retpoline-testing

Tests for a future article.

HTML result view exported from: https://openbenchmarking.org/result/1801075-AL-RETPOLINE03&sor.

ProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay DriverOpenCLCompilerFile-SystemScreen ResolutionEPYC 76012 x Xeon Gold 6138 noretpoline Retpoline Retpoline + GCC no retpoline Retpoline Retpoline + GCCAMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads)TYAN B8026T70AE24HRAMD Device 1450129024MB280GB INTEL SSDPE21D280GAASPEED ASPEED FamilyVE228Broadcom Limited NetXtreme BCM5720 Gigabit PCIeUbuntu 17.104.14.0-phx-retpoline (x86_64)GNOME Shell 3.26.1modesetting 1.19.5OpenCL 1.2 pocl 1.0 LLVM 5.0.0GCC 7.2.0 + Clang 5.0.0-3 + LLVM 5.0.0ext41920x10804.14.0-phx-retpoline-gcc-retpo (x86_64)2 x Intel Xeon Gold 6138 @ 3.70GHz (40 Cores / 80 Threads)TYAN S7106Intel Device 202096256MB256GB Samsung SSD 850 + 2000GB Seagate ST2000DM006-2DM1 + 2 x 120GB TOSHIBA-TR150Intel I210 Gigabit Connection4.14.0-phx-retpoline (x86_64)GCC 7.2.04.14.0-phx-retpoline-gcc-retpo (x86_64)OpenBenchmarking.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++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none --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 --with-tune=generic --without-cuda-driver -v Disk Details- EPYC 7601: noretpoline: NONE / data=ordered,errors=remount-ro,relatime,rw- EPYC 7601: Retpoline: NONE / data=ordered,errors=remount-ro,relatime,rw- EPYC 7601: Retpoline + GCC: NONE / data=ordered,errors=remount-ro,relatime,rw- 2 x Xeon Gold 6138: no retpoline: CFQ / data=ordered,errors=remount-ro,relatime,rw- 2 x Xeon Gold 6138: Retpoline: CFQ / data=ordered,errors=remount-ro,relatime,rw- 2 x Xeon Gold 6138: Retpoline + GCC: CFQ / data=ordered,errors=remount-ro,relatime,rwProcessor Details- EPYC 7601: noretpoline: Scaling Governor: acpi-cpufreq ondemand- EPYC 7601: Retpoline: Scaling Governor: acpi-cpufreq ondemand- EPYC 7601: Retpoline + GCC: Scaling Governor: acpi-cpufreq ondemand- 2 x Xeon Gold 6138: no retpoline: Scaling Governor: intel_pstate powersave- 2 x Xeon Gold 6138: Retpoline: Scaling Governor: intel_pstate powersave- 2 x Xeon Gold 6138: Retpoline + GCC: Scaling Governor: intel_pstate powersaveSystem Details- Python 2.7.14.

fio: Rand Read - Libaio - No - Yes - 2MB - Default Test Directoryfio: Rand Read - Libaio - No - Yes - 4KB - Default Test Directoryfio: Rand Write - Libaio - No - Yes - 2MB - Default Test Directoryfio: Rand Write - Libaio - No - Yes - 4KB - Default Test Directoryfio: Seq Read - Libaio - No - Yes - 2MB - Default Test Directoryfio: Seq Read - Libaio - No - Yes - 4KB - Default Test Directoryfio: Seq Write - Libaio - No - Yes - 2MB - Default Test Directoryfio: Seq Write - Libaio - No - Yes - 4KB - Default Test Directoryfs-mark: 1000 Files, 1MB Sizefs-mark: 4000 Files, 32 Sub Dirs, 1MB Sizecompilebench: Compilecompilebench: Initial Createt-test1: 1t-test1: 2parboil: OpenMP CUTCPparboil: OpenMP MRI Griddingrodinia: OpenMP LavaMDrodinia: OpenMP CFD Solverlzbench: XZ 0 - Compressionlzbench: Zstd 1 - Compressioncachebench: Readcachebench: Writecachebench: Read / Modify / Writejohn-the-ripper: Blowfishebizzy: build-apache: Time To Compilebuild-linux-kernel: Time To Compilec-ray: Total Timestockfish: Total Timecompress-lzma: 256MB File Compressionglibc-bench: ffsglibc-bench: sqrtglibc-bench: pthread_oncetjbench: Decompression Throughputredis: LPOPredis: SADDredis: LPUSHredis: GETredis: SETpybench: Total For Average Test Timesapache: Static Web Page Servingscikit-learn: pgbench: Buffer Test - Normal Load - Read Onlypgbench: Buffer Test - Normal Load - Read WriteEPYC 76012 x Xeon Gold 6138 noretpoline Retpoline Retpoline + GCC no retpoline Retpoline Retpoline + GCC2509.831176.201972.701059.732511.771183.531971.101081.23597.93581.871696.87410.8537.7614.382.71283.1631.6510.97243352205.7321622.7722924.2435458109323031.7138.303.504507329.574.834.704.83140.831520530.251233207.561117859.501394712.791131889.69179416587.3434.002615.601165.972157.601046.932618.671160.832158.031091.50639.77629.271691.37406.4136.6814.462.73283.2531.8610.76243362215.2421556.9822918.9635714107358732.2138.443.454507328.954.834.704.83140.991394465.791293172.501147686.871362255.921161527.67180116838.7633.912637.271128.572158.271048.772618.201193.802156.871072.57619.67631.301693.10406.9138.1814.662.70288.3631.8711.08243352214.2521835.7622900.813571595269132.1238.523.484501329.004.834.704.83140.871230732.381249051.081123252.421352327.021162399.91179516716.4834.47543.05383.14523.89338.58519.38457.26514.75415.49128.67111.021496.78533.7065.9123.122.35407.9128.509.73323612973.0824954.9625777.794965294632126.2430.253.163437281.843.2612.303.87145.811423431.251589703.691357638.811627717.711564339.04130722258.72184.75599258.142136.64543.18384.79514.67341.01518.49454.44526.12413.25133.73136.671694.04508.9266.5623.292.38413.8528.709.93303692872.0923882.8825061.334839995477326.3130.463.153654281.253.269.953.80154.021486377.461573307.831382896.041738340.061525833.21130621347.64185.68597483.032467.38543.37388.19512.18341.98518.48456.92527.34416.94129.63111.171637.02478.7664.4523.042.38405.2528.4310.54313702839.7923753.7924928.445108995067126.3230.633.183613282.043.269.964.49146.511347143.441604656.601454926.311574731.421399642.27131518757.65186.08577641.993976.32OpenBenchmarking.org

Flexible IO Tester

Type: Random Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 2.1.13Type: Random Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryRetpoline + GCCRetpolinenoretpolineno retpoline6001200180024003000SE +/- 22.27, N = 3SE +/- 0.45, N = 3SE +/- 0.24, N = 3SE +/- 0.10, N = 3SE +/- 0.14, N = 3SE +/- 1.17, N = 32637.272615.602509.83543.37543.18543.051. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Random Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 2.1.13Type: Random Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryRetpoline + GCCRetpolinenoretpolineno retpoline30060090012001500SE +/- 11.00, N = 3SE +/- 0.33, N = 3SE +/- 0.67, N = 31315130412512622622621. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Random Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 2.1.13Type: Random Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectorynoretpolineRetpolineRetpoline + GCCno retpoline30060090012001500SE +/- 17.24, N = 3SE +/- 13.29, N = 3SE +/- 1.94, N = 3SE +/- 2.02, N = 3SE +/- 1.24, N = 3SE +/- 4.29, N = 31176.201165.971128.57388.19384.79383.141. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Random Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 2.1.13Type: Random Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectorynoretpolineRetpolineRetpoline + GCCno retpoline60K120K180K240K300KSE +/- 4402.78, N = 3SE +/- 3375.56, N = 3SE +/- 591.14, N = 3SE +/- 504.90, N = 3SE +/- 309.68, N = 3SE +/- 1071.43, N = 33010372984282889519704396194957821. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Random Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 2.1.13Type: Random Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryRetpoline + GCCRetpolinenoretpolineno retpoline5001000150020002500SE +/- 3.18, N = 3SE +/- 3.16, N = 3SE +/- 1.43, N = 3SE +/- 0.94, N = 3SE +/- 2.04, N = 3SE +/- 2.16, N = 32158.272157.601972.70523.89514.67512.181. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Random Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 2.1.13Type: Random Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryRetpoline + GCCRetpolinenoretpolineno retpoline2004006008001000SE +/- 1.67, N = 3SE +/- 1.53, N = 3SE +/- 0.67, N = 3SE +/- 0.58, N = 3SE +/- 1.20, N = 3SE +/- 1.20, N = 3107610759832522482461. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Random Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 2.1.13Type: Random Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectorynoretpolineRetpoline + GCCRetpolineno retpoline2004006008001000SE +/- 11.43, N = 3SE +/- 11.31, N = 3SE +/- 17.28, N = 6SE +/- 1.80, N = 3SE +/- 1.78, N = 3SE +/- 1.94, N = 31059.731048.771046.93341.98341.01338.581. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Random Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 2.1.13Type: Random Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectorynoretpolineRetpoline + GCCRetpolineno retpoline60K120K180K240K300KSE +/- 2932.09, N = 3SE +/- 2976.09, N = 3SE +/- 4419.58, N = 6SE +/- 451.27, N = 3SE +/- 444.91, N = 3SE +/- 484.83, N = 32712732685422680418549285249846411. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Sequential Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 2.1.13Type: Sequential Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryRetpolineRetpoline + GCCnoretpolineno retpoline6001200180024003000SE +/- 0.03, N = 3SE +/- 0.10, N = 3SE +/- 0.03, N = 3SE +/- 0.72, N = 3SE +/- 0.25, N = 3SE +/- 0.05, N = 32618.672618.202511.77519.38518.49518.481. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Sequential Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 2.1.13Type: Sequential Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryRetpoline + GCCRetpolinenoretpolineno retpoline30060090012001500SE +/- 0.33, N = 3SE +/- 0.33, N = 31306130612522502502501. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Sequential Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 2.1.13Type: Sequential Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryRetpoline + GCCnoretpolineRetpolineno retpoline30060090012001500SE +/- 17.58, N = 3SE +/- 18.11, N = 3SE +/- 6.24, N = 3SE +/- 0.98, N = 3SE +/- 1.52, N = 3SE +/- 0.14, N = 31193.801183.531160.83457.26456.92454.441. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Sequential Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 2.1.13Type: Sequential Read - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryRetpoline + GCCnoretpolineRetpolineno retpoline70K140K210K280K350KSE +/- 4452.90, N = 3SE +/- 4646.93, N = 3SE +/- 1598.66, N = 3SE +/- 246.03, N = 3SE +/- 379.04, N = 3SE +/- 33.79, N = 33056653029502971441143111142271136081. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Sequential Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 2.1.13Type: Sequential Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryRetpolineRetpoline + GCCnoretpolineno retpoline5001000150020002500SE +/- 3.77, N = 3SE +/- 4.03, N = 3SE +/- 1.26, N = 3SE +/- 0.39, N = 3SE +/- 0.90, N = 3SE +/- 8.78, N = 42158.032156.871971.10527.34526.12514.751. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Sequential Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 2.1.13Type: Sequential Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryRetpoline + GCCRetpolinenoretpolineno retpoline2004006008001000SE +/- 1.86, N = 3SE +/- 1.86, N = 3SE +/- 0.58, N = 3SE +/- 4.34, N = 4107510759822542532481. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Sequential Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 2.1.13Type: Sequential Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryRetpolinenoretpolineRetpoline + GCCno retpoline2004006008001000SE +/- 17.15, N = 6SE +/- 14.34, N = 3SE +/- 7.38, N = 3SE +/- 1.44, N = 3SE +/- 2.46, N = 3SE +/- 0.97, N = 31091.501081.231072.57416.94415.49413.251. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

Flexible IO Tester

Type: Sequential Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 2.1.13Type: Sequential Write - IO Engine: Libaio - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryRetpolinenoretpolineRetpoline + GCCno retpoline60K120K180K240K300KSE +/- 4389.73, N = 6SE +/- 3663.70, N = 3SE +/- 1888.66, N = 3SE +/- 361.30, N = 3SE +/- 614.11, N = 3SE +/- 241.21, N = 32794042767782745301042321038681033081. (CC) gcc options: -rdynamic -std=gnu99 -O3 -ffast-math -include -lrt -laio -lz -lm -lpthread -ldl

FS-Mark

Test: 1000 Files, 1MB Size

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 1000 Files, 1MB SizeRetpolineRetpoline + GCCnoretpolineno retpoline140280420560700SE +/- 1.90, N = 3SE +/- 3.59, N = 3SE +/- 5.75, N = 3SE +/- 0.20, N = 3SE +/- 2.15, N = 6SE +/- 2.17, N = 3639.77619.67597.93133.73129.63128.671. (CC) gcc options: -static

FS-Mark

Test: 4000 Files, 32 Sub Dirs, 1MB Size

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 4000 Files, 32 Sub Dirs, 1MB SizeRetpoline + GCCRetpolinenoretpolineno retpoline140280420560700SE +/- 4.00, N = 3SE +/- 1.74, N = 3SE +/- 1.13, N = 3SE +/- 0.54, N = 3SE +/- 7.23, N = 6SE +/- 2.97, N = 6631.30629.27581.87136.67111.17111.021. (CC) gcc options: -static

Compile Bench

Test: Compile

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: CompilenoretpolineRetpolineRetpoline + GCCno retpoline400800120016002000SE +/- 9.70, N = 3SE +/- 4.84, N = 3SE +/- 3.64, N = 3SE +/- 62.71, N = 6SE +/- 40.27, N = 6SE +/- 7.13, N = 31696.871693.101691.371694.041637.021496.78

Compile Bench

Test: Initial Create

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Initial Createno retpolineRetpolineRetpoline + GCCnoretpoline120240360480600SE +/- 1.88, N = 3SE +/- 25.41, N = 3SE +/- 27.64, N = 3SE +/- 2.54, N = 3SE +/- 4.20, N = 3SE +/- 3.88, N = 3533.70508.92478.76410.85406.91406.41

t-test1

Threads: 1

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgSeconds, Fewer Is Bettert-test1 2017-01-13Threads: 1RetpolinenoretpolineRetpoline + GCCno retpoline1530456075SE +/- 0.10, N = 3SE +/- 0.09, N = 3SE +/- 0.12, N = 3SE +/- 0.14, N = 3SE +/- 0.15, N = 3SE +/- 0.09, N = 336.6837.7638.1864.4565.9166.561. (CC) gcc options: -pthread

t-test1

Threads: 2

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgSeconds, Fewer Is Bettert-test1 2017-01-13Threads: 2noretpolineRetpolineRetpoline + GCCno retpoline612182430SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 0.13, N = 3SE +/- 0.06, N = 314.3814.4614.6623.0423.1223.291. (CC) gcc options: -pthread

Parboil

Test: OpenMP CUTCP

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP CUTCPno retpolineRetpolineRetpoline + GCCnoretpoline0.61431.22861.84292.45723.0715SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 62.352.382.382.702.712.731. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Parboil

Test: OpenMP MRI Gridding

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP MRI GriddingnoretpolineRetpolineRetpoline + GCCno retpoline90180270360450SE +/- 0.38, N = 3SE +/- 0.14, N = 3SE +/- 1.39, N = 3SE +/- 6.86, N = 4SE +/- 6.83, N = 3SE +/- 6.92, N = 3283.16283.25288.36405.25407.91413.851. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Rodinia

Test: OpenMP LavaMD

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP LavaMDRetpoline + GCCno retpolineRetpolinenoretpoline714212835SE +/- 0.08, N = 3SE +/- 0.11, N = 3SE +/- 0.16, N = 3SE +/- 0.14, N = 3SE +/- 0.08, N = 3SE +/- 0.14, N = 328.4328.5028.7031.6531.8631.871. (CXX) g++ options: -O2 -lOpenCL

Rodinia

Test: OpenMP CFD Solver

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP CFD Solverno retpolineRetpolineRetpoline + GCCnoretpoline3691215SE +/- 0.16, N = 3SE +/- 0.15, N = 5SE +/- 0.30, N = 6SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.11, N = 39.739.9310.5410.7610.9711.081. (CXX) g++ options: -O2 -lOpenCL

lzbench

Test: XZ 0 - Process: Compression

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgMB/s, More Is Betterlzbench 2017-08-08Test: XZ 0 - Process: Compressionno retpolineRetpoline + GCCRetpolinenoretpoline714212835SE +/- 0.72, N = 6SE +/- 0.50, N = 6SE +/- 0.56, N = 63231302424241. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

lzbench

Test: XZ 0 - Process: Decompression

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgMB/s, More Is Betterlzbench 2017-08-08Test: XZ 0 - Process: Decompressionno retpolineRetpolineRetpoline + GCCnoretpoline20406080100SE +/- 1.03, N = 6SE +/- 2.00, N = 6SE +/- 1.77, N = 68987867777761. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

lzbench

Test: Zstd 1 - Process: Compression

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgMB/s, More Is Betterlzbench 2017-08-08Test: Zstd 1 - Process: CompressionRetpoline + GCCRetpolineno retpolinenoretpoline80160240320400SE +/- 2.08, N = 3SE +/- 2.52, N = 3SE +/- 0.88, N = 3SE +/- 1.33, N = 3SE +/- 1.20, N = 3SE +/- 1.33, N = 33703693613363353351. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

lzbench

Test: Zstd 1 - Process: Decompression

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgMB/s, More Is Betterlzbench 2017-08-08Test: Zstd 1 - Process: DecompressionRetpoline + GCCRetpolineno retpolinenoretpoline2004006008001000SE +/- 3.00, N = 3SE +/- 4.18, N = 3SE +/- 4.58, N = 3SE +/- 4.51, N = 3SE +/- 4.33, N = 3SE +/- 3.93, N = 39839829759119119101. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

CacheBench

Test: Read

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgMB/s, More Is BetterCacheBenchTest: Readno retpolineRetpolineRetpoline + GCCnoretpoline6001200180024003000SE +/- 7.30, N = 3SE +/- 12.31, N = 3SE +/- 11.39, N = 3SE +/- 0.02, N = 3SE +/- 0.98, N = 3SE +/- 9.38, N = 32973.082872.092839.792215.242214.252205.731. (CC) gcc options: -lrt

CacheBench

Test: Write

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgMB/s, More Is BetterCacheBenchTest: Writeno retpolineRetpolineRetpoline + GCCnoretpoline5K10K15K20K25KSE +/- 34.79, N = 3SE +/- 197.71, N = 3SE +/- 82.06, N = 3SE +/- 25.70, N = 3SE +/- 53.65, N = 3SE +/- 133.19, N = 324954.9623882.8823753.7921835.7621622.7721556.981. (CC) gcc options: -lrt

CacheBench

Test: Read / Modify / Write

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgMB/s, More Is BetterCacheBenchTest: Read / Modify / Writeno retpolineRetpolineRetpoline + GCCnoretpoline6K12K18K24K30KSE +/- 112.30, N = 3SE +/- 271.49, N = 3SE +/- 226.86, N = 3SE +/- 5.63, N = 3SE +/- 9.68, N = 3SE +/- 12.91, N = 325777.7925061.3324928.4422924.2422918.9622900.811. (CC) gcc options: -lrt

John The Ripper

Test: Blowfish

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.8.0Test: BlowfishRetpoline + GCCno retpolineRetpolinenoretpoline11K22K33K44K55KSE +/- 357.46, N = 3SE +/- 562.61, N = 3SE +/- 1882.95, N = 6SE +/- 232.00, N = 3SE +/- 192.26, N = 3SE +/- 400.35, N = 35108949652483993571535714354581. (CC) gcc options: -fopenmp -lcrypt

ebizzy

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgRecords/s, More Is Betterebizzy 0.3noretpolineRetpolineRetpoline + GCCno retpoline200K400K600K800K1000KSE +/- 19950.45, N = 3SE +/- 26430.87, N = 6SE +/- 23463.01, N = 6SE +/- 13516.78, N = 6SE +/- 13692.48, N = 5SE +/- 7835.23, N = 3109323010735879526919547739506719463211. (CC) gcc options: -pthread -lpthread -O3 -march=native

Timed Apache Compilation

Time To Compile

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Apache Compilation 2.4.7Time To Compileno retpolineRetpolineRetpoline + GCCnoretpoline714212835SE +/- 0.06, N = 3SE +/- 0.14, N = 3SE +/- 0.10, N = 3SE +/- 0.14, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 326.2426.3126.3231.7132.1232.21

Timed Linux Kernel Compilation

Time To Compile

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 4.13Time To Compileno retpolineRetpolineRetpoline + GCCnoretpoline918273645SE +/- 0.75, N = 6SE +/- 0.77, N = 6SE +/- 0.80, N = 6SE +/- 0.68, N = 6SE +/- 0.76, N = 6SE +/- 0.63, N = 630.2530.4630.6338.3038.4438.52

C-Ray

Total Time

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total TimeRetpolineno retpolineRetpoline + GCCnoretpoline0.78751.5752.36253.153.9375SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 33.153.163.183.453.483.501. (CC) gcc options: -lm -lpthread -O3

Stockfish

Total Time

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgms, Fewer Is BetterStockfish 2014-11-26Total Timeno retpolineRetpoline + GCCRetpolinenoretpoline10002000300040005000SE +/- 37.22, N = 3SE +/- 210.25, N = 6SE +/- 202.73, N = 6SE +/- 2.65, N = 3SE +/- 3.48, N = 33437361336544501450745071. (CXX) g++ options: -lpthread -fno-exceptions -fno-rtti -ansi -pedantic -O3 -msse -msse3 -mpopcnt -flto

LZMA Compression

256MB File Compression

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgSeconds, Fewer Is BetterLZMA Compression256MB File CompressionRetpolineno retpolineRetpoline + GCCnoretpoline70140210280350SE +/- 0.60, N = 3SE +/- 0.34, N = 3SE +/- 2.80, N = 3SE +/- 0.22, N = 3SE +/- 0.42, N = 3SE +/- 0.18, N = 3281.25281.84282.04328.95329.00329.571. (CXX) g++ options: -O2

glibc bench

Benchmark: ffs

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: ffsno retpolineRetpolineRetpoline + GCCnoretpoline1.08682.17363.26044.34725.434SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 33.263.263.264.834.834.83

glibc bench

Benchmark: sqrt

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: sqrtnoretpolineRetpolineRetpoline + GCCno retpoline3691215SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.62, N = 6SE +/- 0.62, N = 6SE +/- 0.00, N = 34.704.704.709.959.9612.30

glibc bench

Benchmark: pthread_once

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgnanoseconds, Fewer Is Betterglibc bench 1.0Benchmark: pthread_onceRetpolineno retpolineRetpoline + GCCnoretpoline1.08682.17363.26044.34725.434SE +/- 0.00, N = 3SE +/- 0.06, N = 3SE +/- 0.24, N = 6SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 33.803.874.494.834.834.83

libjpeg-turbo tjbench

Test: Decompression Throughput

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgMegapixels/sec, More Is Betterlibjpeg-turbo tjbench 1.5.1Test: Decompression ThroughputRetpolineRetpoline + GCCno retpolinenoretpoline306090120150SE +/- 1.27, N = 3SE +/- 6.27, N = 6SE +/- 5.29, N = 6SE +/- 0.00, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 3154.02146.51145.81140.99140.87140.831. (CC) gcc options: -O3 -lm

Redis

Test: LPOP

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 3.0.1Test: LPOPnoretpolineRetpolineno retpolineRetpoline + GCC300K600K900K1200K1500KSE +/- 1539.75, N = 3SE +/- 61195.38, N = 6SE +/- 11491.58, N = 3SE +/- 77843.26, N = 6SE +/- 36207.27, N = 6SE +/- 25234.38, N = 61520530.251394465.791230732.381486377.461423431.251347143.441. (CC) gcc options: -ggdb -rdynamic -lm -pthread

Redis

Test: SADD

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 3.0.1Test: SADDRetpoline + GCCno retpolineRetpolinenoretpoline300K600K900K1200K1500KSE +/- 62798.70, N = 6SE +/- 68212.97, N = 6SE +/- 95230.31, N = 6SE +/- 51667.16, N = 6SE +/- 7598.74, N = 3SE +/- 19184.10, N = 41604656.601589703.691573307.831293172.501249051.081233207.561. (CC) gcc options: -ggdb -rdynamic -lm -pthread

Redis

Test: LPUSH

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 3.0.1Test: LPUSHRetpoline + GCCRetpolineno retpolinenoretpoline300K600K900K1200K1500KSE +/- 49927.23, N = 6SE +/- 51060.27, N = 6SE +/- 57810.96, N = 6SE +/- 3427.20, N = 3SE +/- 6595.02, N = 3SE +/- 8338.77, N = 31454926.311382896.041357638.811147686.871123252.421117859.501. (CC) gcc options: -ggdb -rdynamic -lm -pthread

Redis

Test: GET

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 3.0.1Test: GETRetpolineno retpolineRetpoline + GCCnoretpoline400K800K1200K1600K2000KSE +/- 96355.24, N = 6SE +/- 50934.82, N = 6SE +/- 80851.67, N = 6SE +/- 2966.06, N = 3SE +/- 27266.47, N = 3SE +/- 28137.42, N = 61738340.061627717.711574731.421394712.791362255.921352327.021. (CC) gcc options: -ggdb -rdynamic -lm -pthread

Redis

Test: SET

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 3.0.1Test: SETno retpolineRetpolineRetpoline + GCCnoretpoline300K600K900K1200K1500KSE +/- 28059.92, N = 3SE +/- 90449.67, N = 6SE +/- 55833.41, N = 6SE +/- 5879.00, N = 3SE +/- 7115.17, N = 3SE +/- 26854.76, N = 61564339.041525833.211399642.271162399.911161527.671131889.691. (CC) gcc options: -ggdb -rdynamic -lm -pthread

PyBench

Total For Average Test Times

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2008-08-14Total For Average Test TimesRetpolineno retpolineRetpoline + GCCnoretpoline400800120016002000SE +/- 1.53, N = 3SE +/- 1.76, N = 3SE +/- 7.97, N = 3SE +/- 11.33, N = 3130613071315179417951801

Apache Benchmark

Static Web Page Serving

2 x Xeon Gold 6138EPYC 7601OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.7Static Web Page Servingno retpolineRetpolineRetpoline + GCCnoretpoline5K10K15K20K25KSE +/- 139.49, N = 3SE +/- 63.48, N = 3SE +/- 51.11, N = 3SE +/- 97.15, N = 3SE +/- 296.19, N = 3SE +/- 41.45, N = 322258.7221347.6418757.6516838.7616716.4816587.341. (CC) gcc options: -shared -fPIC -O2 -pthread

Scikit-Learn

EPYC 76012 x Xeon Gold 6138OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 0.17.1RetpolinenoretpolineRetpoline + GCCno retpoline4080120160200SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.48, N = 3SE +/- 0.99, N = 3SE +/- 0.29, N = 3SE +/- 2.58, N = 333.9134.0034.47184.75185.68186.08

PostgreSQL pgbench

Scaling: Buffer Test - Test: Normal Load - Mode: Read Only

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 10.0Scaling: Buffer Test - Test: Normal Load - Mode: Read Onlyno retpolineRetpolineRetpoline + GCC130K260K390K520K650KSE +/- 2613.28, N = 3SE +/- 1869.94, N = 3SE +/- 4346.70, N = 3599258.14597483.03577641.991. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -fPIC -lpgcommon -lpgport -lpthread -lrt -lcrypt -ldl -lm

PostgreSQL pgbench

Scaling: Buffer Test - Test: Normal Load - Mode: Read Write

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 10.0Scaling: Buffer Test - Test: Normal Load - Mode: Read WriteRetpoline + GCCRetpolineno retpoline9001800270036004500SE +/- 15.09, N = 3SE +/- 49.01, N = 6SE +/- 40.05, N = 33976.322467.382136.641. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -fPIC -lpgcommon -lpgport -lpthread -lrt -lcrypt -ldl -lm

Phoronix Test Suite v10.8.4