Talos II Dual 22-Core POWER9 Spectre Benchmarks

POWER9 Spectre mitigation benchmarks by Michael Larabel.

HTML result view exported from: https://openbenchmarking.org/result/1811098-SK-TALOSIIDU15&sor.

Talos II Dual 22-Core POWER9 Spectre BenchmarksProcessorMotherboardMemoryDiskGraphicsMonitorNetworkOSKernelCompilerFile-SystemScreen ResolutionKernel ProtectionDefault Kernel + User ProtectionNo ProtectionPOWER9 altivec supported @ 3.80GHz (44 Cores / 176 Threads)PowerNV T2P9D01 REV 1.0165536MBSamsung SSD 960 EVO 500GBASPEED ASPEED FamilyVE228Broadcom NetXtreme BCM5719 Gigabit PCIeUbuntu 18.104.18.0-10-generic (ppc64le)GCC 8.2.0ext41920x1080OpenBenchmarking.orgCompiler Details- --build=powerpc64le-linux-gnu --disable-libphobos --disable-multilib --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-secureplt --enable-shared --enable-targets=powerpcle-linux --enable-threads=posix --host=powerpc64le-linux-gnu --program-prefix=powerpc64le-linux-gnu- --target=powerpc64le-linux-gnu --with-cpu=power8 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-long-double-128 -v Processor Details- Scaling Governor: powernv-cpufreq ondemandPython Details- Python 2.7.15+ + Python 3.6.7Security Details- Kernel Protection: __user pointer sanitization ori31 speculation barrier enabled + Indirect branch serialisation (kernel only) + Kernel entry/exit barrier (eieio) - Default Kernel + User Protection: __user pointer sanitization ori31 speculation barrier enabled + Indirect branch cache disabled + Kernel entry/exit barrier (eieio) Disk Details- Default Kernel + User Protection, No Protection: NONE / errors=remount-ro,relatime,rw

Talos II Dual 22-Core POWER9 Spectre Benchmarksmpcbench: Multi-Precision Benchmarkrodinia: OpenMP LavaMDrodinia: OpenMP CFD Solverrodinia: OpenMP Streamclusterffte: N=256, 1D Complex FFT Routinego-benchmark: httpgo-benchmark: jsongo-benchmark: buildgo-benchmark: garbagebork: File Encryption Timebyte: Dhrystone 2stockfish: Total Timebuild-imagemagick: Time To Compilebuild-llvm: Time To Compilec-ray: Total Time - 4K, 16 Rays Per Pixelpovray: Trace Timeprimesieve: 1e12 Prime Number Generationrust-prime: Prime Number Test To 200,000,000aobench: 2048 x 2048 - Total Timecython-bench: openssl: RSA 4096-bit Performancecompilebench: Compilecompilebench: Initial Createcompilebench: Read Compiled Treebuild-linux-kernel: Time To Compilehackbench: 32 - Processstress-ng: Cryptostress-ng: Forkingstress-ng: Hsearchstress-ng: Lsearchstress-ng: Tsearchstress-ng: CPU Stressstress-ng: Semaphoresstress-ng: Vector Mathstress-ng: Memory Copyingstress-ng: Socket Activitystress-ng: Glibc C String Functionsstress-ng: Glibc Qsort Data Sortingstress-ng: System V Message Passingpybench: Total For Average Test Timesphpbench: PHP Benchmark SuiteKernel ProtectionDefault Kernel + User ProtectionNo Protection537018.3713.2529.6067701928510423881928049160231121124522.22262585137979922727.7717419.0524.9217.493.7259.5140.6674172533346182350.6627.178149585921869313868891232134716688416834027094137082436864922041939330044400718.7614.7929.771341717489864429899607203123370022.51267558217913510330.0318518.9325.9917.473.5164.6166.6273942324296153855.4628.148075593601641382028591227934751788154831726272129058725362936174090159731543316.7512.9328.571568381381347127993756394110682522.16266187188301184627.2817217.8321.8516.383.3259.5238.8873902539348188750.0123.898298605852016603898981240245663888699836828717137922636966070121816329783OpenBenchmarking.org

GNU MPC

Multi-Precision Benchmark

OpenBenchmarking.orgGlobal Score, More Is BetterGNU MPC 1.1.0Multi-Precision BenchmarkNo ProtectionKernel ProtectionDefault Kernel + User Protection12002400360048006000SE +/- 3.33, N = 3SE +/- 3.33, N = 35433537040071. (CC) gcc options: -m64 -mtune=power8 -O3 -MT -MD -MP -MF

Rodinia

Test: OpenMP LavaMD

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP LavaMDNo ProtectionKernel ProtectionDefault Kernel + User Protection510152025SE +/- 0.20, N = 3SE +/- 0.31, N = 4SE +/- 0.26, N = 316.7518.3718.761. (CXX) g++ options: -O2 -lOpenCL

Rodinia

Test: OpenMP CFD Solver

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP CFD SolverNo ProtectionKernel ProtectionDefault Kernel + User Protection48121620SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.08, N = 312.9313.2514.791. (CXX) g++ options: -O2 -lOpenCL

Rodinia

Test: OpenMP Streamcluster

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP StreamclusterNo ProtectionKernel ProtectionDefault Kernel + User Protection714212835SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 328.5729.6029.771. (CXX) g++ options: -O2 -lOpenCL

FFTE

Test: N=256, 1D Complex FFT Routine

OpenBenchmarking.orgMFLOPS, More Is BetterFFTE 6.0Test: N=256, 1D Complex FFT RoutineKernel Protection15003000450060007500SE +/- 0.43, N = 367701. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

Go Benchmarks

Test: http

OpenBenchmarking.orgNanoseconds/Operation, Fewer Is BetterGo BenchmarksTest: httpDefault Kernel + User ProtectionNo ProtectionKernel Protection400K800K1200K1600K2000KSE +/- 107637.64, N = 12SE +/- 209033.95, N = 9SE +/- 220954.01, N = 10134171715683811928510

Go Benchmarks

Test: json

OpenBenchmarking.orgNanoseconds/Operation, Fewer Is BetterGo BenchmarksTest: jsonNo ProtectionKernel ProtectionDefault Kernel + User Protection1000K2000K3000K4000K5000KSE +/- 54441.22, N = 3SE +/- 3340.75, N = 3SE +/- 87285.98, N = 12381347142388194898644

Go Benchmarks

Test: build

OpenBenchmarking.orgNanoseconds/Operation, Fewer Is BetterGo BenchmarksTest: buildNo ProtectionKernel ProtectionDefault Kernel + User Protection6000M12000M18000M24000M30000MSE +/- 318891587.38, N = 12SE +/- 94097418.28, N = 3SE +/- 573457278.59, N = 3279937563942804916023129899607203

Go Benchmarks

Test: garbage

OpenBenchmarking.orgNanoseconds/Operation, Fewer Is BetterGo BenchmarksTest: garbageNo ProtectionKernel ProtectionDefault Kernel + User Protection300K600K900K1200K1500KSE +/- 3498.26, N = 3SE +/- 12454.68, N = 12SE +/- 12597.46, N = 12110682512112451233700

Bork File Encrypter

File Encryption Time

OpenBenchmarking.orgSeconds, Fewer Is BetterBork File Encrypter 1.4File Encryption TimeNo ProtectionKernel ProtectionDefault Kernel + User Protection510152025SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 322.1622.2222.51

BYTE Unix Benchmark

Computational Test: Dhrystone 2

OpenBenchmarking.orgLPS, More Is BetterBYTE Unix Benchmark 3.6Computational Test: Dhrystone 2Default Kernel + User ProtectionNo ProtectionKernel Protection6M12M18M24M30MSE +/- 13055.64, N = 3SE +/- 109377.13, N = 3SE +/- 494133.76, N = 3267558212661871826258513

Stockfish

Total Time

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 9Total TimeNo ProtectionKernel ProtectionDefault Kernel + User Protection20M40M60M80M100MSE +/- 743288.42, N = 3SE +/- 1474158.28, N = 3SE +/- 662670.61, N = 38301184679799227791351031. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++11 -pedantic -O3 -flto

Timed ImageMagick Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed ImageMagick Compilation 6.9.0Time To CompileNo ProtectionKernel ProtectionDefault Kernel + User Protection714212835SE +/- 0.12, N = 3SE +/- 0.40, N = 5SE +/- 0.44, N = 427.2827.7730.03

Timed LLVM Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 6.0.1Time To CompileNo ProtectionKernel ProtectionDefault Kernel + User Protection4080120160200172174185

C-Ray

Total Time - 4K, 16 Rays Per Pixel

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Time - 4K, 16 Rays Per PixelNo ProtectionDefault Kernel + User ProtectionKernel Protection510152025SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 317.8318.9319.051. (CC) gcc options: -lm -lpthread -O3

POV-Ray

Trace Time

OpenBenchmarking.orgSeconds, Fewer Is BetterPOV-Ray 3.7.0.7Trace TimeNo ProtectionKernel ProtectionDefault Kernel + User Protection612182430SE +/- 0.15, N = 3SE +/- 0.13, N = 3SE +/- 0.09, N = 321.8524.9225.991. (CXX) g++ options: -pipe -O3 -ffast-math -pthread -R/usr/lib -lSDL -lX11 -lIlmImf -lImath -lHalf -lIex -lIexMath -lIlmThread -lpthread -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system

Primesieve

1e12 Prime Number Generation

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 7.11e12 Prime Number GenerationNo ProtectionDefault Kernel + User ProtectionKernel Protection48121620SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 316.3817.4717.491. (CXX) g++ options: -O3 -lpthread

Rust Prime Benchmark

Prime Number Test To 200,000,000

OpenBenchmarking.orgSeconds, Fewer Is BetterRust Prime BenchmarkPrime Number Test To 200,000,000No ProtectionDefault Kernel + User ProtectionKernel Protection0.8371.6742.5113.3484.185SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.07, N = 33.323.513.721. (CC) gcc options: -m64 -pie -nodefaultlibs

AOBench

Size: 2048 x 2048 - Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterAOBenchSize: 2048 x 2048 - Total TimeKernel ProtectionNo ProtectionDefault Kernel + User Protection1428425670SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 359.5159.5264.611. (CC) gcc options: -lm -O3

Cython benchmark

OpenBenchmarking.orgSeconds, Fewer Is BetterCython benchmark 0.27No ProtectionKernel ProtectionDefault Kernel + User Protection1530456075SE +/- 0.03, N = 3SE +/- 0.65, N = 3SE +/- 0.04, N = 338.8840.6666.62

OpenSSL

RSA 4096-bit Performance

OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.1.1RSA 4096-bit PerformanceKernel ProtectionDefault Kernel + User ProtectionNo Protection16003200480064008000SE +/- 23.66, N = 3SE +/- 51.34, N = 3SE +/- 26.62, N = 37417739473901. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl

Compile Bench

Test: Compile

OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: CompileNo ProtectionKernel ProtectionDefault Kernel + User Protection5001000150020002500SE +/- 11.40, N = 3SE +/- 9.03, N = 3SE +/- 8.20, N = 3253925332324

Compile Bench

Test: Initial Create

OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Initial CreateNo ProtectionKernel ProtectionDefault Kernel + User Protection80160240320400SE +/- 1.84, N = 3SE +/- 0.18, N = 3SE +/- 1.65, N = 3348346296

Compile Bench

Test: Read Compiled Tree

OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Read Compiled TreeNo ProtectionKernel ProtectionDefault Kernel + User Protection400800120016002000SE +/- 0.59, N = 3SE +/- 7.11, N = 3SE +/- 4.49, N = 3188718231538

Timed Linux Kernel Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 4.18Time To CompileNo ProtectionKernel ProtectionDefault Kernel + User Protection1224364860SE +/- 0.71, N = 5SE +/- 0.56, N = 9SE +/- 0.92, N = 450.0150.6655.46

Hackbench

Count: 32 - Type: Process

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 32 - Type: ProcessNo ProtectionKernel ProtectionDefault Kernel + User Protection714212835SE +/- 0.04, N = 3SE +/- 0.21, N = 3SE +/- 0.02, N = 323.8927.1728.141. (CC) gcc options: -lpthread

Stress-NG

Test: Crypto

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: CryptoNo ProtectionKernel ProtectionDefault Kernel + User Protection2K4K6K8K10KSE +/- 11.73, N = 3SE +/- 66.57, N = 3SE +/- 41.59, N = 38298814980751. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: Forking

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: ForkingNo ProtectionDefault Kernel + User ProtectionKernel Protection13K26K39K52K65KSE +/- 278.00, N = 3SE +/- 106.66, N = 3SE +/- 643.64, N = 36058559360585921. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: Hsearch

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: HsearchNo ProtectionKernel ProtectionDefault Kernel + User Protection40K80K120K160K200KSE +/- 1283.68, N = 3SE +/- 4467.03, N = 12SE +/- 3274.73, N = 122016601869311641381. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: Lsearch

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: LsearchNo ProtectionKernel ProtectionDefault Kernel + User Protection80160240320400SE +/- 0.70, N = 3SE +/- 1.09, N = 3SE +/- 1.34, N = 33893862021. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: Tsearch

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: TsearchNo ProtectionKernel ProtectionDefault Kernel + User Protection2004006008001000SE +/- 2.73, N = 3SE +/- 2.77, N = 3SE +/- 5.85, N = 38988898591. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: CPU Stress

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: CPU StressNo ProtectionKernel ProtectionDefault Kernel + User Protection3K6K9K12K15KSE +/- 9.31, N = 3SE +/- 18.47, N = 3SE +/- 7.41, N = 31240212321122791. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: Semaphores

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: SemaphoresNo ProtectionDefault Kernel + User ProtectionKernel Protection100K200K300K400K500KSE +/- 8664.98, N = 3SE +/- 4183.35, N = 3SE +/- 5258.60, N = 34566383475173471661. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: Vector Math

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: Vector MathNo ProtectionKernel ProtectionDefault Kernel + User Protection20K40K60K80K100KSE +/- 225.56, N = 3SE +/- 212.84, N = 3SE +/- 114.43, N = 38869988416881541. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: Memory Copying

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: Memory CopyingNo ProtectionKernel ProtectionDefault Kernel + User Protection2K4K6K8K10KSE +/- 154.42, N = 3SE +/- 154.58, N = 3SE +/- 122.58, N = 38368834083171. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: Socket Activity

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: Socket ActivityNo ProtectionKernel ProtectionDefault Kernel + User Protection6K12K18K24K30KSE +/- 211.09, N = 3SE +/- 103.85, N = 3SE +/- 143.39, N = 32871727094262721. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: Glibc C String Functions

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: Glibc C String FunctionsNo ProtectionKernel ProtectionDefault Kernel + User Protection300K600K900K1200K1500KSE +/- 14521.20, N = 3SE +/- 8505.49, N = 3SE +/- 6341.64, N = 31379226137082412905871. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: Glibc Qsort Data Sorting

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: Glibc Qsort Data SortingNo ProtectionKernel ProtectionDefault Kernel + User Protection80160240320400SE +/- 1.48, N = 3SE +/- 0.71, N = 3SE +/- 0.49, N = 33693682531. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

Stress-NG

Test: System V Message Passing

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.07.26Test: System V Message PassingNo ProtectionKernel ProtectionDefault Kernel + User Protection1.4M2.8M4.2M5.6M7MSE +/- 92751.45, N = 3SE +/- 32592.10, N = 3SE +/- 63949.45, N = 36607012649220462936171. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc

PyBench

Total For Average Test Times

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesNo ProtectionKernel ProtectionDefault Kernel + User Protection9001800270036004500SE +/- 3.06, N = 3SE +/- 5.24, N = 3181619394090

PHPBench

PHP Benchmark Suite

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteKernel ProtectionNo ProtectionDefault Kernel + User Protection70K140K210K280K350KSE +/- 554.58, N = 3SE +/- 3520.86, N = 3SE +/- 1702.94, N = 3330044329783159731


Phoronix Test Suite v10.8.4