eMAG

ARMv8 Cortex-A72 testing with a SolidRun CEX7 (EDK II BIOS) and MSI NVIDIA GeForce GT 1030 on Fedora 33 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2102210-FI-2012272NE18&grs&rdt.

eMAGProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkAudioOSKernelDesktopDisplay ServerDisplay DriverCompilerFile-SystemScreen Resolution1234HoneyComb LX2KAmpere eMAG ARMv8 @ 3.00GHz (32 Cores)AmpereComputing OSPREY (4.8.19 BIOS)Applied Micro Circuits X-Gene126GB256GB Samsung SSD 860ASPEEDVE228Intel I210Ubuntu 20.045.7.0-050700-generic (aarch64)GNOME Shell 3.36.3X Server 1.20.8modesetting 1.20.8GCC 9.3.0ext41920x1080ARMv8 Cortex-A72 (16 Cores)SolidRun CEX7 (EDK II BIOS)32GB128GB Generic + 8GB SL08G + 63GB DF4064MSI NVIDIA GeForce GT 1030NVIDIA GP108 HD AudioFedora 335.10.10-00042-gbfa806f5daa5-dirty (aarch64)X Server 1.20.10GCC 10.2.1 20201125 + Clang 11.0.0 + CUDA 11.2btrfsOpenBenchmarking.orgCompiler Details- 1: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v - 2: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v - 3: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v - 4: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v - HoneyComb LX2K: --build=aarch64-redhat-linux --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-initfini-array --enable-languages=c,c++,fortran,objc,obj-c++,ada,go,lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-gcc-major-version-only --with-isl --with-linker-hash-style=gnu Processor Details- 1, 2, 3, 4: Scaling Governor: cppc_cpufreq ondemandPython Details- 1: Python 3.8.2- 2: Python 3.8.2- 3: Python 3.8.2- 4: Python 3.8.2- HoneyComb LX2K: Python 3.9.1Security Details- 1: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Mitigation of PTI + spec_store_bypass: Vulnerable + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Vulnerable + tsx_async_abort: Not affected- 2: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Mitigation of PTI + spec_store_bypass: Vulnerable + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Vulnerable + tsx_async_abort: Not affected- 3: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Mitigation of PTI + spec_store_bypass: Vulnerable + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Vulnerable + tsx_async_abort: Not affected- 4: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Mitigation of PTI + spec_store_bypass: Vulnerable + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Vulnerable + tsx_async_abort: Not affected- HoneyComb LX2K: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Not affected + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Branch predictor hardening + srbds: Not affected + tsx_async_abort: Not affected

eMAGclomp: Static OMP Speeduponednn: IP Shapes 3D - u8s8f32 - CPUencode-ape: WAV To APEonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUasmfish: 1024 Hash Memory, 26 Depthcoremark: CoreMark Size 666 - Iterations Per Secondonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUstockfish: Total Timemafft: Multiple Sequence Alignment - LSU RNAonednn: IP Shapes 1D - u8s8f32 - CPUencode-opus: WAV To Opus Encodesimdjson: PartialTweetssimdjson: Kostyasimdjson: DistinctUserIDavifenc: 2simdjson: LargeRandavifenc: 0onednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUbuild-eigen: Time To Compileavifenc: 8rav1e: 10avifenc: 10numpy: x264: H.264 Video Encodingrav1e: 1rav1e: 6rav1e: 5tscp: AI Chess Performanceespeak: Text-To-Speech Synthesisonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: Recurrent Neural Network Inference - u8s8f32 - CPUonednn: Recurrent Neural Network Training - u8s8f32 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUonednn: Recurrent Neural Network Training - f32 - CPUonednn: Deconvolution Batch shapes_3d - f32 - CPUonednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: Convolution Batch Shapes Auto - f32 - CPUonednn: IP Shapes 3D - f32 - CPUonednn: IP Shapes 1D - f32 - CPU1234HoneyComb LX2K7.1418.33096.13029827.833037962385397.370363114.066113.0201569146935.641183.91748.3080.550.480.56250.1650.23404.737184.796357.47422.3020.42521.88391.6532.130.0840.2230.18851590387.83136.614817140.119.817616255.930113.817700.431838.260.3671191.83793.472523.682529.01487.2421.22139.91931520.333135767385035.207466113.511112.3101541712735.344184.87748.2660.550.480.56250.7050.23403.850185.965357.14022.3320.41921.83291.8632.670.0840.2210.18751590384.90238.370616556.721.007716777.030971.017313.530600.563.4163194.62098.280422.800833.22447.0424.64130673.7385080.514289113.112112.5501561760736.355183.3930.550.480.560.23185.1620.42032.780.0840.2220.18751571038.494417065.520.484716436.830446.516864.331978.255.9390173.53976.351522.847032.28407.2419.392112.613112.92435.999183.1370.550.480.570.23183.15351570930770.059.7828173.05084.369024.446830.89251.697.138837.00274962.716459128193571.808864196.134192.123951445725.173129.05037.7350.70.610.71310.0320.28485.908220.110315.89724.4600.44523.18397.0933.560.0810.2240.18651092884.42746.007838288.226.146638077.374589.338360.174820.5118.206620.757122.32321.512753.3212OpenBenchmarking.org

CLOMP

Static OMP Speedup

OpenBenchmarking.orgSpeedup, More Is BetterCLOMP 1.2Static OMP Speedup1234HoneyComb LX2K246810SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 37.17.27.07.21.61. (CC) gcc options: -fopenmp -O3 -lm

oneDNN

Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU1234HoneyComb LX2K90180270360450SE +/- 2.55, N = 3SE +/- 0.17, N = 3SE +/- 1.42, N = 3SE +/- 1.34, N = 3SE +/- 0.09, N = 3418.33421.22424.64419.3997.14MIN: 379.92MIN: 376.78MIN: 380.56MIN: 371.23-O2 - MIN: 96.31. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

Monkey Audio Encoding

WAV To APE

OpenBenchmarking.orgSeconds, Fewer Is BetterMonkey Audio Encoding 3.99.6WAV To APE12HoneyComb LX2K20406080100SE +/- 0.05, N = 5SE +/- 0.06, N = 5SE +/- 0.59, N = 596.1339.9237.001. (CXX) g++ options: -O3 -pedantic -rdynamic -lrt

oneDNN

Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU123HoneyComb LX2K16K32K48K64K80KSE +/- 438.76, N = 12SE +/- 470.00, N = 3SE +/- 360.00, N = 3SE +/- 348.88, N = 329827.831520.330673.774962.7MIN: 23657MIN: 24350.2MIN: 24396.5-O2 - MIN: 74064.51. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

asmFish

1024 Hash Memory, 26 Depth

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 Depth12HoneyComb LX2K7M14M21M28M35MSE +/- 299925.52, N = 3SE +/- 393577.81, N = 3SE +/- 201778.29, N = 3330379623313576716459128

Coremark

CoreMark Size 666 - Iterations Per Second

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per Second123HoneyComb LX2K80K160K240K320K400KSE +/- 656.44, N = 3SE +/- 798.90, N = 3SE +/- 664.33, N = 3SE +/- 23.74, N = 3385397.37385035.21385080.51193571.811. (CC) gcc options: -O2 -lrt" -lrt

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU1234HoneyComb LX2K4080120160200SE +/- 0.92, N = 3SE +/- 0.48, N = 3SE +/- 1.41, N = 3SE +/- 1.75, N = 3SE +/- 2.17, N = 3114.07113.51113.11112.61196.13MIN: 93.84MIN: 91.33MIN: 92.39MIN: 90.72-O2 - MIN: 179.491. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU1234HoneyComb LX2K4080120160200SE +/- 0.34, N = 3SE +/- 0.05, N = 3SE +/- 0.11, N = 3SE +/- 0.27, N = 3SE +/- 0.26, N = 3113.02112.31112.55112.92192.12MIN: 98.24MIN: 103.46MIN: 102.72MIN: 104.76-O2 - MIN: 190.321. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

Stockfish

Total Time

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 12Total Time123HoneyComb LX2K3M6M9M12M15MSE +/- 125793.16, N = 3SE +/- 184845.46, N = 6SE +/- 177988.27, N = 15SE +/- 173726.98, N = 315691469154171271561760795144571. (CXX) g++ options: -lpthread -fno-exceptions -std=c++17 -pedantic -O3 -flto -flto=jobserver

Timed MAFFT Alignment

Multiple Sequence Alignment - LSU RNA

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MAFFT Alignment 7.471Multiple Sequence Alignment - LSU RNA1234HoneyComb LX2K816243240SE +/- 0.31, N = 3SE +/- 0.11, N = 3SE +/- 0.16, N = 3SE +/- 0.49, N = 3SE +/- 0.40, N = 335.6435.3436.3636.0025.171. (CC) gcc options: -std=c99 -O3 -lm -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU1234HoneyComb LX2K4080120160200SE +/- 1.68, N = 3SE +/- 2.81, N = 3SE +/- 1.41, N = 3SE +/- 0.18, N = 3SE +/- 0.47, N = 3183.92184.88183.39183.14129.05MIN: 127.82MIN: 135.46MIN: 133.71MIN: 130.48-O2 - MIN: 125.461. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

Opus Codec Encoding

WAV To Opus Encode

OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus Encode12HoneyComb LX2K1122334455SE +/- 0.03, N = 5SE +/- 0.01, N = 5SE +/- 0.02, N = 548.3148.2737.741. (CXX) g++ options: -fvisibility=hidden -logg -lm

simdjson

Throughput Test: PartialTweets

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: PartialTweets1234HoneyComb LX2K0.15750.3150.47250.630.7875SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.550.550.550.550.70-O3-O3-O3-O3-O21. (CXX) g++ options: -pthread

simdjson

Throughput Test: Kostya

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: Kostya1234HoneyComb LX2K0.13730.27460.41190.54920.6865SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.480.480.480.480.61-O3-O3-O3-O3-O21. (CXX) g++ options: -pthread

simdjson

Throughput Test: DistinctUserID

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: DistinctUserID1234HoneyComb LX2K0.15980.31960.47940.63920.799SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.560.560.560.570.71-O3-O3-O3-O3-O21. (CXX) g++ options: -pthread

libavif avifenc

Encoder Speed: 2

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.7.3Encoder Speed: 212HoneyComb LX2K70140210280350SE +/- 0.06, N = 3SE +/- 0.40, N = 3SE +/- 0.10, N = 3250.17250.71310.031. (CXX) g++ options: -O3 -fPIC

simdjson

Throughput Test: LargeRandom

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: LargeRandom1234HoneyComb LX2K0.0630.1260.1890.2520.315SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.230.230.230.230.28-O3-O3-O3-O3-O21. (CXX) g++ options: -pthread

libavif avifenc

Encoder Speed: 0

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.7.3Encoder Speed: 012HoneyComb LX2K110220330440550SE +/- 1.37, N = 3SE +/- 0.42, N = 3SE +/- 0.81, N = 3404.74403.85485.911. (CXX) g++ options: -O3 -fPIC

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU1234HoneyComb LX2K50100150200250SE +/- 0.92, N = 3SE +/- 1.31, N = 3SE +/- 1.64, N = 3SE +/- 2.05, N = 3SE +/- 0.78, N = 3184.80185.97185.16183.15220.11MIN: 121.06MIN: 114.9MIN: 116.12MIN: 111-O2 - MIN: 217.21. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

Timed Eigen Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Eigen Compilation 3.3.9Time To Compile12HoneyComb LX2K80160240320400SE +/- 1.89, N = 3SE +/- 0.65, N = 3SE +/- 0.80, N = 3357.47357.14315.90

libavif avifenc

Encoder Speed: 8

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.7.3Encoder Speed: 812HoneyComb LX2K612182430SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 322.3022.3324.461. (CXX) g++ options: -O3 -fPIC

rav1e

Speed: 10

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 10123HoneyComb LX2K0.10010.20020.30030.40040.5005SE +/- 0.000, N = 3SE +/- 0.001, N = 3SE +/- 0.000, N = 3SE +/- 0.001, N = 30.4250.4190.4200.445

libavif avifenc

Encoder Speed: 10

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.7.3Encoder Speed: 1012HoneyComb LX2K612182430SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 321.8821.8323.181. (CXX) g++ options: -O3 -fPIC

Numpy Benchmark

OpenBenchmarking.orgScore, More Is BetterNumpy Benchmark12HoneyComb LX2K20406080100SE +/- 0.30, N = 3SE +/- 0.27, N = 3SE +/- 0.62, N = 391.6591.8697.09

x264

H.264 Video Encoding

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2019-12-17H.264 Video Encoding123HoneyComb LX2K816243240SE +/- 0.38, N = 5SE +/- 0.30, N = 3SE +/- 0.51, N = 3SE +/- 0.06, N = 332.1332.6732.7833.56-lavformat -lavcodec -lavutil -lswscale-lavformat -lavcodec -lavutil -lswscale-lavformat -lavcodec -lavutil -lswscale1. (CC) gcc options: -ldl -lm -lpthread

rav1e

Speed: 1

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 1123HoneyComb LX2K0.01890.03780.05670.07560.0945SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 30.0840.0840.0840.081

rav1e

Speed: 6

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 6123HoneyComb LX2K0.05040.10080.15120.20160.252SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 30.2230.2210.2220.224

rav1e

Speed: 5

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4 AlphaSpeed: 5123HoneyComb LX2K0.04230.08460.12690.16920.2115SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 30.1880.1870.1870.186

TSCP

AI Chess Performance

OpenBenchmarking.orgNodes Per Second, More Is BetterTSCP 1.81AI Chess Performance1234HoneyComb LX2K110K220K330K440K550KSE +/- 327.56, N = 5SE +/- 328.25, N = 5SE +/- 264.73, N = 5SE +/- 616.26, N = 55159035159035157105157095109281. (CC) gcc options: -O3 -march=native

eSpeak-NG Speech Engine

Text-To-Speech Synthesis

OpenBenchmarking.orgSeconds, Fewer Is BettereSpeak-NG Speech Engine 20200907Text-To-Speech Synthesis12HoneyComb LX2K20406080100SE +/- 0.12, N = 4SE +/- 0.21, N = 4SE +/- 4.47, N = 487.8384.9084.43-lpthread -lm1. (CC) gcc options: -O2 -std=c99

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU123HoneyComb LX2K1020304050SE +/- 1.48, N = 15SE +/- 0.94, N = 15SE +/- 0.82, N = 15SE +/- 0.15, N = 336.6138.3738.4946.01MIN: 19.74MIN: 19.74MIN: 19.79-O2 - MIN: 45.371. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU123HoneyComb LX2K8K16K24K32K40KSE +/- 244.89, N = 3SE +/- 327.46, N = 12SE +/- 506.17, N = 9SE +/- 226.16, N = 317140.116556.717065.538288.2MIN: 13604.9MIN: 12882.3MIN: 13244.1-O2 - MIN: 37745.71. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU123HoneyComb LX2K612182430SE +/- 1.04, N = 15SE +/- 0.93, N = 15SE +/- 0.93, N = 15SE +/- 0.00, N = 319.8221.0120.4826.15MIN: 8.13MIN: 8.13MIN: 8.14-O2 - MIN: 25.911. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU123HoneyComb LX2K8K16K24K32K40KSE +/- 188.21, N = 3SE +/- 307.63, N = 12SE +/- 378.61, N = 9SE +/- 169.67, N = 316255.916777.016436.838077.3MIN: 13038.5MIN: 12597.8MIN: 13044.4-O2 - MIN: 37721.41. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU123HoneyComb LX2K16K32K48K64K80KSE +/- 165.59, N = 3SE +/- 500.69, N = 12SE +/- 588.64, N = 12SE +/- 201.98, N = 330113.830971.030446.574589.3MIN: 24215.3MIN: 23442.3MIN: 23892.7-O2 - MIN: 74079.31. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU123HoneyComb LX2K8K16K24K32K40KSE +/- 343.55, N = 11SE +/- 395.99, N = 12SE +/- 324.82, N = 12SE +/- 219.25, N = 317700.417313.516864.338360.1MIN: 13035MIN: 12871.9MIN: 12759.4-O2 - MIN: 37645.21. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU1234HoneyComb LX2K16K32K48K64K80KSE +/- 1283.37, N = 9SE +/- 724.02, N = 10SE +/- 171.48, N = 3SE +/- 405.43, N = 12SE +/- 258.40, N = 331838.230600.531978.230770.074820.5MIN: 23896.8MIN: 23693.3MIN: 25646MIN: 24555.1-O2 - MIN: 74172.11. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU1234HoneyComb LX2K306090120150SE +/- 6.33, N = 15SE +/- 7.31, N = 15SE +/- 7.43, N = 12SE +/- 6.08, N = 15SE +/- 2.32, N = 360.3763.4255.9459.78118.21MIN: 26.98MIN: 26.98MIN: 26.97MIN: 26.97-O2 - MIN: 114.11. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU1234HoneyComb LX2K130260390520650SE +/- 15.29, N = 12SE +/- 13.95, N = 15SE +/- 12.91, N = 15SE +/- 12.10, N = 15SE +/- 10.39, N = 3191.84194.62173.54173.05620.76MIN: 115.28MIN: 115.45MIN: 116.21MIN: 116-O2 - MIN: 597.771. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU1234HoneyComb LX2K306090120150SE +/- 7.37, N = 15SE +/- 7.70, N = 12SE +/- 6.62, N = 15SE +/- 8.19, N = 15SE +/- 0.02, N = 393.4798.2876.3584.37122.32MIN: 30.81MIN: 30.76MIN: 30.78MIN: 30.78-O2 - MIN: 121.031. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU1234HoneyComb LX2K612182430SE +/- 0.67, N = 15SE +/- 0.68, N = 12SE +/- 0.69, N = 15SE +/- 1.03, N = 15SE +/- 0.02, N = 323.6822.8022.8524.4521.51MIN: 17.03MIN: 17.06MIN: 17.02MIN: 17.03-O2 - MIN: 21.161. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

oneDNN

Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU1234HoneyComb LX2K1224364860SE +/- 1.27, N = 15SE +/- 1.03, N = 15SE +/- 1.36, N = 15SE +/- 1.62, N = 12SE +/- 1.13, N = 329.0133.2232.2830.8953.32MIN: 15.13MIN: 15.35MIN: 15.14MIN: 15.09-O2 - MIN: 50.831. (CXX) g++ options: -O3 -std=c++11 -fopenmp -mcpu=native -fPIC -pie -lpthread

Phoronix Test Suite v10.8.5