Intel Xeon Gold FGKASLR Linux Testing

Tests for a future article by Michael Larabel.

HTML result view exported from: https://openbenchmarking.org/result/2006275-NE-XEONGOLDF45&rdt&grr.

Intel Xeon Gold FGKASLR Linux TestingProcessorMotherboardChipsetMemoryDiskGraphicsNetworkOSKernelDesktopDisplay ServerDisplay DriverCompilerFile-SystemFGKASLR EnabledJust KASLRNo KASLR/FGKASLR2 x Intel Xeon Gold 5220R @ 3.90GHz (36 Cores / 72 Threads)TYAN S7106 (V2.01.B40 BIOS)Intel Sky Lake-E DMI3 Registers94GB500GB Samsung SSD 860ASPEED2 x Intel I210 + 2 x QLogic cLOM8214 1/10GbEUbuntu 20.045.8.0-rc2-phx-fgkaslr (x86_64) 20200624GNOME Shell 3.36.1X Server 1.20.8modesetting 1.20.8GCC 9.3.0ext4OpenBenchmarking.orgEnvironment Details- CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native"Compiler 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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,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 Disk Details- MQ-DEADLINE / errors=remount-ro,relatime,rwProcessor Details- Scaling Governor: intel_pstate powersave - CPU Microcode: 0x5002f01Java Details- OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-3ubuntu1)Python Details- Python 3.8.2Security Details- itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Intel Xeon Gold FGKASLR Linux Testingleveldb: Seq Fillleveldb: Seq Fillleveldb: Rand Deleteblogbench: Readrocksdb: Seq Fillbuild-llvm: Time To Compilepgbench: Buffer Test - Normal Load - Read Onlystress-ng: Context Switchingfio: Rand Write - IO_uring - No - Yes - 4KB - Default Test Directoryfio: Rand Write - IO_uring - Yes - No - 4KB - Default Test Directoryleveldb: Rand Fillleveldb: Rand Fillleveldb: Overwriteleveldb: Overwritemlpack: scikit_icaapache: Static Web Page Servingsqlite-speedtest: Timed Time - Size 1,000rocksdb: Rand Fillsockperf: Latency Ping Pongsockperf: Throughputethr: UDP - Bandwidth - 64stress-ng: Semaphoresosbench: Create Processesosbench: Create Filesosbench: Memory Allocationsosbench: Launch Programsosbench: Create ThreadsFGKASLR EnabledJust KASLRNo KASLR/FGKASLR859.5529.3825.4212262213183628284.715583335.8032835164254.027954061400850.9019.4840.8079.575.1423262.3866.1661798624.545499722301327.333333334379605.1632.76880015.65583895.58796947.98730218.009345824.7749.7770.1132302683187803282.859589351.4549407205230.498333363367809.4109.8800.3529.975.0223111.2966.4691835354.523508984314693.733333334457335.2832.92004314.87382194.65670648.57937517.846425817.6559.7768.5672341662189127285.347592722.4420566369133.358550071625801.9679.9801.5019.973.8724595.6066.9501860454.476506506316768.666666674456094.9531.68424014.65370594.28676046.85322517.757416OpenBenchmarking.org

LevelDB

Benchmark: Sequential Fill

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Sequential FillFGKASLR EnabledJust KASLRNo KASLR/FGKASLR2004006008001000SE +/- 1.96, N = 3SE +/- 2.40, N = 3SE +/- 0.53, N = 3859.55824.77817.661. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

LevelDB

Benchmark: Sequential Fill

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Sequential FillFGKASLR EnabledJust KASLRNo KASLR/FGKASLR3691215SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 39.39.79.71. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

LevelDB

Benchmark: Random Delete

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random DeleteFGKASLR EnabledJust KASLRNo KASLR/FGKASLR2004006008001000SE +/- 2.40, N = 3SE +/- 0.57, N = 3SE +/- 2.61, N = 3825.42770.11768.571. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

BlogBench

Test: Read

OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: ReadFGKASLR EnabledJust KASLRNo KASLR/FGKASLR500K1000K1500K2000K2500KSE +/- 10482.62, N = 3SE +/- 34384.59, N = 4SE +/- 16967.16, N = 32262213230268323416621. (CC) gcc options: -O3 -march=native -pthread

Facebook RocksDB

Test: Sequential Fill

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Sequential FillFGKASLR EnabledJust KASLRNo KASLR/FGKASLR40K80K120K160K200KSE +/- 184.88, N = 3SE +/- 211.79, N = 3SE +/- 81.95, N = 31836281878031891271. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

Timed LLVM Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 10.0Time To CompileFGKASLR EnabledJust KASLRNo KASLR/FGKASLR60120180240300SE +/- 3.69, N = 3SE +/- 2.87, N = 3SE +/- 1.49, N = 3284.72282.86285.35

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Normal Load - Mode: Read OnlyFGKASLR EnabledJust KASLRNo KASLR/FGKASLR130K260K390K520K650KSE +/- 1841.14, N = 3SE +/- 2925.75, N = 3SE +/- 1757.77, N = 3583335.80589351.45592722.441. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

Stress-NG

Test: Context Switching

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Context SwitchingFGKASLR EnabledJust KASLRNo KASLR/FGKASLR1.5M3M4.5M6M7.5MSE +/- 47027.87, N = 15SE +/- 167659.86, N = 12SE +/- 92758.51, N = 155164254.027205230.496369133.351. (CC) gcc options: -O3 -march=native -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

Flexible IO Tester

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

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.18Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryFGKASLR EnabledJust KASLRNo KASLR/FGKASLR20K40K60K80K100KSE +/- 1926.90, N = 15SE +/- 922.47, N = 12SE +/- 404.15, N = 37954083333855001. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native -ll -lcurl -lssl -lcrypto -lnuma -libverbs -lrt -laio -lz -lpthread -lm -ldl

Flexible IO Tester

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

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.18Type: Random Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test DirectoryFGKASLR EnabledJust KASLRNo KASLR/FGKASLR15K30K45K60K75KSE +/- 1875.80, N = 12SE +/- 66.67, N = 3SE +/- 2930.93, N = 126140063367716251. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native -ll -lcurl -lssl -lcrypto -lnuma -libverbs -lrt -laio -lz -lpthread -lm -ldl

LevelDB

Benchmark: Random Fill

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random FillFGKASLR EnabledJust KASLRNo KASLR/FGKASLR2004006008001000SE +/- 4.95, N = 3SE +/- 4.31, N = 3SE +/- 3.91, N = 3850.90809.41801.971. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

LevelDB

Benchmark: Random Fill

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Random FillFGKASLR EnabledJust KASLRNo KASLR/FGKASLR3691215SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 39.49.89.91. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

LevelDB

Benchmark: Overwrite

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: OverwriteFGKASLR EnabledJust KASLRNo KASLR/FGKASLR2004006008001000SE +/- 2.58, N = 3SE +/- 4.48, N = 3SE +/- 2.57, N = 3840.81800.35801.501. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

LevelDB

Benchmark: Overwrite

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: OverwriteFGKASLR EnabledJust KASLRNo KASLR/FGKASLR3691215SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 39.59.99.91. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

Mlpack Benchmark

Benchmark: scikit_ica

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_icaFGKASLR EnabledJust KASLRNo KASLR/FGKASLR20406080100SE +/- 0.42, N = 3SE +/- 0.60, N = 3SE +/- 0.52, N = 375.1475.0273.87

Apache Benchmark

Static Web Page Serving

OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.29Static Web Page ServingFGKASLR EnabledJust KASLRNo KASLR/FGKASLR5K10K15K20K25KSE +/- 382.53, N = 3SE +/- 244.89, N = 8SE +/- 68.62, N = 323262.3823111.2924595.601. (CC) gcc options: -shared -fPIC -pthread -O3 -march=native

SQLite Speedtest

Timed Time - Size 1,000

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000FGKASLR EnabledJust KASLRNo KASLR/FGKASLR1530456075SE +/- 0.42, N = 3SE +/- 0.77, N = 3SE +/- 0.22, N = 366.1766.4766.951. (CC) gcc options: -O3 -march=native -ldl -lz -lpthread

Facebook RocksDB

Test: Random Fill

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random FillFGKASLR EnabledJust KASLRNo KASLR/FGKASLR40K80K120K160K200KSE +/- 95.98, N = 3SE +/- 545.65, N = 3SE +/- 618.48, N = 31798621835351860451. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

Sockperf

Test: Latency Ping Pong

OpenBenchmarking.orgusec, Fewer Is BetterSockperf 3.4Test: Latency Ping PongFGKASLR EnabledJust KASLRNo KASLR/FGKASLR1.02262.04523.06784.09045.113SE +/- 0.075, N = 25SE +/- 0.057, N = 25SE +/- 0.054, N = 54.5454.5234.4761. (CXX) g++ options: --param -O3 -march=native -rdynamic -ldl -lpthread

Sockperf

Test: Throughput

OpenBenchmarking.orgMessages Per Second, More Is BetterSockperf 3.4Test: ThroughputFGKASLR EnabledJust KASLRNo KASLR/FGKASLR110K220K330K440K550KSE +/- 1497.16, N = 5SE +/- 4529.27, N = 25SE +/- 4418.39, N = 54997225089845065061. (CXX) g++ options: --param -O3 -march=native -rdynamic -ldl -lpthread

Ethr

Server Address: localhost - Protocol: UDP - Test: Bandwidth - Threads: 64

OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: UDP - Test: Bandwidth - Threads: 64FGKASLR EnabledJust KASLRNo KASLR/FGKASLR70K140K210K280K350KSE +/- 964.27, N = 3SE +/- 3845.85, N = 5SE +/- 3488.31, N = 3301327.33314693.73316768.67MIN: 297190 / MAX: 305370MIN: 300710 / MAX: 332790MIN: 307340 / MAX: 332660

Stress-NG

Test: Semaphores

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: SemaphoresFGKASLR EnabledJust KASLRNo KASLR/FGKASLR1000K2000K3000K4000K5000KSE +/- 7103.35, N = 3SE +/- 10281.87, N = 3SE +/- 25941.43, N = 34379605.164457335.284456094.951. (CC) gcc options: -O3 -march=native -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

OSBench

Test: Create Processes

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create ProcessesFGKASLR EnabledJust KASLRNo KASLR/FGKASLR816243240SE +/- 0.26, N = 13SE +/- 0.06, N = 3SE +/- 0.16, N = 332.7732.9231.681. (CC) gcc options: -O3 -march=native -lm

OSBench

Test: Create Files

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create FilesFGKASLR EnabledJust KASLRNo KASLR/FGKASLR48121620SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 315.6614.8714.651. (CC) gcc options: -O3 -march=native -lm

OSBench

Test: Memory Allocations

OpenBenchmarking.orgNs Per Event, Fewer Is BetterOSBenchTest: Memory AllocationsFGKASLR EnabledJust KASLRNo KASLR/FGKASLR20406080100SE +/- 0.12, N = 3SE +/- 0.39, N = 3SE +/- 0.06, N = 395.5994.6694.291. (CC) gcc options: -O3 -march=native -lm

OSBench

Test: Launch Programs

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Launch ProgramsFGKASLR EnabledJust KASLRNo KASLR/FGKASLR1122334455SE +/- 0.61, N = 3SE +/- 0.45, N = 3SE +/- 0.62, N = 347.9948.5846.851. (CC) gcc options: -O3 -march=native -lm

OSBench

Test: Create Threads

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create ThreadsFGKASLR EnabledJust KASLRNo KASLR/FGKASLR48121620SE +/- 0.11, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 318.0117.8517.761. (CC) gcc options: -O3 -march=native -lm

Phoronix Test Suite v10.8.5