LLVM 3.4 Clang Compiler AMD Kaveri Benchmarking Benchmarks by Michael Larabel for a future article on Phoronix.com looking at AMD Kaveri A10-7850K compiler performance on GCC 4.8 and GCC 4.9 compilers and LLVM Clang 3.4 on Kaveri. GCC 4.9.0 20140126: Processor: AMD A10-7850K APU with Radeon R7 @ 3.70GHz (4 Cores), Motherboard: Gigabyte F2A88XM-D3H, Chipset: AMD Device 1422, Memory: 7168MB, Disk: 120GB KINGSTON SV300S3, Graphics: AMD Kaveri 1024MB, Audio: ATI R6xx HDMI, Monitor: TSB-TV, Network: Realtek RTL8111/8168/8411 OS: Ubuntu 14.04, Kernel: 3.13.0-5-generic (x86_64), Desktop: Unity 7.1.2, Display Driver: radeon 7.2.99, Compiler: GCC 4.9.0 20140126, File-System: ext4, Screen Resolution: 1920x1080 LLVM Clang 3.4: Processor: AMD A10-7850K APU with Radeon R7 @ 3.70GHz (4 Cores), Motherboard: Gigabyte F2A88XM-D3H, Chipset: AMD Device 1422, Memory: 7168MB, Disk: 120GB KINGSTON SV300S3, Graphics: AMD Kaveri 1024MB, Audio: ATI R6xx HDMI, Monitor: TSB-TV, Network: Realtek RTL8111/8168/8411 OS: Ubuntu 14.04, Kernel: 3.13.0-5-generic (x86_64), Desktop: Unity 7.1.2, Display Driver: radeon 7.2.99, Compiler: Clang 3.4 + LLVM 3.4, File-System: ext4, Screen Resolution: 1920x1080 PolyBench-C 3.2 Test: 3 Matrix Multiplications Seconds < Lower Is Better GCC 4.9.0 20140126 . 126.07 |================================================== LLVM Clang 3.4 ..... 122.80 |================================================= LAMMPS Molecular Dynamics Simulator 1.0 Test: Rhodopsin Protein Loop Time < Lower Is Better GCC 4.9.0 20140126 . 59.71 |=================================================== LLVM Clang 3.4 ..... 55.30 |=============================================== Timed HMMer Search 2.3.2 Pfam Database Search Seconds < Lower Is Better GCC 4.9.0 20140126 . 19.34 |================================================== LLVM Clang 3.4 ..... 19.87 |=================================================== BLAKE2 20130131 Phoronix Test Suite v5.0.0m0 Cycles Per Byte < Lower Is Better GCC 4.9.0 20140126 . 6.80 |==================================================== LLVM Clang 3.4 ..... 6.78 |==================================================== Fhourstones 3.1 Complex Connect-4 Solving Kpos / sec > Higher Is Better GCC 4.9.0 20140126 . 9569.50 |================================================= LLVM Clang 3.4 ..... 9557.57 |================================================= SciMark 2.0 Computational Test: Composite Mflops > Higher Is Better GCC 4.9.0 20140126 . 641.01 |=========================================== LLVM Clang 3.4 ..... 738.28 |================================================== SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better GCC 4.9.0 20140126 . 420.02 |================================================== LLVM Clang 3.4 ..... 401.33 |================================================ SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better GCC 4.9.0 20140126 . 70.69 |=================================================== LLVM Clang 3.4 ..... 67.72 |================================================= SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better GCC 4.9.0 20140126 . 865.54 |================================================= LLVM Clang 3.4 ..... 885.69 |================================================== SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better GCC 4.9.0 20140126 . 1164.81 |============================================= LLVM Clang 3.4 ..... 1257.71 |================================================= SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better GCC 4.9.0 20140126 . 683.99 |=============================== LLVM Clang 3.4 ..... 1078.97 |================================================= TSCP 1.81 AI Chess Performance Nodes Per Second > Higher Is Better GCC 4.9.0 20140126 . 737919 |================================================== LLVM Clang 3.4 ..... 605783 |========================================= x264 2014-01-09 H.264 Video Encoding Frames Per Second > Higher Is Better GCC 4.9.0 20140126 . 82.37 |=================================================== LLVM Clang 3.4 ..... 81.91 |=================================================== Himeno Benchmark 3.0 Poisson Pressure Solver MFLOPS > Higher Is Better GCC 4.9.0 20140126 . 903.78 |================================================== LLVM Clang 3.4 ..... 886.78 |================================================= Timed Apache Compilation 2.4.7 Time To Compile Seconds < Lower Is Better GCC 4.9.0 20140126 . 58.93 |=================================================== LLVM Clang 3.4 ..... 37.98 |================================= Timed PHP Compilation 5.2.9 Time To Compile Seconds < Lower Is Better GCC 4.9.0 20140126 . 58.39 |=================================================== LLVM Clang 3.4 ..... 33.22 |============================= C-Ray 1.1 Total Time Seconds < Lower Is Better GCC 4.9.0 20140126 . 40.50 |=============================== LLVM Clang 3.4 ..... 67.08 |=================================================== Bullet Physics Engine 2.81 Test: Raytests Seconds < Lower Is Better GCC 4.9.0 20140126 . 4.52 |============================================== LLVM Clang 3.4 ..... 5.15 |==================================================== Bullet Physics Engine 2.81 Test: 3000 Fall Seconds < Lower Is Better GCC 4.9.0 20140126 . 8.56 |================================================ LLVM Clang 3.4 ..... 9.26 |==================================================== Bullet Physics Engine 2.81 Test: 1000 Stack Seconds < Lower Is Better GCC 4.9.0 20140126 . 9.43 |============================================ LLVM Clang 3.4 ..... 11.02 |=================================================== Bullet Physics Engine 2.81 Test: 1000 Convex Seconds < Lower Is Better GCC 4.9.0 20140126 . 7.20 |========================================== LLVM Clang 3.4 ..... 8.98 |==================================================== Bullet Physics Engine 2.81 Test: Prim Trimesh Seconds < Lower Is Better GCC 4.9.0 20140126 . 1.54 |================================================= LLVM Clang 3.4 ..... 1.65 |==================================================== Bullet Physics Engine 2.81 Test: Convex Trimesh Seconds < Lower Is Better GCC 4.9.0 20140126 . 1.76 |============================================== LLVM Clang 3.4 ..... 2.00 |==================================================== Crafty 23.4 Elapsed Time Seconds < Lower Is Better GCC 4.9.0 20140126 . 103.35 |================================================= LLVM Clang 3.4 ..... 105.01 |================================================== FFmpeg 2.1.1 H.264 HD To NTSC DV Seconds < Lower Is Better GCC 4.9.0 20140126 . 21.37 |=================================================== LLVM Clang 3.4 ..... 21.41 |=================================================== Hierarchical INTegration 1.0 Test: FLOAT QUIPs > Higher Is Better GCC 4.9.0 20140126 . 239745129.57 |============================================ LLVM Clang 3.4 ..... 145204373.31 |=========================== Apache Benchmark 2.4.7 Static Web Page Serving Requests Per Second > Higher Is Better GCC 4.9.0 20140126 . 17507.22 |================================================ LLVM Clang 3.4 ..... 17173.10 |===============================================