LLVM Clang 3.4 AMD APU Benchmarks

Benchmarks by Michael Larabel for a future article on Phoronix.com. Quick look at LLVM Clang 3.3 vs. Clang 3.4 compiler performance. More tests forthcoming.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 1312033-SO-LLVMCLANG98
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
Allow Limiting Results To Certain Suite(s)

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
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
Toggle/Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Clang 3.3
December 03 2013
 
Clang 3.5 SVN
December 03 2013
 
Invert Behavior (Only Show Selected Data)
 
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 3.4 AMD APU BenchmarksOpenBenchmarking.orgPhoronix Test SuiteAMD A10-6800K APU @ 4.70GHz (4 Cores)MSI FM2-A85XA-G65 (MS-7793) v1.0AMD Family 15h8192MB64GB OCZ AGILITYSapphire AMD Radeon HD 6950 2048MBRealtek ALC892SyncMasterRealtek RTL8111/8168/8411Ubuntu 13.103.13.0-999-generic (x86_64)Unity 7.1.2X Server 1.14.3radeon 7.2.993.1 Mesa 10.1.0-devel (git-5b331f6 saucy-oibaf-ppa) Gallium 0.4Clang 3.3-5ubuntu4Clang 3.5-1~exp1ext42560x1600ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilersFile-SystemScreen ResolutionLLVM Clang 3.4 AMD APU Benchmarks PerformanceSystem Logs

Clang 3.3 vs. Clang 3.5 SVN ComparisonPhoronix Test SuiteBaseline+6.4%+6.4%+12.8%+12.8%+19.2%+19.2%+25.6%+25.6%12.4%10.6%6%5.4%2.5%2.1%KASUMI25.7%WAV To MP314.7%WAV To FLACD.L.M.FF.F.T7.5%P.T.S.v.8.5P.P.SG.I.R.1.S4.3%Monte Carlo3.9%CAST-2562.8%CompositeTotal TimeBotanLAME MP3 EncodingFLAC Audio EncodingSciMarkSciMarkBLAKE2Himeno BenchmarkSmallptSciMarkBotanSciMarkC-RayClang 3.3Clang 3.5 SVN

LLVM Clang 3.4 AMD APU Benchmarksblake2: Phoronix Test Suite v4.8.5botan: Tigerbotan: KASUMIbotan: AES-256botan: Twofishbotan: CAST-256botan: X9.19-MACc-ray: Total Timeencode-flac: WAV To FLAChimeno: Poisson Pressure Solverencode-mp3: WAV To MP3scimark2: Compositescimark2: Monte Carloscimark2: Fast Fourier Transformscimark2: Sparse Matrix Multiplyscimark2: Dense LU Matrix Factorizationscimark2: Jacobi Successive Over-Relaxationsmallpt: Global Illumination Renderer; 100 SamplesClang 3.3Clang 3.5 SVN9.66361.1679.39180.17217.19115.6275.1763.667.09770.3216.36830.17493.1174.161016.391198.511368.692119.11360.7863.17178.18218.46112.4575.8662.346.31811.8018.76850.62474.5969.011015.771325.061368.69220OpenBenchmarking.org

BLAKE2

OpenBenchmarking.orgCycles Per Byte, Fewer Is BetterBLAKE2 20121223Phoronix Test Suite v4.8.5Clang 3.3Clang 3.5 SVN3691215SE +/- 0.03, N = 3SE +/- 0.00, N = 39.669.111. (CC) gcc options: -std=gnu99 -O3 -march=native -lcrypto -lz

Botan

OpenBenchmarking.orgMbytes/s, More Is BetterBotan 1.10.3Test: TigerClang 3.3Clang 3.5 SVN80160240320400361.16360.781. (CXX) g++ options: -m64 -ldl -lpthread -lrt -O2

OpenBenchmarking.orgMbytes/s, More Is BetterBotan 1.10.3Test: KASUMIClang 3.3Clang 3.5 SVN2040608010079.3963.171. (CXX) g++ options: -m64 -ldl -lpthread -lrt -O2

OpenBenchmarking.orgMbytes/s, More Is BetterBotan 1.10.3Test: AES-256Clang 3.3Clang 3.5 SVN4080120160200180.17178.181. (CXX) g++ options: -m64 -ldl -lpthread -lrt -O2

OpenBenchmarking.orgMbytes/s, More Is BetterBotan 1.10.3Test: TwofishClang 3.3Clang 3.5 SVN50100150200250217.19218.461. (CXX) g++ options: -m64 -ldl -lpthread -lrt -O2

OpenBenchmarking.orgMbytes/s, More Is BetterBotan 1.10.3Test: CAST-256Clang 3.3Clang 3.5 SVN306090120150115.62112.451. (CXX) g++ options: -m64 -ldl -lpthread -lrt -O2

OpenBenchmarking.orgMbytes/s, More Is BetterBotan 1.10.3Test: X9.19-MACClang 3.3Clang 3.5 SVN2040608010075.1775.861. (CXX) g++ options: -m64 -ldl -lpthread -lrt -O2

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 TimeClang 3.3Clang 3.5 SVN1428425670SE +/- 1.09, N = 6SE +/- 1.12, N = 663.6662.341. (CC) gcc options: -lm -lpthread -O3 -march=native

FLAC Audio Encoding

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.0WAV To FLACClang 3.3Clang 3.5 SVN246810SE +/- 0.00, N = 5SE +/- 0.00, N = 57.096.311. (CXX) g++ options: -O3 -march=native -fvisibility=hidden -logg -lm

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 SolverClang 3.3Clang 3.5 SVN2004006008001000SE +/- 8.56, N = 3SE +/- 2.25, N = 3770.32811.801. (CC) gcc options: -O3 -march=native

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 MP3Clang 3.3Clang 3.5 SVN510152025SE +/- 0.00, N = 5SE +/- 0.01, N = 516.3618.761. (CC) gcc options: -pipe -O3 -march=native -lm

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: CompositeClang 3.3Clang 3.5 SVN2004006008001000SE +/- 0.55, N = 4SE +/- 0.52, N = 4830.17850.621. (CXX) g++ options: -O3 -march=native

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Monte CarloClang 3.3Clang 3.5 SVN110220330440550SE +/- 0.57, N = 4SE +/- 0.53, N = 4493.11474.591. (CXX) g++ options: -O3 -march=native

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Fast Fourier TransformClang 3.3Clang 3.5 SVN1632486480SE +/- 0.64, N = 4SE +/- 0.23, N = 274.1669.011. (CXX) g++ options: -O3 -march=native

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Sparse Matrix MultiplyClang 3.3Clang 3.5 SVN2004006008001000SE +/- 1.46, N = 4SE +/- 2.79, N = 41016.391015.771. (CXX) g++ options: -O3 -march=native

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Dense LU Matrix FactorizationClang 3.3Clang 3.5 SVN30060090012001500SE +/- 1.56, N = 4SE +/- 1.64, N = 41198.511325.061. (CXX) g++ options: -O3 -march=native

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Jacobi Successive Over-RelaxationClang 3.3Clang 3.5 SVN30060090012001500SE +/- 0.00, N = 4SE +/- 0.00, N = 41368.691368.691. (CXX) g++ options: -O3 -march=native

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 SamplesClang 3.3Clang 3.5 SVN50100150200250SE +/- 0.33, N = 3SE +/- 0.33, N = 32112201. (CXX) g++ options: -fopenmp -O3 -march=native