LLVM Clang 3.1 GCC 4.7 Intel Core i7 Benchmarks

Benchmarking of GCC 4.6.3, GCC 4.7.0, LLVM Clang 3.0, and LLVM Clang 3.1 SVN compilers for a future article on Phoronix.com. Compilers built all test binaries with CFLAGS/CXXFLAGS of -O3 -march=native from this Intel Core i7 Clarksfield notebook. Ubuntu 12.04 LTS x86_64 was used with the Linux 3.2 kernel.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 1204215-SU-LLVMCLANG23
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts

Limit displaying results to tests within:

Audio Encoding 2 Tests
Bioinformatics 4 Tests
C/C++ Compiler Tests 17 Tests
Compression Tests 2 Tests
CPU Massive 22 Tests
Creator Workloads 9 Tests
Cryptography 2 Tests
Encoding 5 Tests
Fortran Tests 2 Tests
HPC - High Performance Computing 5 Tests
Common Kernel Benchmarks 3 Tests
MPI Benchmarks 2 Tests
Multi-Core 13 Tests
Raytracing 2 Tests
Renderers 4 Tests
Scientific Computing 5 Tests
Server 4 Tests
Server CPU Tests 7 Tests
Single-Threaded 7 Tests
Video Encoding 3 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
GCC 4.6.3
April 21 2012
 
GCC 4.7.0
April 21 2012
 
LLVM-Clang 3.0
April 21 2012
 
LLVM-Clang 3.1 SVN
April 21 2012
 
Open64 5.0
April 21 2012
 
Invert Hiding All Results Option
 

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


LLVM Clang 3.1 GCC 4.7 Intel Core i7 BenchmarksOpenBenchmarking.orgPhoronix Test SuiteIntel Core i7 720Q @ 1.60GHz (8 Cores)LENOVO 4318CTOIntel Core DMI4096MB160GB INTEL SSDSA2M160NVIDIA Quadro FX 880M 1024MB (405/324MHz)Conexant CX20585Intel 82577LM Gigabit Connection + Intel Centrino Ultimate-N 6300Ubuntu 12.043.4.0-999-generic (x86_64)Unity 5.10.0X Server 1.11.3nouveau 0.0.162.1 Mesa 8.0.2 Gallium 0.4GCC 4.6.3GCC 4.7.0Clang 3.0 + LLVM 3.0Clang 3.1 (SVN 155295) + LLVM 3.1svnOpen64 5.0ext41600x900ProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilersFile-SystemScreen ResolutionLLVM Clang 3.1 GCC 4.7 Intel Core I7 Benchmarks PerformanceSystem Logs- GCC 4.6.3: --enable-checking=release --enable-languages=c,c++ --enable-lto- GCC 4.7.0: --enable-checking=release --enable-languages=c,c++ --enable-lto- LLVM-Clang 3.0: Optimized build; Built Apr 21 2012 (12:40:14); Host: x86_64-unknown-linux-gnu; Host CPU: corei7 - LLVM-Clang 3.1 SVN: Optimized build; Built Apr 21 2012 (14:54:30); Default target: x86_64-unknown-linux-gnu; Host CPU: corei7 - Scaling Governor: ondemand- Compiz was running on this system.

LLVM Clang 3.1 GCC 4.7 Intel Core i7 Benchmarksffte: N=64, 1D Complex FFT Routinehmmer: Pfam Database Searchmafft: Multiple Sequence Alignmentmrbayes: Primate Phylogeny Analysisgmpbench: Total Timefhourstones: Complex Connect-4 Solvingbyte: Dhrystone 2john-the-ripper: Blowfishttsiod-renderer: Phong Rendering With Soft-Shadow Mappingvpxenc: vpxencx264: H.264 Video Encodinghimeno: Poisson Pressure Solvercompress-7zip: Compress Speed Testbuild-php: Time To Compilec-ray: Total Timecompress-pbzip2: 256MB File Compressionsmallpt: Global Illumination Renderer; 100 Samplescrafty: Elapsed Timeencode-flac: WAV To FLACencode-mp3: WAV To MP3ffmpeg: AVI To NTSC VCDminion: Bibdminion: Gracefulminion: Solitaireminion: Quasigroupnero2d: Total Timetachyon: Total Timeopenssl: RSA 4096-bit Performancepgbench: TPC-B Transactions Per Secondnginx: Static Web Page Servingapache: Static Web Page ServingGCC 4.6.3GCC 4.7.0LLVM-Clang 3.0LLVM-Clang 3.1 SVNOpen64 5.03276.1926.5213.7337.0717398643.4018737961.27221563.329.9358.461060.59884950.42139.0119.3469109.889.9723.5815.87206.67111.35203.00255.83640.1736.0542.232375.1815482.0814290.993156.9226.6313.6032.381751.208631.4318075520.17221159.509.7658.261142.97889054.7491.1822.3068108.299.3223.1015.73217.34122.41211.57253.16638.6037.2142.082405.9114956.3014322.053270.1933.0216.141002.879108118.20234223.86110.30224.90277.6915559.9614867.103197.6832.6314.7340.098551.906629.071053.29916647.49109.1123.21251107.8510.8523.8116.09219.58110.04217.96262.5036.1742.082426.9115767.2114834.663267.81OpenBenchmarking.org

FFTE

FFTE is a package by Daisuke Takahashi to compute Discrete Fourier Transforms of 1-, 2- and 3- dimensional sequences of length (2^p)*(3^q)*(5^r). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterFFTE 5.0Test: N=64, 1D Complex FFT RoutineOpen64 5.0LLVM-Clang 3.1 SVNLLVM-Clang 3.0GCC 4.7.0GCC 4.6.37001400210028003500SE +/- 6.76, N = 3SE +/- 29.94, N = 3SE +/- 7.30, N = 3SE +/- 48.14, N = 3SE +/- 11.98, N = 33267.813197.683270.193156.923276.191. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp -pthread -lmpi_f90 -lmpi_f77 -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil -lm

Timed HMMer Search

This test searches through the Pfam database of profile hidden markov models. The search finds the domain structure of Drosophila Sevenless protein. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 2.3.2Pfam Database SearchLLVM-Clang 3.1 SVNLLVM-Clang 3.0GCC 4.7.0GCC 4.6.3816243240SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 332.6333.0226.6326.521. (CC) gcc options: -O3 -march=native -pthread -lhmmer -lsquid -lm

Timed MAFFT Alignment

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MAFFT Alignment 6.864Multiple Sequence AlignmentLLVM-Clang 3.1 SVNLLVM-Clang 3.0GCC 4.7.0GCC 4.6.348121620SE +/- 0.07, N = 3SE +/- 0.28, N = 4SE +/- 0.04, N = 3SE +/- 0.18, N = 614.7316.1413.6013.731. (CC) gcc options: -O3 -lm -lpthread

Timed MrBayes Analysis

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.1.2Primate Phylogeny AnalysisLLVM-Clang 3.1 SVNGCC 4.7.0GCC 4.6.3918273645SE +/- 0.25, N = 3SE +/- 0.09, N = 3SE +/- 0.03, N = 340.0932.3837.07

GMPbench

OpenBenchmarking.orgGMPbench Score, More Is BetterGMPbench 0.2Total TimeGCC 4.7.0GCC 4.6.34008001200160020001751.201739.001. (CC) gcc options: -O3 -march=native

Fhourstones

OpenBenchmarking.orgKpos / sec, More Is BetterFhourstones 3.1Complex Connect-4 SolvingLLVM-Clang 3.1 SVNGCC 4.7.0GCC 4.6.32K4K6K8K10KSE +/- 5.30, N = 3SE +/- 5.11, N = 3SE +/- 22.37, N = 38551.908631.438643.401. (CC) gcc options: -O3

BYTE Unix Benchmark

OpenBenchmarking.orgLPS, More Is BetterBYTE Unix Benchmark 3.6Computational Test: Dhrystone 2GCC 4.7.0GCC 4.6.34M8M12M16M20MSE +/- 14064.68, N = 3SE +/- 36800.75, N = 318075520.1718737961.271. (CC) gcc options: -O3 -march=native

John The Ripper

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.7.9Test: BlowfishLLVM-Clang 3.1 SVNGCC 4.7.0GCC 4.6.35001000150020002500SE +/- 1.53, N = 3SE +/- 1.67, N = 3SE +/- 1.67, N = 3662221122151. (CC) gcc options: -fopenmp -lcrypt

TTSIOD 3D Renderer

OpenBenchmarking.orgFPS, More Is BetterTTSIOD 3D Renderer 2.2wPhong Rendering With Soft-Shadow MappingGCC 4.7.0GCC 4.6.31428425670SE +/- 0.12, N = 3SE +/- 0.04, N = 359.5063.321. (CXX) g++ options: -O3 -march=native -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -lstdc++

VP8 libvpx Encoding

OpenBenchmarking.orgFrames Per Second, More Is BetterVP8 libvpx Encoding 0.9.7-p1vpxencLLVM-Clang 3.1 SVNGCC 4.7.0GCC 4.6.33691215SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 39.079.769.931. (CC) gcc options: -m64 -lvpx -lm -lpthread

x264

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2011-12-06H.264 Video EncodingGCC 4.7.0GCC 4.6.31326395265SE +/- 0.12, N = 3SE +/- 0.05, N = 358.2658.46

Himeno Benchmark

The Himeno benchmark is a linear solver of pressure Poisson using a point-Jacobi method. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverLLVM-Clang 3.1 SVNLLVM-Clang 3.0GCC 4.7.0GCC 4.6.32004006008001000SE +/- 9.10, N = 3SE +/- 5.12, N = 3SE +/- 2.04, N = 3SE +/- 16.23, N = 31053.291002.871142.971060.591. (CC) gcc options: -O3 -march=native

7-Zip Compression

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 9.20.1Compress Speed TestLLVM-Clang 3.1 SVNLLVM-Clang 3.0GCC 4.7.0GCC 4.6.32K4K6K8K10KSE +/- 7.55, N = 3SE +/- 11.50, N = 3SE +/- 42.67, N = 3SE +/- 65.12, N = 391669108889088491. (CXX) g++ options: -pipe -lpthread

Timed PHP Compilation

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 5.2.9Time To CompileLLVM-Clang 3.1 SVNGCC 4.7.0GCC 4.6.31224364860SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.07, N = 347.4954.7450.42-lpthread1. (CC) gcc options: -O3 -march=native -pedantic -ldl -lz -lm

C-Ray

This is a test of C-Ray, a simple raytracer designed to test the floating-point CPU performance. This test is multi-threaded (16 threads per core), will shoot 8 rays per pixel for anti-aliasing, and will generate a 1600 x 1200 image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total TimeLLVM-Clang 3.1 SVNLLVM-Clang 3.0GCC 4.7.0GCC 4.6.3306090120150SE +/- 4.36, N = 6SE +/- 3.60, N = 6SE +/- 3.00, N = 6SE +/- 12.48, N = 6109.11118.2091.18139.011. (CC) gcc options: -lm -lpthread -O3 -march=native

Parallel BZIP2 Compression

OpenBenchmarking.orgSeconds, Fewer Is BetterParallel BZIP2 Compression 1.0.5256MB File CompressionLLVM-Clang 3.1 SVNGCC 4.7.0GCC 4.6.3612182430SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 323.2122.3019.341. (CXX) g++ options: -O2 -pthread -lpthread -lbz2

Smallpt

Smallpt is a C++ global illumination renderer written in less than 100 lines of code. Global illumination is done via unbiased Monte Carlo path tracing and there is multi-threading support via the OpenMP library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSmallpt 1.0Global Illumination Renderer; 100 SamplesLLVM-Clang 3.1 SVNLLVM-Clang 3.0GCC 4.7.0GCC 4.6.350100150200250SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 325123468691. (CXX) g++ options: -fopenmp -O3 -march=native

Crafty

OpenBenchmarking.orgSeconds, Fewer Is BetterCrafty 23.4Elapsed TimeLLVM-Clang 3.1 SVNGCC 4.7.0GCC 4.6.320406080100SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.11, N = 3107.85108.29109.881. (CC) gcc options: -lstdc++ -lm

FLAC Audio Encoding

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.2.1WAV To FLACLLVM-Clang 3.1 SVNGCC 4.7.0GCC 4.6.33691215SE +/- 0.02, N = 5SE +/- 0.02, N = 5SE +/- 0.01, N = 510.859.329.971. (CXX) g++ options: -O3 -march=native -lm

LAME MP3 Encoding

LAME is an MP3 encoder licensed under the LGPL. This test measures the time required to encode a WAV file to MP3 format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.99.3WAV To MP3LLVM-Clang 3.1 SVNGCC 4.7.0GCC 4.6.3612182430SE +/- 0.04, N = 5SE +/- 0.02, N = 5SE +/- 0.04, N = 523.8123.1023.58

FFmpeg

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 0.10AVI To NTSC VCDLLVM-Clang 3.1 SVNGCC 4.7.0GCC 4.6.348121620SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 316.0915.7315.871. (CC) gcc options: -lavdevice -lavfilter -lavformat -lavcodec -lswresample -lswscale -lavutil -ldl -lasound -lSDL -lm -pthread -lbz2

Minion

OpenBenchmarking.orgSeconds, Fewer Is BetterMinion 0.12Benchmark: BibdLLVM-Clang 3.1 SVNLLVM-Clang 3.0GCC 4.7.0GCC 4.6.350100150200250SE +/- 0.37, N = 3SE +/- 0.43, N = 3SE +/- 0.27, N = 3SE +/- 0.63, N = 3219.58223.86217.34206.67-mdynamic-no-pic -lz -lbz2-mdynamic-no-pic -lz -lbz21. (CXX) g++ options: -O3 -fomit-frame-pointer -rdynamic -lboost_iostreams-mt

OpenBenchmarking.orgSeconds, Fewer Is BetterMinion 0.12Benchmark: GracefulLLVM-Clang 3.1 SVNLLVM-Clang 3.0GCC 4.7.0GCC 4.6.3306090120150SE +/- 0.31, N = 3SE +/- 0.08, N = 3SE +/- 0.33, N = 3SE +/- 0.27, N = 3110.04110.30122.41111.35-mdynamic-no-pic -lz -lbz2-mdynamic-no-pic -lz -lbz21. (CXX) g++ options: -O3 -fomit-frame-pointer -rdynamic -lboost_iostreams-mt

OpenBenchmarking.orgSeconds, Fewer Is BetterMinion 0.12Benchmark: SolitaireLLVM-Clang 3.1 SVNLLVM-Clang 3.0GCC 4.7.0GCC 4.6.350100150200250SE +/- 0.41, N = 3SE +/- 0.44, N = 3SE +/- 0.22, N = 3SE +/- 0.36, N = 3217.96224.90211.57203.00-mdynamic-no-pic -lz -lbz2-mdynamic-no-pic -lz -lbz21. (CXX) g++ options: -O3 -fomit-frame-pointer -rdynamic -lboost_iostreams-mt

OpenBenchmarking.orgSeconds, Fewer Is BetterMinion 0.12Benchmark: QuasigroupLLVM-Clang 3.1 SVNLLVM-Clang 3.0GCC 4.7.0GCC 4.6.360120180240300SE +/- 0.21, N = 3SE +/- 0.27, N = 3SE +/- 0.04, N = 3SE +/- 0.25, N = 3262.50277.69253.16255.83-mdynamic-no-pic -lz -lbz2-mdynamic-no-pic -lz -lbz21. (CXX) g++ options: -O3 -fomit-frame-pointer -rdynamic -lboost_iostreams-mt

Open FMM Nero2D

OpenBenchmarking.orgSeconds, Fewer Is BetterOpen FMM Nero2D 2.0.2Total TimeGCC 4.7.0GCC 4.6.3140280420560700638.60640.171. (CXX) g++ options: -O3 -march=native -lfftw3 -llapack -lblas -lgfortran -lm

Tachyon

OpenBenchmarking.orgSeconds, Fewer Is BetterTachyon 0.98.9Total TimeLLVM-Clang 3.1 SVNGCC 4.7.0GCC 4.6.3918273645SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 336.1737.2136.051. (CC) gcc options: -m32 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread

OpenSSL

OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.0.0eRSA 4096-bit PerformanceLLVM-Clang 3.1 SVNGCC 4.7.0GCC 4.6.31020304050SE +/- 0.05, N = 4SE +/- 0.03, N = 4SE +/- 0.10, N = 442.0842.0842.231. (CC) gcc options: -m64 -O3 -lssl -lcrypto -ldl

PostgreSQL pgbench

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 8.4.11TPC-B Transactions Per SecondLLVM-Clang 3.1 SVNGCC 4.7.0GCC 4.6.35001000150020002500SE +/- 16.75, N = 3SE +/- 10.83, N = 3SE +/- 12.24, N = 32426.912405.912375.181. (CC) gcc options: -O3 -march=native -fno-strict-aliasing -fwrapv -lpgport -lpq -lcrypt -ldl -lm

NGINX Benchmark

OpenBenchmarking.orgRequests Per Second, More Is BetterNGINX Benchmark 1.0.11Static Web Page ServingLLVM-Clang 3.1 SVNLLVM-Clang 3.0GCC 4.7.0GCC 4.6.33K6K9K12K15KSE +/- 141.05, N = 3SE +/- 184.13, N = 3SE +/- 54.72, N = 3SE +/- 89.02, N = 315767.2115559.9614956.3015482.081. (CC) gcc options: -lpthread -lcrypt -lcrypto -lz

Apache Benchmark

OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.2.21Static Web Page ServingLLVM-Clang 3.1 SVNLLVM-Clang 3.0GCC 4.7.0GCC 4.6.33K6K9K12K15KSE +/- 131.26, N = 3SE +/- 62.06, N = 3SE +/- 59.62, N = 3SE +/- 14.58, N = 314834.6614867.1014322.0514290.991. (CC) gcc options: -pthread -O3 -march=native -lm -lexpat -lrt -lcrypt -lpthread -ldl