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. LLVM Clang 3.2: Processor: Intel Core i7 990X @ 3.47GHz (12 Cores), Motherboard: MSI X58M (MS-7593) v1.0, Chipset: Intel 5520/5500/X58 + ICH10R, Memory: 12288MB, Disk: 1000GB Western Digital WD1003FBYX-0, Graphics: NVIDIA GeForce 8400 GS 512MB (500/333MHz), Audio: Realtek ALC889, Network: Realtek RTL8111/8168/8411 OS: Ubuntu 13.10, Kernel: 3.10.0-2-generic (x86_64), Desktop: Xfce 4.10, Display Server: X Server 1.14.1, Display Driver: NVIDIA 304.88, OpenGL: 3.3.0 NVIDIA 304.88, Compiler: Clang 3.2 + LLVM 3.2svn, File-System: ext4, Screen Resolution: 1920x1080 LLVM Clang 3.3: Processor: Intel Core i7 990X @ 3.47GHz (12 Cores), Motherboard: MSI X58M (MS-7593) v1.0, Chipset: Intel 5520/5500/X58 + ICH10R, Memory: 12288MB, Disk: 1000GB Western Digital WD1003FBYX-0, Graphics: NVIDIA GeForce 8400 GS 512MB (500/333MHz), Audio: Realtek ALC889, Network: Realtek RTL8111/8168/8411 OS: Ubuntu 13.10, Kernel: 3.10.0-2-generic (x86_64), Desktop: Xfce 4.10, Display Server: X Server 1.14.1, Display Driver: NVIDIA 304.88, OpenGL: 3.3.0 NVIDIA 304.88, Compiler: Clang 3.3 + LLVM 3.3, File-System: ext4, Screen Resolution: 1920x1080 LLVM Clang 3.4 SVN 20130715: Processor: Intel Core i7 990X @ 3.47GHz (12 Cores), Motherboard: MSI X58M (MS-7593) v1.0, Chipset: Intel 5520/5500/X58 + ICH10R, Memory: 12288MB, Disk: 1000GB Western Digital WD1003FBYX-0, Graphics: NVIDIA GeForce 8400 GS 512MB (500/333MHz), Audio: Realtek ALC889, Network: Realtek RTL8111/8168/8411 OS: Ubuntu 13.10, Kernel: 3.10.0-2-generic (x86_64), Desktop: Xfce 4.10, Display Server: X Server 1.14.1, Display Driver: NVIDIA 304.88, OpenGL: 3.3.0 NVIDIA 304.88, Compiler: Clang 3.4 (SVN 186320) + LLVM 3.4svn, File-System: ext4, Screen Resolution: 1920x1080 Timed HMMer Search 2.3.2 Pfam Database Search Seconds < Lower Is Better LLVM Clang 3.2 .............. 11.24 |========================================== LLVM Clang 3.3 .............. 10.01 |===================================== LLVM Clang 3.4 SVN 20130715 . 10.13 |====================================== Timed MAFFT Alignment 6.864 Multiple Sequence Alignment Seconds < Lower Is Better LLVM Clang 3.2 .............. 6.12 |========================================= LLVM Clang 3.3 .............. 6.48 |=========================================== LLVM Clang 3.4 SVN 20130715 . 6.06 |======================================== BLAKE2 20121223 Phoronix Test Suite v4.8.0m2 Cycles Per Byte < Lower Is Better LLVM Clang 3.2 .............. 7.28 |========================================== LLVM Clang 3.3 .............. 7.40 |=========================================== LLVM Clang 3.4 SVN 20130715 . 7.39 |=========================================== SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better LLVM Clang 3.2 .............. 477.75 |========================================= LLVM Clang 3.3 .............. 478.82 |========================================= LLVM Clang 3.4 SVN 20130715 . 477.23 |========================================= SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better LLVM Clang 3.2 .............. 289.81 |========================================= LLVM Clang 3.3 .............. 289.32 |========================================= LLVM Clang 3.4 SVN 20130715 . 288.10 |========================================= SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better LLVM Clang 3.2 .............. 1694.32 |======================================== LLVM Clang 3.3 .............. 1701.36 |======================================== LLVM Clang 3.4 SVN 20130715 . 1701.37 |======================================== SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better LLVM Clang 3.2 .............. 1628.51 |===================== LLVM Clang 3.3 .............. 1628.51 |===================== LLVM Clang 3.4 SVN 20130715 . 3137.25 |======================================== SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better LLVM Clang 3.2 .............. 1385.71 |======================================== LLVM Clang 3.3 .............. 1385.71 |======================================== LLVM Clang 3.4 SVN 20130715 . 1392.00 |======================================== x264 2013-06-08 H.264 Video Encoding Frames Per Second > Higher Is Better LLVM Clang 3.2 .............. 177.42 |========================================= LLVM Clang 3.3 .............. 176.15 |========================================= LLVM Clang 3.4 SVN 20130715 . 177.77 |========================================= GraphicsMagick 1.3.16 Operation: Blur Iterations Per Minute > Higher Is Better LLVM Clang 3.2 .............. 66 |============================================ LLVM Clang 3.3 .............. 67 |============================================= LLVM Clang 3.4 SVN 20130715 . 67 |============================================= GraphicsMagick 1.3.16 Operation: Sharpen Iterations Per Minute > Higher Is Better LLVM Clang 3.2 .............. 44 |============================================= LLVM Clang 3.3 .............. 44 |============================================= LLVM Clang 3.4 SVN 20130715 . 44 |============================================= GraphicsMagick 1.3.16 Operation: Resizing Iterations Per Minute > Higher Is Better LLVM Clang 3.2 .............. 82 |============================================ LLVM Clang 3.3 .............. 83 |============================================ LLVM Clang 3.4 SVN 20130715 . 84 |============================================= GraphicsMagick 1.3.16 Operation: HWB Color Space Iterations Per Minute > Higher Is Better LLVM Clang 3.2 .............. 100 |============================================ LLVM Clang 3.3 .............. 101 |============================================ LLVM Clang 3.4 SVN 20130715 . 101 |============================================ GraphicsMagick 1.3.16 Operation: Local Adaptive Thresholding Iterations Per Minute > Higher Is Better LLVM Clang 3.2 .............. 40 |============================================= LLVM Clang 3.3 .............. 39 |============================================ LLVM Clang 3.4 SVN 20130715 . 40 |============================================= Himeno Benchmark 3.0 Poisson Pressure Solver MFLOPS > Higher Is Better LLVM Clang 3.2 .............. 1393.66 |======================================= LLVM Clang 3.3 .............. 1413.54 |======================================= LLVM Clang 3.4 SVN 20130715 . 1438.19 |======================================== Timed ImageMagick Compilation 6.8.1-10 Time To Compile Seconds < Lower Is Better LLVM Clang 3.2 .............. 23.28 |====================================== LLVM Clang 3.3 .............. 24.85 |======================================== LLVM Clang 3.4 SVN 20130715 . 25.89 |========================================== Timed PHP Compilation 5.2.9 Time To Compile Seconds < Lower Is Better LLVM Clang 3.2 .............. 18.96 |====================================== LLVM Clang 3.3 .............. 20.58 |========================================== LLVM Clang 3.4 SVN 20130715 . 20.74 |========================================== C-Ray 1.1 Total Time Seconds < Lower Is Better LLVM Clang 3.2 .............. 23.76 |========================================== LLVM Clang 3.3 .............. 23.76 |========================================== LLVM Clang 3.4 SVN 20130715 . 23.25 |========================================= Primesieve 4.2 1e12 Prime Number Generation Seconds < Lower Is Better LLVM Clang 3.3 .............. 443.31 |========================================= LLVM Clang 3.4 SVN 20130715 . 432.84 |======================================== Smallpt 1.0 Global Illumination Renderer; 100 Samples Seconds < Lower Is Better LLVM Clang 3.2 .............. 231 |============================================ LLVM Clang 3.3 .............. 227 |=========================================== LLVM Clang 3.4 SVN 20130715 . 230 |============================================ FLAC Audio Encoding 1.3.0 WAV To FLAC Seconds < Lower Is Better LLVM Clang 3.3 .............. 8.14 |=========================================== LLVM Clang 3.4 SVN 20130715 . 7.08 |===================================== FFmpeg 2.0 H.264 HD To NTSC DV Seconds < Lower Is Better LLVM Clang 3.2 .............. 20.92 |========================================= LLVM Clang 3.3 .............. 21.05 |========================================= LLVM Clang 3.4 SVN 20130715 . 21.33 |========================================== N-Queens 1.0 Elapsed Time Seconds < Lower Is Better LLVM Clang 3.2 .............. 274.20 |========================================= LLVM Clang 3.3 .............. 274.20 |========================================= LLVM Clang 3.4 SVN 20130715 . 274.76 |========================================= PostgreSQL pgbench 8.4.11 TPC-B Transactions Per Second TPS > Higher Is Better LLVM Clang 3.3 .............. 276.47 |========================================= LLVM Clang 3.4 SVN 20130715 . 275.12 |========================================= Apache Benchmark 2.4.3 Static Web Page Serving Requests Per Second > Higher Is Better LLVM Clang 3.2 .............. 18271.82 |======================================= LLVM Clang 3.3 .............. 17678.96 |===================================== LLVM Clang 3.4 SVN 20130715 . 18491.40 |=======================================