIN: 8.76MIN: 8.85MIN: 8.861. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl


Phoronix Test Suite v10.8.5