LLVM 3.4 Compiler Benchmarking Comparison

Intel Core i7 990XGulftown processor testing of LLVM 3.2, 3.3, and 3.4 SVN. Benchmarking by Michael Larabel for a future article on Phoronix.com.

HTML result view exported from: https://openbenchmarking.org/result/1307158-SO-LLVM34COM27.

LLVM 3.4 Compiler Benchmarking ComparisonProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 20130715Intel Core i7 990X @ 3.47GHz (12 Cores)MSI X58M (MS-7593) v1.0Intel 5520/5500/X58 + ICH10R12288MB1000GB Western Digital WD1003FBYX-0NVIDIA GeForce 8400 GS 512MB (500/333MHz)Realtek ALC889Realtek RTL8111/8168/8411Ubuntu 13.103.10.0-2-generic (x86_64)Xfce 4.10X Server 1.14.1NVIDIA 304.883.3.0 NVIDIA 304.88Clang 3.2 + LLVM 3.2svnext41920x1080Clang 3.3 + LLVM 3.3Clang 3.4 (SVN 186320) + LLVM 3.4svnOpenBenchmarking.orgCompiler Details- LLVM Clang 3.2: Optimized build; Built Jul 15 2013 (09:25:52); Default target: x86_64-unknown-linux-gnu; Host CPU: corei7- LLVM Clang 3.3: Optimized build; Built Jul 15 2013 (09:12:51); Default target: x86_64-unknown-linux-gnu; Host CPU: corei7- LLVM Clang 3.4 SVN 20130715: Optimized build; Built Jul 15 2013 (09:43:39); Default target: x86_64-unknown-linux-gnu; Host CPU: corei7 Processor Details- Scaling Governor: acpi-cpufreq ondemand

LLVM 3.4 Compiler Benchmarking Comparisonhmmer: Pfam Database Searchmafft: Multiple Sequence Alignmentblake2: Phoronix Test Suite v4.8.0m2scimark2: Monte Carloscimark2: Fast Fourier Transformscimark2: Sparse Matrix Multiplyscimark2: Dense LU Matrix Factorizationscimark2: Jacobi Successive Over-Relaxationx264: H.264 Video Encodinggraphics-magick: Blurgraphics-magick: Sharpengraphics-magick: Resizinggraphics-magick: HWB Color Spacegraphics-magick: Local Adaptive Thresholdinghimeno: Poisson Pressure Solverbuild-imagemagick: Time To Compilebuild-php: Time To Compilec-ray: Total Timeprimesieve: 1e12 Prime Number Generationsmallpt: Global Illumination Renderer; 100 Samplesencode-flac: WAV To FLACffmpeg: H.264 HD To NTSC DVn-queens: Elapsed Timepgbench: TPC-B Transactions Per Secondapache: Static Web Page ServingLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 2013071511.246.127.28477.75289.811694.321628.511385.71177.42664482100401393.6623.2818.9623.7623120.92274.2018271.8210.016.487.40478.82289.321701.361628.511385.71176.15674483101391413.5424.8520.5823.76443.312278.1421.05274.20276.4717678.9610.136.067.39477.23288.101701.373137.251392.00177.77674484101401438.1925.8920.7423.25432.842307.0821.33274.76275.1218491.40OpenBenchmarking.org

Timed HMMer Search

Pfam Database Search

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 2.3.2Pfam Database SearchLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 201307153691215SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 311.2410.0110.131. (CC) gcc options: -O3 -march=native -pthread -lhmmer -lsquid -lm

Timed MAFFT Alignment

Multiple Sequence Alignment

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MAFFT Alignment 6.864Multiple Sequence AlignmentLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 20130715246810SE +/- 0.16, N = 6SE +/- 0.15, N = 6SE +/- 0.14, N = 66.126.486.061. (CC) gcc options: -O3 -lm -lpthread

BLAKE2

Phoronix Test Suite v4.8.0m2

OpenBenchmarking.orgCycles Per Byte, Fewer Is BetterBLAKE2 20121223Phoronix Test Suite v4.8.0m2LLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 20130715246810SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 37.287.407.391. (CC) gcc options: -std=gnu99 -O3 -march=native -lcrypto -lz

SciMark

Computational Test: Monte Carlo

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Monte CarloLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 20130715100200300400500SE +/- 0.53, N = 4SE +/- 0.53, N = 4SE +/- 0.87, N = 4477.75478.82477.231. (CXX) g++ options: -O3 -march=native

SciMark

Computational Test: Fast Fourier Transform

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Fast Fourier TransformLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 2013071560120180240300SE +/- 0.25, N = 4SE +/- 0.47, N = 4SE +/- 0.69, N = 4289.81289.32288.101. (CXX) g++ options: -O3 -march=native

SciMark

Computational Test: Sparse Matrix Multiply

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Sparse Matrix MultiplyLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 20130715400800120016002000SE +/- 1.76, N = 4SE +/- 1.77, N = 4SE +/- 3.38, N = 41694.321701.361701.371. (CXX) g++ options: -O3 -march=native

SciMark

Computational Test: Dense LU Matrix Factorization

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Dense LU Matrix FactorizationLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 201307157001400210028003500SE +/- 1.43, N = 4SE +/- 1.43, N = 4SE +/- 0.00, N = 41628.511628.513137.251. (CXX) g++ options: -O3 -march=native

SciMark

Computational Test: Jacobi Successive Over-Relaxation

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Jacobi Successive Over-RelaxationLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 2013071530060090012001500SE +/- 1.56, N = 4SE +/- 1.56, N = 4SE +/- 1.57, N = 41385.711385.711392.001. (CXX) g++ options: -O3 -march=native

x264

H.264 Video Encoding

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2013-06-08H.264 Video EncodingLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 201307154080120160200SE +/- 0.51, N = 5SE +/- 0.20, N = 5SE +/- 0.52, N = 5177.42176.15177.771. (CC) gcc options: -ldl -m64 -lm -lpthread -O3 -ffast-math -march=native -std=gnu99 -fomit-frame-pointer -fno-tree-vectorize

GraphicsMagick

Operation: Blur

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.16Operation: BlurLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 201307151530456075SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 36667671. (CC) gcc options: -O3 -march=native -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread

GraphicsMagick

Operation: Sharpen

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.16Operation: SharpenLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 201307151020304050SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 34444441. (CC) gcc options: -O3 -march=native -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread

GraphicsMagick

Operation: Resizing

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.16Operation: ResizingLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 2013071520406080100SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 38283841. (CC) gcc options: -O3 -march=native -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread

GraphicsMagick

Operation: HWB Color Space

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.16Operation: HWB Color SpaceLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 2013071520406080100SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31001011011. (CC) gcc options: -O3 -march=native -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread

GraphicsMagick

Operation: Local Adaptive Thresholding

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.16Operation: Local Adaptive ThresholdingLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 20130715918273645SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 34039401. (CC) gcc options: -O3 -march=native -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread

Himeno Benchmark

Poisson Pressure Solver

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 2013071530060090012001500SE +/- 0.98, N = 3SE +/- 1.32, N = 3SE +/- 0.56, N = 31393.661413.541438.191. (CC) gcc options: -O3 -march=native

Timed ImageMagick Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed ImageMagick Compilation 6.8.1-10Time To CompileLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 20130715612182430SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.20, N = 323.2824.8525.89

Timed PHP Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 5.2.9Time To CompileLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 20130715510152025SE +/- 0.10, N = 3SE +/- 0.19, N = 3SE +/- 0.08, N = 318.9620.5820.74-lpthread1. (CC) gcc options: -O3 -march=native -pedantic -ldl -lz -lm

C-Ray

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total TimeLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 20130715612182430SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 323.7623.7623.251. (CC) gcc options: -lm -lpthread -O3 -march=native

Primesieve

1e12 Prime Number Generation

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 4.21e12 Prime Number GenerationLLVM Clang 3.3LLVM Clang 3.4 SVN 20130715100200300400500SE +/- 0.24, N = 3SE +/- 0.27, N = 3443.31432.841. (CXX) g++ options: -O2

Smallpt

Global Illumination Renderer; 100 Samples

OpenBenchmarking.orgSeconds, Fewer Is BetterSmallpt 1.0Global Illumination Renderer; 100 SamplesLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 2013071550100150200250SE +/- 0.88, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 32312272301. (CXX) g++ options: -fopenmp -O3 -march=native

FLAC Audio Encoding

WAV To FLAC

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.0WAV To FLACLLVM Clang 3.3LLVM Clang 3.4 SVN 20130715246810SE +/- 0.01, N = 5SE +/- 0.00, N = 58.147.081. (CXX) g++ options: -O3 -march=native -fvisibility=hidden -logg -lm

FFmpeg

H.264 HD To NTSC DV

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 2.0H.264 HD To NTSC DVLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 20130715510152025SE +/- 0.09, N = 3SE +/- 0.02, N = 3SE +/- 0.07, N = 320.9221.0521.331. (CC) gcc options: -lavdevice -lavfilter -lavformat -lavcodec -lswresample -lswscale -lavutil -ldl -lXv -lX11 -lXext -lasound -lSDL -lm -pthread -O3 -march=native -std=c99 -fomit-frame-pointer -fno-math-errno -fno-signed-zeros -Qunused-arguments -MMD

N-Queens

Elapsed Time

OpenBenchmarking.orgSeconds, Fewer Is BetterN-Queens 1.0Elapsed TimeLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 2013071560120180240300SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3274.20274.20274.761. (CC) gcc options: -static -fopenmp -O3 -march=native

PostgreSQL pgbench

TPC-B Transactions Per Second

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 8.4.11TPC-B Transactions Per SecondLLVM Clang 3.3LLVM Clang 3.4 SVN 2013071560120180240300SE +/- 1.02, N = 3276.47275.121. (CC) gcc options: -O3 -march=native -fno-strict-aliasing -fwrapv -lpgport -lpq -lcrypt -ldl -lm

Apache Benchmark

Static Web Page Serving

OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.3Static Web Page ServingLLVM Clang 3.2LLVM Clang 3.3LLVM Clang 3.4 SVN 201307154K8K12K16K20KSE +/- 285.79, N = 3SE +/- 263.14, N = 4SE +/- 28.55, N = 318271.8217678.9618491.401. (CC) gcc options: -shared -fPIC -pthread -O3 -march=native

Phoronix Test Suite v10.8.4