LLVM 3.5 RC4 Benchmarking

Benchmarks by Michael Larabel for a future article on Phoronix.com reviewing the LLVM Clang compiler performance.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 1408312-PTS-LLVM35RC08
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

Limit displaying results to tests within:

Audio Encoding 2 Tests
Timed Code Compilation 2 Tests
C/C++ Compiler Tests 9 Tests
CPU Massive 8 Tests
Creator Workloads 4 Tests
Encoding 2 Tests
Multi-Core 4 Tests
Programmer / Developer System Benchmarks 2 Tests
Server CPU Tests 3 Tests
Single-Threaded 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
LLVM Clang 3.4
August 31 2014
 
LLVM Clang 3.5 RC4
August 31 2014
 
Invert Hiding All Results Option
 
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 3.5 RC4 BenchmarkingOpenBenchmarking.orgPhoronix Test SuiteIntel Core i7-3517UE @ 2.80GHz (4 Cores)CompuLab Intense-PCIntel 3rd Gen Core DRAM8192MB500GB Hitachi HCC54755Intel HD 4000 (1000MHz)Intel 7 /C210MX279Intel 82579LM Gigabit Connection + Realtek RTL8188CE 802.11b/g/nUbuntu 14.043.17.0-999-generic (x86_64)Unity 7.2.2X Server 1.15.1intel 2.99.9143.3 Mesa 10.4.0-devel (git-eb4541e 2014-08-21 trusty-oibaf-ppa+gallium-nine)Clang 3.4 + LLVM 3.4Clang 3.5.0 + LLVM 3.5.0ext41920x1080ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilersFile-SystemScreen ResolutionLLVM 3.5 RC4 Benchmarking PerformanceSystem Logs- LLVM Clang 3.4: Optimized build; Built Aug 31 2014 (14:41:12); Default target: x86_64-unknown-linux-gnu; Host CPU: core-avx-i - LLVM Clang 3.5 RC4: Optimized build; Built Aug 31 2014 (14:18:02); Default target: x86_64-unknown-linux-gnu; Host CPU: core-avx-i - Scaling Governor: intel_pstate performance

LLVM Clang 3.4 vs. LLVM Clang 3.5 RC4 ComparisonPhoronix Test SuiteBaseline+19.3%+19.3%+38.6%+38.6%+57.9%+57.9%+77.2%+77.2%77.3%8.4%6.1%4.3%4.1%3%SharpenWAV To MP324.8%Time To Compile10.5%S.M.MTotal TimeResizing4.6%Monte CarloP.P.S4.2%CompositeTime To Compile3.8%WAV To FLAC3.6%D.L.M.FGraphicsMagickLAME MP3 EncodingTimed PHP CompilationSciMarkC-RayGraphicsMagickSciMarkHimeno BenchmarkSciMarkTimed Apache CompilationFLAC Audio EncodingSciMarkLLVM Clang 3.4LLVM Clang 3.5 RC4

LLVM 3.5 RC4 Benchmarkingscimark2: Compositescimark2: Monte Carloscimark2: Fast Fourier Transformscimark2: Sparse Matrix Multiplyscimark2: Dense LU Matrix Factorizationscimark2: Jacobi Successive Over-Relaxationgraphics-magick: Blurgraphics-magick: Sharpengraphics-magick: Resizinggraphics-magick: HWB Color Spacegraphics-magick: Local Adaptive Thresholdinghimeno: Poisson Pressure Solverbuild-apache: Time To Compilebuild-php: Time To Compilec-ray: Total Timeencode-flac: WAV To FLACencode-mp3: WAV To MP3apache: Static Web Page ServingLLVM Clang 3.4LLVM Clang 3.5 RC41152.03426.23194.931652.572309.721176.70602268106631202.3145.7541.6593.177.1816.9117819.651199.70444.58193.311791.412378.161191.08593965105631153.7647.4946.0487.827.4421.10OpenBenchmarking.org

SciMark

This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: CompositeLLVM Clang 3.4LLVM Clang 3.5 RC430060090012001500SE +/- 11.44, N = 4SE +/- 7.88, N = 41152.031199.701. (CXX) g++ options: -O3 -march=native
OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: CompositeLLVM Clang 3.4LLVM Clang 3.5 RC42004006008001000Min: 1118.17 / Avg: 1152.03 / Max: 1168.51Min: 1177.33 / Avg: 1199.7 / Max: 1213.841. (CXX) g++ options: -O3 -march=native

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Monte CarloLLVM Clang 3.4LLVM Clang 3.5 RC4100200300400500SE +/- 9.80, N = 4SE +/- 0.03, N = 4426.23444.581. (CXX) g++ options: -O3 -march=native
OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Monte CarloLLVM Clang 3.4LLVM Clang 3.5 RC480160240320400Min: 397.31 / Avg: 426.23 / Max: 440.75Min: 444.49 / Avg: 444.58 / Max: 444.651. (CXX) g++ options: -O3 -march=native

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Fast Fourier TransformLLVM Clang 3.4LLVM Clang 3.5 RC44080120160200SE +/- 2.94, N = 4SE +/- 0.56, N = 4194.93193.311. (CXX) g++ options: -O3 -march=native
OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Fast Fourier TransformLLVM Clang 3.4LLVM Clang 3.5 RC44080120160200Min: 191.16 / Avg: 194.93 / Max: 203.61Min: 191.83 / Avg: 193.31 / Max: 194.511. (CXX) g++ options: -O3 -march=native

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Sparse Matrix MultiplyLLVM Clang 3.4LLVM Clang 3.5 RC4400800120016002000SE +/- 40.26, N = 4SE +/- 9.48, N = 41652.571791.411. (CXX) g++ options: -O3 -march=native
OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Sparse Matrix MultiplyLLVM Clang 3.4LLVM Clang 3.5 RC430060090012001500Min: 1537.36 / Avg: 1652.57 / Max: 1711.07Min: 1776.26 / Avg: 1791.41 / Max: 1818.971. (CXX) g++ options: -O3 -march=native

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Dense LU Matrix FactorizationLLVM Clang 3.4LLVM Clang 3.5 RC45001000150020002500SE +/- 22.69, N = 4SE +/- 36.75, N = 42309.722378.161. (CXX) g++ options: -O3 -march=native
OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Dense LU Matrix FactorizationLLVM Clang 3.4LLVM Clang 3.5 RC4400800120016002000Min: 2241.93 / Avg: 2309.72 / Max: 2338.13Min: 2268.69 / Avg: 2378.16 / Max: 2421.181. (CXX) g++ options: -O3 -march=native

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Jacobi Successive Over-RelaxationLLVM Clang 3.4LLVM Clang 3.5 RC430060090012001500SE +/- 2.57, N = 4SE +/- 0.17, N = 41176.701191.081. (CXX) g++ options: -O3 -march=native
OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Jacobi Successive Over-RelaxationLLVM Clang 3.4LLVM Clang 3.5 RC42004006008001000Min: 1169.88 / Avg: 1176.7 / Max: 1181.5Min: 1190.79 / Avg: 1191.08 / Max: 1191.541. (CXX) g++ options: -O3 -march=native

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests to stress the system's CPU. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: BlurLLVM Clang 3.4LLVM Clang 3.5 RC41326395265SE +/- 0.33, N = 3SE +/- 0.00, N = 360591. (CC) gcc options: -O3 -march=native -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread
OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: BlurLLVM Clang 3.4LLVM Clang 3.5 RC41224364860Min: 59 / Avg: 59.67 / Max: 60Min: 59 / Avg: 59 / Max: 591. (CC) gcc options: -O3 -march=native -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: SharpenLLVM Clang 3.4LLVM Clang 3.5 RC4918273645SE +/- 0.00, N = 3SE +/- 0.00, N = 322391. (CC) gcc options: -O3 -march=native -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread
OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: SharpenLLVM Clang 3.4LLVM Clang 3.5 RC4816243240Min: 22 / Avg: 22 / Max: 22Min: 39 / Avg: 39 / Max: 391. (CC) gcc options: -O3 -march=native -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: ResizingLLVM Clang 3.4LLVM Clang 3.5 RC41530456075SE +/- 0.33, N = 3SE +/- 0.00, N = 368651. (CC) gcc options: -O3 -march=native -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread
OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: ResizingLLVM Clang 3.4LLVM Clang 3.5 RC41326395265Min: 68 / Avg: 68.33 / Max: 69Min: 65 / Avg: 65 / Max: 651. (CC) gcc options: -O3 -march=native -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: HWB Color SpaceLLVM Clang 3.4LLVM Clang 3.5 RC420406080100SE +/- 0.00, N = 3SE +/- 0.33, N = 31061051. (CC) gcc options: -O3 -march=native -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread
OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: HWB Color SpaceLLVM Clang 3.4LLVM Clang 3.5 RC420406080100Min: 106 / Avg: 106 / Max: 106Min: 105 / Avg: 105.33 / Max: 1061. (CC) gcc options: -O3 -march=native -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: Local Adaptive ThresholdingLLVM Clang 3.4LLVM Clang 3.5 RC41428425670SE +/- 0.00, N = 3SE +/- 0.00, N = 363631. (CC) gcc options: -O3 -march=native -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread
OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: Local Adaptive ThresholdingLLVM Clang 3.4LLVM Clang 3.5 RC41224364860Min: 63 / Avg: 63 / Max: 63Min: 63 / Avg: 63 / Max: 631. (CC) gcc options: -O3 -march=native -pthread -ljbig -lwebp -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -lpthread

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 3.4LLVM Clang 3.5 RC430060090012001500SE +/- 0.97, N = 3SE +/- 2.98, N = 31202.311153.761. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverLLVM Clang 3.4LLVM Clang 3.5 RC42004006008001000Min: 1200.54 / Avg: 1202.31 / Max: 1203.89Min: 1150.33 / Avg: 1153.76 / Max: 1159.691. (CC) gcc options: -O3 -march=native

Timed Apache Compilation

This test times how long it takes to build the Apache HTTP Server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Apache Compilation 2.4.7Time To CompileLLVM Clang 3.4LLVM Clang 3.5 RC41122334455SE +/- 0.19, N = 3SE +/- 0.25, N = 345.7547.49
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Apache Compilation 2.4.7Time To CompileLLVM Clang 3.4LLVM Clang 3.5 RC41020304050Min: 45.44 / Avg: 45.75 / Max: 46.11Min: 47.11 / Avg: 47.49 / Max: 47.97

Timed PHP Compilation

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 5.2.9Time To CompileLLVM Clang 3.4LLVM Clang 3.5 RC41020304050SE +/- 0.02, N = 3SE +/- 0.08, N = 341.6546.041. (CC) gcc options: -O3 -march=native -pedantic -ldl -lz -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 5.2.9Time To CompileLLVM Clang 3.4LLVM Clang 3.5 RC4918273645Min: 41.61 / Avg: 41.65 / Max: 41.68Min: 45.9 / Avg: 46.04 / Max: 46.181. (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 3.4LLVM Clang 3.5 RC420406080100SE +/- 0.03, N = 3SE +/- 0.00, N = 393.1787.821. (CC) gcc options: -lm -lpthread -O3 -march=native
OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total TimeLLVM Clang 3.4LLVM Clang 3.5 RC420406080100Min: 93.11 / Avg: 93.17 / Max: 93.23Min: 87.82 / Avg: 87.82 / Max: 87.831. (CC) gcc options: -lm -lpthread -O3 -march=native

FLAC Audio Encoding

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.0WAV To FLACLLVM Clang 3.4LLVM Clang 3.5 RC4246810SE +/- 0.02, N = 5SE +/- 0.00, N = 57.187.441. (CXX) g++ options: -O3 -march=native -fvisibility=hidden -logg -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.0WAV To FLACLLVM Clang 3.4LLVM Clang 3.5 RC43691215Min: 7.13 / Avg: 7.18 / Max: 7.21Min: 7.43 / Avg: 7.44 / Max: 7.451. (CXX) g++ options: -O3 -march=native -fvisibility=hidden -logg -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 3.4LLVM Clang 3.5 RC4510152025SE +/- 0.05, N = 5SE +/- 0.01, N = 516.9121.101. (CC) gcc options: -pipe -O3 -march=native -lncurses -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.99.3WAV To MP3LLVM Clang 3.4LLVM Clang 3.5 RC4510152025Min: 16.83 / Avg: 16.91 / Max: 17.09Min: 21.07 / Avg: 21.1 / Max: 21.121. (CC) gcc options: -pipe -O3 -march=native -lncurses -lm

Apache Benchmark

This is a test of ab, which is the Apache benchmark program. This test profile measures how many requests per second a given system can sustain when carrying out 1,000,000 requests with 100 requests being carried out concurrently. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.7Static Web Page ServingLLVM Clang 3.44K8K12K16K20KSE +/- 43.80, N = 317819.651. (CC) gcc options: -shared -fPIC -pthread -O3 -march=native