sysbench oneDNN Ryzen 9 5950X

AMD Ryzen 9 5950X 16-Core testing with a ASUS ROG CROSSHAIR VIII HERO (WI-FI) (3204 BIOS) and llvmpipe 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 2103140-PTS-SYSBENCH79
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

Limit displaying results to tests within:

CPU Massive 2 Tests
Multi-Core 2 Tests
Server CPU Tests 2 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
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
Performance Per
Dollar
Date
Run
  Test
  Duration
1
March 13 2021
  40 Minutes
2
March 13 2021
  39 Minutes
3
March 13 2021
  39 Minutes
4
March 13 2021
  40 Minutes
5
March 13 2021
  40 Minutes
Invert Hiding All Results Option
  39 Minutes

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

Kernel Notes: Transparent Huge Pages: madvise
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: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009
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 Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

2

3

4

5

Processor: AMD Ryzen 9 5950X 16-Core @ 3.40GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR VIII HERO (WI-FI) (3204 BIOS), Chipset: AMD Starship/Matisse, Memory: 32GB, Disk: 2000GB Corsair Force MP600 + 2000GB, Graphics: llvmpipe, Audio: AMD Device ab28, Network: Realtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200

OS: Ubuntu 20.10, Kernel: 5.10.23-051023-generic (x86_64), Desktop: GNOME Shell 3.38.2, Display Server: X Server 1.20.9, OpenGL: 4.5 Mesa 21.1.0-devel (git-684f97d 2021-03-12 groovy-oibaf-ppa) (LLVM 11.0.1 256 bits), Vulkan: 1.0.168, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160

sysbench oneDNN Ryzen 9 5950XProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen Resolution12345AMD Ryzen 9 5950X 16-Core @ 3.40GHz (16 Cores / 32 Threads)ASUS ROG CROSSHAIR VIII HERO (WI-FI) (3204 BIOS)AMD Starship/Matisse32GB2000GB Corsair Force MP600 + 2000GBllvmpipeAMD Device ab28Realtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200Ubuntu 20.105.10.23-051023-generic (x86_64)GNOME Shell 3.38.2X Server 1.20.94.5 Mesa 21.1.0-devel (git-684f97d 2021-03-12 groovy-oibaf-ppa) (LLVM 11.0.1 256 bits)1.0.168GCC 10.2.0ext43840x2160OpenBenchmarking.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++,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: 0xa201009Security 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

12345Result OverviewPhoronix Test Suite100%103%106%109%oneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNoneDNNSysbenchoneDNNSysbenchoneDNND.B.s - f32 - CPUIP Shapes 3D - f32 - CPUR.N.N.I - f32 - CPUR.N.N.T - f32 - CPUIP Shapes 3D - u8s8f32 - CPUR.N.N.I - bf16bf16bf16 - CPUC.B.S.A - u8s8f32 - CPUR.N.N.T - u8s8f32 - CPUD.B.s - u8s8f32 - CPUR.N.N.I - u8s8f32 - CPUM.M.B.S.T - f32 - CPUC.B.S.A - f32 - CPUM.M.B.S.T - u8s8f32 - CPUIP Shapes 1D - f32 - CPUD.B.s - u8s8f32 - CPUIP Shapes 1D - u8s8f32 - CPUCPUD.B.s - f32 - CPURAM / MemoryR.N.N.T - bf16bf16bf16 - CPU

sysbench oneDNN Ryzen 9 5950Xsysbench: CPUsysbench: RAM / Memoryonednn: IP Shapes 1D - f32 - CPUonednn: IP Shapes 3D - f32 - CPUonednn: IP Shapes 1D - u8s8f32 - CPUonednn: IP Shapes 3D - u8s8f32 - 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: 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 - CPU1234590905.9514807.783.958468.682560.8348990.46177017.03634.946623.5767418.56951.068521.677742713.471755.702725.571746.000.6396422708.581740.781.4182491133.1214841.973.950238.611220.8318550.45650717.04475.140193.5673518.64721.062781.674652692.661722.242705.251727.540.6387072700.711715.761.4102591295.9614820.943.934898.612850.8337770.45450717.18834.814433.5689218.53441.058531.676382676.921727.292695.281739.580.6339262706.441713.751.4072491225.8614845.303.943318.811450.8304060.45838417.09674.586983.5621118.72031.056881.679592729.231721.032695.621736.050.6360192702.761716.201.4065491042.9414868.163.92958.896360.8326450.45964617.09554.703123.5715418.79881.057751.669732691.731714.952708.381733.880.6368442699.181736.711.41415OpenBenchmarking.org

Sysbench

This is a benchmark of Sysbench with the built-in CPU and memory sub-tests. Sysbench is a scriptable multi-threaded benchmark tool based on LuaJIT. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEvents Per Second, More Is BetterSysbench 1.0.20Test: CPU1234520K40K60K80K100KSE +/- 88.94, N = 3SE +/- 77.39, N = 3SE +/- 84.52, N = 3SE +/- 87.78, N = 3SE +/- 86.89, N = 390905.9591133.1291295.9691225.8691042.941. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm
OpenBenchmarking.orgEvents Per Second, More Is BetterSysbench 1.0.20Test: CPU1234516K32K48K64K80KMin: 90808.1 / Avg: 90905.95 / Max: 91083.53Min: 91043.29 / Avg: 91133.12 / Max: 91287.18Min: 91175.63 / Avg: 91295.96 / Max: 91458.95Min: 91118.24 / Avg: 91225.86 / Max: 91399.8Min: 90950.94 / Avg: 91042.94 / Max: 91216.611. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm

OpenBenchmarking.orgMiB/sec, More Is BetterSysbench 1.0.20Test: RAM / Memory123453K6K9K12K15KSE +/- 13.87, N = 3SE +/- 20.27, N = 3SE +/- 31.43, N = 3SE +/- 17.12, N = 3SE +/- 0.96, N = 314807.7814841.9714820.9414845.3014868.161. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm
OpenBenchmarking.orgMiB/sec, More Is BetterSysbench 1.0.20Test: RAM / Memory123453K6K9K12K15KMin: 14789.09 / Avg: 14807.78 / Max: 14834.89Min: 14802.52 / Avg: 14841.97 / Max: 14869.74Min: 14758.77 / Avg: 14820.94 / Max: 14860.08Min: 14811.12 / Avg: 14845.3 / Max: 14864.06Min: 14866.26 / Avg: 14868.16 / Max: 14869.41. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU123450.89071.78142.67213.56284.4535SE +/- 0.01694, N = 3SE +/- 0.00752, N = 3SE +/- 0.01623, N = 3SE +/- 0.01060, N = 3SE +/- 0.03409, N = 33.958463.950233.934893.943313.92950MIN: 3.72MIN: 3.76MIN: 3.72MIN: 3.74MIN: 3.721. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU12345246810Min: 3.93 / Avg: 3.96 / Max: 3.99Min: 3.94 / Avg: 3.95 / Max: 3.96Min: 3.91 / Avg: 3.93 / Max: 3.97Min: 3.92 / Avg: 3.94 / Max: 3.96Min: 3.86 / Avg: 3.93 / Max: 3.981. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU12345246810SE +/- 0.02162, N = 3SE +/- 0.03495, N = 3SE +/- 0.05813, N = 3SE +/- 0.05920, N = 3SE +/- 0.03275, N = 38.682568.611228.612858.811458.89636MIN: 8.14MIN: 8.21MIN: 8.13MIN: 8.33MIN: 8.51. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU123453691215Min: 8.64 / Avg: 8.68 / Max: 8.71Min: 8.57 / Avg: 8.61 / Max: 8.68Min: 8.53 / Avg: 8.61 / Max: 8.73Min: 8.73 / Avg: 8.81 / Max: 8.93Min: 8.85 / Avg: 8.9 / Max: 8.961. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU123450.18790.37580.56370.75160.9395SE +/- 0.001499, N = 3SE +/- 0.002629, N = 3SE +/- 0.003750, N = 3SE +/- 0.001726, N = 3SE +/- 0.002188, N = 30.8348990.8318550.8337770.8304060.832645MIN: 0.75MIN: 0.75MIN: 0.76MIN: 0.75MIN: 0.751. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU12345246810Min: 0.83 / Avg: 0.83 / Max: 0.84Min: 0.83 / Avg: 0.83 / Max: 0.84Min: 0.83 / Avg: 0.83 / Max: 0.84Min: 0.83 / Avg: 0.83 / Max: 0.83Min: 0.83 / Avg: 0.83 / Max: 0.841. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU123450.10390.20780.31170.41560.5195SE +/- 0.004852, N = 3SE +/- 0.002258, N = 3SE +/- 0.002018, N = 3SE +/- 0.002373, N = 3SE +/- 0.001381, N = 30.4617700.4565070.4545070.4583840.459646MIN: 0.42MIN: 0.42MIN: 0.41MIN: 0.43MIN: 0.421. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU1234512345Min: 0.45 / Avg: 0.46 / Max: 0.47Min: 0.45 / Avg: 0.46 / Max: 0.46Min: 0.45 / Avg: 0.45 / Max: 0.46Min: 0.45 / Avg: 0.46 / Max: 0.46Min: 0.46 / Avg: 0.46 / Max: 0.461. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU1234548121620SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.18, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 317.0417.0417.1917.1017.10MIN: 16.37MIN: 16.63MIN: 16.53MIN: 16.53MIN: 16.561. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU1234548121620Min: 17.01 / Avg: 17.04 / Max: 17.07Min: 17.03 / Avg: 17.04 / Max: 17.07Min: 17 / Avg: 17.19 / Max: 17.56Min: 17.08 / Avg: 17.1 / Max: 17.12Min: 17.07 / Avg: 17.1 / Max: 17.131. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU123451.15652.3133.46954.6265.7825SE +/- 0.26241, N = 15SE +/- 0.32359, N = 12SE +/- 0.25214, N = 12SE +/- 0.27001, N = 15SE +/- 0.23987, N = 154.946625.140194.814434.586984.70312MIN: 2.89MIN: 2.91MIN: 2.87MIN: 2.9MIN: 2.911. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU12345246810Min: 3.47 / Avg: 4.95 / Max: 6.38Min: 3.47 / Avg: 5.14 / Max: 6.4Min: 3.18 / Avg: 4.81 / Max: 6.38Min: 3.29 / Avg: 4.59 / Max: 6.39Min: 3.26 / Avg: 4.7 / Max: 6.361. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU123450.80481.60962.41443.21924.024SE +/- 0.00466, N = 3SE +/- 0.00034, N = 3SE +/- 0.00441, N = 3SE +/- 0.00236, N = 3SE +/- 0.00116, N = 33.576743.567353.568923.562113.57154MIN: 3.43MIN: 3.43MIN: 3.42MIN: 3.41MIN: 3.441. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU12345246810Min: 3.57 / Avg: 3.58 / Max: 3.58Min: 3.57 / Avg: 3.57 / Max: 3.57Min: 3.56 / Avg: 3.57 / Max: 3.57Min: 3.56 / Avg: 3.56 / Max: 3.57Min: 3.57 / Avg: 3.57 / Max: 3.571. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU12345510152025SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.04, N = 318.5718.6518.5318.7218.80MIN: 18.16MIN: 18.1MIN: 18.14MIN: 18.33MIN: 18.291. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU12345510152025Min: 18.55 / Avg: 18.57 / Max: 18.59Min: 18.56 / Avg: 18.65 / Max: 18.73Min: 18.51 / Avg: 18.53 / Max: 18.55Min: 18.71 / Avg: 18.72 / Max: 18.73Min: 18.73 / Avg: 18.8 / Max: 18.851. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU123450.24040.48080.72120.96161.202SE +/- 0.00181, N = 3SE +/- 0.00465, N = 3SE +/- 0.00161, N = 3SE +/- 0.00105, N = 3SE +/- 0.00119, N = 31.068521.062781.058531.056881.05775MIN: 0.97MIN: 0.98MIN: 0.97MIN: 0.97MIN: 0.971. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU12345246810Min: 1.07 / Avg: 1.07 / Max: 1.07Min: 1.05 / Avg: 1.06 / Max: 1.07Min: 1.06 / Avg: 1.06 / Max: 1.06Min: 1.06 / Avg: 1.06 / Max: 1.06Min: 1.06 / Avg: 1.06 / Max: 1.061. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU123450.37790.75581.13371.51161.8895SE +/- 0.00014, N = 3SE +/- 0.00176, N = 3SE +/- 0.00123, N = 3SE +/- 0.00545, N = 3SE +/- 0.00358, N = 31.677741.674651.676381.679591.66973MIN: 1.62MIN: 1.6MIN: 1.59MIN: 1.59MIN: 1.591. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU12345246810Min: 1.68 / Avg: 1.68 / Max: 1.68Min: 1.67 / Avg: 1.67 / Max: 1.68Min: 1.67 / Avg: 1.68 / Max: 1.68Min: 1.67 / Avg: 1.68 / Max: 1.69Min: 1.66 / Avg: 1.67 / Max: 1.671. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU123456001200180024003000SE +/- 4.23, N = 3SE +/- 1.98, N = 3SE +/- 12.00, N = 3SE +/- 9.17, N = 3SE +/- 9.51, N = 32713.472692.662676.922729.232691.73MIN: 2695.46MIN: 2678.17MIN: 2645.36MIN: 2694.95MIN: 2662.921. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU123455001000150020002500Min: 2707.39 / Avg: 2713.47 / Max: 2721.6Min: 2689.94 / Avg: 2692.66 / Max: 2696.52Min: 2656.16 / Avg: 2676.92 / Max: 2697.72Min: 2710.92 / Avg: 2729.23 / Max: 2739.38Min: 2673.02 / Avg: 2691.73 / Max: 2704.021. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU12345400800120016002000SE +/- 11.93, N = 3SE +/- 7.22, N = 3SE +/- 13.99, N = 3SE +/- 5.23, N = 3SE +/- 4.01, N = 31755.701722.241727.291721.031714.95MIN: 1720.03MIN: 1693.61MIN: 1698.14MIN: 1705.22MIN: 1698.741. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU1234530060090012001500Min: 1733.08 / Avg: 1755.7 / Max: 1773.6Min: 1710.75 / Avg: 1722.24 / Max: 1735.55Min: 1707.74 / Avg: 1727.29 / Max: 1754.4Min: 1715.73 / Avg: 1721.03 / Max: 1731.49Min: 1709.68 / Avg: 1714.95 / Max: 1722.811. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU123456001200180024003000SE +/- 10.35, N = 3SE +/- 8.00, N = 3SE +/- 10.44, N = 3SE +/- 11.59, N = 3SE +/- 10.27, N = 32725.572705.252695.282695.622708.38MIN: 2696.49MIN: 2682.75MIN: 2671.19MIN: 2673.04MIN: 2680.821. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU123455001000150020002500Min: 2706.42 / Avg: 2725.57 / Max: 2741.97Min: 2691.93 / Avg: 2705.25 / Max: 2719.58Min: 2680.64 / Avg: 2695.28 / Max: 2715.5Min: 2683.52 / Avg: 2695.62 / Max: 2718.78Min: 2691.36 / Avg: 2708.38 / Max: 2726.841. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU12345400800120016002000SE +/- 12.65, N = 3SE +/- 0.49, N = 3SE +/- 12.37, N = 3SE +/- 9.45, N = 3SE +/- 6.52, N = 31746.001727.541739.581736.051733.88MIN: 1710.33MIN: 1714.36MIN: 1717.31MIN: 1713.46MIN: 1708.51. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU1234530060090012001500Min: 1724.52 / Avg: 1746 / Max: 1768.31Min: 1726.99 / Avg: 1727.54 / Max: 1728.52Min: 1726.68 / Avg: 1739.58 / Max: 1764.32Min: 1722.39 / Avg: 1736.05 / Max: 1754.19Min: 1723.61 / Avg: 1733.88 / Max: 1745.961. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU123450.14390.28780.43170.57560.7195SE +/- 0.000949, N = 3SE +/- 0.000454, N = 3SE +/- 0.000478, N = 3SE +/- 0.000994, N = 3SE +/- 0.000627, N = 30.6396420.6387070.6339260.6360190.636844MIN: 0.61MIN: 0.61MIN: 0.6MIN: 0.61MIN: 0.611. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU12345246810Min: 0.64 / Avg: 0.64 / Max: 0.64Min: 0.64 / Avg: 0.64 / Max: 0.64Min: 0.63 / Avg: 0.63 / Max: 0.63Min: 0.63 / Avg: 0.64 / Max: 0.64Min: 0.64 / Avg: 0.64 / Max: 0.641. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU123456001200180024003000SE +/- 4.11, N = 3SE +/- 5.61, N = 3SE +/- 11.04, N = 3SE +/- 7.92, N = 3SE +/- 4.86, N = 32708.582700.712706.442702.762699.18MIN: 2692.02MIN: 2685.06MIN: 2680.57MIN: 2678.89MIN: 2678.941. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU123455001000150020002500Min: 2703.76 / Avg: 2708.58 / Max: 2716.76Min: 2694.98 / Avg: 2700.71 / Max: 2711.94Min: 2693.74 / Avg: 2706.44 / Max: 2728.43Min: 2688.93 / Avg: 2702.76 / Max: 2716.38Min: 2692.3 / Avg: 2699.18 / Max: 2708.581. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU12345400800120016002000SE +/- 8.56, N = 3SE +/- 0.54, N = 3SE +/- 2.22, N = 3SE +/- 2.84, N = 3SE +/- 9.03, N = 31740.781715.761713.751716.201736.71MIN: 1706.77MIN: 1703.59MIN: 1692.03MIN: 1701.07MIN: 1699.51. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU1234530060090012001500Min: 1726.7 / Avg: 1740.78 / Max: 1756.26Min: 1714.7 / Avg: 1715.76 / Max: 1716.45Min: 1709.32 / Avg: 1713.75 / Max: 1716.26Min: 1710.54 / Avg: 1716.2 / Max: 1719.36Min: 1719.04 / Avg: 1736.71 / Max: 1748.771. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU123450.31910.63820.95731.27641.5955SE +/- 0.00092, N = 3SE +/- 0.00138, N = 3SE +/- 0.00206, N = 3SE +/- 0.00102, N = 3SE +/- 0.00141, N = 31.418241.410251.407241.406541.41415MIN: 1.31MIN: 1.31MIN: 1.3MIN: 1.3MIN: 1.321. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU12345246810Min: 1.42 / Avg: 1.42 / Max: 1.42Min: 1.41 / Avg: 1.41 / Max: 1.41Min: 1.4 / Avg: 1.41 / Max: 1.41Min: 1.4 / Avg: 1.41 / Max: 1.41Min: 1.41 / Avg: 1.41 / Max: 1.421. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

1

Kernel Notes: Transparent Huge Pages: madvise
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: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009
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 Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 13 March 2021 16:11 by user pts.

2

Kernel Notes: Transparent Huge Pages: madvise
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: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009
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 Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 13 March 2021 17:15 by user pts.

3

Kernel Notes: Transparent Huge Pages: madvise
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: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009
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 Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 13 March 2021 18:00 by user pts.

4

Kernel Notes: Transparent Huge Pages: madvise
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: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009
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 Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 13 March 2021 19:43 by user pts.

5

Processor: AMD Ryzen 9 5950X 16-Core @ 3.40GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR VIII HERO (WI-FI) (3204 BIOS), Chipset: AMD Starship/Matisse, Memory: 32GB, Disk: 2000GB Corsair Force MP600 + 2000GB, Graphics: llvmpipe, Audio: AMD Device ab28, Network: Realtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200

OS: Ubuntu 20.10, Kernel: 5.10.23-051023-generic (x86_64), Desktop: GNOME Shell 3.38.2, Display Server: X Server 1.20.9, OpenGL: 4.5 Mesa 21.1.0-devel (git-684f97d 2021-03-12 groovy-oibaf-ppa) (LLVM 11.0.1 256 bits), Vulkan: 1.0.168, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
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: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009
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 Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 13 March 2021 20:30 by user pts.