LLVM 3.2 Clang Intel Linux Compiler Intel Core i7-3960X Sandy Bridge Extreme Edition computer running Ubuntu 12.10 and comparing builds of LLVM/Clang 3.1 and LLVM/Clang 3.2 SVN. LLVM and Clang both times were built in release mode of --disable-assertions and --enable-optimized. Compiler flags passed to tests included -O3 and then each -march=nocona and -march=corei7-avx to see if there's any x86_64 compiler tuning optimizations when using corei7-avx for Intel Sandy Bridge CPUs. Benchmarking for a future article on Phoronix.com.
HTML result view exported from: https://openbenchmarking.org/result/1210095-RA-LLVM32CLA70&sro&grs .
Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Compiler File-System Screen Resolution nocona corei7-avx nocona corei7-avx LLVM-Clang 3.1 LLVM-Clang 3.1 LLVM-Clang 3.2 SVN LLVM-Clang 3.2 SVN Intel Core i7-3960X @ 3.30GHz (12 Cores) Intel DX79SI Intel Xeon E5/Core 8192MB 64GB OCZ VERTEX AMD Radeon HD 4650 512MB Realtek ALC892 DELL S2409W Intel 82579LM Gigabit Connection Ubuntu 12.10 3.6.0-030600-generic (x86_64) Unity 6.8.0 X Server 1.13.0 radeon 6.99.99 3.0 Mesa 9.0-devel Gallium 0.4 Clang 3.1 + LLVM 3.1 ext4 1920x1080 Clang 3.2 (SVN 165433) + LLVM 3.2svn OpenBenchmarking.org Compiler Details - nocona: LLVM-Clang 3.1, corei7-avx: LLVM-Clang 3.1: Optimized build; Built Oct 9 2012 (06:32:52); Default target: x86_64-unknown-linux-gnu; Host CPU: corei7-avx Processor Details - Scaling Governor: ondemand System Details - Compiz was running on this system.
graphics-magick: Sharpen build-php: Time To Compile c-ray: Total Time graphics-magick: Resizing graphics-magick: Blur graphics-magick: Local Adaptive Thresholding smallpt: Global Illumination Renderer; 100 Samples pgbench: TPC-B Transactions Per Second apache: Static Web Page Serving graphics-magick: HWB Color Space himeno: Poisson Pressure Solver encode-flac: WAV To FLAC hmmer: Pfam Database Search nocona corei7-avx nocona corei7-avx LLVM-Clang 3.1 LLVM-Clang 3.1 LLVM-Clang 3.2 SVN LLVM-Clang 3.2 SVN 53 20.57 26.34 104 87 45 147 366.36 16863.32 121 1575.76 6.41 15.41 52 20.57 26.30 104 87 49 146 360.51 16850.87 121 1582.01 6.41 15.42 53 15.47 21.62 107 78 48 157 370.42 16835.24 123 1561.02 6.40 15.45 31 15.69 20.99 89 82 45 152 377.98 16277.25 124 1545.92 6.36 15.37 OpenBenchmarking.org
GraphicsMagick Operation: Sharpen corei7-avx nocona OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.16 Operation: Sharpen LLVM-Clang 3.1 LLVM-Clang 3.2 SVN 12 24 36 48 60 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 52 31 53 53 1. (CC) gcc options: -march=nocona -O3 -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread
Timed PHP Compilation Time To Compile corei7-avx nocona OpenBenchmarking.org Seconds, Fewer Is Better Timed PHP Compilation 5.2.9 Time To Compile LLVM-Clang 3.1 LLVM-Clang 3.2 SVN 5 10 15 20 25 SE +/- 0.06, N = 3 SE +/- 0.22, N = 6 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 20.57 15.69 20.57 15.47 1. (CC) gcc options: -march=nocona -O3 -pedantic -ldl -lpthread -lz -lm
C-Ray Total Time corei7-avx nocona OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time LLVM-Clang 3.1 LLVM-Clang 3.2 SVN 6 12 18 24 30 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 26.30 20.99 26.34 21.62 1. (CC) gcc options: -lm -lpthread -O3 -march=nocona
GraphicsMagick Operation: Resizing corei7-avx nocona OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.16 Operation: Resizing LLVM-Clang 3.1 LLVM-Clang 3.2 SVN 20 40 60 80 100 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 104 89 104 107 1. (CC) gcc options: -march=nocona -O3 -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread
GraphicsMagick Operation: Blur corei7-avx nocona OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.16 Operation: Blur LLVM-Clang 3.1 LLVM-Clang 3.2 SVN 20 40 60 80 100 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 SE +/- 0.33, N = 3 SE +/- 0.00, N = 3 87 82 87 78 1. (CC) gcc options: -march=nocona -O3 -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread
GraphicsMagick Operation: Local Adaptive Thresholding corei7-avx nocona OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.16 Operation: Local Adaptive Thresholding LLVM-Clang 3.1 LLVM-Clang 3.2 SVN 11 22 33 44 55 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 49 45 45 48 1. (CC) gcc options: -march=nocona -O3 -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread
Smallpt Global Illumination Renderer; 100 Samples corei7-avx nocona OpenBenchmarking.org Seconds, Fewer Is Better Smallpt 1.0 Global Illumination Renderer; 100 Samples LLVM-Clang 3.1 LLVM-Clang 3.2 SVN 30 60 90 120 150 SE +/- 0.33, N = 3 SE +/- 0.58, N = 3 SE +/- 0.33, N = 3 SE +/- 0.88, N = 3 146 152 147 157 1. (CXX) g++ options: -fopenmp -march=nocona -O3
PostgreSQL pgbench TPC-B Transactions Per Second corei7-avx nocona OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 8.4.11 TPC-B Transactions Per Second LLVM-Clang 3.1 LLVM-Clang 3.2 SVN 80 160 240 320 400 SE +/- 4.65, N = 3 SE +/- 0.65, N = 3 SE +/- 6.05, N = 3 SE +/- 6.91, N = 3 360.51 377.98 366.36 370.42 1. (CC) gcc options: -march=nocona -O3 -fno-strict-aliasing -fwrapv -lpgport -lpq -lcrypt -ldl -lm
Apache Benchmark Static Web Page Serving corei7-avx nocona OpenBenchmarking.org Requests Per Second, More Is Better Apache Benchmark 2.4.3 Static Web Page Serving LLVM-Clang 3.1 LLVM-Clang 3.2 SVN 4K 8K 12K 16K 20K SE +/- 43.37, N = 3 SE +/- 299.58, N = 3 SE +/- 71.80, N = 3 SE +/- 26.63, N = 3 16850.87 16277.25 16863.32 16835.24 1. (CC) gcc options: -shared -fPIC -pthread -march=nocona -O3
GraphicsMagick Operation: HWB Color Space corei7-avx nocona OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.16 Operation: HWB Color Space LLVM-Clang 3.1 LLVM-Clang 3.2 SVN 30 60 90 120 150 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 121 124 121 123 1. (CC) gcc options: -march=nocona -O3 -pthread -lXext -lSM -lICE -lX11 -lz -lm -lpthread
Himeno Benchmark Poisson Pressure Solver corei7-avx nocona OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver LLVM-Clang 3.1 LLVM-Clang 3.2 SVN 300 600 900 1200 1500 SE +/- 2.49, N = 3 SE +/- 3.54, N = 3 SE +/- 1.69, N = 3 SE +/- 2.19, N = 3 1582.01 1545.92 1575.76 1561.02 1. (CC) gcc options: -O3 -march=nocona
FLAC Audio Encoding WAV To FLAC corei7-avx nocona OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.2.1 WAV To FLAC LLVM-Clang 3.1 LLVM-Clang 3.2 SVN 2 4 6 8 10 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 SE +/- 0.01, N = 5 6.41 6.36 6.41 6.40 1. (CXX) g++ options: -march=nocona -O3 -logg -lm
Timed HMMer Search Pfam Database Search corei7-avx nocona OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 2.3.2 Pfam Database Search LLVM-Clang 3.1 LLVM-Clang 3.2 SVN 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 15.42 15.37 15.41 15.45 1. (CC) gcc options: -march=nocona -O3 -pthread -lhmmer -lsquid -lm
Phoronix Test Suite v10.8.5