LLVM LVI Mitigation Test Cascadelake-R 2 x Intel Xeon Gold 5220R "Cascade Lake Refresh" performance when looking at LVI mitigation with LLVM Clang 11 Git master as of 3 April. Benchmarks by Michael Larabel for a future article.
HTML result view exported from: https://openbenchmarking.org/result/2004056-NI-LLVMMITIG88&grr&rdt .
LLVM LVI Mitigation Test Cascadelake-R Processor Motherboard Chipset Memory Disk Graphics Network OS Kernel Desktop Display Server Display Driver Compiler File-System Screen Resolution Defaults LVI Mitigated 2 x Intel Xeon Gold 5220R @ 3.90GHz (36 Cores / 72 Threads) TYAN S7106 (V2.01.B40 BIOS) Intel Sky Lake-E DMI3 Registers 94GB 500GB Samsung SSD 860 ASPEED 2 x Intel I210 + 2 x QLogic cLOM8214 1/10GbE Ubuntu 20.04 5.6.2-050602-generic (x86_64) GNOME Shell 3.36.0 X Server 1.20.7 modesetting 1.20.7 Clang 11.0.0 ext4 1024x768 OpenBenchmarking.org Environment Details - Defaults: CXXFLAGS=-O3 CFLAGS=-O3 - LVI Mitigated: CXXFLAGS="-O3 -mlvi-cfi" CFLAGS="-O3 -mlvi-cfi" Compiler Details - Optimized build; Default target: x86_64-unknown-linux-gnu; Host CPU: cascadelake Processor Details - Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c Security 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 + tsx_async_abort: Mitigation of TSX disabled
LLVM LVI Mitigation Test Cascadelake-R cpp-perf-bench: Rand Numbers pgbench: Buffer Test - Normal Load - Read Write cryptopp: Integer + Elliptic Curve Public Key Algorithms dav1d: Chimera 1080p 10-bit tachyon: Total Time graphics-magick: Rotate pgbench: Buffer Test - Normal Load - Read Only mrbayes: Primate Phylogeny Analysis vpxenc: Speed 0 aom-av1: Speed 4 Realtime sqlite-speedtest: Timed Time - Size 1,000 cpp-perf-bench: Stepanov Vector cpp-perf-bench: Atol cryptopp: Unkeyed Algorithms graphics-magick: Noise-Gaussian graphics-magick: Swirl graphics-magick: HWB Color Space john-the-ripper: MD5 aom-av1: Speed 5 Two-Pass vpxenc: Speed 5 tungsten: Non-Exponential himeno: Poisson Pressure Solver aom-av1: Speed 8 Realtime dav1d: Chimera 1080p c-ray: Total Time - 4K, 16 Rays Per Pixel cpp-perf-bench: Stepanov Abstraction dav1d: Summer Nature 1080p cpp-perf-bench: Ctype john-the-ripper: Blowfish tungsten: Water Caustic compress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 scimark2: Composite cpp-perf-bench: Function Objects dav1d: Summer Nature 4K smhasher: MeowHash encode-flac: WAV To FLAC x265: H.265 1080p Video Encoding smhasher: Spooky32 smhasher: Spooky32 smhasher: fasthash32 smhasher: fasthash32 smhasher: t1ha2_atonce smhasher: t1ha2_atonce smhasher: t1ha0_aes_avx2 smhasher: t1ha0_aes_avx2 x264: H.264 Video Encoding smhasher: wyhash encode-mp3: WAV To MP3 hmmer: Pfam Database Search tungsten: Hair tscp: AI Chess Performance toybrot: C++ Threads toybrot: C++ Tasks scimark2: Monte Carlo Defaults LVI Mitigated 1416.241 3861.621572 4011.065557 62.83 35.8320 599 611319.004416 133.121 5.97 0.63 69.964 82.710 68.458 287.794090 23 47 182 166626 1.00 15.83 7.78929 3413.025283 17.46 307.41 30.251 33.453 322.11 30.646 1589 23.1136 21.421 2108.90 15.586 177.57 40.484 12.118 35.39 31.588 21600.26 25.359 9541.48 24.624 14180.11 25.133 71100.99 114.11 18.810 12.010 10.443 9.70799 1333602 1895 1889 644.25 1495.886 3739.862598 2899.348618 61.70 35.9463 591 454563.067842 144.848 4.59 0.52 100.949 86.422 69.080 247.524504 17 29 99 163426 0.83 12.90 13.41184 3392.372641 13.86 283.87 37.565 34.108 292.60 31.431 1588 23.6695 23.178 2038.44 28.205 161.22 67.900 12.334 27.04 58.590 21591.41 59.232 9533.29 45.948 14136.05 45.787 70559.67 94.01 42.148 13.258 13.943 10.4535 1277910 2028 2025 217.75 OpenBenchmarking.org
CppPerformanceBenchmarks Test: Random Numbers OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Random Numbers Defaults LVI Mitigated 300 600 900 1200 1500 SE +/- 0.10, N = 3 SE +/- 0.08, N = 3 1416.24 1495.89 -mlvi-cfi 1. (CXX) g++ options: -O3 -std=c++11
PostgreSQL pgbench Scaling: Buffer Test - Test: Normal Load - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 12.0 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write Defaults LVI Mitigated 800 1600 2400 3200 4000 SE +/- 51.74, N = 3 SE +/- 79.30, N = 12 3861.62 3739.86 -mlvi-cfi 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
Crypto++ Test: Integer + Elliptic Curve Public Key Algorithms OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Integer + Elliptic Curve Public Key Algorithms Defaults LVI Mitigated 900 1800 2700 3600 4500 SE +/- 0.75, N = 3 SE +/- 1.90, N = 3 4011.07 2899.35 -mlvi-cfi 1. (CXX) g++ options: -O3 -fPIC -pthread -pipe
dav1d Video Input: Chimera 1080p 10-bit OpenBenchmarking.org FPS, More Is Better dav1d 0.6.0 Video Input: Chimera 1080p 10-bit Defaults LVI Mitigated 14 28 42 56 70 SE +/- 0.39, N = 3 SE +/- 0.27, N = 3 62.83 61.70 MIN: 48.65 / MAX: 93.25 -mlvi-cfi - MIN: 47.18 / MAX: 99.37 1. (CC) gcc options: -O3 -pthread
Tachyon Total Time OpenBenchmarking.org Seconds, Fewer Is Better Tachyon 0.99b6 Total Time Defaults LVI Mitigated 8 16 24 32 40 SE +/- 0.37, N = 15 SE +/- 0.37, N = 15 35.83 35.95 1. (CC) gcc options: -m64 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread
GraphicsMagick Operation: Rotate OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Rotate Defaults LVI Mitigated 130 260 390 520 650 SE +/- 3.53, N = 3 SE +/- 4.68, N = 15 599 591 -mlvi-cfi 1. (CC) gcc options: -O3 -pthread -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lz -lm -lpthread
PostgreSQL pgbench Scaling: Buffer Test - Test: Normal Load - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 12.0 Scaling: Buffer Test - Test: Normal Load - Mode: Read Only Defaults LVI Mitigated 130K 260K 390K 520K 650K SE +/- 4734.70, N = 3 SE +/- 722.27, N = 3 611319.00 454563.07 -mlvi-cfi 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
Timed MrBayes Analysis Primate Phylogeny Analysis OpenBenchmarking.org Seconds, Fewer Is Better Timed MrBayes Analysis 3.2.7 Primate Phylogeny Analysis Defaults LVI Mitigated 30 60 90 120 150 SE +/- 0.53, N = 3 SE +/- 0.02, N = 3 133.12 144.85 -mlvi-cfi 1. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -maes -mavx -mfma -mavx2 -mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mrdrnd -mbmi -mbmi2 -madx -mmpx -O3 -std=c99 -pedantic -lm
VP9 libvpx Encoding Speed: Speed 0 OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.8.2 Speed: Speed 0 Defaults LVI Mitigated 1.3433 2.6866 4.0299 5.3732 6.7165 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 5.97 4.59 -mlvi-cfi 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=c++11
AOM AV1 Encoder Mode: Speed 4 Realtime OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2020-01-10 Encoder Mode: Speed 4 Realtime Defaults LVI Mitigated 0.1418 0.2836 0.4254 0.5672 0.709 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.63 0.52 -mlvi-cfi 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
SQLite Speedtest Timed Time - Size 1,000 OpenBenchmarking.org Seconds, Fewer Is Better SQLite Speedtest 3.30 Timed Time - Size 1,000 Defaults LVI Mitigated 20 40 60 80 100 SE +/- 0.14, N = 3 SE +/- 0.17, N = 3 69.96 100.95 -mlvi-cfi 1. (CC) gcc options: -O3 -ldl -lz -lpthread
CppPerformanceBenchmarks Test: Stepanov Vector OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Stepanov Vector Defaults LVI Mitigated 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 82.71 86.42 -mlvi-cfi 1. (CXX) g++ options: -O3 -std=c++11
CppPerformanceBenchmarks Test: Atol OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Atol Defaults LVI Mitigated 15 30 45 60 75 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 68.46 69.08 -mlvi-cfi 1. (CXX) g++ options: -O3 -std=c++11
Crypto++ Test: Unkeyed Algorithms OpenBenchmarking.org MiB/second, More Is Better Crypto++ 8.2 Test: Unkeyed Algorithms Defaults LVI Mitigated 60 120 180 240 300 SE +/- 0.14, N = 3 SE +/- 0.02, N = 3 287.79 247.52 -mlvi-cfi 1. (CXX) g++ options: -O3 -fPIC -pthread -pipe
GraphicsMagick Operation: Noise-Gaussian OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Noise-Gaussian Defaults LVI Mitigated 6 12 18 24 30 23 17 -mlvi-cfi 1. (CC) gcc options: -O3 -pthread -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lz -lm -lpthread
GraphicsMagick Operation: Swirl OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: Swirl Defaults LVI Mitigated 11 22 33 44 55 47 29 -mlvi-cfi 1. (CC) gcc options: -O3 -pthread -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lz -lm -lpthread
GraphicsMagick Operation: HWB Color Space OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.33 Operation: HWB Color Space Defaults LVI Mitigated 40 80 120 160 200 SE +/- 1.33, N = 3 SE +/- 0.33, N = 3 182 99 -mlvi-cfi 1. (CC) gcc options: -O3 -pthread -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lz -lm -lpthread
John The Ripper Test: MD5 OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.9.0-jumbo-1 Test: MD5 Defaults LVI Mitigated 40K 80K 120K 160K 200K SE +/- 107.36, N = 3 SE +/- 332.58, N = 3 166626 163426 -mlvi-cfi 1. (CC) gcc options: -m64 -lssl -lcrypto -pthread -lm -lz -ldl -lcrypt -mavx512f -O3 -march=native -std=gnu89 -Qunused-arguments -funroll-loops
AOM AV1 Encoder Mode: Speed 5 Two-Pass OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2020-01-10 Encoder Mode: Speed 5 Two-Pass Defaults LVI Mitigated 0.225 0.45 0.675 0.9 1.125 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.00 0.83 -mlvi-cfi 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
VP9 libvpx Encoding Speed: Speed 5 OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.8.2 Speed: Speed 5 Defaults LVI Mitigated 4 8 12 16 20 SE +/- 0.22, N = 4 SE +/- 0.11, N = 3 15.83 12.90 -mlvi-cfi 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=c++11
Tungsten Renderer Scene: Non-Exponential OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Non-Exponential Defaults LVI Mitigated 3 6 9 12 15 SE +/- 0.34842, N = 12 SE +/- 0.47787, N = 15 7.78929 13.41184 -mlvi-cfi 1. (CXX) g++ options: -O3 -std=c++0x -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512pf -mno-avx512er -mno-avx512ifma -mno-avx512vbmi -rdynamic -ljpeg -lpthread -ldl
Himeno Benchmark Poisson Pressure Solver OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver Defaults LVI Mitigated 700 1400 2100 2800 3500 SE +/- 0.21, N = 3 SE +/- 4.95, N = 3 3413.03 3392.37 -mlvi-cfi 1. (CC) gcc options: -O3 -mavx2
AOM AV1 Encoder Mode: Speed 8 Realtime OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2020-01-10 Encoder Mode: Speed 8 Realtime Defaults LVI Mitigated 4 8 12 16 20 SE +/- 0.15, N = 3 SE +/- 0.05, N = 3 17.46 13.86 -mlvi-cfi 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
dav1d Video Input: Chimera 1080p OpenBenchmarking.org FPS, More Is Better dav1d 0.6.0 Video Input: Chimera 1080p Defaults LVI Mitigated 70 140 210 280 350 SE +/- 3.72, N = 3 SE +/- 0.41, N = 3 307.41 283.87 MIN: 189.85 / MAX: 395.11 -mlvi-cfi - MIN: 206.72 / MAX: 353.03 1. (CC) gcc options: -O3 -pthread
C-Ray Total Time - 4K, 16 Rays Per Pixel OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel Defaults LVI Mitigated 9 18 27 36 45 SE +/- 0.09, N = 3 SE +/- 0.12, N = 3 30.25 37.57 -mlvi-cfi 1. (CC) gcc options: -lm -lpthread -O3
CppPerformanceBenchmarks Test: Stepanov Abstraction OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Stepanov Abstraction Defaults LVI Mitigated 8 16 24 32 40 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 33.45 34.11 -mlvi-cfi 1. (CXX) g++ options: -O3 -std=c++11
dav1d Video Input: Summer Nature 1080p OpenBenchmarking.org FPS, More Is Better dav1d 0.6.0 Video Input: Summer Nature 1080p Defaults LVI Mitigated 70 140 210 280 350 SE +/- 3.79, N = 15 SE +/- 2.76, N = 3 322.11 292.60 MIN: 86.27 / MAX: 371.88 -mlvi-cfi - MIN: 176.3 / MAX: 324.99 1. (CC) gcc options: -O3 -pthread
CppPerformanceBenchmarks Test: Ctype OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Ctype Defaults LVI Mitigated 7 14 21 28 35 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 30.65 31.43 -mlvi-cfi 1. (CXX) g++ options: -O3 -std=c++11
John The Ripper Test: Blowfish OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.9.0-jumbo-1 Test: Blowfish Defaults LVI Mitigated 300 600 900 1200 1500 1589 1588 -mlvi-cfi 1. (CC) gcc options: -m64 -lssl -lcrypto -pthread -lm -lz -ldl -lcrypt -mavx512f -O3 -march=native -std=gnu89 -Qunused-arguments -funroll-loops
Tungsten Renderer Scene: Water Caustic OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Water Caustic Defaults LVI Mitigated 6 12 18 24 30 SE +/- 0.13, N = 3 SE +/- 0.05, N = 3 23.11 23.67 -mlvi-cfi 1. (CXX) g++ options: -O3 -std=c++0x -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512pf -mno-avx512er -mno-avx512ifma -mno-avx512vbmi -rdynamic -ljpeg -lpthread -ldl
XZ Compression Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 OpenBenchmarking.org Seconds, Fewer Is Better XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 Defaults LVI Mitigated 6 12 18 24 30 SE +/- 0.29, N = 4 SE +/- 0.39, N = 3 21.42 23.18 -mlvi-cfi 1. (CC) gcc options: -pthread -fvisibility=hidden -O3
SciMark Computational Test: Composite OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Composite Defaults LVI Mitigated 500 1000 1500 2000 2500 SE +/- 0.71, N = 3 SE +/- 0.84, N = 3 2108.90 2038.44 -mlvi-cfi 1. (CC) gcc options: -O3 -lm
CppPerformanceBenchmarks Test: Function Objects OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Function Objects Defaults LVI Mitigated 7 14 21 28 35 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 15.59 28.21 -mlvi-cfi 1. (CXX) g++ options: -O3 -std=c++11
dav1d Video Input: Summer Nature 4K OpenBenchmarking.org FPS, More Is Better dav1d 0.6.0 Video Input: Summer Nature 4K Defaults LVI Mitigated 40 80 120 160 200 SE +/- 1.65, N = 3 SE +/- 0.83, N = 3 177.57 161.22 MIN: 112.26 / MAX: 191.68 -mlvi-cfi - MIN: 102.76 / MAX: 173.49 1. (CC) gcc options: -O3 -pthread
SMHasher Hash: MeowHash OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2020-02-29 Hash: MeowHash Defaults LVI Mitigated 15 30 45 60 75 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 40.48 67.90 -mlvi-cfi 1. (CXX) g++ options: -O3 -march=native -lpthread
FLAC Audio Encoding WAV To FLAC OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.3.2 WAV To FLAC Defaults LVI Mitigated 3 6 9 12 15 SE +/- 0.02, N = 5 SE +/- 0.03, N = 5 12.12 12.33 -mlvi-cfi 1. (CXX) g++ options: -O3 -lm
x265 H.265 1080p Video Encoding OpenBenchmarking.org Frames Per Second, More Is Better x265 3.1.2 H.265 1080p Video Encoding Defaults LVI Mitigated 8 16 24 32 40 SE +/- 0.17, N = 3 SE +/- 0.06, N = 3 35.39 27.04 -mlvi-cfi 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
SMHasher Hash: Spooky32 OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2020-02-29 Hash: Spooky32 Defaults LVI Mitigated 13 26 39 52 65 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 31.59 58.59 -mlvi-cfi 1. (CXX) g++ options: -O3 -march=native -lpthread
SMHasher Hash: Spooky32 OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2020-02-29 Hash: Spooky32 Defaults LVI Mitigated 5K 10K 15K 20K 25K SE +/- 5.67, N = 3 SE +/- 1.27, N = 3 21600.26 21591.41 -mlvi-cfi 1. (CXX) g++ options: -O3 -march=native -lpthread
SMHasher Hash: fasthash32 OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2020-02-29 Hash: fasthash32 Defaults LVI Mitigated 13 26 39 52 65 SE +/- 0.00, N = 3 SE +/- 1.34, N = 3 25.36 59.23 -mlvi-cfi 1. (CXX) g++ options: -O3 -march=native -lpthread
SMHasher Hash: fasthash32 OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2020-02-29 Hash: fasthash32 Defaults LVI Mitigated 2K 4K 6K 8K 10K SE +/- 0.05, N = 3 SE +/- 0.00, N = 3 9541.48 9533.29 -mlvi-cfi 1. (CXX) g++ options: -O3 -march=native -lpthread
SMHasher Hash: t1ha2_atonce OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2020-02-29 Hash: t1ha2_atonce Defaults LVI Mitigated 10 20 30 40 50 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 24.62 45.95 -mlvi-cfi 1. (CXX) g++ options: -O3 -march=native -lpthread
SMHasher Hash: t1ha2_atonce OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2020-02-29 Hash: t1ha2_atonce Defaults LVI Mitigated 3K 6K 9K 12K 15K SE +/- 0.22, N = 3 SE +/- 0.71, N = 3 14180.11 14136.05 -mlvi-cfi 1. (CXX) g++ options: -O3 -march=native -lpthread
SMHasher Hash: t1ha0_aes_avx2 OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2020-02-29 Hash: t1ha0_aes_avx2 Defaults LVI Mitigated 10 20 30 40 50 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 25.13 45.79 -mlvi-cfi 1. (CXX) g++ options: -O3 -march=native -lpthread
SMHasher Hash: t1ha0_aes_avx2 OpenBenchmarking.org MiB/sec, More Is Better SMHasher 2020-02-29 Hash: t1ha0_aes_avx2 Defaults LVI Mitigated 15K 30K 45K 60K 75K SE +/- 421.92, N = 3 SE +/- 540.12, N = 3 71100.99 70559.67 -mlvi-cfi 1. (CXX) g++ options: -O3 -march=native -lpthread
x264 H.264 Video Encoding OpenBenchmarking.org Frames Per Second, More Is Better x264 2019-12-17 H.264 Video Encoding Defaults LVI Mitigated 30 60 90 120 150 SE +/- 2.04, N = 12 SE +/- 0.37, N = 3 114.11 94.01 -mlvi-cfi 1. (CC) gcc options: -ldl -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -mstack-alignment=64 -fPIC -fomit-frame-pointer -fno-tree-vectorize
SMHasher Hash: wyhash OpenBenchmarking.org cycles/hash, Fewer Is Better SMHasher 2020-02-29 Hash: wyhash Defaults LVI Mitigated 10 20 30 40 50 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 18.81 42.15 -mlvi-cfi 1. (CXX) g++ options: -O3 -march=native -lpthread
LAME MP3 Encoding WAV To MP3 OpenBenchmarking.org Seconds, Fewer Is Better LAME MP3 Encoding 3.100 WAV To MP3 Defaults LVI Mitigated 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 12.01 13.26 -mlvi-cfi 1. (CC) gcc options: -O3 -pipe -lncurses -lm
Timed HMMer Search Pfam Database Search OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 2.3.2 Pfam Database Search Defaults LVI Mitigated 4 8 12 16 20 SE +/- 0.11, N = 3 SE +/- 0.03, N = 3 10.44 13.94 -mlvi-cfi 1. (CC) gcc options: -O3 -pthread -lhmmer -lsquid -lm
Tungsten Renderer Scene: Hair OpenBenchmarking.org Seconds, Fewer Is Better Tungsten Renderer 0.2.2 Scene: Hair Defaults LVI Mitigated 3 6 9 12 15 SE +/- 0.06856, N = 3 SE +/- 0.04000, N = 3 9.70799 10.45350 -mlvi-cfi 1. (CXX) g++ options: -O3 -std=c++0x -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512pf -mno-avx512er -mno-avx512ifma -mno-avx512vbmi -rdynamic -ljpeg -lpthread -ldl
TSCP AI Chess Performance OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance Defaults LVI Mitigated 300K 600K 900K 1200K 1500K SE +/- 937.62, N = 5 1333602 1277910 -mlvi-cfi 1. (CC) gcc options: -O3 -march=native
toyBrot Fractal Generator Implementation: C++ Threads OpenBenchmarking.org ms, Fewer Is Better toyBrot Fractal Generator Implementation: C++ Threads Defaults LVI Mitigated 400 800 1200 1600 2000 SE +/- 10.11, N = 3 SE +/- 23.26, N = 3 1895 2028 -mlvi-cfi 1. (CXX) g++ options: -O3 -lpthread
toyBrot Fractal Generator Implementation: C++ Tasks OpenBenchmarking.org ms, Fewer Is Better toyBrot Fractal Generator Implementation: C++ Tasks Defaults LVI Mitigated 400 800 1200 1600 2000 SE +/- 5.17, N = 3 SE +/- 20.03, N = 3 1889 2025 -mlvi-cfi 1. (CXX) g++ options: -O3 -lpthread
SciMark Computational Test: Monte Carlo OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Monte Carlo Defaults LVI Mitigated 140 280 420 560 700 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 644.25 217.75 -mlvi-cfi 1. (CC) gcc options: -O3 -lm
Phoronix Test Suite v10.8.5