MMIO Stale Data Mitigation Benchmarks

Tests for a future article. Intel Core i9-11900K testing with a ASUS ROG MAXIMUS XIII HERO (1402 BIOS) and ASUS Intel RKL GT1 31GB on Ubuntu 21.10 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2206160-PTS-MMIOSTAL02&rdt&grs.

MMIO Stale Data Mitigation BenchmarksProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen ResolutionClear buffers, SMT vulnerablemmio_stale_data=offIntel Core i9-11900K @ 5.10GHz (8 Cores / 16 Threads)ASUS ROG MAXIMUS XIII HERO (1402 BIOS)Intel Tiger Lake-H32GB2000GB Corsair Force MP600ASUS Intel RocketLake-S GT1 [UHD 750] (1300MHz)Intel Tiger Lake-H HD AudioMX2792 x Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411Ubuntu 21.105.19.0-rc2-mmio-stale-data-patched (x86_64)GNOME Shell 40.5X Server 1.20.134.6 Mesa 22.2.0-devel (git-fd27d51 2022-06-15 impish-oibaf-ppa)1.3.217GCC 11.2.0ext41920x1080ASUS Intel RKL GT1 31GB (1300MHz)OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-gcn/usr --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-build-config=bootstrap-lto-lean --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- NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Details- Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x53 - Thermald 2.4.6Java Details- OpenJDK Runtime Environment (build 11.0.15+10-Ubuntu-0ubuntu0.21.10.1)Python Details- Python 3.9.7Security Details- Clear buffers, SMT vulnerable: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + spec_store_bypass: Mitigation of SSB disabled via prctl + 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: Not affected - mmio_stale_data=off: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Vulnerable + spec_store_bypass: Mitigation of SSB disabled via prctl + 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: Not affected

MMIO Stale Data Mitigation Benchmarksredis: GETstress-ng: Context Switchingior: 4MB - Default Test Directorystress-ng: Atomicapache: 20nettle: sha512onnx: bertsquad-12 - CPU - Standardcompilebench: Compilenettle: chachapostmark: Disk Transaction Performancestress-ng: CPU Cachetensorflow-lite: Inception ResNet V2stress-ng: IO_uringonnx: super-resolution-10 - CPU - Standardnginx: 200memtier-benchmark: Redisnginx: 100sockperf: Throughputdacapobench: Tradebeansrenaissance: Finagle HTTP Requestsapache: 1apache: 100sockperf: Latency Ping Pongcompilebench: Initial Createnettle: aes256stress-ng: CPU Stresssvt-av1: Preset 12 - Bosphorus 1080pnginx: 20redis: SETsqlite-speedtest: Timed Time - Size 1,000tensorflow-lite: Inception V4rawtherapee: Total Benchmark Timerocksdb: Read Rand Write Randpgbench: 100 - 50 - Read Only - Average Latencysvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080pgimp: auto-levelsstress-ng: Mallocrenaissance: In-Memory Database Shootoutapache: 200darktable: Boat - CPU-onlycompilebench: Read Compiled Treesvt-vp9: PSNR/SSIM Optimized - Bosphorus 4Koctave-benchmark: gimp: resizesvt-vp9: VMAF Optimized - Bosphorus 4Krocksdb: Read While Writingbuild-llvm: Unix Makefilessvt-av1: Preset 10 - Bosphorus 4Kdacapobench: Tradesoaptensorflow-lite: Mobilenet Floatgimp: rotatesvt-vp9: VMAF Optimized - Bosphorus 1080psvt-av1: Preset 12 - Bosphorus 4Kgimp: unsharp-maskstress-ng: Cryptostress-ng: System V Message Passingstress-ng: Forkingnpb: SP.Cstress-ng: Socket Activitysvt-hevc: 7 - Bosphorus 1080pstress-ng: SENDFILEsvt-hevc: 10 - Bosphorus 4Krocksdb: Update Randsvt-hevc: 7 - Bosphorus 4Kdarktable: Server Room - CPU-onlybuild-linux-kernel: defconfignpb: IS.Dstress-ng: Matrix Mathtensorflow-lite: SqueezeNetnginx: 1pgbench: 100 - 50 - Read Writerocksdb: Rand Readstress-ng: MMAPpgbench: 100 - 50 - Read Write - Average Latencysvt-av1: Preset 10 - Bosphorus 1080pstress-ng: NUMAstress-ng: Semaphoresblender: Fishy Cat - CPU-Onlystress-ng: Glibc Qsort Data Sortingnpb: BT.Ctensorflow-lite: NASNet Mobilesvt-hevc: 10 - Bosphorus 1080pstress-ng: MEMFDblender: Pabellon Barcelona - CPU-Onlystress-ng: Memory Copyingdacapobench: Jythonstress-ng: x86_64 RdRandonnx: GPT-2 - CPU - Standardpgbench: 100 - 50 - Read Onlyblender: BMW27 - CPU-Onlynettle: poly1305-aesstress-ng: Glibc C String Functionsstress-ng: Vector Mathonnx: fcn-resnet101-11 - CPU - Standarddarktable: Server Rack - CPU-onlydarktable: Masskrug - CPU-onlyonnx: ArcFace ResNet-100 - CPU - Standardonnx: yolov4 - CPU - Standardstress-ng: Futextensorflow-lite: Mobilenet Quantdacapobench: H2npb: EP.Csockperf: Latency Under LoadClear buffers, SMT vulnerablemmio_stale_data=off4502788.712430837.521774.60339665.9892526.25721.609952573.571466.738524209.4726868.125726.277126497111.333068855.25501994.9483949320352734.820094.59106471.072.350771.5512018.9923175.10425.403501759.23212061.3339.61626947.252.13920940590.135201.318.7907727010.122324.289131.244.0694673.4662.555.1815.76157.182242464615.36280.70129911432.148.387193.67109.18610.61313464.6511271215.8270423.806620.499216.66140.77243060.7385.4466920845.983.34378.918999.6057560.791813.30111694.603145053591223217.771.590231.46624.611718628.66172.36185.3123989.467077.56282.53741.66451.461613.78299477956.306963372478130.684466.631716938.5758932.061030.2174.44420715322542009.312756.9025881663.9614.3654042373.3113841930.871604.93352590.3795187.42739.8810202635.021501.738723214.3227460.125255.716997506101.813017015.03510403.1082604520682699.319834.64107861.662.380762.2212158.1522936.55429.784506778.753183488.0839.97126719.852.57620781800.134199.838.7267783046.992308.089743.414.0954702.8062.215.2095.79256.932233150617.88981.03030021437.318.417192.98108.80210.65013419.9311307266.3070631.436602.499193.96140.47242573.0285.6167052145.893.34979.0571001.3157656.071816.11111539.673149153660896217.491.588231.75324.641716750.63172.20185.1424007.457082.13282.71742.02451.671614.44299577931.966961372390130.654465.801717173.4958931.541030.2174.44419635092532900.282716.2125711614.2619.160OpenBenchmarking.org

Redis

Test: GET

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: GETClear buffers, SMT vulnerablemmio_stale_data=off1000K2000K3000K4000K5000KSE +/- 24337.32, N = 3SE +/- 44687.51, N = 44502788.704042373.311. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

Stress-NG

Test: Context Switching

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Context SwitchingClear buffers, SMT vulnerablemmio_stale_data=off3M6M9M12M15MSE +/- 160660.40, N = 3SE +/- 204019.40, N = 1512430837.5213841930.871. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

IOR

Block Size: 4MB - Disk Target: Default Test Directory

OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 4MB - Disk Target: Default Test DirectoryClear buffers, SMT vulnerablemmio_stale_data=off400800120016002000SE +/- 12.73, N = 15SE +/- 24.03, N = 151774.601604.93MIN: 392.17 / MAX: 2478.06MIN: 820.91 / MAX: 2498.451. (CC) gcc options: -O2 -lm -lmpi

Stress-NG

Test: Atomic

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: AtomicClear buffers, SMT vulnerablemmio_stale_data=off80K160K240K320K400KSE +/- 4514.52, N = 3SE +/- 3752.59, N = 3339665.98352590.371. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Apache HTTP Server

Concurrent Requests: 20

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 20Clear buffers, SMT vulnerablemmio_stale_data=off20K40K60K80K100KSE +/- 339.98, N = 3SE +/- 274.39, N = 392526.2595187.421. (CC) gcc options: -shared -fPIC -O2

Nettle

Test: sha512

OpenBenchmarking.orgMbyte/s, More Is BetterNettle 3.8Test: sha512Clear buffers, SMT vulnerablemmio_stale_data=off160320480640800SE +/- 3.21, N = 3SE +/- 7.38, N = 3721.60739.881. (CC) gcc options: -O2 -ggdb3 -lnettle -lm -lcrypto

ONNX Runtime

Model: bertsquad-12 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: bertsquad-12 - Device: CPU - Executor: StandardClear buffers, SMT vulnerablemmio_stale_data=off2004006008001000SE +/- 12.86, N = 12SE +/- 4.51, N = 399510201. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

Compile Bench

Test: Compile

OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: CompileClear buffers, SMT vulnerablemmio_stale_data=off6001200180024003000SE +/- 31.35, N = 15SE +/- 25.45, N = 62573.572635.02

Nettle

Test: chacha

OpenBenchmarking.orgMbyte/s, More Is BetterNettle 3.8Test: chachaClear buffers, SMT vulnerablemmio_stale_data=off30060090012001500SE +/- 5.10, N = 3SE +/- 5.54, N = 31466.731501.73MIN: 654.37 / MAX: 4535.7MIN: 654.5 / MAX: 4611.781. (CC) gcc options: -O2 -ggdb3 -lnettle -lm -lcrypto

PostMark

Disk Transaction Performance

OpenBenchmarking.orgTPS, More Is BetterPostMark 1.51Disk Transaction PerformanceClear buffers, SMT vulnerablemmio_stale_data=off2K4K6K8K10KSE +/- 95.67, N = 3SE +/- 102.67, N = 3852487231. (CC) gcc options: -O3

Stress-NG

Test: CPU Cache

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: CPU CacheClear buffers, SMT vulnerablemmio_stale_data=off50100150200250SE +/- 2.51, N = 15SE +/- 3.07, N = 3209.47214.321. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

TensorFlow Lite

Model: Inception ResNet V2

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Inception ResNet V2Clear buffers, SMT vulnerablemmio_stale_data=off6K12K18K24K30KSE +/- 14.80, N = 3SE +/- 86.07, N = 326868.127460.1

Stress-NG

Test: IO_uring

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: IO_uringClear buffers, SMT vulnerablemmio_stale_data=off6K12K18K24K30KSE +/- 64.12, N = 3SE +/- 255.70, N = 625726.2725255.711. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

ONNX Runtime

Model: super-resolution-10 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: super-resolution-10 - Device: CPU - Executor: StandardClear buffers, SMT vulnerablemmio_stale_data=off15003000450060007500SE +/- 77.89, N = 4SE +/- 75.32, N = 12712669971. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

nginx

Concurrent Requests: 200

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 200Clear buffers, SMT vulnerablemmio_stale_data=off110K220K330K440K550KSE +/- 446.08, N = 3SE +/- 251.45, N = 3497111.33506101.811. (CC) gcc options: -lcrypt -lz -O3 -march=native

Memtier_benchmark

Protocol: Redis

OpenBenchmarking.orgOps/sec, More Is BetterMemtier_benchmark 1.3Protocol: RedisClear buffers, SMT vulnerablemmio_stale_data=off700K1400K2100K2800K3500KSE +/- 27004.28, N = 8SE +/- 18522.57, N = 33068855.253017015.031. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

nginx

Concurrent Requests: 100

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 100Clear buffers, SMT vulnerablemmio_stale_data=off110K220K330K440K550KSE +/- 1049.35, N = 3SE +/- 224.71, N = 3501994.94510403.101. (CC) gcc options: -lcrypt -lz -O3 -march=native

Sockperf

Test: Throughput

OpenBenchmarking.orgMessages Per Second, More Is BetterSockperf 3.7Test: ThroughputClear buffers, SMT vulnerablemmio_stale_data=off200K400K600K800K1000KSE +/- 7370.73, N = 5SE +/- 6450.15, N = 258394938260451. (CXX) g++ options: --param -O3 -rdynamic

DaCapo Benchmark

Java Test: Tradebeans

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradebeansClear buffers, SMT vulnerablemmio_stale_data=off400800120016002000SE +/- 22.05, N = 20SE +/- 21.71, N = 2020352068

Renaissance

Test: Finagle HTTP Requests

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Finagle HTTP RequestsClear buffers, SMT vulnerablemmio_stale_data=off6001200180024003000SE +/- 8.40, N = 3SE +/- 28.48, N = 32734.82699.3MIN: 2523.04 / MAX: 2994.67MIN: 2469.47 / MAX: 3073.31

Apache HTTP Server

Concurrent Requests: 1

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 1Clear buffers, SMT vulnerablemmio_stale_data=off4K8K12K16K20KSE +/- 20.97, N = 3SE +/- 94.85, N = 320094.5919834.641. (CC) gcc options: -shared -fPIC -O2

Apache HTTP Server

Concurrent Requests: 100

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 100Clear buffers, SMT vulnerablemmio_stale_data=off20K40K60K80K100KSE +/- 105.45, N = 3SE +/- 294.96, N = 3106471.07107861.661. (CC) gcc options: -shared -fPIC -O2

Sockperf

Test: Latency Ping Pong

OpenBenchmarking.orgusec, Fewer Is BetterSockperf 3.7Test: Latency Ping PongClear buffers, SMT vulnerablemmio_stale_data=off0.53551.0711.60652.1422.6775SE +/- 0.015, N = 25SE +/- 0.025, N = 52.3502.3801. (CXX) g++ options: --param -O3 -rdynamic

Compile Bench

Test: Initial Create

OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Initial CreateClear buffers, SMT vulnerablemmio_stale_data=off170340510680850SE +/- 4.90, N = 3SE +/- 11.46, N = 3771.55762.22

Nettle

Test: aes256

OpenBenchmarking.orgMbyte/s, More Is BetterNettle 3.8Test: aes256Clear buffers, SMT vulnerablemmio_stale_data=off3K6K9K12K15KSE +/- 128.94, N = 3SE +/- 13.25, N = 312018.9912158.15MIN: 7702.46 / MAX: 20580.16MIN: 7916.5 / MAX: 20586.61. (CC) gcc options: -O2 -ggdb3 -lnettle -lm -lcrypto

Stress-NG

Test: CPU Stress

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: CPU StressClear buffers, SMT vulnerablemmio_stale_data=off5K10K15K20K25KSE +/- 102.57, N = 3SE +/- 36.93, N = 323175.1022936.551. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

SVT-AV1

Encoder Mode: Preset 12 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.0Encoder Mode: Preset 12 - Input: Bosphorus 1080pClear buffers, SMT vulnerablemmio_stale_data=off90180270360450SE +/- 0.28, N = 3SE +/- 0.85, N = 3425.40429.781. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

nginx

Concurrent Requests: 20

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 20Clear buffers, SMT vulnerablemmio_stale_data=off110K220K330K440K550KSE +/- 579.06, N = 3SE +/- 662.40, N = 3501759.20506778.751. (CC) gcc options: -lcrypt -lz -O3 -march=native

Redis

Test: SET

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: SETClear buffers, SMT vulnerablemmio_stale_data=off700K1400K2100K2800K3500KSE +/- 7906.99, N = 3SE +/- 10443.72, N = 33212061.333183488.081. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

SQLite Speedtest

Timed Time - Size 1,000

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000Clear buffers, SMT vulnerablemmio_stale_data=off918273645SE +/- 0.03, N = 3SE +/- 0.14, N = 339.6239.971. (CC) gcc options: -O2 -lz

TensorFlow Lite

Model: Inception V4

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Inception V4Clear buffers, SMT vulnerablemmio_stale_data=off6K12K18K24K30KSE +/- 148.28, N = 3SE +/- 43.81, N = 326947.226719.8

RawTherapee

Total Benchmark Time

OpenBenchmarking.orgSeconds, Fewer Is BetterRawTherapeeTotal Benchmark TimeClear buffers, SMT vulnerablemmio_stale_data=off1224364860SE +/- 0.02, N = 3SE +/- 0.03, N = 352.1452.581. RawTherapee, version 5.8, command line.

Facebook RocksDB

Test: Read Random Write Random

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.0.1Test: Read Random Write RandomClear buffers, SMT vulnerablemmio_stale_data=off400K800K1200K1600K2000KSE +/- 4401.84, N = 3SE +/- 13179.82, N = 3209405920781801. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 50 - Mode: Read Only - Average Latency

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 14.0Scaling Factor: 100 - Clients: 50 - Mode: Read Only - Average LatencyClear buffers, SMT vulnerablemmio_stale_data=off0.03040.06080.09120.12160.152SE +/- 0.001, N = 3SE +/- 0.001, N = 30.1350.1341. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

SVT-VP9

Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pClear buffers, SMT vulnerablemmio_stale_data=off4080120160200SE +/- 0.16, N = 3SE +/- 0.19, N = 3201.31199.831. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

GIMP

Test: auto-levels

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.24Test: auto-levelsClear buffers, SMT vulnerablemmio_stale_data=off246810SE +/- 0.039, N = 3SE +/- 0.017, N = 38.7908.726

Stress-NG

Test: Malloc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: MallocClear buffers, SMT vulnerablemmio_stale_data=off2M4M6M8M10MSE +/- 53293.72, N = 3SE +/- 27882.04, N = 37727010.127783046.991. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Renaissance

Test: In-Memory Database Shootout

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: In-Memory Database ShootoutClear buffers, SMT vulnerablemmio_stale_data=off5001000150020002500SE +/- 12.64, N = 3SE +/- 7.82, N = 32324.22308.0MIN: 2160.37 / MAX: 2574.03MIN: 2135.4 / MAX: 2564.1

Apache HTTP Server

Concurrent Requests: 200

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 200Clear buffers, SMT vulnerablemmio_stale_data=off20K40K60K80K100KSE +/- 325.91, N = 3SE +/- 196.56, N = 389131.2489743.411. (CC) gcc options: -shared -fPIC -O2

Darktable

Test: Boat - Acceleration: CPU-only

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.6.0Test: Boat - Acceleration: CPU-onlyClear buffers, SMT vulnerablemmio_stale_data=off0.92141.84282.76423.68564.607SE +/- 0.003, N = 3SE +/- 0.005, N = 34.0694.095

Compile Bench

Test: Read Compiled Tree

OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Read Compiled TreeClear buffers, SMT vulnerablemmio_stale_data=off10002000300040005000SE +/- 0.00, N = 3SE +/- 29.34, N = 34673.464702.80

SVT-VP9

Tuning: PSNR/SSIM Optimized - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 4KClear buffers, SMT vulnerablemmio_stale_data=off1428425670SE +/- 0.02, N = 3SE +/- 0.07, N = 362.5562.211. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

GNU Octave Benchmark

OpenBenchmarking.orgSeconds, Fewer Is BetterGNU Octave Benchmark 6.2.0Clear buffers, SMT vulnerablemmio_stale_data=off1.1722.3443.5164.6885.86SE +/- 0.042, N = 9SE +/- 0.024, N = 55.1815.209

GIMP

Test: resize

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.24Test: resizeClear buffers, SMT vulnerablemmio_stale_data=off1.30322.60643.90965.21286.516SE +/- 0.036, N = 13SE +/- 0.060, N = 55.7615.792

SVT-VP9

Tuning: VMAF Optimized - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 4KClear buffers, SMT vulnerablemmio_stale_data=off1326395265SE +/- 0.42, N = 11SE +/- 0.65, N = 357.1856.931. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

Facebook RocksDB

Test: Read While Writing

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.0.1Test: Read While WritingClear buffers, SMT vulnerablemmio_stale_data=off500K1000K1500K2000K2500KSE +/- 9766.85, N = 3SE +/- 17624.42, N = 3224246422331501. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Timed LLVM Compilation

Build System: Unix Makefiles

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 13.0Build System: Unix MakefilesClear buffers, SMT vulnerablemmio_stale_data=off130260390520650SE +/- 1.36, N = 3SE +/- 0.84, N = 3615.36617.89

SVT-AV1

Encoder Mode: Preset 10 - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.0Encoder Mode: Preset 10 - Input: Bosphorus 4KClear buffers, SMT vulnerablemmio_stale_data=off20406080100SE +/- 0.74, N = 12SE +/- 0.60, N = 1280.7081.031. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

DaCapo Benchmark

Java Test: Tradesoap

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradesoapClear buffers, SMT vulnerablemmio_stale_data=off6001200180024003000SE +/- 17.52, N = 3SE +/- 22.55, N = 2029913002

TensorFlow Lite

Model: Mobilenet Float

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Mobilenet FloatClear buffers, SMT vulnerablemmio_stale_data=off30060090012001500SE +/- 3.29, N = 3SE +/- 3.71, N = 31432.141437.31

GIMP

Test: rotate

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.24Test: rotateClear buffers, SMT vulnerablemmio_stale_data=off246810SE +/- 0.010, N = 3SE +/- 0.019, N = 38.3878.417

SVT-VP9

Tuning: VMAF Optimized - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 1080pClear buffers, SMT vulnerablemmio_stale_data=off4080120160200SE +/- 2.14, N = 5SE +/- 2.00, N = 5193.67192.981. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

SVT-AV1

Encoder Mode: Preset 12 - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.0Encoder Mode: Preset 12 - Input: Bosphorus 4KClear buffers, SMT vulnerablemmio_stale_data=off20406080100SE +/- 0.08, N = 3SE +/- 0.40, N = 3109.19108.801. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

GIMP

Test: unsharp-mask

OpenBenchmarking.orgSeconds, Fewer Is BetterGIMP 2.10.24Test: unsharp-maskClear buffers, SMT vulnerablemmio_stale_data=off3691215SE +/- 0.02, N = 3SE +/- 0.01, N = 310.6110.65

Stress-NG

Test: Crypto

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: CryptoClear buffers, SMT vulnerablemmio_stale_data=off3K6K9K12K15KSE +/- 47.88, N = 3SE +/- 29.69, N = 313464.6513419.931. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Stress-NG

Test: System V Message Passing

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: System V Message PassingClear buffers, SMT vulnerablemmio_stale_data=off2M4M6M8M10MSE +/- 41159.70, N = 3SE +/- 3240.59, N = 311271215.8211307266.301. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Stress-NG

Test: Forking

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: ForkingClear buffers, SMT vulnerablemmio_stale_data=off15K30K45K60K75KSE +/- 348.18, N = 3SE +/- 172.90, N = 370423.8070631.431. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

NAS Parallel Benchmarks

Test / Class: SP.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CClear buffers, SMT vulnerablemmio_stale_data=off14002800420056007000SE +/- 13.70, N = 3SE +/- 2.93, N = 36620.496602.491. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.0

Stress-NG

Test: Socket Activity

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Socket ActivityClear buffers, SMT vulnerablemmio_stale_data=off2K4K6K8K10KSE +/- 5.15, N = 3SE +/- 5.40, N = 39216.669193.961. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

SVT-HEVC

Tuning: 7 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080pClear buffers, SMT vulnerablemmio_stale_data=off306090120150SE +/- 1.26, N = 3SE +/- 1.49, N = 3140.77140.471. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

Stress-NG

Test: SENDFILE

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: SENDFILEClear buffers, SMT vulnerablemmio_stale_data=off50K100K150K200K250KSE +/- 1107.84, N = 3SE +/- 888.45, N = 3243060.73242573.021. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

SVT-HEVC

Tuning: 10 - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 4KClear buffers, SMT vulnerablemmio_stale_data=off20406080100SE +/- 0.05, N = 3SE +/- 0.13, N = 385.4485.611. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

Facebook RocksDB

Test: Update Random

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.0.1Test: Update RandomClear buffers, SMT vulnerablemmio_stale_data=off140K280K420K560K700KSE +/- 384.85, N = 3SE +/- 888.10, N = 36692086705211. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

SVT-HEVC

Tuning: 7 - Input: Bosphorus 4K

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 4KClear buffers, SMT vulnerablemmio_stale_data=off1020304050SE +/- 0.34, N = 3SE +/- 0.31, N = 345.9845.891. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

Darktable

Test: Server Room - Acceleration: CPU-only

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.6.0Test: Server Room - Acceleration: CPU-onlyClear buffers, SMT vulnerablemmio_stale_data=off0.75351.5072.26053.0143.7675SE +/- 0.004, N = 3SE +/- 0.003, N = 33.3433.349

Timed Linux Kernel Compilation

Build: defconfig

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.16Build: defconfigClear buffers, SMT vulnerablemmio_stale_data=off20406080100SE +/- 0.12, N = 3SE +/- 0.08, N = 378.9279.06

NAS Parallel Benchmarks

Test / Class: IS.D

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.DClear buffers, SMT vulnerablemmio_stale_data=off2004006008001000SE +/- 0.82, N = 3SE +/- 1.53, N = 3999.601001.311. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.0

Stress-NG

Test: Matrix Math

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Matrix MathClear buffers, SMT vulnerablemmio_stale_data=off12K24K36K48K60KSE +/- 53.26, N = 3SE +/- 16.25, N = 357560.7957656.071. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

TensorFlow Lite

Model: SqueezeNet

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: SqueezeNetClear buffers, SMT vulnerablemmio_stale_data=off400800120016002000SE +/- 7.65, N = 3SE +/- 7.43, N = 31813.301816.11

nginx

Concurrent Requests: 1

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 1Clear buffers, SMT vulnerablemmio_stale_data=off20K40K60K80K100KSE +/- 160.88, N = 3SE +/- 361.48, N = 3111694.60111539.671. (CC) gcc options: -lcrypt -lz -O3 -march=native

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 50 - Mode: Read Write

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 14.0Scaling Factor: 100 - Clients: 50 - Mode: Read WriteClear buffers, SMT vulnerablemmio_stale_data=off7K14K21K28K35KSE +/- 67.52, N = 3SE +/- 100.77, N = 331450314911. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Facebook RocksDB

Test: Random Read

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.0.1Test: Random ReadClear buffers, SMT vulnerablemmio_stale_data=off11M22M33M44M55MSE +/- 381547.95, N = 3SE +/- 449997.89, N = 353591223536608961. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Stress-NG

Test: MMAP

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: MMAPClear buffers, SMT vulnerablemmio_stale_data=off50100150200250SE +/- 0.48, N = 3SE +/- 0.85, N = 3217.77217.491. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 50 - Mode: Read Write - Average Latency

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 14.0Scaling Factor: 100 - Clients: 50 - Mode: Read Write - Average LatencyClear buffers, SMT vulnerablemmio_stale_data=off0.35780.71561.07341.43121.789SE +/- 0.003, N = 3SE +/- 0.005, N = 31.5901.5881. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

SVT-AV1

Encoder Mode: Preset 10 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.0Encoder Mode: Preset 10 - Input: Bosphorus 1080pClear buffers, SMT vulnerablemmio_stale_data=off50100150200250SE +/- 3.26, N = 3SE +/- 2.47, N = 5231.47231.751. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

Stress-NG

Test: NUMA

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: NUMAClear buffers, SMT vulnerablemmio_stale_data=off612182430SE +/- 0.14, N = 3SE +/- 0.17, N = 324.6124.641. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Stress-NG

Test: Semaphores

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: SemaphoresClear buffers, SMT vulnerablemmio_stale_data=off400K800K1200K1600K2000KSE +/- 5573.05, N = 3SE +/- 10137.25, N = 31718628.661716750.631. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Blender

Blend File: Fishy Cat - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.2Blend File: Fishy Cat - Compute: CPU-OnlyClear buffers, SMT vulnerablemmio_stale_data=off4080120160200SE +/- 0.06, N = 3SE +/- 0.06, N = 3172.36172.20

Stress-NG

Test: Glibc Qsort Data Sorting

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Glibc Qsort Data SortingClear buffers, SMT vulnerablemmio_stale_data=off4080120160200SE +/- 0.42, N = 3SE +/- 0.24, N = 3185.31185.141. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

NAS Parallel Benchmarks

Test / Class: BT.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.CClear buffers, SMT vulnerablemmio_stale_data=off5K10K15K20K25KSE +/- 38.49, N = 3SE +/- 35.59, N = 323989.4624007.451. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.0

TensorFlow Lite

Model: NASNet Mobile

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: NASNet MobileClear buffers, SMT vulnerablemmio_stale_data=off15003000450060007500SE +/- 21.37, N = 3SE +/- 14.99, N = 37077.567082.13

SVT-HEVC

Tuning: 10 - Input: Bosphorus 1080p

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pClear buffers, SMT vulnerablemmio_stale_data=off60120180240300SE +/- 0.42, N = 3SE +/- 0.47, N = 3282.53282.711. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

Stress-NG

Test: MEMFD

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: MEMFDClear buffers, SMT vulnerablemmio_stale_data=off160320480640800SE +/- 0.76, N = 3SE +/- 0.59, N = 3741.66742.021. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Blender

Blend File: Pabellon Barcelona - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.2Blend File: Pabellon Barcelona - Compute: CPU-OnlyClear buffers, SMT vulnerablemmio_stale_data=off100200300400500SE +/- 0.43, N = 3SE +/- 0.49, N = 3451.46451.67

Stress-NG

Test: Memory Copying

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Memory CopyingClear buffers, SMT vulnerablemmio_stale_data=off30060090012001500SE +/- 0.83, N = 3SE +/- 0.75, N = 31613.781614.441. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

DaCapo Benchmark

Java Test: Jython

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonClear buffers, SMT vulnerablemmio_stale_data=off6001200180024003000SE +/- 9.16, N = 4SE +/- 20.30, N = 429942995

Stress-NG

Test: x86_64 RdRand

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: x86_64 RdRandClear buffers, SMT vulnerablemmio_stale_data=off20K40K60K80K100KSE +/- 24.81, N = 3SE +/- 10.17, N = 377956.3077931.961. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

ONNX Runtime

Model: GPT-2 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPU - Executor: StandardClear buffers, SMT vulnerablemmio_stale_data=off15003000450060007500SE +/- 2.20, N = 3SE +/- 0.29, N = 3696369611. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 50 - Mode: Read Only

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 14.0Scaling Factor: 100 - Clients: 50 - Mode: Read OnlyClear buffers, SMT vulnerablemmio_stale_data=off80K160K240K320K400KSE +/- 1809.62, N = 3SE +/- 3420.46, N = 33724783723901. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Blender

Blend File: BMW27 - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.2Blend File: BMW27 - Compute: CPU-OnlyClear buffers, SMT vulnerablemmio_stale_data=off306090120150SE +/- 0.09, N = 3SE +/- 0.03, N = 3130.68130.65

Nettle

Test: poly1305-aes

OpenBenchmarking.orgMbyte/s, More Is BetterNettle 3.8Test: poly1305-aesClear buffers, SMT vulnerablemmio_stale_data=off10002000300040005000SE +/- 1.30, N = 3SE +/- 1.15, N = 34466.634465.801. (CC) gcc options: -O2 -ggdb3 -lnettle -lm -lcrypto

Stress-NG

Test: Glibc C String Functions

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Glibc C String FunctionsClear buffers, SMT vulnerablemmio_stale_data=off400K800K1200K1600K2000KSE +/- 10949.44, N = 3SE +/- 9914.06, N = 31716938.571717173.491. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Stress-NG

Test: Vector Math

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Vector MathClear buffers, SMT vulnerablemmio_stale_data=off13K26K39K52K65KSE +/- 11.95, N = 3SE +/- 17.22, N = 358932.0658931.541. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

ONNX Runtime

Model: fcn-resnet101-11 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: fcn-resnet101-11 - Device: CPU - Executor: StandardClear buffers, SMT vulnerablemmio_stale_data=off20406080100SE +/- 0.17, N = 3SE +/- 0.17, N = 31031031. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

Darktable

Test: Server Rack - Acceleration: CPU-only

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.6.0Test: Server Rack - Acceleration: CPU-onlyClear buffers, SMT vulnerablemmio_stale_data=off0.04880.09760.14640.19520.244SE +/- 0.001, N = 3SE +/- 0.000, N = 30.2170.217

Darktable

Test: Masskrug - Acceleration: CPU-only

OpenBenchmarking.orgSeconds, Fewer Is BetterDarktable 3.6.0Test: Masskrug - Acceleration: CPU-onlyClear buffers, SMT vulnerablemmio_stale_data=off0.99991.99982.99973.99964.9995SE +/- 0.009, N = 3SE +/- 0.006, N = 34.4444.444

ONNX Runtime

Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardClear buffers, SMT vulnerablemmio_stale_data=off400800120016002000SE +/- 3.03, N = 3SE +/- 44.46, N = 12207119631. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

ONNX Runtime

Model: yolov4 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: yolov4 - Device: CPU - Executor: StandardClear buffers, SMT vulnerablemmio_stale_data=off120240360480600SE +/- 1.30, N = 3SE +/- 10.67, N = 125325091. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

Stress-NG

Test: Futex

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: FutexClear buffers, SMT vulnerablemmio_stale_data=off500K1000K1500K2000K2500KSE +/- 26117.34, N = 5SE +/- 49539.09, N = 152542009.312532900.281. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

TensorFlow Lite

Model: Mobilenet Quant

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Mobilenet QuantClear buffers, SMT vulnerablemmio_stale_data=off6001200180024003000SE +/- 56.11, N = 12SE +/- 58.68, N = 152756.902716.21

DaCapo Benchmark

Java Test: H2

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2Clear buffers, SMT vulnerablemmio_stale_data=off6001200180024003000SE +/- 46.91, N = 20SE +/- 65.73, N = 2025882571

NAS Parallel Benchmarks

Test / Class: EP.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.CClear buffers, SMT vulnerablemmio_stale_data=off400800120016002000SE +/- 23.94, N = 3SE +/- 27.41, N = 151663.961614.261. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.0

Sockperf

Test: Latency Under Load

OpenBenchmarking.orgusec, Fewer Is BetterSockperf 3.7Test: Latency Under LoadClear buffers, SMT vulnerablemmio_stale_data=off510152025SE +/- 1.49, N = 25SE +/- 0.10, N = 514.3719.161. (CXX) g++ options: --param -O3 -rdynamic


Phoronix Test Suite v10.8.4