tigerlake oneDNN 2.0

Intel Core i7-1165G7 testing with a Dell 0GG9PT (1.0.3 BIOS) and Intel UHD 3GB on Ubuntu 20.10 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2012100-HA-TIGERLAKE50
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
1
December 09 2020
  56 Minutes
2
December 10 2020
  1 Hour, 19 Minutes
3
December 10 2020
  56 Minutes
4
December 10 2020
  55 Minutes
Invert Hiding All Results Option
  1 Hour, 1 Minute

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):

1

Compiler Notes: --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 Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x60 - Thermald 2.3
Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

2

3

4

Processor: Intel Core i7-1165G7 @ 4.70GHz (4 Cores / 8 Threads), Motherboard: Dell 0GG9PT (1.0.3 BIOS), Chipset: Intel Tiger Lake-LP, Memory: 16GB, Disk: Kioxia KBG40ZNS256G NVMe 256GB, Graphics: Intel UHD 3GB (1300MHz), Audio: Realtek ALC289, Network: Intel Wi-Fi 6 AX201

OS: Ubuntu 20.10, Kernel: 5.10.0-051000rc6-generic (x86_64) 20201129, Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.6 Mesa 21.0.0-devel (git-b3504f4 2020-11-29 groovy-oibaf-ppa), Vulkan: 1.2.145, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 1920x1200

tigerlake oneDNN 2.0ProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLVulkanCompilerFile-SystemScreen Resolution1234Intel Core i7-1165G7 @ 4.70GHz (4 Cores / 8 Threads)Dell 0GG9PT (1.0.3 BIOS)Intel Tiger Lake-LP16GBKioxia KBG40ZNS256G NVMe 256GBIntel UHD 3GB (1300MHz)Realtek ALC289Intel Wi-Fi 6 AX201Ubuntu 20.105.10.0-051000rc6-generic (x86_64) 20201129GNOME Shell 3.38.1X Server 1.20.9modesetting 1.20.94.6 Mesa 21.0.0-devel (git-b3504f4 2020-11-29 groovy-oibaf-ppa)1.2.145GCC 10.2.0ext41920x1200OpenBenchmarking.orgCompiler 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: intel_pstate powersave - CPU Microcode: 0x60 - Thermald 2.3Security 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

1234Result OverviewPhoronix Test Suite100%102%105%107%110%oneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNR.N.N.T - f32 - CPUM.M.B.S.T - u8s8f32 - CPUIP Shapes 3D - u8s8f32 - CPUM.M.B.S.T - f32 - CPUR.N.N.I - bf16bf16bf16 - CPUD.B.s - bf16bf16bf16 - CPUIP Shapes 3D - f32 - CPUC.B.S.A - u8s8f32 - CPUD.B.s - bf16bf16bf16 - CPUIP Shapes 1D - u8s8f32 - CPUR.N.N.I - u8s8f32 - CPUC.B.S.A - bf16bf16bf16 - CPUIP Shapes 1D - f32 - CPUM.M.B.S.T - bf16bf16bf16 - CPUD.B.s - f32 - CPUIP Shapes 1D - bf16bf16bf16 - CPUIP Shapes 3D - bf16bf16bf16 - CPUC.B.S.A - f32 - CPUR.N.N.I - f32 - CPUR.N.N.T - u8s8f32 - CPUD.B.s - u8s8f32 - CPUR.N.N.T - bf16bf16bf16 - CPUD.B.s - f32 - CPUD.B.s - u8s8f32 - CPU

tigerlake oneDNN 2.0onednn: IP Shapes 1D - f32 - CPUonednn: IP Shapes 3D - f32 - CPUonednn: IP Shapes 1D - u8s8f32 - CPUonednn: IP Shapes 3D - u8s8f32 - CPUonednn: IP Shapes 1D - bf16bf16bf16 - CPUonednn: IP Shapes 3D - bf16bf16bf16 - CPUonednn: Convolution Batch Shapes Auto - f32 - CPUonednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: Deconvolution Batch shapes_3d - f32 - CPUonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUonednn: Recurrent Neural Network Training - f32 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUonednn: Recurrent Neural Network Training - u8s8f32 - CPUonednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPUonednn: Deconvolution Batch shapes_1d - bf16bf16bf16 - CPUonednn: Deconvolution Batch shapes_3d - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - u8s8f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - bf16bf16bf16 - CPU123412.277657.076012.692193.1429126.36748.6131314.075414.832312.113010.083143.158892.874459288.614793.389287.6052.110961.883842.81384791.265.328869278.984805.802.9958512.678312.16246.982252.660393.0837126.22488.5871713.992314.755112.050510.000713.145632.8755810128.694784.919258.9451.768861.638642.28474735.505.237149221.604736.762.9379012.562812.224877.039232.665063.0835226.15688.5527013.993514.768312.0583610.008733.149772.869279261.914778.319271.6651.878161.000442.67294768.705.315269258.574798.912.9823812.679612.158187.069392.663663.1141426.21608.6188713.967514.776012.006299.950063.135822.867839221.824757.379219.3651.543261.837342.40754777.415.288239213.544773.782.9570712.6133OpenBenchmarking.org

oneDNN

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU12343691215SE +/- 0.47, N = 12SE +/- 0.47, N = 12SE +/- 0.48, N = 12SE +/- 0.48, N = 1212.2812.1612.2212.16MIN: 5.86MIN: 6.19MIN: 5.92MIN: 6.151. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU123448121620Min: 7.06 / Avg: 12.28 / Max: 12.84Min: 6.99 / Avg: 12.16 / Max: 12.95Min: 6.96 / Avg: 12.22 / Max: 13.03Min: 6.92 / Avg: 12.16 / Max: 12.821. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU1234246810SE +/- 0.11413, N = 3SE +/- 0.10165, N = 4SE +/- 0.08723, N = 5SE +/- 0.10291, N = 47.076016.982257.039237.06939MIN: 5.64MIN: 5.64MIN: 5.64MIN: 5.641. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU12343691215Min: 6.86 / Avg: 7.08 / Max: 7.26Min: 6.68 / Avg: 6.98 / Max: 7.11Min: 6.7 / Avg: 7.04 / Max: 7.21Min: 6.77 / Avg: 7.07 / Max: 7.231. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU12340.60571.21141.81712.42283.0285SE +/- 0.02299, N = 12SE +/- 0.02163, N = 13SE +/- 0.02319, N = 12SE +/- 0.02230, N = 122.692192.660392.665062.66366MIN: 1.52MIN: 1.49MIN: 1.51MIN: 1.521. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU1234246810Min: 2.44 / Avg: 2.69 / Max: 2.73Min: 2.4 / Avg: 2.66 / Max: 2.7Min: 2.41 / Avg: 2.67 / Max: 2.7Min: 2.42 / Avg: 2.66 / Max: 2.711. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU12340.70721.41442.12162.82883.536SE +/- 0.03162, N = 15SE +/- 0.04011, N = 12SE +/- 0.03418, N = 14SE +/- 0.03984, N = 123.142913.083713.083523.11414MIN: 2.15MIN: 2.14MIN: 2.14MIN: 2.141. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU1234246810Min: 2.76 / Avg: 3.14 / Max: 3.25Min: 2.68 / Avg: 3.08 / Max: 3.19Min: 2.71 / Avg: 3.08 / Max: 3.17Min: 2.69 / Avg: 3.11 / Max: 3.211. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU1234612182430SE +/- 0.51, N = 12SE +/- 0.52, N = 12SE +/- 0.52, N = 12SE +/- 0.53, N = 1226.3726.2226.1626.22MIN: 18.67MIN: 18.54MIN: 18.56MIN: 18.691. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU1234612182430Min: 20.78 / Avg: 26.37 / Max: 26.96Min: 20.47 / Avg: 26.22 / Max: 26.85Min: 20.44 / Avg: 26.16 / Max: 26.78Min: 20.43 / Avg: 26.22 / Max: 26.851. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU1234246810SE +/- 0.11949, N = 12SE +/- 0.12441, N = 12SE +/- 0.13060, N = 12SE +/- 0.12648, N = 128.613138.587178.552708.61887MIN: 4.82MIN: 4.9MIN: 4.89MIN: 4.891. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU12343691215Min: 7.33 / Avg: 8.61 / Max: 8.86Min: 7.25 / Avg: 8.59 / Max: 8.91Min: 7.13 / Avg: 8.55 / Max: 8.77Min: 7.24 / Avg: 8.62 / Max: 8.841. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU123448121620SE +/- 0.15, N = 12SE +/- 0.15, N = 12SE +/- 0.14, N = 12SE +/- 0.15, N = 1214.0813.9913.9913.97MIN: 8.28MIN: 8.27MIN: 8.32MIN: 8.261. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU123448121620Min: 12.43 / Avg: 14.08 / Max: 14.26Min: 12.32 / Avg: 13.99 / Max: 14.24Min: 12.45 / Avg: 13.99 / Max: 14.29Min: 12.27 / Avg: 13.97 / Max: 14.21. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU123448121620SE +/- 0.15, N = 3SE +/- 0.13, N = 3SE +/- 0.15, N = 3SE +/- 0.10, N = 314.8314.7614.7714.78MIN: 12.23MIN: 12.2MIN: 12.23MIN: 12.181. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU123448121620Min: 14.55 / Avg: 14.83 / Max: 15.05Min: 14.5 / Avg: 14.76 / Max: 14.89Min: 14.48 / Avg: 14.77 / Max: 14.97Min: 14.58 / Avg: 14.78 / Max: 14.921. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU12343691215SE +/- 0.18, N = 14SE +/- 0.16, N = 15SE +/- 0.16, N = 15SE +/- 0.17, N = 1512.1112.0512.0612.01MIN: 9.3MIN: 9.23MIN: 9.25MIN: 9.291. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU123448121620Min: 9.97 / Avg: 12.11 / Max: 12.43Min: 10.02 / Avg: 12.05 / Max: 12.29Min: 9.91 / Avg: 12.06 / Max: 12.33Min: 9.87 / Avg: 12.01 / Max: 12.251. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU12343691215SE +/- 0.08502, N = 13SE +/- 0.08538, N = 13SE +/- 0.09129, N = 10SE +/- 0.08167, N = 1210.0831410.0007110.008739.95006MIN: 8.02MIN: 7.96MIN: 8.02MIN: 81. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU12343691215Min: 9.08 / Avg: 10.08 / Max: 10.28Min: 8.99 / Avg: 10 / Max: 10.16Min: 9.19 / Avg: 10.01 / Max: 10.14Min: 9.06 / Avg: 9.95 / Max: 10.081. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU12340.71081.42162.13242.84323.554SE +/- 0.02155, N = 3SE +/- 0.02446, N = 3SE +/- 0.02100, N = 3SE +/- 0.02404, N = 33.158893.145633.149773.13582MIN: 2.6MIN: 2.56MIN: 2.56MIN: 2.521. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU1234246810Min: 3.12 / Avg: 3.16 / Max: 3.18Min: 3.1 / Avg: 3.15 / Max: 3.17Min: 3.11 / Avg: 3.15 / Max: 3.17Min: 3.09 / Avg: 3.14 / Max: 3.161. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU12340.6471.2941.9412.5883.235SE +/- 0.04516, N = 14SE +/- 0.04485, N = 14SE +/- 0.04145, N = 15SE +/- 0.04074, N = 152.874452.875582.869272.86783MIN: 2.2MIN: 2.2MIN: 2.2MIN: 2.21. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU1234246810Min: 2.33 / Avg: 2.87 / Max: 2.95Min: 2.32 / Avg: 2.88 / Max: 2.95Min: 2.33 / Avg: 2.87 / Max: 2.94Min: 2.34 / Avg: 2.87 / Max: 2.941. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU12342K4K6K8K10KSE +/- 10.13, N = 3SE +/- 838.11, N = 15SE +/- 13.30, N = 3SE +/- 25.29, N = 39288.6110128.699261.919221.82MIN: 9211.6MIN: 9177.53MIN: 9188.63MIN: 9131.621. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU12342K4K6K8K10KMin: 9269.35 / Avg: 9288.61 / Max: 9303.7Min: 9229.62 / Avg: 10128.69 / Max: 21861.5Min: 9247.19 / Avg: 9261.91 / Max: 9288.47Min: 9182.57 / Avg: 9221.82 / Max: 9269.061. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU123410002000300040005000SE +/- 3.63, N = 3SE +/- 1.10, N = 3SE +/- 11.08, N = 3SE +/- 3.42, N = 34793.384784.914778.314757.37MIN: 4739.2MIN: 4730.95MIN: 4711.53MIN: 4705.951. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU12348001600240032004000Min: 4789.44 / Avg: 4793.38 / Max: 4800.63Min: 4783.28 / Avg: 4784.91 / Max: 4786.99Min: 4756.8 / Avg: 4778.31 / Max: 4793.69Min: 4753.66 / Avg: 4757.37 / Max: 4764.211. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU12342K4K6K8K10KSE +/- 12.30, N = 3SE +/- 16.53, N = 3SE +/- 3.49, N = 3SE +/- 8.71, N = 39287.609258.949271.669219.36MIN: 9203.45MIN: 9178.94MIN: 9203.74MIN: 9154.531. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU123416003200480064008000Min: 9263.47 / Avg: 9287.6 / Max: 9303.86Min: 9229.46 / Avg: 9258.94 / Max: 9286.62Min: 9265.36 / Avg: 9271.66 / Max: 9277.41Min: 9201.94 / Avg: 9219.36 / Max: 9228.431. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU12341224364860SE +/- 0.77, N = 12SE +/- 0.76, N = 12SE +/- 0.79, N = 12SE +/- 0.81, N = 1252.1151.7751.8851.54MIN: 37.03MIN: 37.02MIN: 36.99MIN: 36.971. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU12341020304050Min: 43.61 / Avg: 52.11 / Max: 53.05Min: 43.44 / Avg: 51.77 / Max: 52.81Min: 43.24 / Avg: 51.88 / Max: 52.91Min: 42.63 / Avg: 51.54 / Max: 52.591. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU12341428425670SE +/- 0.54, N = 3SE +/- 0.78, N = 3SE +/- 0.68, N = 3SE +/- 0.66, N = 361.8861.6461.0061.84MIN: 51.24MIN: 51.14MIN: 49.68MIN: 51.041. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU12341224364860Min: 60.85 / Avg: 61.88 / Max: 62.64Min: 60.09 / Avg: 61.64 / Max: 62.6Min: 59.65 / Avg: 61 / Max: 61.83Min: 60.54 / Avg: 61.84 / Max: 62.671. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU12341020304050SE +/- 0.55, N = 15SE +/- 0.50, N = 15SE +/- 0.54, N = 15SE +/- 0.51, N = 1542.8142.2842.6742.41MIN: 37.08MIN: 37.05MIN: 37.1MIN: 37.071. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU1234918273645Min: 37.46 / Avg: 42.81 / Max: 43.77Min: 37.53 / Avg: 42.28 / Max: 43.17Min: 37.47 / Avg: 42.67 / Max: 43.69Min: 37.5 / Avg: 42.41 / Max: 43.351. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU123410002000300040005000SE +/- 5.01, N = 3SE +/- 8.32, N = 3SE +/- 10.39, N = 3SE +/- 11.63, N = 34791.264735.504768.704777.41MIN: 4737.32MIN: 4681.84MIN: 4703.52MIN: 4716.671. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU12348001600240032004000Min: 4785.38 / Avg: 4791.26 / Max: 4801.22Min: 4723.68 / Avg: 4735.5 / Max: 4751.56Min: 4749.83 / Avg: 4768.7 / Max: 4785.69Min: 4764.82 / Avg: 4777.41 / Max: 4800.641. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU12341.1992.3983.5974.7965.995SE +/- 0.02481, N = 3SE +/- 0.02123, N = 3SE +/- 0.01827, N = 3SE +/- 0.02650, N = 35.328865.237145.315265.28823MIN: 3.85MIN: 3.82MIN: 3.85MIN: 3.841. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU1234246810Min: 5.3 / Avg: 5.33 / Max: 5.38Min: 5.21 / Avg: 5.24 / Max: 5.28Min: 5.29 / Avg: 5.32 / Max: 5.35Min: 5.26 / Avg: 5.29 / Max: 5.341. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU12342K4K6K8K10KSE +/- 22.31, N = 3SE +/- 15.21, N = 3SE +/- 11.73, N = 3SE +/- 16.01, N = 39278.989221.609258.579213.54MIN: 9189.3MIN: 9153.44MIN: 9188.24MIN: 9114.951. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU123416003200480064008000Min: 9236.96 / Avg: 9278.98 / Max: 9312.97Min: 9196.56 / Avg: 9221.6 / Max: 9249.09Min: 9236.65 / Avg: 9258.57 / Max: 9276.79Min: 9187.65 / Avg: 9213.54 / Max: 9242.791. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU123410002000300040005000SE +/- 1.89, N = 3SE +/- 3.67, N = 3SE +/- 4.47, N = 3SE +/- 19.93, N = 34805.804736.764798.914773.78MIN: 4747.23MIN: 4687.27MIN: 4736.17MIN: 46941. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU12348001600240032004000Min: 4802.51 / Avg: 4805.8 / Max: 4809.04Min: 4729.44 / Avg: 4736.76 / Max: 4740.98Min: 4792.76 / Avg: 4798.91 / Max: 4807.61Min: 4742.99 / Avg: 4773.78 / Max: 4811.11. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU12340.67411.34822.02232.69643.3705SE +/- 0.01083, N = 3SE +/- 0.00606, N = 3SE +/- 0.01125, N = 3SE +/- 0.00577, N = 32.995852.937902.982382.95707MIN: 2.11MIN: 2.07MIN: 2.05MIN: 2.081. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU1234246810Min: 2.98 / Avg: 3 / Max: 3.02Min: 2.93 / Avg: 2.94 / Max: 2.95Min: 2.97 / Avg: 2.98 / Max: 3Min: 2.95 / Avg: 2.96 / Max: 2.971. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU12343691215SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 312.6812.5612.6812.61MIN: 11.4MIN: 11.16MIN: 11.16MIN: 11.151. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU123448121620Min: 12.62 / Avg: 12.68 / Max: 12.78Min: 12.53 / Avg: 12.56 / Max: 12.6Min: 12.64 / Avg: 12.68 / Max: 12.75Min: 12.56 / Avg: 12.61 / Max: 12.651. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

1

Compiler Notes: --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 Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x60 - Thermald 2.3
Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 9 December 2020 20:20 by user phoronix.

2

Compiler Notes: --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 Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x60 - Thermald 2.3
Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 10 December 2020 05:40 by user phoronix.

3

Compiler Notes: --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 Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x60 - Thermald 2.3
Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 10 December 2020 07:55 by user phoronix.

4

Processor: Intel Core i7-1165G7 @ 4.70GHz (4 Cores / 8 Threads), Motherboard: Dell 0GG9PT (1.0.3 BIOS), Chipset: Intel Tiger Lake-LP, Memory: 16GB, Disk: Kioxia KBG40ZNS256G NVMe 256GB, Graphics: Intel UHD 3GB (1300MHz), Audio: Realtek ALC289, Network: Intel Wi-Fi 6 AX201

OS: Ubuntu 20.10, Kernel: 5.10.0-051000rc6-generic (x86_64) 20201129, Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.6 Mesa 21.0.0-devel (git-b3504f4 2020-11-29 groovy-oibaf-ppa), Vulkan: 1.2.145, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 1920x1200

Compiler Notes: --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 Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x60 - Thermald 2.3
Security Notes: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 10 December 2020 09:23 by user phoronix.